Multi-Drug Resistance in Cancer (Methods in Molecular Biology, v596)

Methods

in

Molecular Biology™

Series Editor John M. Walker School of Life Sciences University of Hertfordshire Hatfield, Hertfordshire, AL10 9AB, UK



For other titles published in this series, go to www.springer.com/series/7651

Multi-Drug Resistance in Cancer Edited by

Jun Zhou Department of Genetics and Cell Biology, College of Life Sciences, Nankai University, Tianjin, China

Editor Jun Zhou Department of Genetics and Cell Biology College of Life Sciences Nankai University Tianjin China

ISSN 1064-3745 e-ISSN 1940-6029 ISBN 978-1-60761-415-9 e-ISBN 978-1-60761-416-6 DOI 10.1007/978-1-60761-416-6 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2009938934 © Humana Press, a part of Springer Science+Business Media, LLC 2010 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Humana Press, c/o Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or ­dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. While the advice and information in this book are believed to be true and accurate at the date of going to press, ­neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)

sdfsdf

Preface Chemotherapy is one of the major treatment options for cancer patients; however, the efficacy of chemotherapeutic management of cancer is severely limited by multidrug resistance, in that cancer cells become simultaneously resistant to many structurally and mechanistically unrelated drugs. In the past three decades, a number of mechanisms by which cancer cells acquire multidrug resistance have been discovered. In addition, the development of agents or strategies to overcome resistance has been the subject of intense study. This book contains comprehensive and up-to-date reviews of multidrug resistance mechanisms, from over-expression of ATP-binding cassette drug transporters such as P-glycoprotein, multidrug resistance-associated proteins, and breast cancer resistance protein to the drug ratio-dependent antagonism and the paradigm of cancer stem cells. The book also includes strategies to overcome multidrug resistance, from the development of compounds that inhibit drug transporter function to the modulation of transporter expression. In addition, this book contains techniques for the detection and imaging of drug transporters, methods for the investigation of drug resistance in animal models, and strategies to evaluate the efficacy of resistance reversal agents. The book intends to provide a state-of-the-art collection of reviews and methods for both basic and clinician investigators who are interested in cancer multidrug resistance mechanisms and reversal strategies. Tianjin, China

Jun Zhou

v

sdfsdf

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v ix

  1  Multidrug Resistance in Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Bruce C. Baguley   2  Multidrug Resistance in Oncology and Beyond: From Imaging of Drug Efflux Pumps to Cellular Drug Targets . . . . . . . . . . . . . . . . . . . . . . . . . . Wouter B. Nagengast, Thijs H. Oude Munnink, Eli C.F. Dijkers, Geke A.P. Hospers, Adrienne H. Brouwers, Carolien P. Schröder, Marjolijn Lub-de Hooge, and Elisabeth G.E. deVries   3  Studying Drug Resistance Using Genetically Engineered Mouse Models for Breast Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sven Rottenberg, Marina Pajic, and Jos Jonkers   4  Mechanisms of Multidrug Resistance in Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . Jean-Pierre Gillet and Michael M. Gottesman   5  Molecular Mechanisms of Drug Resistance in Single-Step and Multi-Step Drug-Selected Cancer Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Anna Maria Calcagno and Suresh V. Ambudkar   6  Pharmacogenetics of ATP-Binding Cassette Transporters and Clinical Implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ingolf Cascorbi and Sierk Haenisch   7  Flow Cytometric Evaluation of Multidrug Resistance Proteins . . . . . . . . . . . . . . . Adorjan Aszalos and Barbara J. Taylor   8  Targeted Chemotherapy in Drug-Resistant Tumors, Noninvasive Imaging of P-Glycoprotein-Mediated Functional Transport in Cancer, and Emerging Role of Pgp in Neurodagenerative Diseases . . . . . . . . . . . . . . . . . . Jothilingam Sivapackiam, Seth T. Gammon, Scott E. Harpstrite, and Vijay Sharma   9  Epigenetic Regulation of Multidrug Resistance 1 Gene Expression: Profiling CpG Methylation Status Using Bisulphite Sequencing . . . . . . . . . . . . . . Emma K. Baker and Assam El-Osta 10  Expression and Function of P-Glycoprotein in Normal Tissues: Effect on Pharmacokinetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Frantisek Staud, Martina Ceckova, Stanislav Micuda, and Petr Pavek 11  Molecular Mechanism of ATP-Dependent Solute Transport by Multidrug Resistance-Associated Protein 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . Xiu-bao Chang 12  Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes . . . . Douglas D. Ross and Takeo Nakanishi

1

vii

15

33 47

77

95 123

141

183

199

223 251

viii

Contents

13  Drug Ratio-Dependent Antagonism: A New Category of Multidrug Resistance and Strategies for Its Circumvention . . . . . . . . . . . . . . . . . . . . . . . . . . Troy O. Harasym, Barry D. Liboiron, and Lawrence D. Mayer 14 Reversing Agents for ATP-Binding Cassette Drug Transporters . . . . . . . . . . . . . . Chow H. Lee 15 Overcoming Multidrug Resistance in Cancer: Clinical Studies of P-Glycoprotein Inhibitors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Helen M. Coley 16  Pharmacokinetic and Pharmacodynamic Implications of P-Glycoprotein Modulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Jeannie M. Padowski and Gary M. Pollack 17  Examination of CYP3A and P-Glycoprotein-Mediated Drug–Drug Interactions Using Animal Models . . . . . . . . . . . . . . . . . . . . . . . . . . Punit H. Marathe and A. David Rodrigues 18  Generating Inhibitors of P-Glycoprotein: Where to, Now? . . . . . . . . . . . . . . . . . Emily Crowley, Christopher A. McDevitt, and Richard Callaghan 19  Immunosuppressors as Multidrug Resistance Reversal Agents . . . . . . . . . . . . . . . Hamid Morjani and Claudie Madoulet 20  Overcoming Multidrug Resistance by RNA Interference . . . . . . . . . . . . . . . . . . . Alexandra Stege, Andrea Krühn, and Hermann Lage 21  Circumventing Tumor Resistance to Chemotherapy by Nanotechnology . . . . . . . Xing-Jie Liang, Chunying Chen, Yuliang Zhao, and Paul C. Wang Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

291 325

341

359

385 405 433 447 467 489

Contributors Suresh V. Ambudkar  •  Laboratory of Cell Biology, Center for Cancer Research, National Cancer Institute, NIH, DHHS, Bethesda, MD, USA Adorjan Aszalos  •  Laboratory of Cell Biology, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA Bruce C. Baguley  •  Auckland Cancer Society Research Centre, The University of Auckland, Auckland, New Zealand Emma K. Baker  •  Epigenetics in Human Health and Disease Laboratory, Baker IDI Heart and Diabetes Institute, Melbourne, VIC, Australia Adrienne H. Brouwers  •  Department of Medical Oncology, University Medical Center Groningen, Groningen, The Netherlands Anna Maria Calcagno  •  Laboratory of Cell Biology, Center for Cancer Research, National Cancer Institute, NIH, DHHS, Bethesda, MD, USA Richard Callaghan  •  Nuffield Department of Clinical Laboratory Sciences, John Radcliffe Hospital, University of Oxford, Oxford, UK Ingolf Cascorbi  •  Institute of Experimental and Clinical Pharmacology, University of Kiel, Kiel, Germany Martina Ceckova  •  Department of Pharmacology and Toxicology, Faculty of Pharmacy in Hradec Kralove, Charles University in Prague, Hradec Kralove, Czech Republic Xiu-bao Chang  •  Mayo Clinic College of Medicine, Mayo Clinic Arizona, Scottsdale, AZ, USA Chunying Chen  •  Key Laboratory of Biomedical Effects of Nanomaterials and Nanosafety, National Center for Nanosciences and Technology of China, Beijing, China Key Laboratory of Biomedical Effects of Nanomaterials and Nanosafety, Institute of High Energy Physics, Chinese Academy of Sciences, Beijing, China Helen M. Coley  •  Division of Biological Sciences, Faculty of Health and Medical Sciences, University of Surrey, Guildford, Surrey, UK Emily Crowley  •  Nuffield Department of Clinical Laboratory Sciences, John Radcliffe Hospital, University of Oxford, Oxford, UK Eli C.F. Dijkers  •  Department of Medical Oncology, University Medical Center Groningen, Groningen, The Netherlands Assam El-Osta  •  Epigenetics in Human Health and Disease Laboratory, Baker IDI Heart and Diabetes Institute, Melbourne, VIC, Australia Seth T. Gammon  •  Molecular Imaging Center, Mallinckrodt Institute of Radiology, Washington University Medical School, St. Louis, MO, USA Jean-Pierre Gillet  •  Laboratory of Cell Biology, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA Michael M. Gottesman  •  Laboratory of Cell Biology, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA Sierk Haenisch  •  Institute of Experimental and Clinical Pharmacology, University of Kiel, Kiel, Germany

ix

x

Contributors

Troy O. Harasym  •  Celator Pharmaceuticals Corp., Vancouver, BC, Canada Scott E. Harpstrite  •  Molecular Imaging Center, Mallinckrodt Institute of Radiology, Washington University Medical School, St. Louis, MO, USA Marjolijn Lub-de Hooge  •  Department of Medical Oncology, University Medical Center Groningen, Groningen, The Netherlands Geke A.P. Hospers  •  Department of Medical Oncology, University Medical Center Groningen, Groningen, The Netherlands Jos Jonkers  •  Division of Molecular Biology, The Netherlands Cancer Institute, Amsterdam, The Netherlands Andrea Krühn  •  Charité Campus Mitte, Institute of Pathology, Berlin, Germany Hermann Lage  •  Charité Campus Mitte, Institute of Pathology, Berlin, Germany Chow H. Lee  •  Chemistry Program, University of Northern British Columbia, Prince George, BC, Canada Xing-Jie Liang  •  Key Laboratory of Biomedical Effects of Nanomaterials and Nanosafety, National Center for Nanosciences and Technology of China, Beijing, China Barry D. Liboiron  •  Celator Pharmaceuticals Corp., Vancouver, BC, Canada Claudie Madoulet  •  Laboratoire de Biochimie, Reims Pharmacy School, Reims Cedex, France Punit H. Marathe  •  Metabolism and Pharmacokinetics, Bristol-Myers Squibb, Princeton, NJ, USA Lawrence D. Mayer  •  Celator Pharmaceuticals Corp., Vancouver, BC, Canada Christopher A. McDevitt  •  Nuffield Department of Clinical Laboratory Sciences, John Radcliffe Hospital, University of Oxford, Oxford, UK Stanislav Micuda  •  Department of Pharmacology and Toxicology, Faculty of Pharmacy in Hradec Kralove, Charles University in Prague, Hradec Kralove, Czech Republic Hamid Morjani  •  MEDyC Unité CNRS UMR6237, Reims Pharmacy School, Reims Cedex, France Thijs H. Oude Munnink  •  Department of Medical Oncology, University Medical Center Groningen, Groningen, The Netherlands Wouter B. Nagengast  •  Department of Medical Oncology, University Medical Center Groningen, Groningen, The Netherlands Takeo Nakanishi  •  Department of Membrane Transport and Biopharmaceutics, Kanazawa University School of Pharmaceutical Sciences, Kanazawa, Japan Jeannie M. Padowski  •  Eshelman School of Pharmacy, The University of North Carolina at Chapel Hill, Chapel Hill, NC, USA Marina Pajic  •  Division of Molecular Biology, The Netherlands Cancer Institute, Amsterdam, The Netherlands Petr Pavek  •  Department of Pharmacology and Toxicology, Faculty of Pharmacy in Hradec Kralove, Charles University in Prague, Hradec Kralove, Czech Republic Gary M. Pollack  •  Eshelman School of Pharmacy, The University of North Carolina at Chapel Hill, Chapel Hill, NC, USA A. David Rodrigues  •  Metabolism and Pharmacokinetics, Bristol-Myers Squibb, Princeton, NJ, USA Douglas D. Ross  •  University of Maryland Greenebaum Cancer Center, University of Maryland School of Medicine, and the Baltimore VA Medical Center, Baltimore, MD, USA Sven Rottenberg  •  Division of Molecular Biology, The Netherlands Cancer Institute, Amsterdam, The Netherlands



Contributors

xi

Carolien P. Schröder  •  Department of Medical Oncology, University Medical Center Groningen, Groningen, The Netherlands Vijay Sharma  •  Molecular Imaging Center, Mallinckrodt Institute of Radiology, Washington University Medical School, St. Louis, MO, USA Jothilingam Sivapackiam  •  Molecular Imaging Center, Mallinckrodt Institute of Radiology, Washington University Medical School, St. Louis, MO, USA Frantisek Staud  •  Department of Pharmacology and Toxicology, Faculty of Pharmacy in Hradec Kralove, Charles University in Prague, Hradec Kralove, Czech Republic Alexandra Stege  •  Charité Campus Mitte, Institute of Pathology, Berlin, Germany Barbara J. Taylor  •  Laboratory of Cancer Biology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA Elisabeth G.E. de Vries  •  Department of Medical Oncology, University Medical Center Groningen, Groningen, The Netherlands Paul C. Wang  •  Laboratory of Molecular Imaging, Department of Radiology, Howard University, Washington, DC, USA Yuliang Zhao  •  Key Laboratory of Biomedical Effects of Nanomaterials and Nanosafety, National Center for Nanosciences and Technology of China, Beijing, China Key Laboratory of Biomedical Effects of Nanomaterials and Nanosafety, Institute of High Energy Physics, Chinese Academy of Sciences, Beijing, China

sdfsdf

Chapter 1 Multidrug Resistance in Cancer Bruce C. Baguley Abstract It is becoming increasingly clear that the proliferation of human tumours is driven by a small proportion of cells, termed tumour stem cells, which have the properties of self-renewal. On analogy with stem cells for normal tissues, there are likely to be multiple mechanisms, involving both intrinsic cellular properties and microenvironmental factors, which enable tumour stem cells to resist potentially genotoxic agents. Intrinsic properties include maintenance of cells in a predominantly non-cycling state, expression of transport proteins such as P-glycoprotein, protection from induced apoptosis or other forms of cell death, and limitation of diffusion of potential cytotoxins from the bloodstream. In addition, tumour stem cells are likely to contain multiple genetic changes that will potentially activate host immune mechanisms, which are designed to respond to such changes, and the methods by which tumours suppress such mechanisms are of great relevance to drug resistance. A number of methods of overcoming intrinsic multidrug resistance of tumours have been developed but methods for overcoming tumour resistance mediated by host cells are still at an early stage and require further research. Key words: Cytokinetics, ABC transporters, Drug diffusion, Apoptosis, Tumour dormancy, Macrophages, Apoptosis, Niche, Microenvironment

1. Introduction Cancer multidrug resistance describes a phenomenon whereby resistance to one anticancer drug is accompanied by resistance to drugs whose structures and mechanisms of action may be completely different. One might consider the following two theoretical examples; in the first, a woman is diagnosed with advanced ovarian cancer. Chemotherapy is commenced using combined carboplatin and paclitaxel and a complete remission is obtained. After an interval of 1 year, an abdominal mass is detected and combination therapy is reinstituted. However, in this case there is no significant reduction in tumour mass and after four cycles, treatment with irinotecan is initiated. No response is obtained and treatment is continued with doxorubicin, again with no response. J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_1, © Humana Press, a part of Springer Science + Business Media, LLC 2010

1

2

Baguley

In the second example, a patient diagnosed with metastatic pancreatic cancer is treated with the drug gemcitabine. Both the primary tumour and a lymph node metastasis continue to grow, and chemotherapy is changed to a combination of 5-fluorouracil and oxaliplatin, but again with no effect on tumour progression. These examples demonstrate the two main types of multidrug resistance, one acquired during treatment and the other preexisting at the time of diagnosis. Early ideas on the nature of multidrug resistance were strongly influenced by studies on multidrug resistance in bacteria, in which resistant strains with an identified genetic basis for a lack of response to multiple antibiotics and/or chemotherapeutic agents can be characterised (1). Experimental models for tumour growth were based particularly on transplantable murine leukaemias where it was assumed that the majority of transplanted tumour cells were capable of forming tumours and that resistant cells could be identified or selected for. The development of stem cell theory for animal and human tissues, with its subsequent extension to tumour tissue, changed this concept by postulating that survival of normal or tumour tissue is controlled not by the whole population but by a very small proportion of the total cells that have the property of selfrenewal. The tumour stem cell model, which has had increasing general acceptance, implies that the resistance properties of the tumour stem cell population will dictate overall response to therapy. An important facet of this model is that the survival properties of the tumour stem cells are determined from the microenvironment of these cells, which is usually referred to as the niche. This model highlights two principal methodological problems in the investigation of resistance; cancer stem cells within a tumour population cannot easily be directly identified and also cannot be understood adequately when they are separated from the niche environment. Stem cells in normal tissues have multiple resistance mechanisms to preserve their integrity in the face of potential mutagenic mechanisms associated with inflammation, infection, and dietary toxins. The host stromal cells in the niche microenvironment, particularly those of fibroblast origin, provide soluble and matrixlinked factors to inhibit cell division and apoptosis while simultaneously preserving a primitive multipotent phenotype (2). This microenvironment has been particularly well characterised for the bone marrow (3), where the low oxygen tension suggests diminished perfusion of oxygen and probably also diminished perfusion of potential toxins and mutagens (4). Stem cells appear to express multiple transport proteins of ATP-binding cassette (ABC) family excluding toxins and mutagens (5). In addition, stem cells express pathways such as NF-kB and bcl-2 that protect them from the induction of apoptosis (6). Thus, the protection of normal stem cells from death is a combination of intrinsic (cellular) and extrinsic (microenvironmental) factors.

Multidrug Resistance in Cancer

3

The aim of this review is to consider mechanisms of cancer multidrug resistance in the context of stem cell theory, and then to consider possible methods to overcome these mechanisms. While most reviews on multidrug resistance have often concentrated on intrinsic multidrug resistance of tumour cells, this review also includes discussion of extrinsic contributions to multidrug resistance. The picture of multidrug resistance that emerges from this discussion is therefore a complex one, and the development of strategies to combat clinical multidrug resistance requires considerable further research and mechanistic insight.

2. Resistance Properties of Tumour Stem Cells 2.1. Cytokinetic Resistance

A salient feature of stem cells in normal tissue is that their specialised microenvironment (niche) maintains proliferation at a low level and therefore protects them from genetic (or epigenetic) damage. As shown in Fig. 1.1, this involves stromal cell production of factors, generally members of the TGFb/BMP superfamily, which act on stem cells through corresponding surface receptors, smad family proteins, and signalling pathways (7) and result in the increased production of cyclin-dependent kinase (cdk) inhibitors such as p15, p16, p21, and p27. Tumour stem cells are likely to be contained in a similar niche microenvironment, but because of genetic or epigenetic changes that reduce or prevent their production their production of cdk inhibitors is defective in many cases. Thus, proliferation of tumour stem cells is inadequately constrained and cells leave the niche to become transit amplifying cells that continue, at least initially, to proliferate.

Fig. 1.1. Simplified examples of control of stem cell proliferation by the niche microenvironment. (a) Control by members of the TGF-b superfamily, thought to be an important mechanism in the stem cell niche of normal tissues. (b) Control by g-interferon, thought to be an important mechanism underlying tumour cell dormancy.

4

Baguley

The loss of control of proliferation of tumour stem cells in the niche should theoretically make them more sensitive than normal tissue stem cells to cytotoxic agents. There is a second potential mechanism, as shown in Fig. 1.1, whereby the proliferation of tumour stem cells may be constrained by the niche microenvironment. The clinical observation that a kidney transplant recipient developed melanoma with the genotype of the kidney donor, even though the donor had been free of melanoma for 16 years (8) as well as other examples (9) point to a mechanism whereby tumour stem cells may be maintained for long periods in a non-proliferating state. A possible explanation for the loss of dormancy following treatment of the recipient patient with immunosuppressive drugs is that such therapy weakens T cell-mediated induction of dormancy. Surgery might in some cases also induce a loss of dormancy (10). A preclinical model for tumour dormancy is provided by a study in which a group of mice were treated with a carcinogen. Under normal conditions, only a small proportion developed tumours, but when tumour-free mice were treated subsequently with antibodies to T cells or to g-interferon a high proportion developed tumours (11). Examination of tumour-free mice before immunosuppressive treatment revealed the presence of microscopic groups of tumour cells associated with T cells. From these observations one can hypothesise that IFNg acts in concert with other cytokines, probably through the induction of STAT1 (12) to constrain tumour cell proliferation (13). Other host cell-mediated mechanisms may also contribute to reduction of tumour cell proliferation, raising the possibility that immune cell-mediated release of cytokines or other factors can augment factors produced by the normal stroma to maintain tumour stem in a non-proliferating and thus drug-resistant state. 2.2. Multidrug Resistance Mechanisms Preventing Drugs from Reaching Target Cells

Regardless of whether they are administered orally, intravenously, intra-arterially, or by other routes, anticancer drugs must diffuse from the bloodstream to individual tumour stem cells. The vascular density of the tumour, which determines the mean diffusion distance from blood supply to the individual tumour cell, will have a major effect on diffusion time for some drugs and can contribute to multidrug resistance. As with the case of normal bone marrow, there is evidence that tumour stem cells may exist in a state of low oxygen tension, suggesting the presence of a perfusion barrier, which could limit the rate of penetration of anticancer drugs (4). Tumour hypoxia may also be either intrinsic (related to vascular geometry) or intermittent (related to temporal changes in tumour blood flow), and both of these states may contribute to drug access and efficacy. Drug diffusion depends not only on tumour architecture and dynamics but also on drug properties such as the molecular weight and degree of protein binding.

Multidrug Resistance in Cancer

5

Albumin is a common binding protein that accounts for the low free drug fraction of many drugs, and a-acid glycoprotein, an acute phase protein commonly elevated in cancer patients, can also have a major effect on free drug concentrations (14). The free drug fraction can be quite low (<5%) for some anticancer drugs, particularly complex organic molecules, and this will affect the rate of extracellular diffusion (15). Drugs move through cancer tissue not only by extracellular diffusion but also by cellular uptake and efflux, with rapid transcellular diffusion enhancing their ability to reach their target. The latter process is controlled by multiple processes such as the rate of passage between the extracellular fluid and the cytoplasm, the degree of binding to macromolecules within the cell, the rate of sequestration into intracellular vesicles, and the rate of cellular efflux. Some anticancer drugs require transporters for uptake/ efflux, and the activity of this transporter may therefore regulate the efficiency of drug delivery; an example is the copper transporter ATP7B in the case of cisplatin, carboplatin, and oxaliplatin (16). Drug efflux transporters of the ABC family, which are generally found both on the plasma membrane and on the membranes of cellular vesicles, generally act to minimise cytoplasmic drug concentrations by promoting efflux from the cell or sequestration into vesicles, thus reducing the efficiency of trans-cellular diffusion of drugs. An example of the latter process is provided by the sequestration of cisplatin into melanosomes of melanoma cells (17). Intracellular protein binding and DNA binding will also influence the rate of trans-cellular diffusion by reducing cytoplasmic free drug concentration. 2.3. Multidrug Resistance Mechanisms Preventing Drugs from Reaching Target Intracellular Concentrations

Studies with tumour cell lines have identified transport proteins, acting either to promote drug efflux from the cell or to sequester it to cellular vesicles that are later eliminated by exocytosis, as a major mechanism for multidrug resistance. The first such transport protein to be identified was P-glycoprotein, a membraneassociated protein typically found on the plasma membrane and is able, like albumin, to bind to a broad variety of small molecules, particularly those containing hydrophobic domains and positively charged areas (18). However, in contrast to albumin, P-glycoprotein is able to carry out an ATP-dependent conformational change that moves the substrate to the exterior of the cell where it is released. At least 48 structurally related transporters, known collectively as the ABC family, are known (19), and the three subfamilies concerned with drug transport are the “B” subfamily that includes P-glycoprotein, the “C” subfamily that includes multidrug resistance-associated protein (MRP) transporters, and the “G” subfamily that includes the ABCG2, MXR, and ABCP proteins. Some of these transporters act directly on the drug while others act on conjugates in concert with cellular

6

Baguley

conjugating enzymes that first link the drug to glutathione, glucuronide, or sulphate. Expression of ABCG2 and MRP1 has been reported in side populations of tumour cell lines with some of the characteristics of stem cells (20), consistent with the hypothesis that tumour stem cells commonly express these drug efflux proteins (21, 22). 2.4. Multidrug Resistance Mechanisms Involving Cell Death Pathways

A major mechanism of cancer stem cell multidrug resistance is thought to involve suppression of pathways that lead to apoptosis or to other forms of cell death. Anticancer drugs commonly induce stress pathways such as p38 kinase (23, 24) or suppress “survival” pathways such as those coordinated by PI3K (phosphatidyl-3-phosphate kinase) and ERK1 (extracellular regulated kinase-1) (25). Modulation of these pathways affects the balance of activity of the bcl-2 family of proteins and their relatives (26), which in turn control the stability of the outer mitochondrial membrane. A change in balance of bcl-2 family members leads to the so-called permeability transition (27), where the outer mitochondrial membrane is disrupted and proteins that are normally stored in the space between the inner and outer mitochondrial membranes are released into the cytoplasm. These proteins include, depending on the cell, cytochrome C, endonuclease G, and apoptosis-inducing factor (AIF) and can act in various ways to induce cell death. Cytochrome C is well known for its ability to induce a cascade of caspase enzymes, which convert the cell to small fragments that can be recognised and ingested by nearby phagocytes. There are many examples of changes in tumour cells that reduce the ability of anticancer drugs to induce cell death pathways. Many tumour cells have either a mutated gene for phosphoinositide 3-kinase (PI3K) or have lost the gene for PTEN, a phosphatase that regulates the activity of PI3K. The resultant loss of regulation of PI3K leads in turn to increased activity of AKT (PKB) and phosphorylation of bad, a member of the bcl-2 family. Loss of unphosphorylated bad protects mitochondria from the permeability transition and thus increases resistance to cell death. Similarly, activating mutations in the RAS or RAF genes lead to activation of the ERK1 enzyme, inactivation by phosphorylation of bid, another member of the bcl-2 family, and protection of mitochondria from the permeability transition. Phosphorylation of bcl-2, following activation of the NF-kB transcription factor and its downstream target twist-1by cytokines and other cellular stresses, can lead to resistance (28). Induced overexpression of bcl-2 itself can also provide a mechanism of resistance (29).

2.5. The Role of Tumour Heterogeneity

Heterogeneity is an important feature that distinguishes tumour stem cell populations from stem cells in normal tissue. If this heterogeneity extends to expression of multidrug resistance

Multidrug Resistance in Cancer

7

mechanisms, the overcoming of one form of cell-mediated multidrug resistance may affect only a proportion of the total stem cell population, leaving the remainder to repopulate the niche. There are a number of mechanisms for the generation of such heterogeneity; tumour cells contain a number of mutations or epigenetic changes that lead to defects both in their ability to differentiate and in their control of proliferation. Such changes affect not only cell cycle entry but also centrosome replication (30, 31), and centrosome defects lead during mitosis to chromosome instability, tetraploidy, and aneuploidy. Other mechanisms, such as microsatellite instability and defective DNA repair can also contribute to a so-called mutator phenotype, whereby tumour cells may develop large numbers of genetic changes (32). Thus, while normal stem cells are diploid, homogenous, and normally quiescent, tumour stem cells are usually aneuploid, heterogeneous, and have a significant proliferation rate. 2.6. Multidrug Resistance Involving Host Immune Responses

The body has a number of inbuilt mechanisms to prevent the proliferation of potential cancer stem cells. For instance, the acquisition of genetic changes has the effect of inducing cellular stress responses involving the secretion of specific proteins, such as the interleukins IL-6 and IL-8, which in turn can induce senescence of tumour cells (33, 34). The development of genetic changes in cancer stem cells can also lead to potentially cytotoxic T-lymphocyte responses (35). Furthermore, proteins such as HMGB1, when released from stressed or dying cells, can interact with TLR-4 tolllike receptors macrophages and dendritic cells to induce potentially cytotoxic responses against the tumour (36). Tumours also have a relatively high rate of cell turnover, as indicated by the fact that cell cycle times of individual tumour cells are measured in days while the volume doubling times of solid tumours are measured in months (37) and may thus sustain such responses. Against these positive effects, tumours may also facilitate resistance to host immune mechanisms. The continuous release of apoptotic tumour cells as a consequence of cell turnover leads to uptake by tumour macrophages and dendritic cells and a resultant immunosuppressive response (38). Thus, tumours exist in a dynamic balance of progression or regression that is controlled largely by host-mediated effects. The application of cancer therapeutic agents will affect this balance and in some cases can increase immunosuppression, for instance, by depleting immune cells or by increasing the production of potentially immunosuppressive apoptotic tumour cells. As shown in Fig. 1.1, T-cell-mediated release of interferon-g can potentially prevent the proliferation of tumour stem cells and therapy that reduces interferon release could allow continuation and even stimulation of tumour growth. Adequate methods for measurement of individual components of the complex network of host responses have still to be developed.

8

Baguley

3. Methods of Overcoming Multidrug Resistance

3.1. Methods for Overcoming Transport/ Diffusion Multidrug Resistance

There are two general approaches to the overcoming of multidrug resistance in cancer. The first, which has achieved the most attention, is to facilitate anticancer drug-induced death of the tumour stem cell population, while the second is to re-establish host-controlled cell cycle arrest of this population. Each of these approaches involves overcoming of limitations of tumour tissue diffusion and cellular uptake that affect drug efficacy. The expression of drug transporters, which can affect the efficacy of a broad range of anticancer drugs in tumour stem cells, is well established (39, 40), and there are two main approaches to overcoming this resistance (41). The first is to co-administer a drug that inhibits the action of the transporter (42), while the second is to design or utilise anticancer drugs whose activity is not significantly altered by the presence of transporters (43). A variant of this approach is to co-administer a second drug that increases the cellular uptake of the anticancer drug: this has been examined in the case of paclitaxel through the administration of releasable octaarginine transporters (44). There have been many published studies at both the preclinical and clinical levels of drugs that reverse transport-mediated multidrug resistance (42, 43). Such agents are often described as first, second, and third generation inhibitors of transport proteins. First generation inhibitors include verapamil, cyclosporine A, tamoxifen, and calmodulin antagonists but their use involved comparatively high doses and consequent unacceptable side effects because of interaction with other cellular targets. Second generation inhibitors are generally more dose-potent and include dexverapamil, valspodar, and biricodar, but one of the main problems is that inhibition of cytochrome P450 activity leads to pharmacological interactions, resulting in altered pharmacokinetics of the anticancer drugs with which they were co-administered. Third generation inhibitors are designed to overcome these disadvantages and include the drugs tariquidar, zosuquidar, and laniquidar. Clinical trials of these agents are continuing but results so far have been disappointing. The alternative approach, that of designing drugs that are intrinsically insensitive to transport resistance mechanisms, generally requires that the drug is taken up efficiently into cells and is relatively resistant to conjugation with glutathione, glucuronide, or sulphate. In this way, the rate of uptake greatly exceeds the potential rate of transporter-mediated efflux. Cellular uptake of drugs by diffusion is generally controlled by diffusion: lipophilic drugs can enter cells rapidly by diffusion; more hydrophilic drugs such as doxorubicin are taken up slowly, and very hydrophilic

Multidrug Resistance in Cancer

9

drugs such as cytosine arabinoside require specific permeases (43). Examples of lipophilic drugs that are thought to be substrates for drug transporters but whose uptake rates are rapid include the antileukaemia drug amsacrine (45) and the experimental anticancer agent N-[2-(dimethylamino)ethyl]acridine-4carboxamide (46). Both drugs exhibit activity in cultured cells expressing multidrug resistance transporters (47). As mentioned earlier, tumour stem cells may be protected by a diffusion barrier as a consequence of poor vascular organisation, leading to inefficient distribution of anticancer drugs within tissue. A further design principle required for drugs that will overcome this barrier is to design for high efficiency of tissue diffusion as well as a long residence within the tumour tissue, thus allowing time for diffusion to all tumour compartments including those containing tumour stem cells. The use of slow drug infusion rates may facilitate good distribution within regions of tissue that have an inefficient and sometimes intermittent blood supply. 3.2. Methods for Overcoming Resistance to Apoptosis

Both tumour and normal tissue stem cells have multiple mechanisms to ensure survival under adverse conditions, and in the case of tumour stem cells these mechanisms include mutations in pathways that enhance cell survival. Examples include inactivating mutations of the gene for p53 protein, activating mutations of the gene for PI3K, loss of expression of PTEN (a phosphatase controlling PI3K activity), and activating mutations of the genes for the RAS/RAF pathway. As described earlier, these and other proteins regulate signalling pathways leading to reduction of proapoptotic proteins of the bcl-2 family, such that drug-mediated inhibition can lead to an increased probability of the cellular transition to apoptosis. There are now many examples of therapeutic drugs in these categories and some are currently in clinical trial (48–50). In addition, direct inhibition of bcl-2 family members, i.e., at the convergence of these pathways, has been demonstrated; obatoclax interferes with bcl-2 family-mediated resistance and restores sensitivity to several new anticancer drugs (51). The transcription factor NF-kB specifies a number of pathways, including bcl-2 family members, that inhibit apoptosis (52), and the development of inhibitors of NF-kB signalling to overcome resistance is being investigated (53).

3.3. Methods for Suppression of Tumour Stem Cell Proliferation

The multitude of mechanisms involved in the resistance of tumour stem cells to apoptosis, together with the known multiple resistance to normal tissue stem cells to cytotoxic insult, raises the question as to whether complete elimination of a tumour stem cell population is ultimately possible. The results of many current clinical trials of antitumour agents that target the survival pathways have, with a few exceptions, provided extensions of tumour-free or overall survival measured in months rather than years, consistent

10

Baguley

with the hypothesis that these therapies are not having a major impact on the survival of tumour stem cells. As discussed earlier, both preclinical and clinical data are consistent with the concept that tumours can exist in a dormant state for long periods of time, and studies in mice have suggested that the mechanism for the maintenance of this dormancy involves host-mediated suppression of cell proliferation through factors such as interferon-g. Therapies that lead to long-term arrest of tumour stem cells, either directly or by activation of host cell mechanisms therefore deserve serious consideration. As judged from studies of tumour cell lines, many cancers have defects such as loss p16 or p53 function, which compromise control of long-term cycle arrest. Strategies that reactivate p53 function (48) or that reverse epigenetic suppression of p16 (54) are examples of how cycle arrest might be approached. 3.4. Methods for Augmentation of Host Immune Mechanisms

Tumour tissue can be distinguished from normal tissue by an increased proportion of macrophages and dendritic cells, and there is increasing evidence that host cells play an important role in cancer chemotherapeutic response (55). Following exposure to various sorts of stress, including those caused by anticancer drugs, tumour cells release certain proteins to the cytoplasm and from there to the external environment. Here, they interact with tolllike receptors, particularly with TLR4, on macrophages and dendritic cells, triggering responses that include cytokines and reactive oxygen species. The presence of a functioning immune system has been shown to be important for therapeutic outcome in a number of experimental models (56), and key proteins include HMGB1, a high mobility group protein associated with chromatin, and calreticulin (57), a protein associated with the endoplasmic reticulum. Administration of agents that optimise the production or effects of such proteins might therefore be used to augment responses of tumour cells to chemotherapeutic agents. Existing cytotoxic agents appear to vary in their ability to induce this response, with one of the most effective being the anthracycline derivative doxorubicin. This drug induces DNA damage, which in turn activates PARP (poly ADP ribose polymerase). HMGB1 is ADP-ribosylated by PARP, allowing it to move to the cytoplasm, to be released from the cell and to interact with TLR4 receptors of host macrophages and host cells. Other stress pathways, probably involving calcium ions, appear to trigger co-translocation of calreticulin in association with the chaperone ERp57 (a disulfide isomerase protein) (57) to the plasma membrane where it can interact with TLR4 receptors. It has been found that tumour cells with low ERp57 expression have a normal apoptotic response to doxorubicin in vitro but fail to respond to it in vivo (57), supporting the importance of this response in the overall antitumour effect of the drug. Tumour macrophages and dendritic

Multidrug Resistance in Cancer

11

cells must in themselves be competent in carrying out these responses to cancer chemotherapy. Release of chemokines within the tumour leads to the recruitment of immature macrophages to tumour tissue but until they are appropriately activated these may lack the ability to induce tumour cell responses.

4. Perspective Ideally, this review should have provided firstly, methods to identify the causes of multidrug resistance in cancer patients and secondly, methods by which such resistance could be overcome. Unfortunately, we are still a long way from realising these goals. The development of adequate methods to study multidrug resistance in human cancer remains a formidable challenge. Although initial studies in this area were dominated by the elucidation of outward drug transport mechanisms using immunohistochemistry and drug uptake/efflux measurements, subsequent studies have uncovered many different mechanisms that may confer simultaneous resistance to anticancer drugs with unrelated chemical structures. The material covered in this review describes several mechanisms of by which tumour stem cells exhibit multidrug resistance. A corollary of the tumour stem cell hypothesis is that the resistance properties of the tumour population as a whole may be different to the resistance properties of the tumour stem cells, so the assessment of resistance by drug responses of primary cultures (58) or by gene expression arrays (59) will not necessarily predict the responses of the small proportion of stem cells. This may explain the limited success of such assays in predicting clinical outcome (60). The contribution of the tumour microenvironment to treatment response is further demonstrated by the following example. Murine methylcholanthrene-induced (MCA/129) tumours grow in BALB/c mice and respond well to radiotherapy (15 Gy). BALB/c mice lacking a single enzyme, acid sphingomyelinase, are phenotypically normal and allow growth of MCA/129 tumours. However, such tumours are resistant to radiotherapy (61). Radiation would be expected to cause the same amount of DNA damage to tumour cells regardless of the host, implying that tumour response is a function of the host rather than the tumour cells. The products of sphingomyelinase activity, ceramides are known to have multiple signalling roles in endothelial cells, macrophages, and fibroblasts (62), supporting the concept that responses of host cells are essential, in combination with radiation-induced effects on the tumour population, for the observed tumour response. While further work is required to extend these observations to other tumour models, it is important to realise

12

Baguley

that the use of xenografts of human tumour cell lines growing in immunodeficient mice may be inappropriate for such studies. T lymphocyte-induced interferon-g is an important factor in macrophage maturation (63), and immunocompetent mice may be used for detailed studies. It is clear that further progress in understanding multidrug resistance in experimental cancer will have to employ a multidisciplinary approach with a combination of technologies to distinguish effects on tumour cells from those on host cells and to assess the relative importance of these two determinants in overall response. We are still at a very early stage with regard to the translation of these studies into clinical practice. The development of appropriate markers for host responses will be a critical factor in such research. References 1. Bolhuis H, Van Veen HW, Poolman B, Driessen AJ, Konings WN (1997) Mechanisms of multidrug transporters. FEMS Microbiol Rev 21:55–84 2. Baguley BC, Marshall ES (2008) The use of human tumour cell lines in the discovery of new cancer chemotherapeutic drugs. Expert Opin Drug Discov 3:153–161 3. Meads MB, Hazlehurst LA, Dalton WS (2008) The bone marrow microenvironment as a tumor sanctuary and contributor to drug resistance. Clin Cancer Res 14:2519–2526 4. Parmar K, Mauch P, Vergilio J, Sackstein R, Down JD (2007) Distribution of hematopoietic stem cells in the bone marrow according to regional hypoxia. Proc Natl Acad Sci USA 104:5431–5436 5. Huls M, Russel FG, Masereeuw R (2009) The role of ABC transporters in tissue defense and organ regeneration. J Pharmacol Exp Ther 328:3–9 6. Turco MC, Romano MF, Petrella A et  al (2004) NF-kappaB/Rel-mediated regulation of apoptosis in hematologic malignancies and normal hematopoietic progenitors. Leukemia 18:11–17 7. Massague J (2008) TGFbeta in Cancer. Cell 134:215–230 8. MacKie RM, Reid R, Junor B (2003) Fatal melanoma transferred in a donated kidney 16 years after melanoma surgery. N Engl J Med 348:567–568 9. Aguirre-Ghiso JA (2007) Models, mechanisms and clinical evidence for cancer dormancy. Nat Rev Cancer 7:834–846

10. Demicheli R, Retsky MW, Hrushesky WJ, Baum M (2007) Tumor dormancy and surgery-driven interruption of dormancy in breast cancer: learning from failures. Nat Clin Pract Oncol 4:699–710 11. Koebel CM, Vermi W, Swann JB et al (2007) Adaptive immunity maintains occult cancer in an equilibrium state. Nature 450:903–907 12. Kortylewski M, Komyod W, Kauffmann ME et  al (2004) Interferon-gamma-mediated growth regulation of melanoma cells: involvement of STAT1-dependent and STAT1independent signals. J Invest Dermatol 122:414–422 13. Muller-Hermelink N, Braumuller H, Pichler B et al (2008) TNFR1 signaling and IFN-gamma signaling determine whether T cells induce tumor dormancy or promote multistage carcinogenesis. Cancer Cell 13:507–518 14. Finlay GJ, Baguley BC (2000) Effects of protein binding on the in vitro activity of antitumour acridine derivatives and related anticancer drugs. Cancer Chemother Pharmacol 45:417–422 15. Hicks KO, Pruijn FB, Baguley BC, Wilson WR (2001) Extravascular transport of the DNA intercalator and topoisomerase poison N-[2-(dimethylamino)ethyl]acridine-4-carboxamide (DACA): diffusion and metabolism in multicellular layers of tumor cells. J Pharmacol Exp Ther 297:1088–1098 16. Nakagawa T, Inoue Y, Kodama H et al (2008) Expression of copper-transporting P-type adenosine triphosphatase (ATP7B) correlates with cisplatin resistance in human non-small

Multidrug Resistance in Cancer

17.

18. 19.

20.

21. 22.

23.

24. 25.

26. 27.

28.

29.

30.

cell lung cancer xenografts. Oncol Rep 20: 265–270 Chen KG, Valencia JC, Lai B et  al (2006) Melanosomal sequestration of cytotoxic drugs contributes to the intractability of malignant melanomas. Proc Natl Acad Sci USA 103:9903–9907 Ling V (1997) Multidrug resistance: molecular mechanisms and clinical relevance. Cancer Chemother Pharmacol 40(Suppl):S3–S8 Gillet JP, Efferth T, Remacle J (2007) Chemotherapy-induced resistance by ATPbinding cassette transporter genes. Biochim Biophys Acta 1775:237–262 Ejendal KF, Hrycyna CA (2002) Multidrug resistance and cancer: the role of the human ABC transporter ABCG2. Curr Protein Pept Sci 3:503–511 Sarkadi B, Ozvegy-Laczka C, Nemet K, Varadi A (2004) ABCG2 – a transporter for all seasons. FEBS Lett 567:116–120 Loebinger MR, Giangreco A, Groot KR et al (2008) Squamous cell cancers contain a side population of stem-like cells that are made chemosensitive by ABC transporter blockade. Br J Cancer 98:380–387 Sharma SV, Gajowniczek P, Way IP et  al (2006) A common signaling cascade may underlie “addiction” to the Src, BCR-ABL, and EGF receptor oncogenes. Cancer Cell 10:425–435 Olson JM, Hallahan AR (2004) p38 MAP kinase: a convergence point in cancer therapy. Trends Mol Med 10:125–129 Hersey P, Zhuang L, Zhang XD (2006) Current strategies in overcoming resistance of cancer cells to apoptosis melanoma as a model. Int Rev Cytol 251:131–158 Danial NN (2007) BCL-2 family proteins: critical checkpoints of apoptotic cell death. Clin Cancer Res 13:7254–7263 Forte M, Bernardi P (2006) The permeability transition and BCL-2 family proteins in apoptosis: co-conspirators or independent agents? Cell Death Differ 13:1287–1290 Pham CG, Bubici C, Zazzeroni F et al (2007) Upregulation of Twist-1 by NF-kappaB blocks cytotoxicity induced by chemotherapeutic drugs. Mol Cell Biol 27:3920–3935 Osford SM, Dallman CL, Johnson PW, Ganesan A, Packham G (2004) Current strategies to target the anti-apoptotic Bcl-2 protein in cancer cells. Curr Med Chem 11: 1031–1039 Kramer A, Lukas J, Bartek J (2004) Checking out the centrosome. Cell Cycle 3:1390–1393

13

31. McDermott KM, Zhang J, Holst CR et  al (2006) p16(INK4a) prevents centrosome dysfunction and genomic instability in primary cells. PLoS Biol 4:e51 32. Loeb LA, Bielas JH, Beckman RA (2008) Cancers exhibit a mutator phenotype: clinical implications. Cancer Res 68:3551–3557 33. Kuilman T, Michaloglou C, Vredeveld LC et  al (2008) Oncogene-induced senescence relayed by an interleukin-dependent inflammatory network. Cell 133:1019–1031 34. Acosta JC, O’Loghlen A, Banito A, Raguz S, Gil J (2008) Control of senescence by CXCR2 and its ligands. Cell Cycle 7:2956–2959 35. Smyth MJ, Godfrey DI, Trapani JA (2001) A fresh look at tumor immunosurveillance and immunotherapy. Nat Immunol 2:293–299 36. Zitvogel L, Apetoh L, Ghiringhelli F, Kroemer G (2008) Immunological aspects of cancer chemotherapy. Nat Rev Immunol 8:59–73 37. Baguley BC (2006) Tumor stem cell niches: a new functional framework for the action of anticancer drugs. Recent Pat Anticancer Drug Discov 1:121–127 38. Fonseca C, Dranoff G (2008) Capitalizing on the immunogenicity of dying tumor cells. Clin Cancer Res 14:1603–1608 39. Hadnagy A, Gaboury L, Beaulieu R, Balicki D (2006) SP analysis may be used to identify cancer stem cell populations. Exp Cell Res 312:3701–3710 40. Keshet GI, Goldstein I, Itzhaki O et al (2008) MDR1 expression identifies human melanoma stem cells. Biochem Biophys Res Commun 368:930–936 41. Ozben T (2006) Mechanisms and strategies to overcome multiple drug resistance in cancer. FEBS Lett 580:2903–2909 42. Szakacs G, Paterson JK, Ludwig JA, BoothGenthe C, Gottesman MM (2006) Targeting multidrug resistance in cancer. Nat Rev Drug Discov 5:219–234 43. Baguley BC (2002) Novel strategies for overcoming multidrug resistance in cancer. BioDrugs 16:97–103 44. Dubikovskaya EA, Thorne SH, Pillow TH, Contag CH, Wender PA (2008) Overcoming multidrug resistance of small-molecule therapeutics through conjugation with releasable octaarginine transporters. Proc Natl Acad Sci USA 105:12128–12133 45. Robbie MA, Baguley BC, Denny WA, Gavin JB, Wilson WR (1988) Mechanism of resistance of noncycling mammalian cells to 4¢-(9acridinylamino)methanesulfon-m-anisidide: comparison of uptake, metabolism, and DNA

14

46.

47.

48.

49.

50.

51.

52. 53.

Baguley breakage in log- and plateau-phase Chinese hamster fibroblast cell cultures. Cancer Res 48:310–319 Haldane A, Finlay GJ, Hay MP, Denny WA, Baguley BC (1999) Cellular uptake of N-[2(dimethylamino)ethyl]acridine-4-carboxamide (DACA). Anticancer Drug Des 14:275–280 Davey RA, Su GM, Hargrave RM et al (1997) The potential of N-[2-(dimethylamino)ethyl] acridine-4-carboxamide to circumvent three multidrug-resistance phenotypes in vitro. Cancer Chemother Pharmacol 39:424–430 Foster BA, Coffey HA, Morin MJ, Rastinejad F (1999) Pharmacological rescue of mutant p53 conformation and function. Science 286:2507–2510 Sebolt-Leopold JS (2004) MEK inhibitors: a therapeutic approach to targeting the RasMAP kinase pathway in tumors. Curr Pharm Des 10:1907–1914 Serra V, Markman B, Scaltriti M et al (2008) NVP-BEZ235, a dual PI3K/mTOR inhibitor, prevents PI3K signaling and inhibits the growth of cancer cells with activating PI3K mutations. Cancer Res 68:8022–8030 Nguyen M, Marcellus RC, Roulston A et  al (2007) Small molecule obatoclax (GX15– 070) antagonizes MCL-1 and overcomes MCL-1-mediated resistance to apoptosis. Proc Natl Acad Sci USA 104:19512–19517 Nakanishi C, Toi M (2005) Nuclear factorkappaB inhibitors as sensitizers to anticancer drugs. Nat Rev Cancer 5:297–309 Wang W, McLeod HL, Cassidy J (2003) Disulfiram-mediated inhibition of NF-kappaB activity enhances cytotoxicity of 5-fluorouracil in human colorectal cancer cell lines. Int J Cancer 104:504–511

54. Claus R, Lubbert M (2003) Epigenetic targets in hematopoietic malignancies. Oncogene 22:6489–6496 55. Chen R, Alvero AB, Silasi DA, Steffensen KD, Mor G (2008) Cancers take their Toll – the function and regulation of Toll-like receptors in cancer cells. Oncogene 27:225–233 56. Hicks AM, Riedlinger G, Willingham MC et  al (2006) Transferable anticancer innate immunity in spontaneous regression/complete resistance mice. Proc Natl Acad Sci USA 103:7753–7758 57. Panaretakis T, Joza N, Modjtahedi N et  al (2008) The co-translocation of ERp57 and calreticulin determines the immunogeni­city of cell death. Cell Death Differ 15:1499–1509 58. Baguley BC, Marshall ES (2004) In vitro modelling of human tumour behaviour in drug discovery programmes. Eur J Cancer 40:794–801 59. Sotiriou C, Piccart MJ (2007) Taking geneexpression profiling to the clinic: when will molecular signatures become relevant to patient care? Nat Rev Cancer 7:545–553 60. Samson DJ, Seidenfeld J, Ziegler K, Aronson N (2004) Chemotherapy sensitivity and resistance assays: a systematic review. J Clin Oncol 22:3618–3630 61. Garcia-Barros M, Paris F, Cordon-Cardo C et al (2003) Tumor response to radiotherapy regulated by endothelial cell apoptosis. Science 300:1155–1159 62. Vit JP, Rosselli F (2003) Role of the ceramidesignaling pathways in ionizing radiationinduced apoptosis. Oncogene 22:8645–8652 63. Duff MD, Mestre J, Maddali S et  al (2007) Analysis of gene expression in the tumor-associated macrophage. J Surg Res 142:119–128

Chapter 2 Multidrug Resistance in Oncology and Beyond: From Imaging of Drug Efflux Pumps to Cellular Drug Targets Wouter B. Nagengast, Thijs H. Oude Munnink, Eli C.F. Dijkers, Geke A.P. Hospers, Adrienne H. Brouwers, Carolien P. Schröder, Marjolijn Lub-de Hooge, and Elisabeth G.E. de Vries Abstract Resistance of tumor cells to several structurally unrelated classes of natural products, including anthracyclines, taxanes, and epipodophyllotoxines, is often referred as multidrug resistance (MDR). This is associated with ATP-binding cassette transporters, which function as drug efflux pumps such as P-glycoprotein (Pgp) and multidrug resistance-associated protein 1 (MRP1). Because of the hypothesis in the early eighties that blockade of these efflux pumps by modulators would improve the effect of chemotherapy, extensive effort has been put to visualize these pumps using nuclear imaging with several specific tracers, using both SPECT and PET techniques. The methods and possibilities to visualize these pumps in both the tumor and the blood–brain barrier will be discussed. Because of the fact that the addition of Pgp or MRP modulators has not shown any clinical benefit in patient outcome, these specific MDR tracers are not routinely used in clinical practice. Evidence emerges that combination of chemotherapeutic drugs involved in MDR with the so-called targeted agents can improve patient outcome. The concept of molecular imaging can also be used to visualize the targets for these agents, such as HER2/neu and angiogenic factors such as vascular endothelial growth factor (VEGF). Potentially visualizing molecular drug targets in the tumor can function as biomarkers to support treatment decision for the individual patient. Key words: Drug efflux pump, Imaging, Tumor, Blood–brain barrier, Drug target

1. Introduction The occurrence of resistance to several structurally unrelated classes of natural products, including anthracyclines, taxanes, and epipodophyllotoxines, is often referred as multidrug resistance (MDR). MDR can be caused by several mechanisms such as increased expression of the ATP-binding cassette (ABC) transporters and drug efflux pumps like P-glycoprotein (Pgp) and multidrug resistance-associated protein 1 (MRP1). In the nineties J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_2, © Humana Press, a part of Springer Science + Business Media, LLC 2010

15

16

Nagengast et al.

the idea was that overcoming MDR would resolve the problem of intrinsic or acquired resistance for these drugs. However, randomized clinical studies in different groups of cancer patients evaluating the effect of Pgp or MRP1 modulators combined with chemotherapy have not shown any benefit in outcome for the patients (1, 2). Increasingly thereafter, the idea was that the ceiling was reached with classical chemotherapeutic drugs and that the use of a different approach, the combination with the socalled targeted agents might be superior to circumvent MDR. Over the last decade numerous targeted anticancer agents, specific for intra- and extracellular tumor targets, antigens located in the extracellular matrix or at the blood vessels of tumors, have been developed. To select patients upfront and determine early treatment response, new tracers and imaging modalities are being developed which information reflects with molecular changes during therapy. These tracers may give insight in the biological behavior of tumors and could help in the follow-up of targeted therapies and chemotherapy. In this chapter we will shortly address the methods and possibilities of imaging classical MDR mechanism. In addition, we will address the option to image targets in the tumor for targeted therapies. We will focus on HER2 and vascular endothelial growth factor (VEGF) imaging, as both can serve as target for drugs that can increase the effect of MDR drugs.

2. MDR Drug Efflux Pump Detection in Tumor and Blood– Brain Barrier 2.1. MDR Detection in Tumor

Various detection assays provide information about the presence of drug efflux pumps in the tumor at the mRNA and protein levels. However, these methods do not provide information about the dynamic function of Pgp and MRP in vivo. For that reason, to study Pgp and MRP-mediated transport, single-photon emission computed tomography (SPECT) and positron emission tomography (PET) have been used. Several substrates for Pgp and MRP have been radiolabeled with different SPECT and PET isotopes and both pre- as well as clinical studies have demonstrated the potential and Achilles’ heel of these tracers. 99 Technetium (99mTc) sestamibi, a SPECT tracer, is a substrate for Pgp and MRP and has been used in both pre- and clinical studies for tumor imaging and for visualization of Pgp-mediated transport after modulation with competing drugs of the Pgp pump (3). Kostakoglu et  al. demonstrated the potential of the tracer by a nice inverse correlation between tumor uptake of (99mTc) sestamibi and Pgp staining by immunohistochemistry (IHC) on tumor biopsies and surgical material in breast and lung cancer patients (4). Hereafter, 99mTc-sestamibi was used in breast

Multidrug Resistance in Oncology and Beyond

17

cancer patients to predict tumor responses during neoadjuvant epirubicin therapy. The tumor clearance rate (cut off £ 204 min) of 99mTc-sestamibi out of the tumor was used as a predictive marker to characterize patients. In 15 out of 17 patients (88%) with rapid tracer clearance the mastectomy specimens showed macroscopic evidence of residual tumor, which contained highly dense and viable tumor cells. This indicates a lack of tumor response to the treatment, again demonstrating a nice correlation between imaging data and ex vivo tumor analysis (5). Though in contrast to this, only 8 out of 22 (36%) patients with prolonged tracer uptake had pathologic evidence of residual tumor or showed only scattered and/or small clusters of tumor cells in a dense hyalinized stromal tissue after surgical resection of the tumor. No relationship between tracer uptake and/or patient outcome with Pgp or MRP expression was investigated in this study, which makes it difficult to interpret whether the clearance rate was Pgp or MRP dependent. Other 99mTc SPECT radiopharmaceuticals, such as 99mTc-tetrofosmin and several 99Tc-Q complexes, are also substrates for Pgp and have been tested clinically. In patients with hepatocellular carcinoma 99mTc-tetrofosmin imaging displayed a very low sensitivity for the detection of carcinomas (6). Unfortunately, tracer uptake was not correlated with Pgp expression by IHC. High Pgp expression could have explained this low sensitivity. This is supported by Wang et al. who demonstrated that a negative 99Tc-methoxyisobutyl isonitrile (MIBI) scan (68 out of 78 patients) correlated with positive Pgp expression in hepatocellular carcinoma, suggesting the ability of MDR assessment in this tumor type (7). This finding was supported by a study in which 99mTc-tetrofosmin imaging revealed lesions that were both Pgp and MRP negative by postoperative Pgp IHC in 33 patients with parathyroid adenoma. Tumors with positive Pgp and/or MRP staining could not be detected with 99m Tc-tetrofosmin imaging indicating a good correlation between imaging data and ex vivo analysis (6). The potential of response to chemotherapy prediction by MDR imaging has been most extensively investigated in lung cancer patients. In 20 patients with non-small cell lung cancer (NSCLC) a low baseline uptake of 99mTc-tetrofosmin was correlated with a poor response to paclitaxel-based chemotherapy. The authors suggested that this was due to high MDR and/or Pgp expression (8). Similar results were seen in patients with small cell lung cancer with 99mTc-tetrofosmin imaging prior to cisplatin and etoposide-based chemotherapy (9). All patients (n = 16) with a negative scan prior to chemotherapy had a poor response to therapy. However, 4 out of 23 patients with a positive 99mTc-tetrofosmin image had also a poor response, which could be due to other resistance mechanisms, which decreases the specificity of this scan, resulting in false positive scans. Furthermore, in patients with a

18

Nagengast et al.

negative 99mTc-tetrofosmin scan (low uptake) a poor correlation was found between tracer uptake and Pgp expression; 6 out of 16 patients with a negative scan did not show any Pgp expression in tumor biopsy material by IHC. Therefore, besides expression of MDR-related proteins other factors could contribute to the clearance of 99mTc-tetrofosmin in these patients. A very good correlation between 99Tc-MIBI imaging and Pgp IHC staining was found in 30 NSCLC (stage IIIb and IV) patients. All patients with a positive 99Tc-MIBI scan had negative Pgp expression and vice versa. All 15 cases with a good response (complete response or partial response assessed 3 months after completion of the treatment by clinical and radiological methods) had a positive 99 Tc-MIBI scan and negative Pgp staining. In patients who did not respond on paclitaxel-based chemotherapy both positive (5 out of 15) and negative (10 out of 15) 99Tc-MIBI scans prior to therapy were seen, which were comparable with 99mTc-tetrofosmin imaging, in false positive imaging results (9, 10). A relative large study was performed in 82 untreated breast cancer patients, which aimed to determine Pgp and MRP expression by visual and quantitative indices of double-phase 99mTc MIBI scintimammography, quantifying an early and late tumor uptake. The early [10  min postinjection of the tracer (pi)] and delayed (3 h pi) tumor to normal tissue ratio (T/N ratio) was assessed to compare the initial uptake of 99mTc MIBI and the wash-out rate of tracer out of the tumor. It was hypothesized that the wash-out rate was affected by the presence of Pgp or MRP. Both the early and delayed T/N ratio of the Pgp-negative and MRP-negative groups was higher than that of the Pgp-positive and MRP-positive groups. However, there were no significant differences in wash-out rate according to Pgp and MRP expression (11). Several other agents, including the PET tracers 11C-colchicine and 11C-verapamil have been evaluated for in vivo quantification of Pgp-mediated transport with PET imaging (3). We analyzed after iv injection the 11C-verapamil kinetics in five cancer patients using PET. One hour after injection, accumulation of 11C in lungs, heart, and tumor was respectively 43.0, 1.3, and 0.9% of the injected verapamil dose. Half-lives of 11C-verapamil in these tissues were 46.2, 73.8, and 23.7  min, respectively (12). This showed that iv administered 11C-verapamil was mainly extracted by the lungs and that efflux of 11C-verapamil administered as bolus out of solid tumor tissue is relatively fast. Specific MRP expression has been less extensively investigated even though given the fact that leukotrienes are specific substrates for MRP. Therefore, N-11C-acetyl-leukotriene E4 provides an opportunity to study MRP function noninvasively (13). Results obtained in MRP2 mutated GY/TR− rats (MRP2 is defective in GY/TR− rats due to MRP2 mutations) demonstrated visualization

Multidrug Resistance in Oncology and Beyond

19

of MRP-mediated transport. This tracer permits the study of MRP transport function abnormalities in vivo, e.g., in Dubin-Johnson patients, who are MRP2 gene deficient. Several studies have demonstrated that the substrate specificity of MRP1 is very similar to that of MRP2. Most substrates are conjugated to, or cotransported with, glutathione (GSH), glucuronide, or sulfate (14). We studied MRP2 in a different way. We analyzed the transporter specificity of the cholescintigraphic agents 99mTc-HIDA and 99mTc-MIBI, which are already used clinically for myocardial perfusion measurements. Secondly, we aimed to deduce MRP and Pgp functions in  vivo from hepatic 99mTc kinetics (14). In vitro transporter specificity was demonstrated measuring the accumulation of radioactivity in the human small cell lung cancer cell lines GLC4, GLC4/ADR150× (MRP1overexpressing/Pgp-negative), and GLC4/Pgp (Pgp-overexpre ssing). 99mTc-HIDA accumulated 5.8-fold less in GLC4/ADR150× cells than in GLC4 or GLC4/Pgp cells. In GLC4/ADR150×, the cellular 99mTc-HIDA content was 3.4-fold higher following exposure to by the MRP1,2 inhibitor MK571 (50  mM), while the MK571 had no measurable effect in GLC4 and GLC4/Pgp cells. 99m Tc-MIBI accumulated less in GLC4/Pgp and GLC4/ADR150× cells than in GLC4 cells. In vivo, bile secretion of 99mTc-HIDA was impaired in GY/TR− compared to than in control rats and not affected by GSH depletion in GY/TR− rats. Hepatic secretion of 99m Tc-HIDA was slower in GY/TR− rats (t1/2 40 min) than in control rats (t1/2 7 min). Bile secretion of 99mTc-MIBI was similar in both rat strains and impaired by GSH depletion in control rats only, indicating compensatory activity of additional transporter(s) in GY/TR- rats. 99mTc-HIDA is transported only by MRP1,2, while 99mTc-MIBI is transported by Pgp and MRP1,2. The results indicate that hepatic Pgp and MRP1,2 function can be assessed in vivo by sequential use of both radiopharmaceuticals (14). Apart from functional imaging, static Pgp expression can also be assessed with the 111In-labeled 15D3 monoclonal anti-Pgp (15). In nude BALB/c mice with subcutaneously growing human uterine sarcoma cell tumors with either high (MES-SA/D×5 1977) or low (MES-SA 1976) Pgp expression was used. Uptake was higher in the high compared to than in the low Pgp expressing tumors. All the aforementioned examples nicely illustrate that in vivo imaging of the function of ABC transporters involved in MDR in tumors is possible. However, the interest in imaging them in cancer patients is fading due to the fact that it is difficult to use this information for standard clinical patient care. This is partly the consequence of the frequent occurrence of false positive scans, which could be due to other responsible mechanisms that lead to poor patient outcome, and new targeted therapies have been currently frequently used in the treatment of patients. Therefore, new

20

Nagengast et al.

targeted therapies have emerged the need for additional new imaging probes that give information on targets involved in tumor resistance of these new therapies. 2.2. MDR Detection in Blood–Brain Barrier

The blood–brain barrier (BBB) is a major impediment for the delivery of several drugs to the brain, including cytotoxic, antiepileptic, and anti-HIV drugs. The BBB is formed by specialized capillary endothelial cells, which have several properties, such as strong tight junctions with a high electrical resistance, absence of fenestrations, and presence of efflux-pumps such Pgp and MRP1. Strategies to pass the BBB have been directed to inhibition of efflux pump function. At present, the concept of efflux pump inhibition has been most extensively studied for Pgp. Studies in rodents have shown that Pgp function can be inhibited with cyclosporine (16–18). Numerous knockout mouse studies have revealed that Pgp limits drug distribution across the mouse BBB. To determine the importance of Pgp at the human BBB, Pgp activity was analyzed with 11C-verapamil as the Pgp substrate and cyclosporine as the competing Pgp substrate. 11C-verapamil was administered to healthy volunteers (six women and six men) as an iv infusion before and at least 1 h after infusion of cyclosporine [2.5 mg × kg(−1) × h(−1)]. The brain uptake of 11C-radioactivity (brain area under the curve [AUCbrain]/blood area under the curve [AUCblood]) was determined in the presence and absence of cyclosporine. The AUCbrain/ AUCblood ratio of 11C-radioactivity was increased by 88% ± 20% in the presence of cyclosporine without affecting 11C-verapamil metabolism or plasma protein binding. The corresponding increases of 11C-verapamil in the brain white and gray matter were comparable. Beside pharmacologic modulation of Pgp by specific substrates such as cyclosporine, other therapies like radiotherapy could potentially also influence Pgp function. Early effects on brain capillary endothelium were studied by Mima et  al. (19). They exposed in rats one brain hemisphere to a single dose of 25  Gy and demonstrated that Pgp expression was lower in the irradiated hemisphere than in the nonirradiated hemisphere 5 days after irradiation. It is therefore hypothesized that brain irradiation could also be used to enhance the delivery of Pgp substrates to the brain. We irradiated the right brain hemisphere of rats with single fractions of radiotherapy to elucidate whether radiation therapy reduced Pgp expression and function in the brain, as measured with the Pgp substrate 11C-carvedilol (20). The right hemispheres received single doses of 2–25  Gy followed by 10 mg/kg of the Pgp substrate cyclosporine iv, with once 15  Gy followed by cyclosporine, or with fractionated irradiation (4 × 5  Gy) followed by cyclosporine 5 days later. Irradiation increased 11C-carvedilol uptake dose-dependently to a maximum of 20% above nonirradiated hemisphere.

Multidrug Resistance in Oncology and Beyond

21

Cyclosporine increased 11C-carvedilol uptake dose-dependently in both hemispheres, but more in the irradiated hemisphere. Fractionated irradiation resulted in a lost Pgp expression 10 days after start of irradiation, which coincided with increased 11C-carvedilol uptake. Pgp expression decreased between day 15 and 20 after single dose irradiation and increased again thereafter. Rat brain irradiation resulted in a temporary decreased Pgp function. These studies indicate that radiotherapy influences both Pgp expression and Pgp function. All the aforementioned examples illustrate that in  vivo imaging of the function of ABC transporters involved in MDR in tumors and BBB is possible. The impact for the clinic is, however, currently minor.

3. Imaging of Molecular Targets for Molecular Targeted Drugs That Could Sensitize Tumor Cells to MDRRelated Chemotherapeutic Drugs

3.1. HER2 Imaging

Over the last decade numerous new anticancer agents have entered the clinic. Many of these agents are specifically designed to target receptors, intracellular proteins, or ligands, which are overexpressed by tumor cells. Two examples that have had a big impact clinically are the monoclonal antibodies trastuzumab (Herceptin®) and bevacizumab (Avastin®), both improving patient outcome when combined with MDR-related chemotherapeutic drugs. Trastuzumab binds to the HER2/neu receptor and bevacizumab binds to VEGF-A. Combining trastuzumab with paclitaxel leads to increased progression-free survival and overall survival in metastatic breast cancer (21). Combining bevacizumab with paclitaxel increases progression-free survival in metastatic breast cancer (22). Despite this progress, there still remain a significant number of patients who do not respond to these agents or become resistant. Potentially, visualizing their specific molecular drug targets in the tumor can function as biomarkers to support treatment decision for the individual patient. HER2 imaging and VEGF imaging are discussed as examples of this new imaging approach. The HER2 antibody trastuzumab potentiates the antitumor effect of the MDR drugs, the taxanes. As a consequence, trastuzumab reduces taxane resistance. For antitumor efficacy of trastuzumab, the HER2 gene has to be amplified, as is the case in about 20–30% of the breast tumors (23). The HER2 tumor expression can vary during the treatment of a patient and can differ across metastatic lesions within a patient (24). Therefore, there is a need for methods that are able to assess the HER2 status repeatedly, preferable in all lesions and noninvasively. Molecular imaging of the HER2 receptor could serve this aim. In addition to this diagnostic purpose, HER2

22

Nagengast et al.

imaging could also be used as an early predictive marker for the tumor response to HER2 targeted or HER2 degrading therapies. HER2 imaging starts with selecting a suitable HER2 targeted molecule. Currently available HER2 targeted molecules include full-length monoclonal antibodies, Fab-fragments, F(ab¢)2fragments, diabodies, minibodies, affibodies, scFv-Fc, and peptides. In order to image the HER2 targeted molecule, and thus the receptor, labeling with a radionuclide tracer or fluorescent dye must first be accomplished. The label should be suitable for the proposed imaging technique: PET, SPECT, MRI, or optical imaging (25). For radio-labeling purposes, the physical half-life of the isotope must suit the biological half-life of the HER2 targeting molecule to reduce tumor-to-background ratios and to allow imaging at the optimal time-point. This implicates that full-length monoclonal antibodies are mostly radiolabeled with long-lived isotopes and the smaller HER2 targeting molecules, which have a more rapid clearance, are radiolabeled with shorter-lived isotopes. Full-length HER2 monoclonal antibodies have been radiolabeled with 125/131I, 111In, and 99mTc for HER2 SPECT/gamma camera imaging (26–32) and with 124I, 86Y, 76Br, and 89Zr for HER2 PET imaging (33–38). The smaller HER2 targeting antibody fragments, proteins, and peptides have been labeled with 111In, 125I, and 99mTc for HER2 SPECT/gamma camera imaging (39–45) and with 18F, 68Ga, 64Cu, 124I, and 76Br for HER2 PET imaging (46–51). We have imaged HER2 in patients using 111In-trastuzumab (Fig.  2.1a). SPECT-imaging with 111In-trastuzumab was able to visualize previously unidentified lesions in 13 out of 15 patients (52). Since PET-imaging provides a higher spatial resolution, a better signal-to-noise ratio, and is more quantitative than SPECT, we currently are evaluating the use of 89Zr-trastuzumab for clinical PET-imaging of the HER2-receptor in metastatic breast cancer. Preliminary data of this clinical evaluation show excellent tumor uptake of 89Zr-trastuzumab and high-resolution images (38). Smith-Jones et al. preclinically studied the HER2 response of the HER2 targeted therapy with 17-allylamino-17-demethoxygeldanamycin (17-AAG; tanespimycin), a Heat Shock Protein 90 (HSP90)-inhibitor (48). HSPs are molecular chaperones involved in maintaining the conformation, stability, cellular localization, and activity of several key oncogenic client proteins (53). Client proteins of HSP90 include the key regulator of VEGF expression HIF-1a, HER2, hormone receptors, and others (53). Treatment with HSP90 inhibitors leads to HER2 degradation. The pharmacodynamics of HER2 degradation induced by 17-AAG was visually and quantitatively evaluated by PET imaging with a 68Ga labeled F(ab’)2 fragment of trastuzumab. Results show that PET imaging with this fragment was able to quantify the HER2 response after 48 h as the pharmocodynamic effect of

Multidrug Resistance in Oncology and Beyond

23

Fig. 2.1. A representative 111In-Trastuzumab image of a metastatic HER2 positive breast cancer lesion. (a) Fused computed tomography (CT) with indium-111 – diethylenetriamine penta-acetic acid anhydride (111In-DTPA) – trastuzumab singlephoton emission tomography (SPECT) image (96 h after tracer injection). Tumor indicated by arrow. Reproduced with permission of The Journal of Clinical Oncology. (b) Coronal 89Zr-bevacizumab MicroPET image and fused 3D 89Zr-bevacizumab/ CT image 144 h post injection. Tumor indicated by arrow.

17-AAG (54). Secondly, the 68Ga labeled F(ab’)2 PET visualized HER2 response to 17-AAG was compared with response measurement by 18FDG-PET (55), which visualizes glucose metabolism and is being used for detection, staging, and response monitoring in breast cancer patients (56). Quantification of the 68 Ga labeled F(ab’)2 PET-data showed a 70% reduction in tracer uptake 24  h after treatment with 17-AAG. With 18FDG-PET there was no significant difference in tumor uptake between 17-AAG treated and control mice in the 3 weeks post treatment.

24

Nagengast et al.

The clinical success of trastuzumab is limited by resistance for this drug, which can among others be due to alterations in receptor– antibody interaction. HER2 imaging could therefore potentially be used to elucidate altered receptor–antibody interaction in trastuzumab-resistant tumors (57). 3.2. VEGF Level Imaging

The development of vascular supply, the so-called angiogenesis, is important in the development and growth of tumors. Without a vascular network, tumors cannot grow above several mm3 in size (58, 59). Newly formed blood vessels supply the tumor with nutrients and oxygen, disposal of metabolic waste products and provide route for metastatic spreading. Compared to normal blood vessels, tumor vasculature is structurally and functionally abnormal, presumably to an imbalance of proangiogenic and antiangiogenic growth factors produced by tumor cells (60, 61). This abnormality causes interstitial hypertension, hypoxia, and acidosis, which makes tumors more aggressive. This is associated with chemo- and radiotherapy resistance (60, 62, 63). One of the most important factors involved in angiogenesis is VEGF. VEGF exists of at least five different isoforms A–E. VEGF-A (hereafter referred as VEGF) is one of the most important growth factors involved in tumor angiogenesis with different splice variants like VEGF 121, 145, 165, 189, and 206. VEGF121 is freely soluble, whereas VEGF165 is secreted, though a significant fraction remains localized to the extracellular matrix, such as VEGF189 and VEGF206 (64). The actions of VEGF are transmitted by the VEGF-receptors VEGFR-1, VEGFR-2 (KDR/Flk-1), being the most abundant in tumor angiogenesis, and VEGFR-3 (Flt-4) (65). The VEGFRs are tyrosine kinase-mediated transmembrane receptors, present on endothelial cells, circulating endothelial progenitor cells and both epithelial and mesenchymal tumor cells (66). The role of these receptors on the surface of tumor cells is not completely elucidated. Physiologically, large amounts of VEGF are stored in blood platelets and only small amounts are freely soluble (67). In tumors there is an unproportional upregulation of VEGF production, which leads to locally high VEGF levels mainly located in the extracellular matrix. All this leads to paracrine effects in the tumor tissue, a proangiogenic status, thus altering the microenvironment of the tumor. This results in increased tumor growth and increased interstitial pressure, which leads to hypoxia and elevated production of ongogenic transcription factors, such as hypoxia inducible factor (HIF) (68). HIF-1a is one of the cellular key regulators of the transcription of VEGF. Furthermore, HIF is a potent transcription factor involved in the regulation and activation of multiple oncogenic pathways, increased glucose uptake, and increased homing of endothelial progenitor cells and pericyte progenitor cells. It is also involved in chemotherapeutic drug resistance, for example, by

Multidrug Resistance in Oncology and Beyond

25

increased transcription of the MDR-1 gene, resulting in increased Pgp expression (69, 70). Despite the fact that the importance of tumor angiogenesis in tumor progression was already postulated decades ago, it was only recently that clinical advancement has been made (71, 72). The first clinical breakthrough came from bevacizumab, a humanized monoclonal antibody that neutralizes all isoforms of VEGF-A. In patients with metastatic breast, the addition of bevacizumab to the MDR-associated drug paclitaxel leads to an increased response rate and increased progression-free survival, thus decreasing drug resistance (22). One of the proposed working mechanisms of this anti-VEGF targeted therapy is that treatment results in vessel normalization and thereby increased perfusion and a decline of interstitial fluid pressure, which coincides with an increased penetration and efficacy of chemotherapeutic drugs (62). Other antiangiogenic strategies that have shown clinical efficacy are tyrosine kinase inhibitors (TKIs), which block the signal transduction of VEGFRs on endothelial and tumor cells. More recently, progress has been made in the inhibition of intracellular targets such as HSP90 and the mammalian target of rapamycin (mTOR) pathway. HSP90 inhibition reduces angiogenesis through HIF-1a inhibition resulting in a reduction of VEGF secretion and other HIF-1a activated genes (73–75). For example, a recent study demonstrated that the inhibition by means of 17-AAG, reduced VEGF secretion by 60% under hypoxic conditions in pancreatic cancer cells leading to e.a. decreased mean vessel density (MVD) in a xenograft model (74). The mTOR pathway plays a key role in regulating cancer cell proliferation, tumor growth, and angiogenesis though altering HIF-1 and VEGF expression by its upstream pathways such as phosphoinositide 3-kinases (PI3Ks), AKT and extracellular signal-regulated kinases (ERKs) (76). Treatment with RAD001, an oral mTOR inhibitor, resulted in decreased VEGF expression and decreased MVD in a transgenic mouse model of ovarian cancer (77). Furthermore, the mTOR pathway is involved in specific drug resistance mechanism. For example, upstream PI3K activation leads to MRP1 expression and subsequent chemoresistance in advanced prostate cancer cells (78). These examples nicely demonstrate the close and complex interaction between classic chemotherapeutic MDR, angiogenesis, and tumor progression. Therefore, new antiangiogenic therapies, such as HSP90 and mTOR inhibition, could facilitate to overcome resistance to classic chemotherapeutics and targeted therapies. To select patients who benefit from this targeted therapy, and to follow up new treatment regimes, molecular imaging of transcription factors and key proteins of these pathways, by using specific molecular imaging probes, is of great interest. VEGF is an interesting example key protein. It is an important downstream

26

Nagengast et al.

protein produced as a result of multiple processes (hypoxiaea), activation of growth factor receptors (EGFR, HER, etc.) and intracellular proteins (HIF-1a, mToR, etc.). Therefore, VEGF could potentially serve as a representative biomarker for early prediction of these targeted therapies. Furthermore, VEGF imaging could give insight in drug resistance. For example, acquired resistance in a breast cancer xenograft model for trastuzumab was associated with increased expression of VEGF (79). Treatment with bevacizumab in this setting, and thus lowering the increased VEGF levels, could overcome resistance for trastuzumab, which resulted in delayed tumor progression. To date, several radiolabeled anti-VEGF antibodies and Fabfragments have been used for the development of VEGF imaging: VG76e, HumMV833, bevacizumab, and ranibizumab (80–82). 125 I- and 124I-labeled VG76e, an IgG1 mouse monoclonal antiVEGF antibody, which recognizes the 121, 165, and 189 isoforms of human VEGF-A, showed specific tumor targeting in a human fibrosarcoma xenograft mice model. Jayson et  al. used 124 I-HuMV833, a humanized monoclonal IgG4k antibody that binds VEGF121 and VEGF165, to perform PET-imaging studies in patients with various progressive solid tumors (82). Tumor uptake of 124I-HuMV833 was highly variable between and within patients. For example, there was high uptake of 124I-HuMV833 in an ovarian tumor and low uptake in a poorly vascularized metastasis in a colon cancer patient (81). These differences are very likely to present the variation VEGF secretion among tumor types and lesions. Radiolabeled bevacizumab showed specific tumor uptake in a human ovarian xenograft model (Fig.  2.1b) (80). MicroPET imaging using 89Zirconium (89Zr) labeled bevacizumab showed clear tumor localization 72 h post injection with maximal uptake 168 h post injection (80). Uptake could be quantified noninvasively, allowing follow-up of VEGF secretion during therapy. Comparable results were seen using 89Zr- and 18Fluor (18F)-labeled ranibizumab, a Fab-fragment binding to all VEGF-A isoforms. Though, due to fast distribution and clearance of the Fab-fragment images could be made earlier; already 3  h post injection of the tracer, though absolute tumor uptake is lower compared to bevacizumab (83). This tracer could be attractive to follow up rapid changes of VEGF secretion following therapy. Preliminary clinical SPECT imaging using 111indium (111In)-labeled bevacizumab revealed tumor lesions in both recurrent melanoma and metastatic colon cancer patients (84). Future studies, both pre- as well as clinically can potentially further elucidate the role of VEGFimaging in the assessment of drug resistance for chemotherapeutics and the response evaluation of new molecular targeted therapies. All this could lead to more patient tailored therapy.

Multidrug Resistance in Oncology and Beyond

27

4. Conclusion To overcome MDR and increase the efficacy of chemotherapy new targeted therapies have to be added. At this moment, blockade of specific growth factor receptors, intracellular targets, and tyrosine kinase signaling has increased the efficacy of classic chemotherapy in several cancer types. However, despite this success a lot of patients do not benefit of the addition of these therapies. To select patients up front and follow up treatment response, new tracers and imaging modalities that represent changes in intra- and extracellular tumor targets, antigens located in the extracellular matrix or at the blood vessels of tumors during therapy might support treatment follow-up. As indicated, growth factor receptors present on the membrane of tumor cells, such as HER2, EGFR, etc., are suitable candidates for this. Another approach is to use a downstream product whose transcription is increased in MDR and by other oncogenic processes. As indicated, VEGF is such a target that could serve as a specific readout modality for MDR and the response of new targeted therapies. Besides intact mAb molecules such as trastuzumab and bevacizumab (molecular weight, 150 kDa), mAb fragments and engineered variants are also used, like F(ab)2, F(ab), Fab, single chain Fv (scFv), and the covalent dimers scFv2, diabodies, and minibodies (molecular weights ranging from 25 to 100 kDa), as well as several types of protein therapeutics based on nontraditional scaffolds, like, for example, domain antibodies, affibodies, nanobodies, and anticalins could be used for this purpose (85). During the development of these tracers, one of the main goals should be to observe whether baseline values and/or changes during therapy correspond with patient outcome and ultimately patient survival. All this will lead to more patient-tailored therapy.

Acknowledgments Supported by a personal grant (W.B. Nagengast) and grant RUG 2007-3739 of the Dutch Cancer Society. References 1. Hait WN, Yang JM (2005) Clinical management of recurrent breast cancer: development of multidrug resistance (MDR) and strategies to circumvent it. Semin Oncol 32: S16–S21

2. Pusztai L, Wagner P, Ibrahim N et al (2005) Phase II study of tariquidar, a selective P-glycoprotein inhibitor, in patients with chemotherapy-resistant, advanced breast carcinoma. Cancer 104:682–691

28

Nagengast et al.

3. Hendrikse NH, Franssen EJ, van der Graaf WT, Vaalburg W, de Vries EG (1999) Visualization of multidrug resistance in vivo. Eur J Nucl Med 26:283–293 4. Kostakoglu L, Elahi N, Kiratli P et al (1997) Clinical validation of the influence of P-glycoprotein on technetium-99m-sestamibi uptake in malignant tumors. J Nucl Med 38:1003–1008 5. Ciarmiello A, Del Vecchio S, Silvestro P et al (1998) Tumor clearance of technetium 99m-sestamibi as a predictor of response to neoadjuvant chemotherapy for locally advanced breast cancer. J Clin Oncol 16:1677–1683 6. Ho YJ, Jeng LB, Yang MD et al (2003) A trial of single photon emission computed tomography of the liver with technetium-99m tetrofosmin to detect hepatocellular carcinoma. Anticancer Res 23:1743–1746 7. Wang H, Chen XP, Qiu FZ (2004) Correlation of expression of multidrug resistance protein and messenger RNA with 99mTc-methoxyisobutyl isonitrile (MIBI) imaging in patients with hepatocellular carcinoma. World J Gastroenterol 10:1281–1285 8. Kao CH, Hsieh JF, Tsai SC et  al (2001) Paclitaxel-Based chemotherapy for non-small cell lung cancer: predicting the response with 99mTc-tetrofosmin chest imaging. J Nucl Med 42:17–20 9. Yeh JJ, Hsu WH, Huang WT et  al (2003) Technetium-99m tetrofosmin SPECT predicts chemotherapy response in small cell lung cancer. Tumour Biol 24:151–155 10. Hsu WH, Yen RF, Kao CH et  al (2002) Predicting chemotherapy response to paclitaxel-based therapy in advanced non-small-cell lung cancer (stage IIIb or IV) with a higher T stage (> T2). Technetium-99m methoxyisobutylisonitrile chest single photon emission computed tomography and P-glycoprotein express ion. Oncology 63:173–179 11. Kim IJ, Bae YT, Kim SJ et  al (2006) Determination and prediction of P-glycoprotein and multidrug-resistance-related protein expression in breast cancer with double-phase technetium-99m sestamibi scintimammography. Visual and quantitative analyses. Oncology 70:403–410 12. Hendrikse NH, de Vries EG, Franssen EJ, Vaalburg W, van der Graaf WT (2001) In vivo measurement of [11C]verapamil kinetics in human tissues. Eur J Clin Pharmacol 56: 827–829 13. Guhlmann A, Krauss K, Oberdorfer F et  al (1995) Noninvasive assessment of hepatobiliary and renal elimination of cysteinyl leukotrienes

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

by positron emission tomography. Hepatology 21:1568–1575 Hendrikse NH, Kuipers F, Meijer C et  al (2004) In vivo imaging of hepatobiliary transport function mediated by multidrug resistance associated protein and P-glycoprotein. Cancer Chemother Pharmacol 54:131–138 van Eerd JE, de Geus-Oei LF, Oyen WJ, Corstens FH, Boerman OC (2006) Scintigraphic imaging of P-glycoprotein expression with a radiolabelled antibody. Eur J Nucl Med Mol Imaging 33:1266–1272 Hendrikse NH, Schinkel AH, de Vries EG et  al (1998) Complete in  vivo reversal of P-glycoprotein pump function in the blood– brain barrier visualized with positron emission tomography. Br J Pharmacol 124:1413–1418 Hendrikse NH, de Vries EG, Eriks-Fluks L et al (1999) A new in  vivo method to study P-glycoprotein transport in tumors and the blood–brain barrier. Cancer Res 59:2411–2416 Syvanen S, Blomquist G, Sprycha M et  al (2006) Duration and degree of cyclosporin induced P-glycoprotein inhibition in the rat blood–brain barrier can be studied with PET. Neuroimage 32:1134–1141 Mima T, Toyonaga S, Mori K, Taniguchi T, Ogawa Y (1999) Early decrease of P-glycoprotein in the endothelium of the rat brain capillaries after moderate dose of irradiation. Neurol Res 21:209–215 Bart J, Nagengast WB, Coppes RP et  al (2007) Irradiation of rat brain reduces P-glycoprotein expression and function. Br J Cancer 97:322–326 Slamon DJ, Leyland-Jones B, Shak S et  al (2001) Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 344:783–792 Miller K, Wang M, Gralow J et  al (2007) Paclitaxel plus bevacizumab versus paclitaxel alone for metastatic breast cancer. N Engl J Med 357:2666–2676 Hayes DF, Thor AD, Dressler LG et al (2007) HER2 and response to paclitaxel in nodepositive breast cancer. N Engl J Med 357: 1496–1506 Zidan J, Dashkovsky I, Stayerman C et  al (2005) Comparison of HER-2 overexpression in primary breast cancer and metastatic sites and its effect on biological targeting therapy of metastatic disease. Br J Cancer 93:552–556 Cai W, Niu G, Chen X (2008) Multimodality imaging of the HER-kinase axis in cancer. Eur J Nucl Med Mol Imaging 35:186–208

Multidrug Resistance in Oncology and Beyond 26. De SK, Slamon D, Anderson SK et al (1992) Radiolabeled antibody targeting of the HER-2/ neu oncoprotein. Cancer Res 52:1916–1923 27. Zalutsky MR, Xu FJ, Yu Y et  al (1999) Radioiodinated antibody targeting of the HER-2/neu oncoprotein: effects of labeling method on cellular processing and tissue distribution. Nucl Med Biol 26:781–790 28. Tsai SW, Sun Y, Williams LE et  al (2000) Biodistribution and radioimmunotherapy of human breast cancer xenografts with radiometallabeled DOTA conjugated anti-HER2/neu antibody 4D5. Bioconjug Chem 11:327–334 29. Lub-de Hooge MN, Kosterink JG, Perik PJ et  al (2004) Preclinical characterisation of 111In-DTPA-trastuzumab. Br J Pharmacol 143:99–106 30. Allan SM, Dean C, Fernando I et  al (1993) Radioimmunolocalisation in breast cancer using the gene product of c-erbB2 as the target antigen. Br J Cancer 67:706–712 31. Persson M, Tolmachev V, Andersson K et  al (2005) [(177)Lu]pertuzumab: experimental studies on targeting of HER-2 positive tumour cells. Eur J Nucl Med Mol Imaging 32:1457–1462 32. Meenakshi A, Kumar RS, Ganesh V, Kumar NS (2002) Preliminary study on radioimmunodiagnosis of experimental tumor models using technetium-99m-labeled anti-C-erbB-2 monoclonal antibody. Tumori 88:507–512 33. Bakir MA, Eccles S, Babich JW et  al (1992) C-erbB2 protein overexpression in breast cancer as a target for PET using iodine-124-labeled monoclonal antibodies. J Nucl Med 33:2154–2160 34. Garmestani K, Milenic DE, Plascjak PS, Brechbiel MW (2002) A new and convenient method for purification of 86Y using a Sr(II) selective resin and comparison of biodistribution of 86Y and 111In labeled Herceptin. Nucl Med Biol 29:599–606 35. Bruskin A, Sivaev I, Persson M et  al (2004) Radiobromination of monoclonal antibody using potassium [76Br] (4 isothiocyanatobenzyl-ammonio)-bromo-decahydro-closododecaborate (Bromo-DABI). Nucl Med Biol 31:205–211 36. Winberg KJ, Persson M, Malmstrom PU, Sjoberg S, Tolmachev V (2004) Radiobromination of anti-HER2/neu/ ErbB-2 monoclonal antibody using the p-isothiocyanatobenzene derivative of the [76Br] undecahydro-bromo-7, 8-dicarba-nido-undecaborate(1-) ion. Nucl Med Biol 31:425–433 37. Mume E, Orlova A, Malmstrom PU et al (2005) Radiobromination of humanized anti-HER2

38. 39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

29

monoclonal antibody trastuzumab using N-succinimidyl 5-bromo-3-pyridinecarboxylate, a potential label for immunoPET. Nucl Med Biol 32:613–622 Dijkers ECF, Lub-de Hooge MN, Kosterink JG et  al (2007) Proc Am Soc Clin Oncol Annu Meet 25(18S):3508 Tang Y, Wang J, Scollard DA et  al (2005) Imaging of HER2/neu-positive BT-474 human breast cancer xenografts in athymic mice using (111)In-trastuzumab (Herceptin) Fab fragments. Nucl Med Biol 32:51–58 Tang Y, Scollard D, Chen P et  al (2005) Imaging of HER2/neu expression in BT-474 human breast cancer xenografts in athymic mice using [(99m)Tc]-HYNIC-trastuzumab (Herceptin) Fab fragments. Nucl Med Commun 26:427–432 Adams GP, McCartney JE, Tai MS et  al (1993) Highly specific in  vivo tumor targeting by monovalent and divalent forms of 741F8 anti-c-erbB-2 single-chain Fv. Cancer Res 53:4026–4034 Adams GP, McCartney JE, Wolf EJ et  al (1995) Enhanced tumor specificity of 741F8–1 (sFv’)2, an anti-c-erbB-2 single-chain Fv dimer, mediated by stable radioiodine conjugation. J Nucl Med 36:2276–2281 Adams GP, Schier R, McCall AM et al (1998) Prolonged in  vivo tumour retention of a human diabody targeting the extracellular domain of human HER2/neu. Br J Cancer 77:1405–1412 Orlova A, Magnusson M, Eriksson TL et  al (2006) Tumor imaging using a picomolar affinity HER2 binding affibody molecule. Cancer Res 66:4339–4348 Tran T, Engfeldt T, Orlova A et  al (2007) (99m)Tc-maEEE-Z(HER2:342), an Affibody molecule-based tracer for the detection of HER2 expression in malignant tumors. Bioconjug Chem 18:1956–1964 Cheng Z, De Jesus OP, Namavari M et  al (2008) Small-animal PET imaging of human epidermal growth factor receptor type 2 expression with site-specific 18F-labeled protein scaffold molecules. J Nucl Med 49:804–813 Kramer-Marek G, Kiesewetter DO, Martiniova L et  al (2008) [18F]FBEM-Z(HER2:342)Affibody molecule-a new molecular tracer for in  vivo monitoring of HER2 expression by positron emission tomography. Eur J Nucl Med Mol Imaging 35:1008–1018 Smith-Jones PM, Solit DB, Akhurst T et  al (2004) Imaging the pharmacodynamics of HER2 degradation in response to Hsp90 inhibitors. Nat Biotechnol 22:701–706

30

Nagengast et al.

49. Olafsen T, Kenanova VE, Sundaresan G et al (2005) Optimizing radiolabeled engineered anti-p185HER2 antibody fragments for in vivo imaging. Cancer Res 65:5907–5916 50. Robinson MK, Doss M, Shaller C et al (2005) Quantitative immuno-positron emission tomography imaging of HER2-positive tumor xenografts with an iodine-124 labeled antiHER2 diabody. Cancer Res 65:1471–1478 51. Mume E, Orlova A, Larsson B et  al (2005) Evaluation of ((4-hydroxyphenyl)ethyl) maleimide for site-specific radiobromination of anti-HER2 affibody. Bioconjug Chem 16:1547–1555 52. Perik PJ, Lub-de Hooge MN, Gietema JA et al (2006) Indium-111-labeled trastuzumab scintigraphy in patients with human epidermal growth factor receptor 2-positive metastatic breast cancer. J Clin Oncol 24:2276–2282 53. Neckers L (2007) Heat shock protein 90: the cancer chaperone. J Biosci 32:517–530 54. Smith-Jones PM, Solit DB, Akhurst T et  al (2004) Imaging the pharmacodynamics of HER2 degradation in response to Hsp90 inhibitors. Nat Biotechnol 22:701–706 55. Smith-Jones PM, Solit D, Afroze F, Rosen N, Larson SM (2006) Early tumor response to Hsp90 therapy using HER2 PET: comparison with 18F-FDG PET. J Nucl Med 47:793–796 56. Eubank WB, Mankoff DA (2005) Evolving role of positron emission tomography in breast cancer imaging. Semin Nucl Med 35:84–99 57. Nahta R, Yu D, Hung MC, Hortobagyi GN, Esteva FJ (2006) Mechanisms of disease: understanding resistance to HER2-targeted therapy in human breast cancer. Nat Clin Pract Oncol 3:269–280 58. Folkman J (1990) What is the evidence that tumors are angiogenesis dependent? J Natl Cancer Inst 82:4–6 59. Folkman J, Klagsbrun M (1987) Angiogenic factors. Science 235:442–447 60. Jain RK (2008) Lessons from multidisciplinary translational trials on anti-angiogenic therapy of cancer. Nat Rev Cancer 8:309–316 61. Varey AH, Rennel ES, Qiu Y et  al (2008) VEGF(165)b, an antiangiogenic VEGF-A isoform, binds and inhibits bevacizumab treatment in experimental colorectal carcinoma: balance of pro- and antiangiogenic VEGF-A isoforms has implications for therapy. Br J Cancer 98:1366–1379 62. Jain RK, Duda DG, Clark JW, Loeffler JS (2006) Lessons from phase III clinical trials on anti-VEGF therapy for cancer. Nat Clin Pract Oncol 3:24–40

63. Willett CG, Boucher Y, di Tomaso E et  al (2004) Direct evidence that the VEGF-specific antibody bevacizumab has antivascular effects in human rectal cancer. Nat Med 10:145–147 64. Park JE, Keller GA, Ferrara N (1993) The vascular endothelial growth factor (VEGF) isoforms: differential deposition into the subepithelial extracellular matrix and bioactivity of extracellular matrix-bound VEGF. Mol Biol Cell 4:1317–1326 65. Mishani E, Abourbeh G, Rozen Y et al (2004) Novel carbon-11 labeled 4-dimethylaminobut-2-enoic acid [4-(phenylamino)-quinazoline-6-yl]-amides: potential PET bioprobes for molecular imaging of EGFR-positive tumors. Nucl Med Biol 31:469–476 66. Holliger P, Hudson PJ (2005) Engineered antibody fragments and the rise of single domains. Nat Biotechnol 23:1126–1136 67. Nguyen DM, Lorang D, Chen GA et  al (2001) Enhancement of paclitaxel-mediated cytotoxicity in lung cancer cells by 17-allylamino geldanamycin: in vitro and in vivo analysis. Ann Thorac Surg 72:371–378 68. Ferrara N (2004) Vascular endothelial growth factor: basic science and clinical progress. Endocr Rev 25:581–611 69. Comerford KM, Wallace TJ, Karhausen J et al (2002) Hypoxia-inducible factor-1-dependent regulation of the multidrug resistance (MDR1) gene. Cancer Res 62:3387–3394 70. Liu L, Ning X, Sun L et al (2008) Hypoxiainducible factor-1 alpha contributes to hypoxia-induced chemoresistance in gastric cancer. Cancer Sci 99:121–128 71. Cai W, Olafsen T, Zhang X et al (2007) PET imaging of colorectal cancer in xenograftbearing mice by use of an 18F-labeled T84.66 anti-carcinoembryonic antigen diabody. J Nucl Med 48:304–310 72. Hurwitz H, Fehrenbacher L, Novotny W et al (2004) Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 350:2335–2342 73. Lang SA, Klein D, Moser C et  al (2007) Inhibition of heat shock protein 90 impairs epidermal growth factor-mediated signaling in gastric cancer cells and reduces tumor growth and vascularization in  vivo. Mol Cancer Ther 6:1123–1132 74. Lang SA, Moser C, Gaumann A et al (2007) Targeting heat shock protein 90 in pancreatic cancer impairs insulin-like growth factor-I receptor signaling, disrupts an interleukin-6/ signal-transducer and activator of transcription 3/hypoxia-inducible factor-1alpha autocrine

Multidrug Resistance in Oncology and Beyond

75.

76.

77.

78.

79.

loop, and reduces orthotopic tumor growth. Clin Cancer Res 13:6459–6468 Eccles SA, Massey A, Raynaud FI et al (2008) NVP-AUY922: a novel heat shock protein 90 inhibitor active against xenograft tumor growth, angiogenesis, and metastasis. Cancer Res 68:2850–2860 Jiang BH, Liu LZ (2008) Role of mTOR in anticancer drug resistance: Perspectives for improved drug treatment. Drug Resist Updat 11:63–76 Mabuchi S, Altomare DA, Connolly DC et al (2007) RAD001 (Everolimus) delays tumor onset and progression in a transgenic mouse model of ovarian cancer. Cancer Res 67:2408–2413 Lee JT Jr, Steelman LS, McCubrey JA (2004) Phosphatidylinositol 3’-kinase activation leads to multidrug resistance protein-1 expression and subsequent chemoresistance in advanced prostate cancer cells. Cancer Res 64: 8397–8404 du Manoir JM, Francia G, Man S et al (2006) Strategies for delaying or treating in vivo acquired resistance to trastuzumab in human breast cancer xenografts. Clin Cancer Res 12:904–916

31

80. Nagengast WB, de Vries EG, Hospers GA et al (2007) In vivo VEGF imaging with radiolabeled bevacizumab in a human ovarian tumor xenograft. J Nucl Med 48:1313–1319 81. Jayson GC, Zweit J, Jackson A et  al (2002) Molecular imaging and biological evaluation of HuMV833 anti-VEGF antibody: implications for trial design of antiangiogenic antibodies. J Natl Cancer Inst 94:1484–1493 82. Collingridge DR, Carroll VA, Glaser M et al (2002) The development of [(124)I]iodinated-VG76e: a novel tracer for imaging vascular endothelial growth factor in vivo using positron emission tomography. Cancer Res 62:5912–5919 83. Nagengast WB, de Vries EG, Warnders FJ et  al (2008) Proc. Am. Assoc. Cancer Res. Ann, Meeting Abstract 3161 84. Nagengast WB, Lub-de Hooge MN, Hospers GA et al (2008) Proc Am Soc Clin Oncol Ann Meet 26(15S):3547 85. Van Dongen GA, Visser GW, Lub-De Hooge MN, De Vries EG, Perk LR (2007) ImmunoPET: a navigator in monoclonal antibody development and applications. Oncologist 12:1379–1389

sdfsdf

Chapter 3 Studying Drug Resistance Using Genetically Engineered Mouse Models for Breast Cancer Sven Rottenberg, Marina Pajic, and Jos Jonkers Abstract The generation of genetically engineered mouse models (GEMMs) that mimic breast cancer in humans provides new tools to investigate mechanisms of drug resistance in vivo. The advantages are manifold: inbred mice do not have the genomic heterogeneity seen in patients; mammary tumors are superficial and therefore easily accessible for measurement and sampling pre- and posttreatment; tumors can be transplanted orthotopically into syngeneic, immunocompetent animals; and tumor cells can be modified in vitro (e.g., gene overexpression, shRNA knockdown, insertional mutagenesis) prior to transplantation. Here, we provide an overview with experimental details of various approaches to study mechanisms of drug resistance in GEMMs for breast cancer. Key words: GEMM, Breast cancer, Orthotopic transplantation, BRCA1, BRCA2, E-cadherin, ATP-binding cassette (ABC) transporter, FACS sorting, Tumor-initiating cells

1. Introduction In humans, breast cancers display a variety of responses to anticancer drugs, ranging from pathological complete response to progressive disease. In combination with local therapy (radiation or surgical removal) systemic therapy is frequently curative at an early stage of the disease. However, once solid tumors are disseminated and therapeutic success relies on chemotherapy only, complete tumor eradication is unlikely, even if tumors are initially very sensitive to drug. Intrinsic or acquired multidrug resistance of tumor cells eventually results in therapy failure. As underlying cause of (multi)drug resistance a range of different mechanisms have been identified using cell lines derived from human tumors or tumor samples from patients. These mechanisms include alterations in drug accumulation/metabolism, in cellular targets, or in DNA damage repair (1–5). Since one of the ways in which drugs kill J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_3, © Humana Press, a part of Springer Science + Business Media, LLC 2010

33

34

Rottenberg, Pajic and Jonkers

cells is the induction of apoptosis/senescence, it has been postulated that abrogation of cell death pathways may also affect the action of anticancer drugs (6–10). The relevance of each of these mechanisms in real human tumors, however, usually remains to be determined. To study the contribution of specific mechanisms to drug resistance in an in vivo setting, mouse models that mimic human cancer are useful. Xenotransplantation models provide an alternative, but thus far these were rarely successful in the study of drug resistance, as reported elsewhere (11, 12). However, Quintana et al. recently reported a significant optimization in the transplantation of melanoma cells, resulting in tumor outgrowth after transplantation of only a few cells (13). Possibly this approach might also yield higher transplantation take rates for breast cancer xenotransplants. Other improvements are orthotopic grafting of human breast cancer cell lines into “humanized” stroma (14) or xenotransplantation of fresh breast tumor samples (15). Whether these advances will result in better models to study chemotherapy resistance remains to be seen. Hence, the study of anticancer drug resistance in “human-like” spontaneous tumors in genetically engineered mouse models (GEMMs) (12) remains an important alternative. First-generation GEMMs are transgenic mice with tissue-specific expression of oncogenes or dominant-negative tumor suppressor genes (16). More advanced, second-generation GEMMs are based on conditional knock-out and knock-in mice with spatiotemporally restricted targeted mutations in tumor suppressor genes or oncogenes (16, 17). Several GEMMs have been developed and validated to recapitulate key features of their human archetype. An overview of available models can be found under http://emice.nci.nih.gov/mouse_models. For the study of anticancer drug resistance in breast cancer we have used GEMMs for BRCA1- and BRCA2-associated hereditary breast cancer (K14cre; Brca1F/F; p53F/F or K14cre; Brca2F/F; p53F/F) as well as models for E-cadherin-mutated lobular breast cancer (K14cre; EcadF/F; p53F/F and WAPcre; EcadF/F; p53F/F) (18–22). In these models, spontaneously developing mammary tumors resemble the natural history of human tumors, share many histological features of their corresponding human tumors, express similar markers and in the case of the BRCA1/BRCA2 models also possess a high degree of genomic instability. In addition, they are convenient for drug studies since growth of the relatively superficial tumors is easily measured, and tumor samples can be taken before, during, and after drug treatment. In this chapter, we outline various techniques aimed at studying intrinsic or acquired drug resistance in GEMMs for breast cancer. A summary of the different procedures is presented in Table 3.1. Several of these approaches may also be useful for other tumor types generated in GEMMs.

Studying Drug Resistance Using Genetically Engineered Mouse

35

Table 3.1 Summary of the different procedures for studying intrinsic or acquired drug resistance in GEMMs for breast cancer Opportunities

Hurdles

1. GEMMs bearing spontaneous mammary tumors

Spontaneous tumor development mimicking cancer development in humans

Maximum tolerable dose (MTD) needs to be established Development of multiple tumors Off-target effects of genetic alterations Extensive breeding

2. Syngeneic wild-type female animals bearing mammary tumors through orthotopic grafting of fragments of spontaneous GEMM mammary tumors

Comparison of various drugs/ drug combination schedules using the same original tumor Similar morphology, genomic profile, and treatment response as original tumor

Stromal irritation by transplantation Heterogeneity between tumor fragments derived from different intratumoral locations Alterations through repeated passages

3. Syngeneic wild-type female animals bearing mammary tumors through orthotopic grafting of suspensions of spontaneous GEMM mammary tumors

Comparison of various drugs/ drug combination schedules using the same original tumor Similar morphology, genomic profile and treatment response as original tumor

Stromal irritation by transplantation Dissociation of tumor cells from their stromal niche

4. Syngeneic wild-type female animals bearing mammary tumors through orthotopic grafting of cell lines derived from spontaneous GEMM mammary tumors

Comparison of different drugs/ schedules using the same original tumor

In vitro selection may result in considerable genetic differences to the bulk of cells in the original tumor Morphologic alterations Stromal irritation by transplantation

5. GEMMs carrying orthotopically transplanted tumors derived from GEMMs

Specific alterations of host cells

See 2.-4.

2. Materials 2.1. Cryopreservation and Orthotopic Transplantation of Tumor Fragments or Tumor Cells

1. Dulbecco’s Modified Eagle’s Medium (DMEM) (Gibco/ Invitrogen). 2. Sterile PBS (Gibco/Invitrogen) and 70% ethanol. 3. Freezing medium: 20% DMSO, 60% FCS, 20% DMEM.

36

Rottenberg, Pajic and Jonkers

4. Sterile scalpels (Swann-Morton, England). 5. Watchmaker tweezers (Cat. no. 30617.C), anatomic forceps (Cat. no. 30582), and surgical scissors (Cat. no. 30640) (Vos en zoons, Amsterdam, The Netherlands). 6. 100-µl syringe (Hamilton Bonaduz AG, Switzerland). 7. Cotton swaps (SKS Science, England). 8. Matrigel (GF-reduced, BD-Biosciences). 9. Equipment for injection (hypnorm/dormicum) or inhalation (isofluran) anesthesia. 10. Painkiller (e.g., buprenorphine). 2.2. Tumor Cell Dissociation and FluorescenceActivated Cell Sorting

1. Dulbecco’s Modified Eagle’s Medium (DMEM), FCS, and 0.05% trypsin-EDTA from Gibco/Invitrogen. 2. Collagenase/hyaluronidase solution (10×), dispase (5  mg/ ml), and Hanks Balanced Salt Solution (HBSS/HEPES balanced) from Stem Cell Technologies (England). 3. DNAseI and Red Blood Cell Lysing Buffer Hybri-Max from Sigma-Aldrich. 4. 40- and 70-µm cell strainer from BD Falcon. 5. Antibodies: biotinylated antibodies directed against leukocytes (rat antimouse CD45 and rat antimouse Ly6G, eBioscience), erythroid cells (rat antimouse TER-119, eBioscience), endothelial cells (rat antimouse CD31, eBioscience) and fibroblasts (rat antimouse CD140A, eBioscience) followed by streptavidinCy5 (Invitrogen) coupling. FITC- or PE-labeled rat antimouse antibodies against CD24 and CD49f are from BD Pharmingen.

3. Methods Once a GEMM has been chosen to study drug resistance, several approaches can be taken (Table 3.1). Besides using the mice in which spontaneous tumors arise, it might be useful to employ one of the outlined transplantation techniques, depending on the question to be addressed. Since we think that it is important to study drug resistance in the presence of an intact immune system, we have only described the orthotopic transplantation of mammary tumors from GEMMs into syngeneic, immunocompetent mice. Nevertheless, transplantations into immunodeficient hosts may be useful if the role of the immune system in the development of drug resistance is investigated, in case viral oncoproteins (e.g., Polyoma middle-T; PyMT) or oncoproteins from a different species (e.g., rat NEU) are expressed, or if the GEMM is on a mixed genetic background.

Studying Drug Resistance Using Genetically Engineered Mouse

3.1. GEMMs Bearing Spontaneous Mammary Tumors

37

Mice are not humans and drug pharmacokinetics may differ significantly between the two species. A clinically relevant dose of a compound may kill a mouse or may be rapidly cleared from the animal body resulting in lower drug plasma levels. Therefore, the maximum tolerable dose (MTD) of single drug or drug combinations needs to be established for the GEMM of interest. Although the MTD of the background strain in which the GEMM is bred gives a first indication, it may differ significantly in the GEMM in case the genetically engineered alterations affect also other tissues besides the mammary gland. For example, different MTDs were observed when DNA-damaging drugs such as doxorubicin or cisplatin were tested in K14cre; Brca1F/F; p53F/F vs. Brca1F/F; p53F/F mice (21). The route of drug administration should mimic the administration of the drug to humans, and here in particular i.v. adminis­tration via the tail vein or repeated daily oral dosing requires an experienced investigator. In xenotransplantation studies, treatments are frequently not performed on established tumors but initiated at a predefined time point after transplantation, well before palpable tumors arise. In contrast, it is not practical to choose a specific time point to start therapy in GEMMs with a tumor latency of several months. An early time point before the average tumor latency day would have to be chosen, but then therapy is directed against early (precancerous) lesions rather than a fully developed tumor. We therefore prefer to initiate therapy in GEMMs once tumors are palpable. At this point tumor diameters may vary between 2 and 5 mm, depending on the level of experience of the researcher and the frequency with which animals are palpated. This variation is suboptimal since such differences in size provide another experimental variable that needs to be taken into consideration. More reasonable criteria to start drug administration are predefined tumor volumes (e.g., 200 mm3), but this requires frequent and careful monitoring of animals, otherwise the right time point might be easily missed. Since mammary tumors are frequently egg-shaped, the ellipsoid formula to estimate the volume of an egg (volume = length × width2/2) can be utilized to determine tumor volume using two-dimensional caliper measurements. If the investigator needs additional information such as tumor vascularization or the presence of necrosis, ultrasound (US) imaging is very helpful and also provides a more accurate measurement of tumor volumes. Although considered a high-throughput technique, US imaging takes about 20 min per animal, including anesthesia. In humans, drugs are repeatedly given with a recovery interval of 21  days, which is usually necessary for the bone marrow to recover. In mice, bone marrow toxicity may not be the most important dose-limiting factor for several anticancer drugs, since animals live in relatively sterile environments and the risk to catch opportunistic infections is lower. Instead, gastrointestinal symptoms

38

Rottenberg, Pajic and Jonkers

appear to be restrictive for several agents such as platinum drugs, melphalan, thiotepa, or topotecan (Gonggrijp and Rottenberg, unpublished). The recovery time between treatments can therefore be shortened to 1 or 2 weeks. Like in humans, several cycles (e.g., 6) may be given initially. In case tumors are sensitive to the drug treatment, a complicating factor may be that relapsing and often still drug-sensitive tumors can no longer be treated after six initial treatments due to accumulating toxicity of the host or, in case of i.v. injections, tissue damage at the injection site complicating access to the tail vein. We have therefore implemented a different treatment schedule, in which after the recovery time a second treatment is only given in case the tumor volume remains ³ 50% of the initial size (21). In case the tumor volume drops below 50%, the next treatment is administered when the tumor relapses to 100% of the original volume. This approach appears to be difficult in those first-generation GEMMs that develop multiple mammary tumors at the same time, e.g., models that use MMTV or WAP promoters to overexpress oncogenes like Neu, PyMT, or Wnt. 3.2. Syngeneic Wild-Type Female Mice Bearing Mammary Tumors Through Orthotopic Grafting of Fragments of Spontaneous GEMM Mammary Tumors

A disadvantage of tumor intervention studies in GEMMs is that for several models a complex breeding schedule needs to be maintained (e.g., heterozygosity of the Cre allele, homozygosity of the floxed tumor suppressor gene(s)). Moreover, it often takes several months before “spontaneous” tumors develop in GEMMs. The engineered mutations in GEMMs may also cause unwanted side effects that increase the amount of animals needed. Examples are GEMMs in which the Keratin 14 promoter is used to drive Cre expression in epithelial tissues. These GEMMs develop skin tumors, in addition to mammary tumors (18–20). Finally, several individual tumors per experimental group are required to compensate for the intrinsic tumor heterogeneity in GEMMs. This heterogeneity is a result of the long latency between the initial genetic alteration (e.g., deletion of a tumor suppressor gene) and the development of the final tumor, during which different mutations accumulate. A solution for these complications is the orthotopic transplantation of small tissue pieces from individual GEMM-derived tumors into several syngeneic wild-type female mice (Fig.  3.1a). This approach also resolves the difficulty that the MTD might change due to side effects of the genetically engineered mutation. Importantly, for a limited number of models it has been shown that morphologies, gene expression profiles, and drug responses of the transplanted tumors are remarkably similar to those of the parental tumor. Thus, orthotopic tumor allograft models may be particularly useful for testing various drug regimens on independent outgrowths of the same tumor (21, 23). Practically, small tumor fragments of original GEMM-derived tumors can be frozen like cell lines. This permits cryopreservation of large numbers of primary tumors, which can be characterized

Studying Drug Resistance Using Genetically Engineered Mouse

39

Fig. 3.1. Strategies for orthotopic transplantation of GEMM-derived mammary tumors to study drug resistance. (a) Upon orthotopic grafting of tumor fragments several drugs, drug combinations, or schedules can be tested on independent outgrowths from the same parental tumor. (b) Crossbreeding of GEMMs for breast cancer with mice deficient in a (candidate) drug resistance gene X permits a comparison of drug responses to results obtained in (a). (c) Mammary tumors generated in GEMMs proficient for gene Y can be transplanted into GEMMs deficient for this gene to test the antitumoral effect of inhibitors of gene product Y, alone or in combination with other anticancer drugs.

40

Rottenberg, Pajic and Jonkers

by immunophenotyping and molecular profiling to allow preselection of tumors with high expression of the drug target prior to transplantation and initiation of the intervention study. Cryopreservation of tumor tissues is also very useful to store tumors that have acquired resistance to specific anticancer drugs and need to be transplanted for subsequent experiments to study cross-resistance to other drugs or to explore pharmacological reversal of resistance (21). Here, a risk is that the drug resistance phenotype is not stable and that tumors become again drug sensitive upon transplantation and outgrowth without drug selection. Thus far, we have only observed that as a major complication with topotecan resistance (Zander, Borst and Rottenberg, unpublished). Another powerful though time-consuming approach is to cross a GEMM with mouse strains deficient for a specific gene, which might contribute to drug resistance. Tumors derived from such a compound model can then be transplanted orthotopically into syngeneic wild-type mice and treatment response can be compared to tumors proficient for the (candidate) drug resistance gene (Fig. 3.1b). 3.2.1. Cryopreservation of Mammary Tumor Fragments

1. Sacrifice animal and harvest the tumor in one piece with sterile instruments. Remove as much fat and fibrous tissue as possible. Dip the tumor in 70% ethanol, then in sterile PBS and transfer to Petri dish on ice. 2. Add about 10 ml of PBS to cover the plate. Cut the tumor into small pieces of 1–2  mm in diameter with a scalpel (see Note 1). 3. Using a 25-ml pipette transfer the tumor piece/PBS mixture into a 50-ml Falcon tube and spin for 1 min at 450 × g. 4. Remove supernatant by pipetting and also remove fragments swimming in the supernatant since these represent (mainly) fat or fibrous tissue. Add 5-ml ice-cold DMEM medium; keep on ice. Add dropwise 5 ml of ice-cold freezing medium and mix by swirling the tube. 5. Using a 25-ml pipette (larger diameter of tip) transfer about 1–1.5  ml containing about 10 tumor pieces into 2-ml cryogenic vials on ice and freeze overnight at −80°C in cell freezing devices. Then store in liquid nitrogen.

3.2.2. Thawing of Cryopreserved Tumor Pieces

1. Place cryogenic vial into 37°C water bath. 2. Once only a small ice block remains transfer tumor fragments into 50-ml Falcon tube filled with ice-cold PBS. 3. Spin for 1  min at 450 × g. Remove supernatant by pipetting and repeat wash with ice-cold PBS. 4. Remove supernatant by pipetting and add 10-ml ice-cold PBS. Keep on ice until transplantation.

Studying Drug Resistance Using Genetically Engineered Mouse 3.2.3. Orthotopic Transplantation of Tumor Fragments into Syngeneic Female Animals

41

This procedure describes the transplantation of tumor fragments into the fourth right mammary fat pad. This procedure is applicable also to other mammary glands. If the investigator does not want endogenous mammary epithelium to be present, the nippleconnected part distal of the mammary lymph node can be removed from the fat pad by electrocauterization before tumor implantation in 21-day-old female mice. 1. Keep thawed or freshly cut tumor pieces in PBS on ice. 2. Anaesthetize animal for at least 30  min (e.g., by hypnorm/ dormicum injection or isoflurane inhalation) and fix the mouse belly up (see Note 2). 3. Shave belly and right inguinal area and desinfect with 70% ethanol. 4. Make a 1-cm cranio-caudal incision in the right inguinal skin and explore the fourth mammary fat pad using cotton swaps and forceps. 5. Using watchmaker’s forceps insert a small pocket into the fat pad and install a tumor fragment of 1–2  mm at this site. Carefully move the fad pad to its original location and avoid that the tumor fragment dislocates from the pocket. 6. Close the skin by stitching or wound clips (remove after 1 week) and apply painkiller (e.g., 0.1 µg buprenorphine/g mouse).

3.3. Syngeneic Wild-Type Female Mice Bearing Mammary Tumors Through Orthotopic Grafting of Suspensions of Spontaneous GEMM Mammary Tumors

The advantage of this approach lies in a more homogeneous distribution of tumor cells before transplantation. Though clonally related, tumor cell nests located in different parts of the same tumor might have acquired differential (epi-)genetic alterations. This intratumor heterogeneity might result in differences between tumor outgrowths produced by transplantation of different tumor fragments from the same parental tumor. Although homogenization of tumor cells by dissociation might reduce heterogeneity among the resulting outgrowths, a potential risk might be that dissociation of tumor cells from their stromal niche affects paracrine signaling pathways that are required for tumor cell proliferation or survival or for maintenance of the differentiation state.

3.3.1. Transplantation of Mechanically Dissociated Tumor Cells Derived from GEMM Mammary Tumors

Varticovski et al. mechanically dissociated mammary tumors from different genetically engineered mouse models (MMTV-PyMT, MMTV-neu, MMTV-wnt1/p53+/−, BRCA1/p53+/−, and C3(1)T-Ag) by mincing, passaging through a 40-µm mesh, and subsequent passaging through 18- to 25-gauge needles (23). These mechanically dissociated tumor cells can also be cryopreserved in 10% DMSO by stepped rate freezing. Upon transplantation of 1 × 106 tumor cells into syngeneic or immunodeficient mice, mammary tumor outgrowths developed, which showed a high morphologic similarity to the original tumor.

42

Rottenberg, Pajic and Jonkers

3.3.2. Tumor Cell Dissociation and Fluorescence-Activated Cell Sorting

Another helpful application is the use of cell surface markers, such as mammary epithelial stem cell markers or putative cancer stem cell markers, for tumor cell dissociation and fluorescence-activated cell sorting (FACS) sorting of dissociated tumor cells in order to investigate whether specific tumor cell subpopulations show increased resistance to anticancer drugs. We provide here a protocol adapted from Stingl et al. (24) to dissociate and purify mouse mammary tumor cells based on the mammary epithelial stem cell markers CD49f and CD24. Such subpopulations may then be used for orthotopic transplantations or for in vitro studies (e.g., cytotoxicity assays or Hoechst dye exclusion). 1. Harvest tumor in a 50-ml Falcon tube, in 3-ml DMEM on ice. Mince tissue thoroughly on ice. 2. Add collagenase/hyaluronidase solution diluted to 1×. 3. Leave in 37°C incubator for 2-h shaking. Every 30  min pipette up and down vigorously (5-ml pipette). 4. At the end of digest, transfer the solution into a new 15-ml Falcon tube, add ~8–9 ml cold HF (HBSS/HEPES + 2%FCS). 5. Pellet the cells at 450 × g for 5  min and discard the supernatant. 6. Resuspend pellet in red blood cell lysing buffer: first add 0.3– 0.4  ml of HF buffer, followed by 2-ml red blood cell lysis buffer; mix gently (with 1-ml filter tip, tip cut off); add 9 ml of HF buffer and centrifuge at 450 × g for 5 min. 7. Discard supernatant and add 2  ml of prewarmed (37°C) 0.05% trypsin-EDTA. Gently pipette up and down (with 1-ml filter tip, tip cut off) for 1–3 min. 8. Add 9-ml cold HF and spin at 450 × g for 5 min. Remove as much of the supernatant as possible. 9. Add 2  ml of prewarmed dispase and 100 µl of 2  mg/ml DNase I; pipette up and down for 2–3 min until clumps are released. 10. Wet the cell strainers (40 and 70 µm) with 1–2 ml HF buffer. 11. Dilute the cell suspension with 5-ml cold HF and filter the cell suspension through a 70-µm strainer followed by filtering through a 40-µm cell strainer into a new 50-ml Falcon tube (add an extra 3–4 ml to wash the remaining cells in the tube and strainers while filtering). Transfer the filtered cells into a new 15-ml Falcon tube and centrifuge at 450 × g for 6 min. 12. Resuspend in HF buffer (about 2–3 ml). 13. Primary antibody incubations are done for 15 min in FACS tubes (5 ml, 12 × 75 mm) at the maximum of 1 × 107 cells/ml. To exclude stromal lineage (Lin+) cells, biotinylated antibodies against CD45, TER119, Ly6G (1:100), CD31, and CD140a

Studying Drug Resistance Using Genetically Engineered Mouse

43

(1:200) are used. AntiCD24-FITC (1:50) and antiCD49fPE (1:100) are from BD Pharmingen. As secondary antibody to detect Lin+ cells Strep-Cy5 (1:50) is applied for 20 min. Single channel controls (AntiCD24-FITC, antiCD49f-PE, or biotinylated antibodies only) are included for adjusting channel compensation. 14. Add 2-µg propidium iodide (PI) per ml for 2  min on ice. Filter through a 70-µm strainer and centrifuge at 450 × g for 5  min, remove the supernatant, and resuspend cells in HF buffer. PI-positive cells are dead and should be gated out during the FACS procedure. 15. For cell sorting we use a BD FACSaria sorter with a 100-µm nozzle and reduced pressure (20 psi). For transplantation of sorted cells proceed as described under Subheading 3.4. 3.4. Syngeneic Wild-Type Female Animals Bearing Mammary Tumors Through Orthotopic Grafting of Cell Lines Derived from Spontaneous GEMM Mammary Tumors

Alteration of genes that confer drug resistance can be carried out in mammary tumor cell lines derived from GEMMs for breast cancer. Instead of the time- and resource-consuming generation of compound mouse models, one can try to transplant modified cell lines orthotopically into syngeneic hosts. The risk is that the in vitro culture conditions may cause selection of tumor cells with altered biological properties and genomic profiles. This should be checked since it would compromise the use of tumor cell lines as models for the in situ tumors from which they were derived. The procedure for orthotopic transplantation of tumor cells into the mammary fat pad of syngeneic mice is similar to what is described under Subheading  3.2.3. Instead of generating a pocket and inserting a tumor fragment, cells are injected in a maximal volume of 100 µl into the fat pad. In our experience grafting is improved if the cells are resuspended in Matrigel. For this purpose 30 µl of ice-cold Matrigel is added to cells (e.g., 1 × 106) diluted in 30-µl DMEM and the mixture is kept on ice until injection. Before exploring the fat pad of the animal, flip the Eppendorf tube, take the cell suspension up a Hamilton syringe, and leave at RT for up to 5 min until injection into the fat pad.

3.5. GEMMs Carrying Orthotopically Transplanted Tumors Derived from Other GEMMs

Orthotopic transplantation of tumor pieces or cells into genetically engineered animals with defined stromal defects might be useful for investigating the effects of stromal elements on therapy response and drug resistance. This includes immune cells, cancerassociated fibroblasts, and endothelial cells (25–27). To study the effect of inhibitors of ATP-binding cassette (ABC)-dependent drug efflux transporters that are expressed in tumor cells, it may also be convenient to graft ABC transporter proficient tumors into syngenic mice that are ABC transporter-deficient (Fig. 3.1c). ABC transporters are expressed in the gut, liver, kidneys, or brain endothelium and are involved in the clearance of several drugs (2).

44

Rottenberg, Pajic and Jonkers

Systematic application of an inhibitor may therefore change the drug pharmacokinetics. If specific inhibitors are used in combination with drugs, it will be easier to evaluate tumor cell-specific inhibition if the tumor is transplanted into a transporter-deficient model, although this requires preliminary adjustment of the drug MTD in the ABC transporter-deficient mouse.

4. Notes 1. In case areas of necrosis (liquefied, calcified, white) or hemorrhage (dark red), which are macroscopically visible, these areas should be removed from the sample. 2. The methods and doses of anesthetics are not elaborated, since these vary between mouse strains and can be deduced from the literature.

Acknowledgments Our work is supported by grants of the Dutch Cancer Society (2006-3566 to Piet Borst, S.R. and J.J.; 2007-3772 to J.J., S.R. and Jan H.M. Schellens) and the European Union (FP6 Integrated Project 037665-CHEMORES to Piet Borst and S. R.). We thank Piet Borst for critical reading of the manuscript. References 1. Gottesman MM (2002) Mechanisms of cancer drug resistance. Annu Rev Med 53:615–627 2. Borst P, Oude Elferink R (2002) Mammalian ABC transporters in health and disease. Annu Rev Biochem 71:537–592 3. Fojo T, Bates S (2003) Strategies for reversing drug resistance. Oncogene 22:7512–7523 4. Szakacs G, Paterson JK, Ludwig JA, BoothGenthe C, Gottesman MM (2006) Targeting multidrug resistance in cancer. Nat Rev Drug Discov 5:219–234 5. Fojo T (2007) Multiple paths to a drug resistance phenotype: mutations, translocations, deletions and amplification of coding genes or promoter regions, epigenetic changes and microRNAs. Drug Resist Updat 10:59–67 6. Lee S, Schmitt CA (2003) Chemotherapy response and resistance. Curr Opin Genet Dev 13:90–96

7. Schmitt CA (2003) Senescence, apoptosis and therapy – cutting the lifelines of cancer. Nat Rev Cancer 3:286–295 8. Shabbits JA, Hu Y, Mayer LD (2003) Tumor chemosensitization strategies based on apoptosis manipulations. Mol Cancer Ther 2:805–813 9. Voorzanger-Rousselot N, Alberti L, Blay JY (2006) CD40L induces multidrug resistance to apoptosis in breast carcinoma and lymphoma cells through caspase independent and dependent pathways. BMC Cancer 6:75 10. Debatin KM (2004) Apoptosis pathways in cancer and cancer therapy. Cancer Immunol Immunother 53:153–159 11. Sharpless NE, DePinho RA (2006) The mighty mouse: genetically engineered mouse models in cancer drug development. Nat Rev Drug Discov 5:741–754 12. Rottenberg S, Jonkers J (2008) Modeling therapy resistance in genetically engineered

Studying Drug Resistance Using Genetically Engineered Mouse mouse cancer models. Drug Resist Updat 11:51–60 13. Quintana E, Shackleton M, Sabel MS et  al (2008) Efficient tumour formation by single human melanoma cells. Nature 456:593–598 14. Kuperwasser C, Chavarria T, Wu M et  al (2004) Reconstruction of functionally normal and malignant human breast tissues in mice. Proc Natl Acad Sci USA 101:4966–4971 15. Marangoni E, Vincent-Salomon A, Auger N et al (2007) A new model of patient tumorderived breast cancer xenografts for preclinical assays. Clin Cancer Res 13:3989–3998 16. Van Dyke T, Jacks T (2002) Cancer modeling in the modern era: progress and challenges. Cell 108:135–144 17. Jonkers J, Berns A (2002) Conditional mouse models of sporadic cancer. Nat Rev Cancer 2:251–265 18. Jonkers J, Meuwissen R, van der Gulden H et  al (2001) Synergistic tumor suppressor activity of BRCA2 and p53 in a conditional mouse model for breast cancer. Nat Genet 29:418–425 19. Derksen PW, Liu X, Saridin F et  al (2006) Somatic inactivation of E-cadherin and p53 in mice leads to metastatic lobular mammary carcinoma through induction of anoikis resistance and angiogenesis. Cancer Cell 10: 437–449 20. Liu X, Holstege H, van der Gulden H et  al (2007) Somatic loss of BRCA1 and p53 in mice induces mammary tumors with pathologic

21.

22.

23.

24. 25.

26. 27.

45

and molecular features of human BRCA1mutated basal-like breast cancer. Proc Natl Acad Sci USA 104:12111–12116 Rottenberg S, Nygren AOH, Pajic M et  al (2007) Selective induction of chemotherapy resistance of mammary tumors in a conditional mouse model for hereditary breast cancer. Proc Natl Acad Sci USA 104:12117–12122 Rottenberg S, Jaspers JE, Kersbergen A et al (2008) High sensitivity of BRCA1-deficient mammary tumors to the PARP inhibitor AZD2281 alone and in combination with platinum drugs. Proc Natl Acad Sci USA 105:17079–17084 Varticovski L, Hollingshead MG, Robles AI et  al (2007) Accelerated preclinical testing using transplanted tumors from genetically engineered mouse breast cancer models. Clin Cancer Res 13:2168–2177 Stingl J, Eirew P, Ricketson I et  al (2006) Purification and unique properties of mammary epithelial stem cells. Nature 439:993–997 de Visser KE (2008) Spontaneous immune responses to sporadic tumors: tumor-promoting, tumor-protective or both? Cancer Immunol Immunother 57:1531–1539 Bergers G, Hanahan D (2008) Modes of resistance to anti-angiogenic therapy. Nat Rev Cancer 8:592–603 Micke P, Ostman A (2005) Exploring the tumour environment: cancer-associated fibroblasts as targets in cancer therapy. Expert Opin Ther Targets 9:1217–1233

sdfsdf

Chapter 4 Mechanisms of Multidrug Resistance in Cancer Jean-Pierre Gillet and Michael M. Gottesman Abstract The development of multidrug resistance (MDR) to chemotherapy remains a major challenge in the treatment of cancer. Resistance exists against every effective anticancer drug and can develop by numerous mechanisms including decreased drug uptake, increased drug efflux, activation of detoxifying systems, activation of DNA repair mechanisms, evasion of drug-induced apoptosis, etc. In the first part of this chapter, we briefly summarize the current knowledge on individual cellular mechanisms responsible for MDR, with a special emphasis on ATP-binding cassette transporters, perhaps the main theme of this textbook. Although extensive work has been done to characterize MDR mechanisms in vitro, the translation of this knowledge to the clinic has not been crowned with success. Therefore, identifying genes and mechanisms critical to the development of MDR in vivo and establishing a reliable method for analyzing clinical samples could help to predict the development of resistance and lead to treatments designed to circumvent it. Our thoughts about translational research needed to achieve significant progress in the understanding of this complex phenomenon are therefore discussed in a third section. The pleotropic response of cancer cells to chemotherapy is summarized in a concluding diagram. Key words: Multidrug resistance, Uptake transport, ABC transporters, Drug metabolism, DNA repair, Vaults, Microenvironment, Translational research

1. Introduction Chemotherapy is the treatment of choice for patients diagnosed in the late stages of locally advanced and metastatic cancers. The main challenge is then to administer a drug dosage that maximizes the efficacy and minimizes the toxicity of the treatment. Unfortunately, in a significant number of patients, the tumor does not respond to the therapeutic agents. This impediment to the clinical cure of cancers results from known and yet-to-be determined mechanisms of resistance to chemotherapy. Resistance, either inherent or acquired, exists against every effective anticancer drug and can develop by multiple mechanisms. The overall mechanisms were recently J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_4, © Humana Press, a part of Springer Science + Business Media, LLC 2010

47

48

Gillet and Gottesman

reviewed in part by Lage (1), Mellor et al. (2), and Mimeault et al. (3). These mechanisms can act individually or synergistically, leading to multidrug resistance (MDR), in which the cell becomes resistant to a variety of structurally and mechanistically unrelated drugs in addition to the drug initially administered. In this chapter, we will first summarize the current knowledge on individual cellular mechanisms responsible for MDR. Since a detailed discussion of all the mechanisms is beyond the scope of this chapter, our aim will be to briefly depict them and provide the reader with the most recent and relevant reviews on the subject. As ATP-binding cassette (ABC) transporters are perhaps the unifying theme of this book, we will emphasize their biology and their role in the MDR process to provide a fairly detailed idea of what we currently know about this family of proteins. Although extensive research has characterized some of the mechanisms involved in MDR in vitro, translating this to the clinic still represents a major challenge. We will therefore briefly summarize our thoughts about translational research that could be done to achieve significant progress in the understanding of this complex phenomenon. Finally, we provide a diagram to summarize the pleotropic response of cancer cells to chemotherapy.

2. Mechanisms Involved in the Resistance of Cells to Chemotherapy 2.1. Drug and Plasma Membrane Interactions 2.1.1. Uptake Transport: Emerging Role of Solute Carriers

Aside from pharmaceutical factors such as drug administration, distribution, metabolism, and excretion, the primary obstacle that prevents a drug from reaching the intracellular compartment is the plasma membrane. Therapeutic agents can react with various molecules, resulting in a complex speciation profile. These species can enter cells by either passive diffusion (4) or facilitated transport (5). Although the exact mechanisms of cellular uptake are poorly understood for most chemotherapeutic drugs, it has been well established that decreased expression of reduced-folate carrier (SLC19A1/hRFC1) and polymorphisms in its gene significantly hamper a patient’s response to methotrexate, a common therapeutic agent (6). The solute carrier (SLC) family comprises approximately 360 uptake transporters classified into 45 gene families, outlined at http://www.bioparadigms.org/ slc/menu.asp. Genes of the SLC superfamily encode passive transporters, ion-coupled transporters, and exchangers (7). We can rationally speculate that observations concerning SLC19A1 could also apply to other members of this large family involved in uptake of anticancer drugs. Transporters of the SLC28 and 29 families mediate equilibrative diffusion of nucleosides across the plasma membrane, contributing to salvage pathways of

Mechanisms of Multidrug Resistance in Cancer

49

nucleotide synthesis (8, 9). It has also been demonstrated that these transporters mediate the cellular uptake of nucleoside analogs used in the treatment of cancers (10). Mackey et  al. have highlighted the role of SLC29A1, A2 and SLC28A1 but not SLC28A2 in gemcitabine transport (11). More recently, the same group reported that SLC29A1 is a predictive marker for overall survival in patients with pancreatic cancer who received gemcitabine (12). One could therefore suggest that tumors with reduced expression of SLC29A1, A2, or SLC28A1 transporters or with mutant transporters might be resistant to this drug. Genetic polymorphisms of these transporters may alter drug pharmacokinetics, triggering interindividual differences in the safety and efficacy of drug therapy. SLCO1B3/OATP1B3 was shown to mediate uptake of paclitaxel (13). Investigation of the functional consequences of mutations in SLCO1B3 showed that paclitaxel pharmacokinetics were not associated with SLCO1B3-344T>G or 699G>A (14). The SLC22A1/OCT1, SLC22A2/OCT2, and SLC22A3/OCT3 transporters may mediate uptake of some platinum anticancer drugs (15). Interestingly, genetic variants of SLC22A1 and SLC22A2 showed altered transport of some of their substrates, such as metformin, which is used in therapy for type 2 diabetes mellitus (16, 17). However, the transport of platinum anticancer drugs by these variant SLC22A1 and A2 transporters has not yet been directly investigated. To date, SLCs have not been intensively focused on as candidate transporters for anticancer drugs. Therefore, the correlation of SLC genetic variants to treatment outcomes still needs to be clarified. An important step toward a better understanding of the role of SLCs in drug transport was made with a recent study released by Okabe et al. who used a bioinformatics approach to identify SLC substrates (18). In that study, mRNA expression of 28 members of the SLCO and SLC22 families in the NCI-60 cell line panel (19) was profiled. By correlating expression profiles with growth inhibitory profiles of 1,429 compounds (including anticancer drugs and drug candidates) tested against the cells (20–23), it was confirmed that SLC22A4 confers sensitivity to doxorubicin in cancer cells (18). 2.1.2. Efflux Transport: Role of ABC and Other Transporters

Drug uptake can also be significantly reduced by ATP-dependent drug efflux pumps. These include ABC transporters (24, 25) such as the extensively studied ABCB1 (26), C1 (27), and G2 (28) transporters. As the main mechanism of resistance discussed in this book, ABC transporters will be thoroughly discussed in Subheading 4.2 of this chapter. ABC transporters are not the only actors in the process of ATP-dependent drug efflux, as other transporters such as RLIP76/RALBP1 (29) and ATP7A/B (30) have also been reported to mediate drug resistance. RLIP76 is a GTPase-activating protein that mediates the export of the GSH-conjugates of chemotherapeutic agents such as melphalan

50

Gillet and Gottesman

(31) as well as doxorubicin, vincristine, and mitomycin-c (32). ATP7A/B are copper efflux pumps that appear to have a role in resistance to platinating agents, as supported by numerous studies reviewed by Hall et al. (15) and Kuo et al. (30). 2.1.3. Lipid Metabolism Affects Biophysical Properties of the Lipid Bilayer as well as Signaling Pathways

Multidrug-resistant cells may have alterations in lipid metabolism (33, 34), which induce modifications in the biophysical properties of the lipid bilayer, consequently influencing drug uptake. Ceramide has been intensively studied concerning its role as the cellular messenger of apoptosis (35). GouazeAndersson et al. recently showed that this lipid has a similar role in the regulation of ABCB1 expression (36). The same group also reported that ceramide plays a major role in acquired resistance to doxorubicin (37). The mechanism they propose is the following: doxorubicin increases ceramide levels by either activation of sphingomyelinase or activation of enzymes of de novo ceramide synthesis, resulting in activation of apoptotic pathways. However, as the natural lipid substrate of glucosylceramide synthase (GCS), ceramide upregulates GCS expression through the Sp1 transcription factor. The authors postulate that this positive feedback cycle is antiapoptotic and drives cellular resistance to ceramide-generating types of chemotherapeutic drugs, speculating further that the characterization of this signal transduction pathway might reveal a way to prevent anthracycline resistance (37).

2.1.4. MDR Mechanisms Related to Drug Entry: A Pleiotropic Phenomenon

The mechanisms mentioned earlier are closely interconnected, and this can be illustrated by the reduced uptake of platinum drugs in cisplatin-resistant cells. Our laboratory has studied extensively the causes underlying resistance to platinum drugs (recently reviewed in Hall et al. (15)). Following a first study reporting the establishment of cisplatin-resistant cell lines and their crossresistance to a wide array of structurally and mechanistically unrelated drugs (38), subsequent extensive work by our laboratory has clearly shown the pleiotropic defect in uptake of cisplatin and unrelated compounds found in cancer cells. Using selected cell lines for cisplatin resistance as models, we have observed decreased expression of SLC19A1 transporter and arsenic-binding proteins (38–40). We later demonstrated that this defective uptake results from DNA hypermethylation (41). This suggests a common mechanism that might also reduce the uptake of platinum drugs, as no ABC transporter efflux pumps have been found to be overexpressed in the models studied. This may be explained by the mislocalization of membrane proteins caused by a defect of plasma membrane protein recycling associated with a cytoskeletal defect (42, 43). Other alterations in the biophysical properties of the lipid bilayer, such as a defective fluid-phase endocytosis (44) and an increase in membrane fluidity have also been detected in cisplatin resistant cells (45). Whether or not this latter defect is a

Mechanisms of Multidrug Resistance in Cancer

51

primary cause leading to cisplatin resistance, or a secondary effect has yet to be determined. These studies clearly demonstrate the pleiotropic phenomenon underlying multidrug resistance mechanisms and their effect on cellular drug entry. 2.2. Drug Metabolism

Once in the intracellular compartment, drug metabolism enzymes are the second line of cellular resistance. This process involves phase I and II enzymes.

2.2.1. Phase I Metabolic Enzymes

Phase I or oxidative metabolism is mediated mainly by cytochrome P450 enzymes (CYPs) and epoxide hydrolases. CYPs belong to a superfamily of hemoproteins comprising 57 genes classified in 18 families and 44 subfamilies based on their degree of sequence homology. An up-to-date database and nomenclature for these enzymes can be found at: http://drnelson.utmem.edu/ CytochromeP450.html. CYPs are localized in mitochondria and the endoplasmic reticulum (ER) and catalyze the monooxygenase reaction by incorporating one atom of molecular oxygen into the substrate and one into water. This reaction also requires a source of electrons, provided by the NADPH cytochrome P450 reductase in the ER and ferredoxin in the mitochondria (46). While mitochondrial CYPs are involved in the metabolism of endogenous substrates, microsomal CYPs metabolize both endogenous and exogenous compounds. Therapeutic drugs are therefore metabolized by microsomal CYP and epoxide hydrolases, which convert highly mutagenic aromatic metabolites (epoxide) created from the CYP metabolism in a metabolite that can be conjugated by the phase II enzymes and then effluxed by transporters such as the members of the ABCC transporter family (46). Although mainly expressed in the liver, extrahepatic expression of CYPs has been shown in both normal and tumor tissues (47, 48). The CYP3A, 2D6, and 2C families metabolize most chemotherapeutic drugs. These enzymes are genetically highly polymorphic, and the expression of some of their variants has been found to predict treatment outcome (49). This is exemplified by the 516G>T polymorphism in CYP2B6, which is related to an increase in cyclophosphamide metabolism (50, 51). However, the same polymorphism was correlated with a threefold decrease in efavirenz metabolism (52–54). Phase I reactions result mainly in drug detoxification. However, these enzymes can be utilized in a prodrugbased strategy (55). Understanding the effect of polymorphisms on CYP enzyme activity is consequently particularly important. This is well illustrated by the data obtained for the 516G>T polymorphism in CYP2B6, which indicate clearly that the effect of polymorphisms can vary dramatically, depending on the drug administered. Studies have highlighted the synergism between CYP enzymes and ABC transporters that occurs when metabolites produced by CYP enzymes, especially CYP3A4, are better substrates for ABCB1 than the parent compound or when ABCB1 prevents the

52

Gillet and Gottesman

saturation of CYP enzymes by necessitating a subsequent entry of the drug into the cell (56). This process increases exposure to CYP enzymes and can dramatically decrease the efficacy of chemotherapeutic treatment. 2.2.2. Phase II Enzymes

Phase II enzymes are involved in conjugation reactions including glutathionylation (57), glucuronidation (58), and sulfation (59). These enzymes include glutathione-S-transferase (GST) (60), UDPglucuronosyltransferases (UGT) (61), sulfotransferases (62), and arylamine N-acetyltransferases (NAT) (63), which transform the reactive species into hydrophilic nontoxic metabolite conjugates. These conjugated metabolites are then effluxed by members of the ABCC family of ABC transporters (27). Genetic polymorphisms in these families of genes have also been associated with overall survival in cancer patients. Ekhart et al. (64) review the influence of genetic polymorphisms not only in these enzymes but also in phase I enzymes and ABC transporters on survival after chemotherapy.

2.3. Drug Sequestration

Recent evidence has led us to consider intracellular drug sequestration as an important mechanism of MDR. Abnormalities in lysosomal function, protein trafficking and secretion have now been clearly identified. It has been shown that cisplatin is sequestered into lysosome, golgi, and secretory compartments and then effluxed from the cell (44, 65–67). A complete profile of the mediators of the intravesicular transport has certainly not yet been developed. We do know that the copper efflux transporters ATP7A/B are colocalized with fluorescent cisplatin analogs in vesicles of the secretory pathway (65, 67, 68). A recent study carried out in our laboratory highlighted the role of melanosomes in cisplatin resistance (69). We showed that cisplatin is sequestered in melanosomes, altering the melanogenic pathway and accelerating extracellular transport of melonosomes that contain cisplatin (69). No transporter responsible for this intracellular transport has yet been identified. However, one could suggest a role of the ABC transporter ABCB5, found to be preferentially expressed in pigment-producing cells (70). ABCA3-mediated intracellular drug sequestration has recently been associated with poor overall survival in acute myeloid leukemia (71, 72). ABCA3 is first localized in lysosomal (73) and late endosomal membranes and then colocalized with daunorubicin (72). Its role in MDR was initially indirectly suggested by Steinbach et  al. when they showed a significant effect on cell viability after a combination of ABCA3 gene expression silencing and doxorubicin treatment (71). Direct evidence of its role as mediator of MDR was reported very recently in a study carried out by Chapuy et  al., in which the authors demonstrated that ABCA3 mediates resistance to daunorubicin, mitoxantrone, etoposide, Ara-C, and vincristine (72).

Mechanisms of Multidrug Resistance in Cancer

53

In addition to drug sequestration in intracellular vesicles, “scavenger” metallothioneins (MTs) play a role in ensnaring drugs within a cell. These molecules are cysteine-rich and have high affinity for metal ions. This specificity, along with their ability to ensnare reactive oxygen species strongly suggests them as significant players in resistance to metal-based therapeutic agents and radiation treatment. However, the disparate literature on their role in cancer drug resistance does not allow unequivocal corroboration of this assumption (reviewed by Theocharis et al. (74) and Thirumoorthy et al. (75)). Although important as part of the pleiotropic phenomena underlying MDR, intracellular drug sequestration does not appear to induce resistance to the same extent as that produced by typical efflux transporters. 2.4. DNA-Damage Response Network

A complex network of interacting pathways has evolved to monitor the integrity of a cell’s DNA and govern proper responses to any genetic damage (76). This network includes sensor complexes that detect DNA breaks. It has been shown that Mre11-Rad50-Nbs1 (MRN) recognizes or senses DNA doublestrand breaks (DSBs), while the RPA-ATRIP complex binds to single strand breaks (SSBs). Kinases such as ATM and ATR are then recruited by MRN and RPA-ATRIP, respectively, and phosphorylate/activate a myriad of other proteins including the checkpoint kinases Chk1 and 2, initiating a cascade that results in cell-cycle arrest and DNA repair (Fig. 4.1) (77). However, if the damage is too extensive, rather than repair itself, the cell will enter one of these states: (1) senescence, which is characterized by an irreversible growth arrest, (2) apoptosis, or (3) necrosis (see Subheading 4.2.4). Many chemotherapeutic drugs have been employed to kill proliferating cells, causing extensive DNA damage that ultimately leads to cell cycle arrest and cell death. However, the efficacy of these therapeutic agents such as platinum drugs (78) and alkylating agents (79) can be significantly reduced by the ability of cells to repair DNA. DNA repair involves an intricate network of repair systems that each target a specific subset of lesions. These pathways include (1) the direct reversal pathway (MGMT, ABH2, ABH3), (2) the mismatch repair (MMR) pathway, (3) the nucleotide excision repair (NER) pathway, (4) the base excision repair (BER) pathway, (5) the homologous recombination (HR) pathway, and (6) the nonhomologous end joining (NHEJ) pathway (80). Figure 4.2 summarizes the role of these pathways in the repair of lesions induced by some chemotherapeutic agents (reproduced from Helleday et al. (81)). Studies have reported an inverse correlation of ERCC1 (NER pathways) with either response to platinum therapy or survival in ovarian (82, 83), non-small cell lung cancer (84), and colorectal cancers (85). It has also been shown that MMR deficiency is

54

Gillet and Gottesman

Fig. 4.1. Chk1 and Chk2 kinases are serine/threonine kinases that are activated by the ATM and ATR kinases in response to DNA damage. The checkpoint kinases are transducers of the DNA damage signal and both phosphorylate a number of substrates involved in the DNA damage response. Chk1 and Chk2 share a number of overlapping substrates, although it is clear that they have distinct roles in directing the response of the cell to DNA damage. Current understanding is that the checkpoint kinases are involved not only in cell cycle regulation but also in other aspects of the cellular response to DNA damage. The G1checkpoint is modulated primarily by the ATM-Chk2-p53 pathway, as expression of ATR, Chk1, and Cdc25a is limited until the cell passes this restriction point. At this point, levels of ATR, Chk1, and Cdc25a all increase. If DNA damage is detected, Chk1/Chk2 are activated, Cdc25a is phosphorylated, and thus, destabilized, resulting in a p53-independent S arrest. In S phase, the same cascade can result in an intra-S arrest in response to stalled replication forks. The G2-M checkpoint prevents entry into mitosis with unrepaired DNA lesions. Initiation of this checkpoint is mediated by the ATM/ATR/Chk1/Chk2 cascades as shown, which ultimately suppresses the promitotic activity of cyclin B/cdc2. Along with their pivotal roles in the modulation of the cell cycle checkpoints, Chk1 and Chk2 are also involved in other aspects of the DNA damage response, including DNA repair, induction of apoptosis, and chromatin remodeling. Reproduced from (77), by permission of the American Association for Cancer Research.

associated with cisplatin resistance (86, 87). The MMR mechanism removes the newly inserted intact base instead of the damaged base, triggering subsequent rounds of futile repairs, which can lead to cell death (88). Furthermore, a role in triggering checkpoint signaling and apoptosis was also suggested (89). Resistance to alkylating agents via direct DNA repair by O(6)-methylguanine methyltransferase (MGMT) has been extensively studied and is considered to be a significant barrier to the successful treatment of patients with malignant glioma (90). There is no doubt that other proteins involved in those mechanisms will be revealed in the future through the characterization of MDR tumors.

Mechanisms of Multidrug Resistance in Cancer

55

Fig. 4.2. Overview of DNA repair pathways involved in repairing toxic DNA lesions formed by cancer treatments. The DNA-damaging agents that are used in cancer treatment induce a diverse spectrum of toxic DNA lesions. These lesions are recognized by a variety of DNA repair pathways that are lesion-specific but are complementary in some respects. (a) Ionizing radiation and radiomimetic drugs induce double-strand breaks (DSBs) that are predominantly repaired by nonhomologous end joining (NHEJ). (b, c) Monofunctional alkylators (b) and bifunctional alkylators (c) induce DNA base modifications, which interfere with DNA synthesis. Lesions produced by some alkylators are processed into toxic lesions in a mismatch repair-dependent manner. The base excision repair (BER) and nucleotide-excision repair (NER) pathways are, together with alkyltransferases (ATs), major repair pathways, whereas other repair pathways repair toxic replication lesions, such as those produced by interstrand crosslinks. (d) Antimetabolites interfere with nucleotide metabolism and DNA synthesis, causing replication lesions that have not yet been characterized. Mismatch repair mediates the toxicity of some antimetabolites (for example, thiopurines). The repair pathways involved in repair of antimetabolite-induced lesions are, apart from BER, poorly characterized. (e) Topoisomerase poisons trap topoisome­ rase I or II in transient cleavage complexes with DNA, thus creating DNA breaks and interfering with replication. (f) Replication inhibitors induce replication fork stalling and collapse, resulting in indirect DSBs. The relative contributions of the major repair pathways to the respective types of DNA damage outlined are indicated by the sizes of the boxes. This is based on the extent of sensitivity of repair-deficient cells to anticancer drugs in each category. ENDO endonucleasemediated repair, FA Fanconi anaemia repair pathway, HR homologous recombination, O2G DNA dioxygenases, RecQ RecQ-mediated repair, SSBR DNA single-strand break repair, TLS translesion synthesis. Reprinted by permission of Macmillan Publishers Ltd., (81), copyright 2008.

56

Gillet and Gottesman

Cellular death is the underlying pharmacological purpose for chemotherapy. Disruption of apoptotic pathways, the hallmark of cancer, is a major obstacle in the success of chemotherapy. Both apoptotic and nonapoptotic mechanisms can lead to the cell death (91). Nonapoptotic mechanisms include autophagy, mitotic catastrophe, necrosis, and senescence (92) (Table  4.1). The apoptotic cascade can be initiated via two routes, involving either the release of cytochrome c from the mitochondria (intrinsic/ mitochondrial pathway), or activation of death receptors (TNF-tumor necrosis factor) in response to ligand binding (93). The activation of these pathways leads to the activation of a family of cysteine proteases, the caspases, that mediate the cleavage of cellular substrates leading to morphological and biochemical changes that characterize apoptosis. Increasing evidence, however, suggests that caspase-independent pathways also exist (94). Although in  vitro studies have shown that nuclear translocation of the mitochondrial flavoprotein apoptosis-inducing factor (AIF) might

2.5. Evasion of Drug-Induced Apoptosis

Table 4.1 Characteristics of different types of cell death Morphological changes Type of cell death

Nucleus

Cell membrane Cytoplasm

Biochemical features

Apoptosis

Chromatin condensation; Blebbing nuclear fragmentation; DNA laddering

Fragmentation Caspase-dependent (formation of apoptotic bodies)

Autophagy

Partial chromatin condensation; no DNA laddering

Blebbing

Increased number of autophagic vesicles

Caspase-independent; increased lysosomal activity

Mitotic Multiple micronuclei; catastrophe nuclear fragmentation

–

–

Caspase-independent (at early stage) abnormal CDK1/ cyclin B activation

Necrosis

Clumping and random degradation of nuclear DNA

Swelling; rupture

Increased vacuolation; organelle degeneration; mitochondrial swelling

–

Senescence

Distinct heterochromatic structure (senescenceassociated heterochromatic foci)

–

Flattening and increased granularity

SA-b-gal activity

CDK1 cycline-dependent kinase 1, MDC monodansylcadaverine, MPM2 mitotic phosphoprotein 2, SA-b-gal senescence-associated b-galactosidase, RB retinoblastoma protein Reprinted by permission of Macmillan Publishers Ltd., (91), copyright 2004

Mechanisms of Multidrug Resistance in Cancer

57

be a key factor in this pathway (94), the mechanisms governing caspase-independent cell death are little known. Chemotherapy-induced resistance occurs primarily via the mitochondrial pathway, regulated by the Bcl2 family of genes. Therefore, alterations in Bcl2 genes, which are involved in maintenance of the homeostasis of pro- and antiapoptotic factors, may considerably hamper the success of treatment (95, 96). Many researchers have investigated the usefulness of Bcl2 family proteins in the prediction of chemotherapy outcome. Sjöström et al. found that none of the apoptosis-related proteins they investigated (Bcl2, Bax, Bcl-xl, Bag1, FAS, FASL) could predict response to drug treatment for breast cancer (97). This observation was confirmed in two other studies, one in which the Bcl-2 status was not predictive for paclitaxel treatment in patients with breast cancer (98) and another involving vinorelbine plus docetaxel treatment in patients with non-small cell lung cancer (NSCLC) (99). On the other hand, some studies did find a correlation between Bcl2 expression status and chemotherapy response in patients with breast cancer (100, 101). This exemplifies the conflicting literature which to date does not allow us to establish any clear connection between defective apoptotic pathways and treatment failure. 2.6. Vaults

Vaults were first described in 1986 (102). They are the largest ribonucleoprotein particles ever described, with a size of ~42 × 75 nm and a mass of ~13 MDa (103). They were given this name because of their morphological resemblance to vaulted ceilings in gothic cathedrals (104). Vault particles are evolutionary highly conserved, although they are not found in Saccharomyces cerevisiae, Caenorhabditis elegans, or Drosophila melanogaster (105). They form a barrel-shaped structure composed of multiple copies of three proteins including the major vault protein (MVP), the vault poly-ADP-ribose polymerase (VPARP), and the telomeraseassociated protein-1 (TEP1) plus an untranslated vault RNA (vRNA) (106). Vaults, as detected by MVP expression, are ubiquitously expressed, with high levels in tissues that are chronically exposed to xenobiotics such as lung and epithelial cells of the intestine, and in macrophages and dendritic cells (107). They are localized in the cytoplasm, where up to 100,000 particles can be found per cell (108). For detailed information, see Mossink et al. (109) and Steiner et al. (110). Several cellular functions have been proposed for vaults (111). The detection of the MVP protein (initially named LRP) in a multidrug-resistant ABCB1 negative non-small cell lung cancer cell line implied a role for vaults in MDR (112). However, 15 years later, their role is still not clear. In vitro studies have yielded conflicting results; some reports show a direct role of vaults in MDR (113–115), whereas others fail to substantiate such involvement (116–118).

58

Gillet and Gottesman

In the clinic, numerous studies have been undertaken to understand the role of vaults in multidrug resistance and whether they could be used as predictive markers of treatment response. Most of the work has been focused on hematological malignancies, but a few solid tumors have also been analyzed. This research is summarized in Table 4.2. As is the case for other MDR mechanisms, the role of vaults is still the subject of debate. Disappointingly, vault knockout mice do not show any phenotypic defects compared with wild-type mice (119). Moreover, no enhancement of chemosensitivity was observed and no compensation by increased activity of ABC transporters was found following drug treatment (120). It was concluded that at least in mice, vaults are not directly involved in drug resistance. However, analysis of both human and mice promoters seems to indicate that vault-associated MDR may be species specific. Indeed, the mouse promoter lacks a DNA segment found in the human sequence containing lentiviral elements and an inverted CCAAT box (Ybox) and therefore cannot bind to the NF-Y transcription factor. It has been shown that activation of the ABCB1 promoter is dependent on both elements (121). These substantial differences between human and mice promoters may explain these puzzling observations and indicate that mouse models may not be the ideal tool to study the role of vaults in MDR.

Table 4.2 Conflicting clinical data on the association of MVP with response to therapy and patient prognosis MVP expression correlated to response to chemotherapy

MVP expression correlated to prognosis

Type of cancer

Yes

No

Yes

No

AML

(191–194)

(195–199)

(200–202)

(195–199)

ALL

(203)

(203–205)

ATL

(206)

MM

(207–209)

Ovarian cancer

(210–212)a

Breast cancer NSCLC

(219, 220)

Bladder cancer

(224)

Sarcoma

(225)

Testicular germ-cell tumors Borderline

a

(207, 208)

(209)

(213)

(210–212)a

(213)

(214–216)

(214)

(217, 218)

a

(221, 222)

(222, 223)

(226, 227)

(225)

(228)

(228, 229)

(226, 227)

Mechanisms of Multidrug Resistance in Cancer

2.7. Impact of the Microenvironment

59

Solid tumors are heterogeneous structures composed of tumor and stromal cells that are embedded in the extracellular matrix and sustained by an abnormal vasculature network. This tumor microenvironment can influence the tumor cells’ propensity to metastasize (122, 123) and can dramatically affect treatment outcome (124–126). For a general review on this, see Tredan et al. (127). Tumor responsiveness to chemotherapy is highly influenced both by the abnormal vasculature and an elevated interstitial fluid pressure, which dramatically impede the penetration of macromolecules into the tumor (125, 128). Tumor vasculature is characterized by a disorganized architecture composed of dilated and convoluted blood vessels. These vessels may also be compressed by tumor cells (129), leading to the disruption of blood flow (130). Furthermore, the lack of lymphatic vessels in the tumor vasculature network contributes to the increase of interstitial fluid pressure, which hinders the delivery of therapeutic agents by convection (131–133). As a result, metabolites are not cleared and sustaining nutrients cannot be delivered, leading to hypoxia and acidification of the extracellular compartment (124). Tumor cells that lack oxygen (hypoxic cells) fulfill their energetic needs via the glycolytic pathway, which ultimately generates lactic acid, leading to intracellular acidification (124). The cells then maintain their pH homeostasis through the expression of proton (H+) pumps, rendering the extracellular environment highly acidic. This feature may severely hamper chemotherapy. Indeed, according to the ion trapping theory, weakly basic drugs (e.g., doxorubicin, mitoxantrone, vincristine, etc.) are ionized in the acidic compartment, consequently hindering their entry into the cell. Conversely, an increased accumulation of weakly acidic drugs (e.g., chlorambucil, cyclophosphamide, camptothecin, etc.) in the cell will be observed as the plasma membrane is permeable to nonionized molecules (134, 135). Hypoxia plays an important role in MDR (124). It dramatically reduces the effectiveness of chemotherapeutic agents (e.g., bleomycin) that require oxidation to become cytotoxic or enhances the cytotoxicity of other agents (e.g., mitomycin C) that must undergo reduction to form active cytotoxic species (136). Hypoxic cells also display reduced rates of cell proliferation and can avoid apoptosis by triggering Bcl2 family genes, rendering them resistant to many therapeutic agents that target proliferative cells. It has been shown that hypoxia upregulates glutathione, GSH, and metallothionein (MT) protein levels, which reduces significantly the effectiveness of alkylating agents and platinum-based drugs (137, 138). The main mediator of the response to hypoxia is the HIF1 transcription factor, which was also shown to induce the expression of ABCB1, ABCC1, and ABCG2 (139–143). Taken together, the tumor microenvironment not only leads to the development of the pleiotropic mechanisms underlying MDR, but also amplifies or intensifies them.

60

Gillet and Gottesman

2.8. The Cancer Stem Cell Paradigm

3. The ATP-Binding Cassette (ABC) Transporters

Cancer-initiating cells, also known as cancer stem cells (CSCs), might underlie the intractable nature of many human cancers, explaining why conventional cancer therapy fails in many patients (144, 145). These cells have the capacity to initiate and sustain the growth of a heterogeneous cancer through self-renewal and differentiation. Moreover, they acquire most of the MDR mechanisms discussed in this section. CSCs can be isolated based on the expression of cell surface markers associated with immature cell types. However, no unique set of markers has been identified that distinguishes them from normal stem cells (144, 145). ABCB1 and ABCG2 are well-characterized ABC transporters expressed in both cancer and normal stem cells (146–148). Besides these transporters, ABCA3 was also detected in neuroblastoma CSCs along with ABCG2 (149). ABCA3 is localized in membranes of lysosomes and endoplasmic reticulum, suggesting a role in drug sequestration (72). More recently, it was suggested that ABCB5 could be a marker of melanoma cancer-initiating cells (150). Although the concept is exciting, knowledge of ABCB5 from the genomic to the proteomic level is rudimentary and does not support this hypothesis (70, 151, 152). Although there has been a great deal of interest in the elegant CSC paradigm, its technological and experimental challenges are colossal. One of these challenges is the isolation of normal and cancer stem cell populations to identify differences in self-renewal mechanisms. If achieved, this could open new avenues to cancer research.

The ATP-binding cassette (ABC) transporter proteins are a large superfamily of membrane proteins comprising 48 members divided into seven different families based on sequence similarities. The nomenclature for human ABC transporter genes is provided at: http://nutrigene.4t.com/humanabc.htm. These proteins are evolutionary highly conserved and there is a high sequence homology among all the members, especially those within a particular family. The functional protein typically contains two nucleotide-binding domains (NBDs) and two transmembrane domains (TMDs) encoded by a single polypeptide. ABC transporters may also be multicomponent units in which different genes encode each domain or half molecule. ABC transporters can have a wide array of cellular roles (24). They regulate local permeability by being expressed in the blood– brain barrier, blood cerebrospinal fluid, blood–testis barrier, and placenta (153). In the liver, gastrointestinal tract, and kidney,

Mechanisms of Multidrug Resistance in Cancer

61

ABC transporters excrete toxins, thereby protecting the organism (154). ABC transporters also play an active role in the immune system by transporting peptides into the endoplasmic reticulum that are identified as antigens by class I HLA molecules (e.g., ABCB2/TAP1, ABCB3/TAP2) (155, 156). Furthermore, they play physiological roles in cellular lipid transport and homeostasis (157). Finally, as expected from the diverse functional roles of ABC genes, the genetic deficiencies associated with their mutation also vary widely, as reviewed in Borst and Elferink (154). The first ABC transporter discovered was ABCB1 in 1976 (158). This transporter, called P-glycoprotein, was found to be expressed in Chinese hamster ovary cells selected for colchicine resistance. The authors also discovered that these cells displayed resistance to a variety of structurally and mechanistically unrelated drugs in addition to colchicine (158). Human P-glycoprotein, the product of the MDR1 or ABCB1 gene, was subsequently shown to confer MDR on drug-sensitive cells (159). More than 30 years later, 15 ABC transporters (ABCA2, ABCA3, ABCB1, ABCB4, ABCB5, ABCB11, ABCC1–6, ABCC11–12, and ABCG2) have been associated with drug resistance (Table 4.3). Of these, ABCB1 (160), ABCC1 (161), and ABCG2 (162) have been the most extensively studied. Yet attempts to translate these transporters into clinical targets have so far been unsuccessful (discussed further in Subheading 4 of this chapter). A lot of progress has been made in understanding the molecular mechanisms of ABCB1 through mutagenesis experiments and the resolution of structures of non-mammalian ABC proteins (163). A number of recent reports have addressed genetic polymorphisms in drug transporters; for an excellent review, see Cascorbi (164). Among the 48 ABC transporters, ABCB1 is one of the most thoroughly studied and characterized, with more than 50 SNPs reported (164–166). The correlation of ABC transporter genetic variants to treatment outcomes is gradually being clarified, yet the overall picture is still puzzling, as much of the published data are conflicting. Nevertheless, the many studies reporting correlations between SNPs and clinical outcome indicate the necessity to pursue further investigations; for a detailed review on the role of polymorphisms in ABCB1 drug transport including the role of synonymous polymorphisms in altering ABCB1 transporter function (167), see Fung and Gottesman (168). The regulation of ABCB1 is poorly understood. It has been suggested that therapeutic agents (169, 170) and endogenous stimuli such as hypoxia, acidosis, free radical formation, or glucose deprivation could also induce its expression through multiple signal transduction pathways (26). Thus, the complex network of ABCB1 regulation ensures rapid emergence of pleiotropic resistance in cancer cells.

62

Gillet and Gottesman

Table 4.3 Established anticancer drugs as substrates of ABC transporters Established anticancer drugs (Nonexhaustive list)

Subfamily

Genes

A/ABC1

ABC A2/ABC2/STGD ABCA3/ABC3

Estramustine, mitoxantrone (230, 231) Daunorubicin, Ara-C, mitoxantrone, etoposide (72)

B/MDR-TAP

ABC B1/MDR1/P-gp/PGY1

Anthracyclines, vinca alkaloids, taxanes, etoposide, teniposide, imatinib, irinotecan, SN-38, bisantrene, colchicine, methotrexate, mitoxantrone, saquinivir, actinomycin D, etc. (231, 232) Daunorubicin, doxorubicin, vincristine, etoposide, mitoxantrone (234) Doxorubicin, camptothecin, 10-OH camptothecin, and 5FU (151, 235) Paclitaxel (236)

ABC B4/MDR3/PFIC3/PGY3 ABCB5 ABC B11/BSEP/PFIC2/SPGP C/MRP

ABC C1/MRP1

ABC C2/MRP2/CMOAT

ABC C3/MRP3/CMOAT2 ABC C4/MRP4/MOATB ABC C5/MRP5/MOATC ABC C6/MRP6/MOATE/PXE ABC C10/MRP7 ABC C11/MRP8 G/WHITE

ABC G2/BCRP/MXR/ABCP

Anthracyclines, vinca alkaloids, methotrexate, antifolate antineoplastic agents, etoposide, imatinib, irinotecan, SN-38, arsenite, colchicine, mitoxantrone, saquinivir, etc. (27, 237) Vinca alkaloids, cisplatin, etoposide, doxorubicin, epirubicin, metotrexatetaxanes, irinotecan, SN-38, topotecan, arsenite, mitoxantrone, saquinivir (238–241) Etoposide, tenoposide, metotrexate (242, 243) 6-Mercaptopurine, 6-thioguanine, irinotecan, SN-38, topotecan, AZT, metotrexate, PMEA (244–247) 6-Mercaptopurine, 6-thioguanine, 5-FU, cisplatin, metotrexate, PMEA, AZT (245, 248) Etoposide, doxorubicin, daunorubicin, teniposide, cisplatin (249) Taxanes, vinca-alkaloids (250, 251) 6-Mercaptopurine, 5-FU, PMEA (252, 253) Mitoxantrone, camptothecin, anthracycline, etoposide, teniposide imatinib, flavopiridol, bisantrene, methotrexate, AZT, etc. (171)

Modified from Gillet et al. (24)

4. Translational Research: From the Lab to the Clinic

Extensive research has characterized some of the mechanisms involved in multidrug resistance in  vitro. However, translating this to the clinic still represents a major challenge, as evidenced by the failure of trials to modulate ABCB1 expression (25) and the disputed role in  vivo of ABC transporters in MDR (24, 171). Briefly, meta-analyses exploring ABC transporter expression

Mechanisms of Multidrug Resistance in Cancer

63

profiles in leukemia have highlighted the disparity of the literature (172–174). In contrast to adult acute myeloid leukemia (AML), for which the role of ABCB1 gene expression in the drug resistance of tumors and prognosis of patients is widely accepted (175, 176), the data for adult acute lymphoblastic leukemia (ALL) are conflicting (176–178). In solid tumors, most attention has been directed to the roles played by ABCB1, ABCC1, and ABCG2 in MDR observed in breast cancer. However, it is difficult to decipher their exact role in the clinical drug resistance observed in this disease (179, 180). Various reasons have been suggested for the conflicting reports on the role of ABC transporters in the clinic. The most obvious one is that ABC transporters other than the three usual suspects (ABCB1, C1, and G2) can also influence treatment outcome. As mentioned in Subheading  4.2, 15 ABC transporters have been associated with drug resistance. Moreover, several recent studies suggest that more than 25 ABC transporters can be involved in chemotherapy-induced resistance (71, 181–184). In the previous section, we discussed the pleiotropic response of cancer cells to treatment. Therefore, another explanation for the disparity of the data reported is that studies focus on only one particular mechanism and neglect to investigate the other MDR mechanisms involved. The presence of normal or inflammatory tissues in tumors can also dramatically influence the gene expression profiles generated. Besides these biological factors, there are emerging issues associated with the “-omic” technologies. The variety of experimental systems developed to characterize cancers at a molecular level, such as cDNA expression profiling, comparative genomic hybridization (array CGH), promoter arrays, SNP arrays, etc., has increased dramatically in the last 5 years. In addition, the variety of platforms proposed for each of those analytical systems is remarkable, and numerous proposed normalization processes also render an integrative computational and analytical approach extremely challenging (185). The establishment of standard analytical methods and the development of systems biology/integrative approaches should help to produce a more unified picture of MDR. That could also lead to progress not only in understanding the mechanisms governing multidrug resistance but also in the translation of this knowledge to clinical practice, especially in personalized medicine. In this regard, the recent development of Oncomine, a bioinformatics initiative aimed at collecting, standardizing, analyzing, and delivering cancer transcriptome data to the scientific community is a step in the right direction (186–188). The knowledge acquired these last three decades by our own laboratory on MDR mechanisms has recently been tentatively translated to the clinic. Annereau et al. developed a high-density microarray platform dedicated to multidrug resistance to address the roles of MDR-linked genes in camptothecin resistance using cancer cell lines (189). However, microarrays require relatively

64

Gillet and Gottesman

large amounts of sample and often have poor probe specificity for families of genes with a high degree of homology, such as those involved in MDR. At the same time, Szakacs et  al. generated a database to identify lead compounds in the early stages of drug discovery that are not ABC transporter substrates (184). This repository is of great importance to help avoid the use of toxic drugs that ultimately will not provide any benefit to cancer patients. In that study, qRT-PCR using SYBR Green chemistry was used to profile the expression of ABC transporters. Although this platform is more reliable and accurate than microarrays, high-throughput analyses are tedious and require multiple pipetting steps, which can introduce variability. The availability of a reliable, sensitive and specific high-throughput platform that would allow the discrimination of highly homologous genes from small amounts of clinical tissue as opposed to in vitro models is fundamental to achieving any significant enhancement in the understanding of clinical multidrug resistance. In a recent study (190), our laboratory demonstrated the superiority of two platforms over established technologies in assessing ABC transporter expression profiles. This was demonstrated by an improved database that allows a more precise identification of compounds whose resistance is mediated by ABC transporters. In addition, it helps to pinpoint the compounds responsible for collateral sensitivity. We are now applying these platforms to identify genes critical to the development of MDR in cancer.

5. Conclusion For many years, MDR has been explained solely by an overexpression of ABCB1 in the tumor. Since its discovery in the late seventies, extensive research has characterized this ABC transporter-mediated MDR mechanism in  vitro. These studies have also highlighted a multiplicity of additional mechanisms governing MDR. Although, these mechanisms can act individually, this concise review underlines the pleiotropic phenomenon of the cell response to drug treatment (Fig.  4.3). Beside the cellular response, we have mentioned the role of the microenvironment on the initiation and maintenance of MDR, which is now supported by strong data. These last years have seen the development of a new paradigm suggesting the role of cancer stem cells in the intractability of cancers. Whether or not cancer stem cells represent one of the mechanism(s) of MDR needs to be unraveled. In this chapter, we have discussed briefly the fundamental need for translational research to confirm MDR mechanisms in human

Mechanisms of Multidrug Resistance in Cancer

65

Fig. 4.3. Pleotropic mechanisms of multidrug resistance. (1) Drug entry. Aside from pharmaceutical factors, the primary obstacle that prevents a drug from reaching the intracellular compartment is the plasma membrane. Therapeutic agents can react with many molecules resulting in a complex speciation profile. These species represented uniformly in this schema as a hexagonal (D) can enter cells by passive diffusion, endocytosis, or facilitated transport (uptake transporters). However, drug uptake can be significantly reduced by ATP-dependent drug efflux pumps (such as the ABC transporters ABCB1 and G2) and alterations in lipid metabolism (ceramide pathway) usually found in multidrug-resistant cells, which induce modifications in the biophysical properties of the lipid bilayer. (2) Drug metabolism. Once in the intracellular compartment, drug metabolism enzymes are the second line of cellular resistance. This process involves three phases. Phase I, or oxidative metabolism is mediated mainly by cytochrome P450 enzymes (CYPs) and epoxide hydrolases. Drug species are metabolized and converted into highly mutagenic aromatic metabolites (epoxide) that can be conjugated by phase II enzymes including GSTs, UGTs, SULTs, and NATS. These conjugated metabolites are then effluxed by transporters, which can be considered as phase III of drug metabolism. (3) Drug sequestration. Drug species can be trapped in subcellular organelles such as lysosomes and endosomes through ATP7A/B, ABCA3, or ABCB5 influx and then expelled from the cell. “Scavenger” metallothioneins ensnare metal ions and reactive oxygen species, leading to resistance to metal-based therapy and radiation. (4) Mechanisms activated after nuclear entry. Drug species (newly activated in the case of a prodrug-based strategy) that evade the above mechanisms of resistance enter the nucleus, where they encounter several mechanisms of resistance. Drug species can be effluxed via vault proteins into the cytoplasm and be either sequestered in intracellular vesicles or effluxed from the cell via ATP-dependent transport. Some drug species remain in the nucleus and form damaging adducts with DNA. A complex network of interacting pathways is then initiated, leading either to cell-cycle arrest and DNA repair or if the damage is extensive, rather than repair itself, the cell will enter one of these states: (1) senescence, (2) apoptosis, or (3) necrosis. (5) Evasion of drug-induced apoptosis. Disruption of apoptotic pathways, the hallmark of cancer, is a major obstacle to the success of chemotherapy. Blockage of apoptosis can result through the inhibitory effect of glycosylceramide and a myriad of pathways. (6) Microenvironment. Hypoxia upregulates the expression of numerous MDR-linked genes such as ABC transporters, Bcl2 family genes, glutathione, MT, etc., mainly through the activation of the transcription factor HIF1. It also dramatically reduces the effectiveness of chemotherapeutic agents that require oxidation to become cytotoxic or enhances the cytotoxicity of other agents that must undergo reduction to form active cytotoxic species. The acidic extracellular compartment also has important effects on the success of chemotherapy. (7) Signal transduction pathways. Cancer cells have altered signal transduction pathways, governed via

66

Gillet and Gottesman

cancers. Conflicting reports in the literature concerning the role of MDR-linked genes in cancer and recent technological development based on the “-omics” underline the urgency of the establishment of standard analytical methods to characterize tumor biology, especially MDR. We have entered the postgenomic era where scientific knowledge along with technological resources allows the development of systems biology. Important achievements in the understanding of clinical multidrug resistance will depend on our ability to accept these challenges.

Acknowledgments This research was supported by the Intramural Research Program of the National Institutes of Health, National Cancer Institute. We would like to thank George Leiman for editorial assistance. References 1. Lage H (2008) An overview of cancer multidrug resistance: a still unsolved problem. Cell Mol Life Sci 65:3145–3167 2. Mellor HR, Callaghan R (2008) Resistance to chemotherapy in cancer: a complex and integrated cellular response. Pharmacology 81:275–300 3. Mimeault M, Hauke R, Batra SK (2008) Recent advances on the molecular mechanisms involved in the drug resistance of cancer cells and novel targeting therapies. Clin Pharmacol Ther 83:673–691 4. Flaten GE, Dhanikula AB, Luthman K, Brandl M (2006) Drug permeability across a phospholipid vesicle based barrier: a novel approach for studying passive diffusion. Eur J Pharm Sci 27:80–90 5. Dobson PD, Kell DB (2008) Carrier-mediated cellular uptake of pharmaceutical drugs: an exception or the rule? Nat Rev Drug Discov 7:205–220 6. Assaraf YG (2007) Molecular basis of antifolate resistance. Cancer Metastasis Rev 26:153–181

7. Hediger MA, Romero MF, Peng JB et  al (2004) The ABCs of solute carriers: physiological, pathological and therapeutic implications of human membrane transport proteins. Pflugers Arch 447:465–468 8. Gray JH, Owen RP, Giacomini KM (2004) The concentrative nucleoside transporter family, SLC28. Pflugers Arch 447:728–734 9. Young JD, Yao SY, Sun L, Cass CE, Baldwin SA (2008) Human equilibrative nucleoside transporter (ENT) family of nucleoside and nucleobase transporter proteins. Xenobiotica 38:995–1021 10. Zhang J, Visser F, King KM et  al (2007) The role of nucleoside transporters in cancer chemotherapy with nucleoside drugs. Cancer Metastasis Rev 26:85–110 11. Mackey JR, Mani RS, Selner M et  al (1998) Functional nucleoside transporters are required for gemcitabine influx and manifestation of toxicity in cancer cell lines. Cancer Res 58: 4349–4357

Fig. 4.3. (continued) integrin receptors, growth factor receptors, frizzled receptors, and smoothened-patched receptors. These altered pathways can lead to the blockage of apoptosis and expression of MDR-linked genes such as those involved in DNA repair and drug-efflux pumps. Last but not least, cancer cells often display chromosomal abnormalities that can lead to the overexpression of antiapoptotic genes, etc. SLCs solute carriers, ABCs ATP-binding cassette transporters, SMase sphingomyelinase, GFR growth factor receptor, Wnt wingless, FZD frizzled, Smo smoothened, SHH sonic hedgehog, PTCH patched, MT metallothionein, GSTs glutathione-S-transferases, UGTs UDP-glucuronosyltransferases, SULTs sulfotransferases, NATs arylamine N-acetyltransferases, GCS glucosylceramide synthase, ABCCs ABC transporters subfamily C

Mechanisms of Multidrug Resistance in Cancer 12. Farrell JJ, Elsaleh H, Garcia M et  al (2009) Human equilibrative nucleoside transporter 1 levels predict response to gemcitabine in patients with pancreatic cancer. Gastroenterology 136: 187–195 13. Smith NF, Acharya MR, Desai N, Figg WD, Sparreboom A (2005) Identification of OATP1B3 as a high-affinity hepatocellular transporter of paclitaxel. Cancer Biol Ther 4:815–818 14. Smith NF, Marsh S, Scott-Horton TJ et  al (2007) Variants in the SLCO1B3 gene: interethnic distribution and association with paclitaxel pharmacokinetics. Clin Pharmacol Ther 81:76–82 15. Hall MD, Okabe M, Shen DW, Liang XJ, Gottesman MM (2008) The role of cellular accumulation in determining sensitivity to platinum-based chemotherapy. Annu Rev Pharmacol Toxicol 48:495–535 16. Wang DS, Jonker JW, Kato Y et  al (2002) Involvement of organic cation transporter 1 in hepatic and intestinal distribution of metformin. J Pharmacol Exp Ther 302:510–515 17. Kimura N, Masuda S, Tanihara Y et al (2005) Metformin is a superior substrate for renal organic cation transporter OCT2 rather than hepatic OCT1. Drug Metab Pharmacokinet 20:379–386 18. Okabe M, Szakacs G, Reimers MA et al (2008) Profiling SLCO and SLC22 genes in the NCI60 cancer cell lines to identify drug uptake transporters. Mol Cancer Ther 7:3081–3091 19. Scherf U, Ross DT, Waltham M et al (2000) A gene expression database for the molecular pharmacology of cancer. Nat Genet 24: 236–244 20. Rabow AA, Shoemaker RH, Sausville EA, Covell DG (2002) Mining the National Cancer Institute’s tumor-screening database: identification of compounds with similar cellular activities. J Med Chem 45:818–840 21. Shi LM, Fan Y, Lee JK et al (2000) Mining and visualizing large anticancer drug discovery databases. J Chem Inf Comput Sci 40:367–379 22. Weinstein JN, Kohn KW, Grever MR et  al (1992) Neural computing in cancer drug development: predicting mechanism of action. Science 258:447–451 23. Weinstein JN, Myers TG, O’Connor PM et al (1997) An information-intensive approach to the molecular pharmacology of cancer. Science 275:343–349 24. Gillet JP, Efferth T, Remacle J (2007) Chemotherapy-induced resistance by ATPbinding cassette transporter genes. Biochim Biophys Acta 1775:237–262

67

25. Szakacs G, Paterson JK, Ludwig JA, BoothGenthe C, Gottesman MM (2006) Targeting multidrug resistance in cancer. Nat Rev Drug Discov 5:219–234 26. Callaghan R, Crowley E, Potter S, Kerr ID (2008) P-glycoprotein: so many ways to turn it on. J Clin Pharmacol 48:365–378 27. Deeley RG, Westlake C, Cole SP (2006) Transmembrane transport of endo- and xenobiotics by mammalian ATP-binding cassette multidrug resistance proteins. Physiol Rev 86:849–899 28. Polgar O, Robey RW, Bates SE (2008) ABCG2: structure, function and role in drug response. Expert Opin Drug Metab Toxicol 4:1–15 29. Awasthi S, Singhal SS, Awasthi YC et al (2008) RLIP76 and Cancer. Clin Cancer Res 14:4372–4377 30. Kuo MT, Chen HH, Song IS, Savaraj N, Ishikawa T (2007) The roles of copper transporters in cisplatin resistance. Cancer Metastasis Rev 26:71–83 31. Drake KJ, Singhal J, Yadav S et  al (2007) RALBP1/RLIP76 mediates multidrug resistance. Int J Oncol 30:139–144 32. Singhal SS, Singhal J, Nair MP et al (2007) Doxorubicin transport by RALBP1 and ABCG2 in lung and breast cancer. Int J Oncol 30:717–725 33. Gouaze-Andersson V, Cabot MC (2006) Glycosphingolipids and drug resistance. Biochim Biophys Acta 1758:2096–2103 34. Hendrich AB, Michalak K (2003) Lipids as a target for drugs modulating multidrug resistance of cancer cells. Curr Drug Targets 4:23–30 35. Thevissen K, Francois IE, Winderickx J, Pannecouque C, Cammue BP (2006) Ceramide involvement in apoptosis and apoptotic diseases. Mini Rev Med Chem 6:699–709 36. Gouaze-Andersson V, Yu JY, Kreitenberg AJ et  al (2007) Ceramide and glucosylceramide upregulate expression of the multidrug resistance gene MDR1 in cancer cells. Biochim Biophys Acta 1771:1407–1417 37. Liu YY, Yu JY, Yin D et al (2008) A role for ceramide in driving cancer cell resistance to doxorubicin. FASEB J 22:2541–2551 38. Shen DW, Akiyama S, Schoenlein P, Pastan I, Gottesman MM (1995) Characterisation of high-level cisplatin-resistant cell lines established from a human hepatoma cell line and human KB adenocarcinoma cells: crossresistance and protein changes. Br J Cancer 71:676–683 39. Shen D, Pastan I, Gottesman MM (1998) Cross-resistance to methotrexate and metals

68

40.

41.

42.

43.

44.

45.

46.

47.

48.

49. 50.

51.

52.

Gillet and Gottesman in human cisplatin-resistant cell lines results from a pleiotropic defect in accumulation of these compounds associated with reduced plasma membrane binding proteins. Cancer Res 58:268–275 Shen DW, Goldenberg S, Pastan I, Gottesman MM (2000) Decreased accumulation of (14C)carboplatin in human cisplatin-resistant cells results from reduced energy-dependent uptake. J Cell Physiol 183:108–116 Shen DW, Su A, Liang XJ, Pai-Panandiker A, Gottesman MM (2004) Reduced expression of small GTPases and hypermethylation of the folate binding protein gene in cisplatinresistant cells. Br J Cancer 91:270–276 Liang XJ, Shen DW, Garfield S, Gottesman MM (2003) Mislocalization of membrane proteins associated with multidrug resistance in cisplatin-resistant cancer cell lines. Cancer Res 63:5909–5916 Liang XJ, Mukherjee S, Shen DW, Maxfield FR, Gottesman MM (2006) Endocytic recycling compartments altered in cisplatin-resistant cancer cells. Cancer Res 66:2346–2353 Chauhan SS, Liang XJ, Su AW et  al (2003) Reduced endocytosis and altered lysosome function in cisplatin-resistant cell lines. Br J Cancer 88:1327–1334 Liang XJ, Yin JJ, Zhou JW et  al (2004) Changes in biophysical parameters of plasma membranes influence cisplatin resistance of sensitive and resistant epidermal carcinoma cells. Exp Cell Res 293:283–291 Brown CM, Reisfeld B, Mayeno AN (2008) Cytochromes P450: a structure-based summary of biotransformations using representative substrates. Drug Metab Rev 40:1–100 Oyama T, Kagawa N, Kunugita N et al (2004) Expression of cytochrome P450 in tumor tissues and its association with cancer development. Front Biosci 9:1967–1976 Purnapatre K, Khattar SK, Saini KS (2008) Cytochrome P450s in the development of target-based anticancer drugs. Cancer Lett 259:1–15 van Schaik RH (2008) CYP450 pharmacogenetics for personalizing cancer therapy. Drug Resist Updat 11:77–98 Xie HJ, Yasar U, Lundgren S et  al (2003) Role of polymorphic human CYP2B6 in cyclophosphamide bioactivation. Pharma­ cogenomics J 3:53–61 Xie H, Griskevicius L, Stahle L et  al (2006) Pharmacogenetics of cyclophosphamide in patients with hematological malignancies. Eur J Pharm Sci 27:54–61 Rodriguez-Novoa S, Barreiro P, Rendon A et al (2005) Influence of 516G>T polymorphisms

53.

54.

55.

56. 57. 58.

59. 60.

61.

62. 63. 64.

65.

at the gene encoding the CYP450–2B6 isoenzyme on efavirenz plasma concentrations in HIV-infected subjects. Clin Infect Dis 40:1358–1361 Rotger M, Colombo S, Furrer H et al (2005) Influence of CYP2B6 polymorphism on plasma and intracellular concentrations and toxicity of efavirenz and nevirapine in HIVinfected patients. Pharmacogenet Genomics 15:1–5 Wang J, Sonnerborg A, Rane A et al (2006) Identification of a novel specific CYP2B6 allele in Africans causing impaired metabolism of the HIV drug efavirenz. Pharmacogenet Genomics 16:191–198 Huttunen KM, Mahonen N, Raunio H, Rautio J (2008) Cytochrome P450-activated prodrugs: targeted drug delivery. Curr Med Chem 15:2346–2365 Pal D, Mitra AK (2006) MDR- and CYP3A4mediated drug-drug interactions. J Neuro­ immune Pharmacol 1:323–339 Ghezzi P, Di Simplicio P (2007) Glutathiony­ lation pathways in drug response. Curr Opin Pharmacol 7:398–403 Iyanagi T (2007) Molecular mechanism of phase I and phase II drug-metabolizing enzymes: implications for detoxification. Int Rev Cytol 260:35–112 Hemmerich S, Verdugo D, Rath VL (2004) Strategies for drug discovery by targeting sulfation pathways. Drug Discov Today 9:967–975 Townsend DM, Findlay VL, Tew KD (2005) Glutathione S-transferases as regulators of kinase pathways and anticancer drug targets. Methods Enzymol 401:287–307 Sorich MJ, Smith PA, Miners JO, Mackenzie PI, McKinnon RA (2008) Recent advances in the in silico modelling of UDP glucuronosyltransferase substrates. Curr Drug Metab 9:60–69 Gamage N, Barnett A, Hempel N et al (2006) Human sulfotransferases and their role in chemical metabolism. Toxicol Sci 90:5–22 Butcher NJ, Tiang J, Minchin RF (2008) Regulation of arylamine N-acetyltransferases. Curr Drug Metab 9:498–504 Ekhart C, Rodenhuis S, Smits PH, Beijnen JH, Huitema AD (2009) An overview of the relations between polymorphisms in drug metabolising enzymes and drug transporters and survival after cancer drug treatment. Cancer Treat Rev 35:18–31 Safaei R, Larson BJ, Cheng TC et al (2005) Abnormal lysosomal trafficking and enhanced exosomal export of cisplatin in drug-resistant human ovarian carcinoma cells. Mol Cancer Ther 4:1595–1604

Mechanisms of Multidrug Resistance in Cancer 66. Liang XJ, Shen DW, Chen KG et  al (2005) Trafficking and localization of platinum complexes in cisplatin-resistant cell lines monitored by fluorescence-labeled platinum. J Cell Physiol 202:635–641 67. Safaei R, Katano K, Larson BJ et  al (2005) Intracellular localization and trafficking of fluorescein-labeled cisplatin in human ovarian carcinoma cells. Clin Cancer Res 11:756–767 68. Kalayda GV, Wagner CH, Buss I, Reedijk J, Jaehde U (2008) Altered localisation of the copper efflux transporters ATP7A and ATP7B associated with cisplatin resistance in human ovarian carcinoma cells. BMC Cancer 8:175 69. Chen KG, Valencia JC, Lai B et  al (2006) Melanosomal sequestration of cytotoxic drugs contributes to the intractability of malignant melanomas. Proc Natl Acad Sci USA 103: 9903–9907 70. Chen KG, Szakacs G, Annereau JP et  al (2005) Principal expression of two mRNA isoforms (ABCB 5alpha and ABCB 5beta ) of the ATP-binding cassette transporter gene ABCB 5 in melanoma cells and melanocytes. Pigment Cell Res 18:102–112 71. Steinbach D, Gillet JP, Sauerbrey A et  al (2006) ABCA3 as a possible cause of drug resistance in childhood acute myeloid leukemia. Clin Cancer Res 12:4357–4363 72. Chapuy B, Koch R, Radunski U et al (2008) Intracellular ABC transporter A3 confers multidrug resistance in leukemia cells by lysosomal drug sequestration. Leukemia 22:1576–1586 73. Cheong N, Madesh M, Gonzales LW et  al (2006) Functional and trafficking defects in ATP binding cassette A3 mutants associated with respiratory distress syndrome. J Biol Chem 281:9791–9800 74. Theocharis SE, Margeli AP, Klijanienko JT, Kouraklis GP (2004) Metallothionein expression in human neoplasia. Histopathology 45: 103–118 75. Thirumoorthy N, Manisenthil Kumar KT, Shyam Sundar A, Panayappan L, Chatterjee M (2007) Metallothionein: an overview. World J Gastroenterol 13:993–996 76. Harper JW, Elledge SJ (2007) The DNA damage response: ten years after. Mol Cell 28:739–745 77. Ashwell S, Zabludoff S (2008) DNA damage detection and repair pathways–recent advances with inhibitors of checkpoint kinases in cancer therapy. Clin Cancer Res 14:4032–4037 78. Martin LP, Hamilton TC, Schilder RJ (2008) Platinum resistance: the role of DNA repair pathways. Clin Cancer Res 14:1291–1295 79. Sarkaria JN, Kitange GJ, James CD et  al (2008) Mechanisms of chemoresistance to

69

alkylating agents in malignant glioma. Clin Cancer Res 14:2900–2908 80. Hakem R (2008) DNA-damage repair; the good, the bad, and the ugly. EMBO J 27: 589–605 81. Helleday T, Petermann E, Lundin C, Hodgson B, Sharma RA (2008) DNA repair pathways as targets for cancer therapy. Nat Rev Cancer 8:193–204 82. Dabholkar M, Bostick-Bruton F, Weber C et al (1992) ERCC1 and ERCC2 expression in malignant tissues from ovarian cancer patients. J Natl Cancer Inst 84:1512–1517 83. Kang S, Ju W, Kim JW et al (2006) Association between excision repair cross-complementation group 1 polymorphism and clinical outcome of platinum-based chemotherapy in patients with epithelial ovarian cancer. Exp Mol Med 38:320–324 84. Olaussen KA, Dunant A, Fouret P et  al (2006) DNA repair by ERCC1 in nonsmall-cell lung cancer and cisplatin-based adjuvant chemotherapy. N Engl J Med 355: 983–991 85. Metzger R, Leichman CG, Danenberg KD et  al (1998) ERCC1 mRNA levels complement thymidylate synthase mRNA levels in predicting response and survival for gastric cancer patients receiving combination cisplatin and fluorouracil chemotherapy. J Clin Oncol 16:309–316 86. Fink D, Nebel S, Norris PS et  al (1998) Enrichment for DNA mismatch repairdeficient cells during treatment with cisplatin. Int J Cancer 77:741–746 87. Strathdee G, MacKean MJ, Illand M, Brown R (1999) A role for methylation of the hMLH1 promoter in loss of hMLH1 expression and drug resistance in ovarian cancer. Oncogene 18:2335–2341 88. Karran P, Marinus MG (1982) Mismatch correction at O6-methylguanine residues in E. coli DNA. Nature 296:868–869 89. Yoshioka K, Yoshioka Y, Hsieh P (2006) ATR kinase activation mediated by MutSalpha and MutLalpha in response to cytotoxic O6-methylguanine adducts. Mol Cell 22: 501–510 90. Verbeek B, Southgate TD, Gilham DE, Margison GP (2008) O6-Methylguanine-DNA methyltransferase inactivation and chemotherapy. Br Med Bull 85:17–33 91. Okada H, Mak TW (2004) Pathways of apoptotic and non-apoptotic death in tumour cells. Nat Rev Cancer 4:592–603 92. Broker LE, Kruyt FA, Giaccone G (2005) Cell death independent of caspases: a review. Clin Cancer Res 11:3155–3162

70

Gillet and Gottesman

93. Falschlehner C, Emmerich CH, Gerlach B, Walczak H (2007) TRAIL signalling: decisions between life and death. Int J Biochem Cell Biol 39:1462–1475 94. Tait SW, Green DR (2008) Caspaseindependent cell death: leaving the set without the final cut. Oncogene 27:6452–6461 95. Rodriguez-Nieto S, Zhivotovsky B (2006) Role of alterations in the apoptotic machinery in sensitivity of cancer cells to treatment. Curr Pharm Des 12:4411–4425 96. Viktorsson K, Lewensohn R, Zhivotovsky B (2005) Apoptotic pathways and therapy resistance in human malignancies. Adv Cancer Res 94:143–196 97. Sjostrom J, Blomqvist C, von Boguslawski K et al (2002) The predictive value of bcl-2, bax, bcl-xL, bag-1, fas, and fasL for chemotherapy response in advanced breast cancer. Clin Cancer Res 8:811–816 98. Malamou-Mitsi V, Gogas H, Dafni U et  al (2006) Evaluation of the prognostic and predictive value of p53 and Bcl-2 in breast cancer patients participating in a randomized study with dose-dense sequential adjuvant chemotherapy. Ann Oncol 17:1504–1511 99. Krug LM, Miller VA, Filippa DA et al (2003) Bcl-2 and bax expression in advanced nonsmall cell lung cancer: lack of correlation with chemotherapy response or survival in patients treated with docetaxel plus vinorelbine. Lung Cancer 39:139–143 100. Bonetti A, Zaninelli M, Leone R et al (1998) Bcl-2 but not p53 expression is associated with resistance to chemotherapy in advanced breast cancer. Clin Cancer Res 4:2331–2336 101. Buchholz TA, Davis DW, McConkey DJ et al (2003) Chemotherapy-induced apoptosis and Bcl-2 levels correlate with breast cancer response to chemotherapy. Cancer J 9:33–41 102. Kedersha NL, Rome LH (1986) Isolation and characterization of a novel ribonucleoprotein particle: large structures contain a single species of small RNA. J Cell Biol 103:699–709 103. Kedersha NL, Heuser JE, Chugani DC, Rome LH (1991) Vaults. III. Vault ribonucleoprotein particles open into flower-like structures with octagonal symmetry. J Cell Biol 112:225–235 104. Rome L, Kedersha N, Chugani D (1991) Unlocking vaults: organelles in search of a function. Trends Cell Biol 1:47–50 105. Kedersha NL, Miquel MC, Bittner D, Rome LH (1990) Vaults. II. Ribonucleoprotein structures are highly conserved among higher and lower eukaryotes. J Cell Biol 110: 895–901

106. van Zon A, Mossink MH, Scheper RJ, Sonneveld P, Wiemer EA (2003) The vault complex. Cell Mol Life Sci 60:1828–1837 107. Izquierdo MA, Scheffer GL, Flens MJ et al (1996) Broad distribution of the multidrug resistance-related vault lung resistance protein in normal human tissues and tumors. Am J Pathol 148:877–887 108. Kickhoefer VA, Rajavel KS, Scheffer GL et al (1998) Vaults are up-regulated in multidrugresistant cancer cell lines. J Biol Chem 273: 8971–8974 109. Mossink MH, van Zon A, Scheper RJ, Sonneveld P, Wiemer EA (2003) Vaults: a ribonucleoprotein particle involved in drug resistance? Oncogene 22:7458–7467 110. Steiner E, Holzmann K, Elbling L, Micksche M, Berger W (2006) Cellular functions of vaults and their involvement in multidrug resistance. Curr Drug Targets 7:923–934 111. Berger W, Steiner E, Grusch M, Elbling L, Micksche M (2009) Vaults and the major vault protein: novel roles in signal pathway regulation and immunity. Cell Mol Life Sci 66:43–61 112. Scheper RJ, Broxterman HJ, Scheffer GL et al (1993) Overexpression of a M(r) 110,000 vesicular protein in non-P-glycoproteinmediated multidrug resistance. Cancer Res 53:1475–1479 113. Kitazono M, Okumura H, Ikeda R et  al (2001) Reversal of LRP-associated drug resistance in colon carcinoma SW-620 cells. Int J Cancer 91:126–131 114. Kitazono M, Sumizawa T, Takebayashi Y et  al (1999) Multidrug resistance and the lung resistance-related protein in human colon carcinoma SW-620 cells. J Natl Cancer Inst 91:1647–1653 115. Laurencot CM, Scheffer GL, Scheper RJ, Shoemaker RH (1997) Increased LRP mRNA expression is associated with the MDR phenotype in intrinsically resistant human cancer cell lines. Int J Cancer 72:1021–1026 116. Scheffer GL, Wijngaard PL, Flens MJ et  al (1995) The drug resistance-related protein LRP is the human major vault protein. Nat Med 1:578–582 117. Siva AC, Raval-Fernandes S, Stephen AG et al (2001) Up-regulation of vaults may be necessary but not sufficient for multidrug resistance. Int J Cancer 92:195–202 118. van Zon A, Mossink MH, Schoester M et al (2004) Efflux kinetics and intracellular distribution of daunorubicin are not affected by major vault protein/lung resistance-related protein (vault) expression. Cancer Res 64: 4887–4892

Mechanisms of Multidrug Resistance in Cancer 119. Mossink MH, van Zon A, Franzel-Luiten E et al (2002) Disruption of the murine major vault protein (MVP/LRP) gene does not induce hypersensitivity to cytostatics. Cancer Res 62:7298–7304 120. Mossink MH, de Groot J, van Zon A et  al (2003) Unimpaired dendritic cell functions in MVP/LRP knockout mice. Immunology 110:58–65 121. Jin S, Scotto KW (1998) Transcriptional regulation of the MDR1 gene by histone acetyltransferase and deacetylase is mediated by NF-Y. Mol Cell Biol 18:4377–4384 122. Li H, Liu H, Wang Z et al (2008) The role of transcription factors Sp1 and YY1 in proximal promoter region in initiation of transcription of the mu opioid receptor gene in human lymphocytes. J Cell Biochem 104:237–250 123. Tlsty TD, Coussens LM (2006) Tumor stroma and regulation of cancer development. Annu Rev Pathol 1:119–150 124. Cosse JP, Michiels C (2008) Tumour hypoxia affects the responsiveness of cancer cells to chemotherapy and promotes cancer progression. Anticancer Agents Med Chem 8:790–797 125. Netti PA, Berk DA, Swartz MA, Grodzinsky AJ, Jain RK (2000) Role of extracellular matrix assembly in interstitial transport in solid tumors. Cancer Res 60:2497–2503 126. Heldin CH, Rubin K, Pietras K, Ostman A (2004) High interstitial fluid pressure – an obstacle in cancer therapy. Nat Rev Cancer 4:806–813 127. Tredan O, Galmarini CM, Patel K, Tannock IF (2007) Drug resistance and the solid tumor microenvironment. J Natl Cancer Inst 99:1441–1454 128. Magzoub M, Jin S, Verkman AS (2008) Enhanced macromolecule diffusion deep in tumors after enzymatic digestion of extracellular matrix collagen and its associated proteoglycan decorin. FASEB J 22:276–284 129. Padera TP, Stoll BR, Tooredman JB et  al (2004) Pathology: cancer cells compress intratumour vessels. Nature 427:695 130. Jain RK (1988) Determinants of tumor blood flow: a review. Cancer Res 48:2641–2658 131. Netti PA, Hamberg LM, Babich JW et  al (1999) Enhancement of fluid filtration across tumor vessels: implication for delivery of macromolecules. Proc Natl Acad Sci USA 96:3137–3142 132. Leu AJ, Berk DA, Lymboussaki A, Alitalo K, Jain RK (2000) Absence of functional lymphatics within a murine sarcoma: a molecular and functional evaluation. Cancer Res 60:4324–4327

71

133. Salnikov AV, Iversen VV, Koisti M et  al (2003) Lowering of tumor interstitial fluid pressure specifically augments efficacy of chemotherapy. FASEB J 17:1756–1758 134. Roos A (1978) Weak acids, weak bases and intracellular pH. Respir Physiol 33:27–30 135. Gerweck LE, Vijayappa S, Kozin S (2006) Tumor pH controls the in  vivo efficacy of weak acid and base chemotherapeutics. Mol Cancer Ther 5:1275–1279 136. Boyer MJ (1997) Bioreductive agents: a clinical update. Oncol Res 9:391–395 137. Lluis JM, Morales A, Blasco C et al (2005) Critical role of mitochondrial glutathione in the survival of hepatocytes during hypoxia. J Biol Chem 280:3224–3232 138. Murphy BJ, Laderoute KR, Chin RJ, Sutherland RM (1994) Metallothionein IIA is up-regulated by hypoxia in human A431 squamous carcinoma cells. Cancer Res 54: 5808–5810 139. Comerford KM, Wallace TJ, Karhausen J et  al (2002) Hypoxia-inducible factor-1dependent regulation of the multidrug resistance (MDR1) gene. Cancer Res 62: 3387–3394 140. Liu L, Ning X, Sun L et al (2008) Hypoxiainducible factor-1 alpha contributes to hypoxia-induced chemoresistance in gastric cancer. Cancer Sci 99:121–128 141. Han HK, Han CY, Cheon EP, Lee J, Kang KW (2007) Role of hypoxia-inducible factoralpha in hepatitis-B-virus X protein-mediated MDR1 activation. Biochem Biophys Res Commun 357:567–573 142. Song X, Liu X, Chi W et al (2006) Hypoxiainduced resistance to cisplatin and doxorubicin in non-small cell lung cancer is inhibited by silencing of HIF-1alpha gene. Cancer Chemother Pharmacol 58:776–784 143. Krishnamurthy P, Ross DD, Nakanishi T et  al (2004) The stem cell marker Bcrp/ ABCG2 enhances hypoxic cell survival through interactions with heme. J Biol Chem 279:24218–24225 144. Alkatout I, Kabelitz D, Kalthoff H, Tiwari S (2008) Prowling wolves in sheep’s clothing: the search for tumor stem cells. Biol Chem 389:799–811 145. Trumpp A, Wiestler OD (2008) Mechanisms of Disease: cancer stem cells–targeting the evil twin. Nat Clin Pract Oncol 5:337–347 146. Kim M, Turnquist H, Jackson J et al (2002) The multidrug resistance transporter ABCG2 (breast cancer resistance protein 1) effluxes Hoechst 33342 and is overexpressed in hematopoietic stem cells. Clin Cancer Res 8:22–28

72

Gillet and Gottesman

147. Zhou S, Schuetz JD, Bunting KD et  al (2001) The ABC transporter Bcrp1/ABCG2 is expressed in a wide variety of stem cells and is a molecular determinant of the side-population phenotype. Nat Med 7: 1028–1034 148. Bunting KD (2002) ABC transporters as phenotypic markers and functional regulators of stem cells. Stem Cells 20:11–20 149. Hirschmann-Jax C, Foster AE, Wulf GG et al (2004) A distinct “side population” of cells with high drug efflux capacity in human tumor cells. Proc Natl Acad Sci USA 101: 14228–14233 150. Schatton T, Murphy GF, Frank NY et  al (2008) Identification of cells initiating human melanomas. Nature 451:345–349 151. Frank NY, Margaryan A, Huang Y et al (2005) ABCB5-mediated doxorubicin transport and chemoresistance in human malignant melanoma. Cancer Res 65:4320–4333 152. Frank NY, Pendse SS, Lapchak PH et  al (2003) Regulation of progenitor cell fusion by ABCB5 P-glycoprotein, a novel human ATP-binding cassette transporter. J Biol Chem 278:47156–47165 153. Fromm MF (2004) Importance of P-glyco­ protein at blood-tissue barriers. Trends Pharmacol Sci 25:423–429 154. Borst P, Elferink RO (2002) Mammalian ABC transporters in health and disease. Annu Rev Biochem 71:537–592 155. Trowsdale J, Hanson I, Mockridge I et  al (1990) Sequences encoded in the class II region of the MHC related to the ‘ABC’ superfamily of transporters. Nature 348: 741–744 156. Abele R, Tampe R (2004) The ABCs of immunology: structure and function of TAP, the transporter associated with antigen processing. Physiology (Bethesda) 19:216–224 157. Takahashi K, Kimura Y, Nagata K et al (2005) ABC proteins: key molecules for lipid homeostasis. Med Mol Morphol 38:2–12 158. Juliano RL, Ling V (1976) A surface glycoprotein modulating drug permeability in Chinese hamster ovary cell mutants. Biochim Biophys Acta 455:152–162 159. Ueda K, Cardarelli C, Gottesman MM, Pastan I (1987) Expression of a full-length cDNA for the human “MDR1” gene confers resistance to colchicine, doxorubicin, and vinblastine. Proc Natl Acad Sci USA 84: 3004–3008 160. Ueda K, Clark DP, Chen CJ et al (1987) The human multidrug resistance (mdr1) gene. cDNA cloning and transcription initiation. J Biol Chem 262:505–508

161. Cole SP, Bhardwaj G, Gerlach JH et al (1992) Overexpression of a transporter gene in a multidrug-resistant human lung cancer cell line. Science 258:1650–1654 162. Doyle LA, Yang W, Abruzzo LV et al (1998) A multidrug resistance transporter from human MCF-7 breast cancer cells. Proc Natl Acad Sci USA 95:15665–15670 163. Ambudkar SV, Kim IW, Sauna ZE (2006) The power of the pump: mechanisms of action of P-glycoprotein (ABCB1). Eur J Pharm Sci 27:392–400 164. Cascorbi I (2006) Role of pharmacogenetics of ATP-binding cassette transporters in the pharmacokinetics of drugs. Pharmacol Ther 112:457–473 165. Choudhuri S, Klaassen CD (2006) Structure, function, expression, genomic organization, and single nucleotide polymorphisms of human ABCB1 (MDR1), ABCC (MRP), and ABCG2 (BCRP) efflux transporters. Int J Toxicol 25:231–259 166. Ieiri I, Takane H, Otsubo K (2004) The MDR1 (ABCB1) gene polymorphism and its clinical implications. Clin Pharmacokinet 43: 553–576 167. Kimchi-Sarfaty C, Oh JM, Kim IW et  al (2007) A “silent” polymorphism in the MDR1 gene changes substrate specificity. Science 315:525–528 168. Fung KL, Gottesman MM (2009) A synonymous polymorphism in a common MDR1 (ABCB1) haplotype shapes protein function. Biochim Biophys Acta 1794:860–871 169. Kohno K, Sato S, Takano H, Matsuo K, Kuwano M (1989) The direct activation of human multidrug resistance gene (MDR1) by anticancer agents. Biochem Biophys Res Commun 165:1415–1421 170. Chaudhary PM, Roninson IB (1993) Induction of multidrug resistance in human cells by transient exposure to different chemotherapeutic drugs. J Natl Cancer Inst 85:632–639 171. Robey RW, Polgar O, Deeken J, To KW, Bates SE (2007) ABCG2: determining its relevance in clinical drug resistance. Cancer Metastasis Rev 26:39–57 172. Hirose M, Hosoi E, Hamano S, Jalili A (2003) Multidrug resistance in hematological malignancy. J Med Invest 50:126–135 173. Ross DD (2000) Novel mechanisms of drug resistance in leukemia. Leukemia 14:467–473 174. van den Heuvel-Eibrink MM, Sonneveld P, Pieters R (2000) The prognostic significance of membrane transport-associated multidrug resistance (MDR) proteins in leukemia. Int J Clin Pharmacol Ther 38:94–110

Mechanisms of Multidrug Resistance in Cancer 175. Marie JP, Legrand O (1999) MDR1/P-GP expression as a prognostic factor in acute leukemias. Adv Exp Med Biol 457:1–9 176. Wuchter C, Leonid K, Ruppert V et al (2000) Clinical significance of P-glycoprotein expression and function for response to induction chemotherapy, relapse rate and overall survival in acute leukemia. Haematologica 85:711–721 177. Tafuri A, Gregorj C, Petrucci MT et  al (2002) MDR1 protein expression is an independent predictor of complete remission in newly diagnosed adult acute lymphoblastic leukemia. Blood 100:974–981 178. Wattel E, Lepelley P, Merlat A et al (1995) Expression of the multidrug resistance P glycoprotein in newly diagnosed adult acute lymphoblastic leukemia: absence of correlation with response to treatment. Leukemia 9:1870–1874 179. Larkin A, O’Driscoll L, Kennedy S et  al (2004) Investigation of MRP-1 protein and MDR-1 P-glycoprotein expression in invasive breast cancer: a prognostic study. Int J Cancer 112:286–294 180. Leonessa F, Clarke R (2003) ATP binding cassette transporters and drug resistance in breast cancer. Endocr Relat Cancer 10:43–73 181. Efferth T, Gillet JP, Sauerbrey A et al (2006) Expression profiling of ATP-binding cassette transporters in childhood T-cell acute lymphoblastic leukemia. Mol Cancer Ther 5: 1986–1994 182. Gillet JP, Schneider J, Bertholet V et  al (2006) Microarray Expression Profiling of ABC Transporters in Human Breast Cancer. Cancer Genom Proteom 3:97–106 183. Gillet JP, Efferth T, Steinbach D et al (2004) Microarray-based detection of multidrug resistance in human tumor cells by expression profiling of ATP-binding cassette transporter genes. Cancer Res 64:8987–8993 184. Szakacs G, Annereau JP, Lababidi S et  al (2004) Predicting drug sensitivity and resistance: profiling ABC transporter genes in cancer cells. Cancer Cell 6:129–137 185. Rhodes DR, Chinnaiyan AM (2005) Integrative analysis of the cancer transcriptome. Nat Genet 37(Suppl):S31–S37 186. Rhodes DR, Kalyana-Sundaram S, Mahavisno V et  al (2007) Oncomine 3.0: genes, pathways, and networks in a collection of 18,000 cancer gene expression profiles. Neoplasia 9:166–180 187. Rhodes DR, Kalyana-Sundaram S, Tomlins SA et al (2007) Molecular concepts analysis links tumors, pathways, mechanisms, and drugs. Neoplasia 9:443–454

73

188. Rhodes DR, Yu J, Shanker K et  al (2004) ONCOMINE: a cancer microarray database and integrated data-mining platform. Neoplasia 6:1–6 189. Annereau JP, Szakacs G, Tucker CJ et  al (2004) Analysis of ATP-binding cassette transporter expression in drug-selected cell lines by a microarray dedicated to multidrug resistance. Mol Pharmacol 66:1397–1405 190. Orina JN, Calcagno AM, Wu C-P, et al (2009) Evaluation of current methods to analyze the expression profiles of ABC transporters yields an improved drug-discovery database. Mol Cancer Ther 8:2057–2066 191. Filipits M, Stranzl T, Pohl G et al (2000) Drug resistance factors in acute myeloid leukemia: a comparative analysis. Leukemia 14:68–76 192. Hart SM, Ganeshaguru K, Scheper RJ et al (1997) Expression of the human major vault protein LRP in acute myeloid leukemia. Exp Hematol 25:1227–1232 193. List AF, Spier CS, Grogan TM et al (1996) Overexpression of the major vault transporter protein lung-resistance protein predicts treatment outcome in acute myeloid leukemia. Blood 87:2464–2469 194. Xu D, Arestrom I, Virtala R et  al (1999) High levels of lung resistance related protein mRNA in leukaemic cells from patients with acute myelogenous leukaemia are associated with inferior response to chemotherapy and prior treatment with mitoxantrone. Br J Haematol 106:627–633 195. Legrand O, Simonin G, Beauchamp-Nicoud A, Zittoun R, Marie JP (1999) Simultaneous activity of MRP1 and Pgp is correlated with in  vitro resistance to daunorubicin and with in vivo resistance in adult acute myeloid leukemia. Blood 94:1046–1056 196. Leith CP, Kopecky KJ, Chen IM et al (1999) Frequency and clinical significance of the expression of the multidrug resistance proteins MDR1/P-glycoprotein, MRP1, and LRP in acute myeloid leukemia: a Southwest Oncology Group Study. Blood 94:1086–1099 197. Schaich M, Soucek S, Thiede C, Ehninger G, Illmer T (2005) MDR1 and MRP1 gene expression are independent predictors for treatment outcome in adult acute myeloid leukaemia. Br J Haematol 128:324–332 198. Sunnaram BL, Gandemer V, Sebillot M et al (2003) LRP overexpression in monocytic lineage. Leuk Res 27:755–759 199. van den Heuvel-Eibrink MM, Wiemer EA, Prins A et al (2002) Increased expression of the breast cancer resistance protein (BCRP) in relapsed or refractory acute myeloid leukemia (AML). Leukemia 16:833–839

74

Gillet and Gottesman

200. Borg AG, Burgess R, Green LM, Scheper RJ, Yin JA (1998) Overexpression of lung-resistance protein and increased P-glycoprotein function in acute myeloid leukaemia cells predict a poor response to chemotherapy and reduced patient survival. Br J Haematol 103: 1083–1091 201. Filipits M, Pohl G, Stranzl T et  al (1998) Expression of the lung resistance protein predicts poor outcome in de novo acute myeloid leukemia. Blood 91:1508–1513 202. Pirker R, Pohl G, Stranzl T et al (1999) The lung resistance protein (LRP) predicts poor outcome in acute myeloid leukemia. Adv Exp Med Biol 457:133–139 203. Oh EJ, Kahng J, Kim Y et al (2003) Expression of functional markers in acute lymphoblastic leukemia. Leuk Res 27:903–908 204. Valera ET, Scrideli CA, Queiroz RG, Mori BM, Tone LG (2004) Multiple drug resistance protein (MDR-1), multidrug resistancerelated protein (MRP) and lung resistance protein (LRP) gene expression in childhood acute lymphoblastic leukemia. Sao Paulo Med J 122:166–171 205. Volm M, Stammler G, Zintl F, Koomagi R, Sauerbrey A (1997) Expression of lung resistance-related protein (LRP) in initial and relapsed childhood acute lymphoblastic leukemia. Anticancer Drugs 8:662–665 206. Ohno N, Tani A, Uozumi K et  al (2001) Expression of functional lung resistance– related protein predicts poor outcome in adult T-cell leukemia. Blood 98:1160–1165 207. Filipits M, Drach J, Pohl G et  al (1999) Expression of the lung resistance protein predicts poor outcome in patients with multiple myeloma. Clin Cancer Res 5:2426–2430 208. Raaijmakers HG, Izquierdo MA, Lokhorst HM et  al (1998) Lung-resistance-related protein expression is a negative predictive factor for response to conventional low but not to intensified dose alkylating chemotherapy in multiple myeloma. Blood 91:1029–1036 209. Rimsza LM, Campbell K, Dalton WS et  al (1999) The major vault protein (MVP), a new multidrug resistance associated protein, is frequently expressed in multiple myeloma. Leuk Lymphoma 34:315–324 210. Brinkhuis M, Izquierdo MA, Baak JP et  al (2002) Expression of multidrug resistanceassociated markers, their relation to quantitative pathologic tumour characteristics and prognosis in advanced ovarian cancer. Anal Cell Pathol 24:17–23 211. Izquierdo MA, van der Zee AG, Vermorken JB et  al (1995) Drug resistance-associated marker Lrp for prediction of response to che-

motherapy and prognoses in advanced ovarian carcinoma. J Natl Cancer Inst 87: 1230–1237 212. Mayr D, Pannekamp U, Baretton GB et  al (2000) Immunohistochemical analysis of drug resistance-associated proteins in ovarian carcinomas. Pathol Res Pract 196:469–475 213. Arts HJ, Katsaros D, de Vries EG et  al (1999) Drug resistance-associated markers P-glycoprotein, multidrug resistanceassociated protein 1, multidrug resistanceassociated protein 2, and lung resistance protein as prognostic factors in ovarian carcinoma. Clin Cancer Res 5:2798–2805 214. Burger H, Foekens JA, Look MP et  al (2003) RNA expression of breast cancer resistance protein, lung resistance-related protein, multidrug resistance-associated proteins 1 and 2, and multidrug resistance gene 1 in breast cancer: correlation with chemotherapeutic response. Clin Cancer Res 9:827–836 215. Linn SC, Pinedo HM, van Ark-Otte J et  al (1997) Expression of drug resistance proteins in breast cancer, in relation to chemotherapy. Int J Cancer 71:787–795 216. Schneider J, Lucas R, Sanchez J et al (2000) Modulation of molecular marker expression by induction chemotherapy in locally advanced breast cancer: correlation with the response to therapy and the expression of MDR1 and LRP. Anticancer Res 20:4373–4377 217. Kanzaki A, Toi M, Nakayama K et al (2001) Expression of multidrug resistance-related transporters in human breast carcinoma. Jpn J Cancer Res 92:452–458 218. Pohl G, Filipits M, Suchomel RW et al (1999) Expression of the lung resistance protein (LRP) in primary breast cancer. Anticancer Res 19:5051–5055 219. Volm M, Mattern J, Koomagi R (1997) Expression of lung resistance-related protein (LRP) in non-small cell lung carcinomas of smokers and non-smokers and its predictive value for doxorubicin resistance. Anticancer Drugs 8:931–936 220. Volm M, Rittgen W (2000) Cellular predictive factors for the drug response of lung cancer. Anticancer Res 20:3449–3458 221. Chiou JF, Liang JA, Hsu WH et  al (2003) Comparing the relationship of Taxol-based chemotherapy response with P-glycoprotein and lung resistance-related protein expression in non-small cell lung cancer. Lung 181: 267–273 222. Dingemans AM, van Ark-Otte J, van der Valk P et al (1996) Expression of the human major vault protein LRP in human lung cancer

Mechanisms of Multidrug Resistance in Cancer samples and normal lung tissues. Ann Oncol 7:625–630 223. Berger W, Setinek U, Hollaus P et al (2005) Multidrug resistance markers P-glycoprotein, multidrug resistance protein 1, and lung resistance protein in non-small cell lung cancer: prognostic implications. J Cancer Res Clin Oncol 131:355–363 224. Diestra JE, Condom E, Del Muro XG et al (2003) Expression of multidrug resistance proteins P-glycoprotein, multidrug resistance protein 1, breast cancer resistance protein and lung resistance related protein in locally advanced bladder cancer treated with neoadjuvant chemotherapy: biological and clinical implications. J Urol 170:1383–1387 225. Uozaki H, Horiuchi H, Ishida T et  al (1997) Overexpression of resistance-related proteins (metallothioneins, glutathioneS-transferase pi, heat shock protein 27, and lung resistance-related protein) in osteosarcoma. Relationship with poor prognosis. Cancer 79:2336–2344 226. Gaumann A, Tews DS, Mentzel T et  al (2003) Expression of drug resistance related proteins in sarcomas of the pulmonary artery and poorly differentiated leiomyosarcomas of other origin. Virchows Arch 442: 529–537 227. Plaat BE, Hollema H, Molenaar WM et  al (2000) Soft tissue leiomyosarcomas and malignant gastrointestinal stromal tumors: differences in clinical outcome and expression of multidrug resistance proteins. J Clin Oncol 18:3211–3220 228. Zurita AJ, Diestra JE, Condom E et al (2003) Lung resistance-related protein as a predictor of clinical outcome in advanced testicular germ-cell tumours. Br J Cancer 88:879–886 229. Mandoky L, Geczi L, Doleschall Z et  al (2004) Expression and prognostic value of the lung resistance-related protein (LRP) in germ cell testicular tumors. Anticancer Res 24:1097–1104 230. Boonstra R, Timmer-Bosscha H, van EchtenArends J et  al (2004) Mitoxantrone resistance in a small cell lung cancer cell line is associated with ABCA2 upregulation. Br J Cancer 90:2411–2417 231. Vulevic B, Chen Z, Boyd JT et  al (2001) Cloning and characterization of human adenosine 5¢-triphosphate-binding cassette, sub-family A, transporter 2 (ABCA2). Cancer Res 61:3339–3347 232. Takara K, Sakaeda T, Okumura K (2006) An update on overcoming MDR1-mediated multidrug resistance in cancer chemotherapy. Curr Pharm Des 12:273–286

75

233. Calcagno AM, Kim IW, Wu CP, Shukla S, Ambudkar SV (2007) ABC drug transporters as molecular targets for the prevention of multidrug resistance and drug-drug interactions. Curr Drug Deliv 4:324–333 234. Smith AJ, van Helvoort A, van Meer G et  al (2000) MDR3 P-glycoprotein, a phosphatidylcholine translocase, transports several cytotoxic drugs and directly interacts with drugs as judged by interference with nucleotide trapping. J Biol Chem 275: 23530–23539 235. Huang Y, Anderle P, Bussey KJ et al (2004) Membrane transporters and channels: role of the transportome in cancer chemosensitivity and chemoresistance. Cancer Res 64: 4294–4301 236. Childs S, Yeh RL, Hui D, Ling V (1998) Taxol resistance mediated by transfection of the liver-specific sister gene of P-glycoprotein. Cancer Res 58:4160–4167 237. Deeley RG, Cole SP (2006) Substrate recognition and transport by multidrug resistance protein 1 (ABCC1). FEBS Lett 580:1103–1111 238. Guminski AD, Balleine RL, Chiew YE et al (2006) MRP2 (ABCC2) and cisplatin sensitivity in hepatocytes and human ovarian carcinoma. Gynecol Oncol 100:239–246 239. Huisman MT, Chhatta AA, van Tellingen O, Beijnen JH, Schinkel AH (2005) MRP2 (ABCC2) transports taxanes and confers paclitaxel resistance and both processes are stimulated by probenecid. Int J Cancer 116:824–829 240. Jedlitschky G, Hoffmann U, Kroemer HK (2006) Structure and function of the MRP2 (ABCC2) protein and its role in drug disposition. Expert Opin Drug Metab Toxicol 2:351–366 241. Materna V, Stege A, Surowiak P, Priebsch A, Lage H (2006) RNA interference-triggered reversal of ABCC2-dependent cisplatin resistance in human cancer cells. Biochem Biophys Res Commun 348:153–157 242. Zelcer N, Saeki T, Reid G, Beijnen JH, Borst P (2001) Characterization of drug transport by the human multidrug resistance protein 3 (ABCC3). J Biol Chem 276:46400–46407 243. Zeng H, Chen ZS, Belinsky MG, Rea PA, Kruh GD (2001) Transport of methotrexate (MTX) and folates by multidrug resistance protein (MRP) 3 and MRP1: effect of polyglutamylation on MTX transport. Cancer Res 61:7225–7232 244. Chen ZS, Lee K, Kruh GD (2001) Transport of cyclic nucleotides and estradiol 17beta-d-glucuronide by multidrug resistance protein 4. Resistance to 6-mercaptopu-

76

Gillet and Gottesman

rine and 6-thioguanine. J Biol Chem 276: 33747–33754 2 45. Ritter CA, Jedlitschky G, Meyer zu Schwabedissen H et al (2005) Cellular export of drugs and signaling molecules by the ATP-binding cassette transporters MRP4 (ABCC4) and MRP5 (ABCC5). Drug Metab Rev 37:253–278 246. Tian Q, Zhang J, Chan SY et  al (2006) Topotecan is a substrate for multidrug resistance associated protein 4. Curr Drug Metab 7:105–118 247. Tian Q, Zhang J, Tan TM et  al (2005) Human multidrug resistance associated protein 4 confers resistance to camptothecins. Pharm Res 22:1837–1853 248. Pratt S, Shepard RL, Kandasamy RA et al (2005) The multidrug resistance protein 5 (ABCC5) confers resistance to 5-fluorouracil and transports its monophosphorylated metabolites. Mol Cancer Ther 4:855–863 249. Belinsky MG, Chen ZS, Shchaveleva I, Zeng H, Kruh GD (2002) Characterization of the drug resistance and transport properties

of multidrug resistance protein 6 (MRP6, ABCC6). Cancer Res 62:6172–6177 250. Naramoto H, Uematsu T, Uchihashi T et al (2007) Multidrug resistance-associated protein 7 expression is involved in cross-resistance to docetaxel in salivary gland adenocarcinoma cell lines. Int J Oncol 30:393–401 251. Hopper-Borge E, Chen ZS, Shchaveleva I, Belinsky MG, Kruh GD (2004) Analysis of the drug resistance profile of multidrug resistance protein 7 (ABCC10): resistance to docetaxel. Cancer Res 64:4927–4930 252. Chen ZS, Guo Y, Belinsky MG, Kotova E, Kruh GD (2005) Transport of bile acids, sulfated steroids, estradiol 17-beta-dglucuronide, and leukotriene C4 by human multidrug resistance protein 8 (ABCC11). Mol Pharmacol 67:545–557 253. Guo Y, Kotova E, Chen ZS et al (2003) MRP8, ATP-binding cassette C11 (ABCC11), is a cyclic nucleotide efflux pump and a resistance factor for fluoropyrimidines 2¢, 3¢-dideoxycytidine and 9¢-(2¢-phosphonylmethoxyethyl) adenine. J Biol Chem 278:29509–29514

Chapter 5 Molecular Mechanisms of Drug Resistance in Single-Step and Multi-Step Drug-Selected Cancer Cells Anna Maria Calcagno and Suresh V. Ambudkar Abstract Multidrug resistance (MDR) remains one of the key determinants in chemotherapeutic success of cancer patients. Often, acquired resistance is mediated by the overexpression of ATP-binding cassette (ABC) drug transporters. To study the mechanisms involved in the MDR phenotype, investigators have generated a variety of in vitro cell culture models using both multi-step and single-step drug selections. Sublines produced from multi-step selections have led to the discovery of several crucial drug transporters including ABCB1, ABCC1, and ABCG2. Additionally, a number of mechanisms causing gene overexpression have been elucidated. To more closely mimic in  vivo conditions, investigators have also established MDR sublines with single-step drug selections. Here, we examine some of the multi-step and single-step selected cell lines generated to elucidate the mechanisms involved in the development of MDR in cancer cells. Key words: Multidrug resistance, Multi-step selection, Single-step selection, ABC transporter, Gene amplification, Epigenetic changes

1. Introduction Cancer is one of the top ten leading causes of death in the world. In the USA alone, one of every four deaths will be due to cancer in 2008 (1). Advancements in early detection and cancer treatments have yielded significant progress, yet cancer deaths still outnumber deaths due to heart disease in people less than 85 years of age in the USA. A major factor in therapeutic failure for cancer involves the development of drug resistance to a variety of structurally unrelated anticancer drugs, also known as multidrug resistance (MDR) (2). MDR can develop in several different ways, with the predominant mechanism being the overexpression of ATP-binding cassette (ABC) drug transporters on the plasma membrane of tumor J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_5, © Humana Press, a part of Springer Science + Business Media, LLC 2010

77

78

Calcagno and Ambudkar

cells. These transporters act as energy-driven pumps (3), and as such, maintain intracellular drug concentrations below toxic levels. Thus, the tumor survives, rendering the treatment ineffective. Tumors can demonstrate intrinsic drug resistance in which the tumor is innately resistant to treatment; this occurs in tumors originating from epithelial cells such as renal or adrenal tumors, which naturally express ABC drug transporters (4). Acquired resistance, on the other hand, arises following therapy, and tumors normally present with the MDR phenotype subsequent to various genetic changes. ABC drug transporters belong to the largest superfamily of transporter proteins (5). Members of this family are recognized by a consensus ATP-binding site from 90 to 110 amino acids in length, which also includes a linker region between two Walker motifs. In addition to two ATP-binding sites, ABC transporters normally possess two transmembrane domains. The 48 human ABC transporter genes are further subdivided into seven subfamilies (A–G) based on similar gene structure, order of the domains, and on sequence homology in their consensus domains (6). ABC drug transporters such as ABCB1, ABCC1, and ABCG2 are expressed in normal and tumor cells and are localized to different plasma membrane surfaces; the normal function of a number of these transporters is to efflux endogenous and xenobiotic metabolites from the cell (7). The substrate specificity for ABCB1, ABCG2, and the various ABCC family members overlaps extensively although the primary sequences of these transporters vary significantly (7). This phenomenon makes treatment of multidrug-resistant cancer unsuccessful in spite of the multitude of drugs available. Reports show that patients with ABCB1-positive tumors are three times more likely to fail a course of therapy than those who have tumors that are ABCB1negative (8). Even more taxing for patients are tumors that express multiple ABC transporters, since overexpression is not mutually exclusive and a tumor can overexpress several MDRlinked ABC transporters in tandem. Although over 12 of these transporters have been linked to MDR (9), little is known about the regulation of these transporters. Often multi-step drug selections have been employed to study the MDR phenotype. Several drawbacks are associated with this technique. The multi-step selections utilize higher concentrations of drug than those found in patients as well as extended periods of exposure. These factors produce pleiotropic effects. To avoid such issues, we recently employed a single-step selection to evaluate ABC transporter regulation. We reported that ABC transporter mRNA expression patterns vary with single vs. multi-step selection with doxorubicin in MCF-7 breast cancer cells (10). In multi-step selections with doxorubicin, ABCB1 is often the dominant ABC transporter causing MDR; we have shown that follow-

Molecular Mechanisms of Drug Resistance in Single-Step and Multi-Step

79

ing a single-step selection using low concentrations of doxorubicin other transporters including ABCC2, ABCC4, and ABCG2 are overexpressed. In this chapter, we will review a number of the multi-step and single-step selected cancer cell lines that have been established and used extensively to investigate MDR (Table 5.1). In addition, we will discuss the mechanisms that have been ascertained in the development of these drug-resistant cell lines.

Table 5.1 List of select multi-step and single-step selected cell lines overexpressing ABC transporters Selection regimen

Cell line

Drug

ABC transporter overexpressed

Multi-step

KB-3-1

Colchicine Doxorubicin Vinblastine

ABCB1 (13) ABCB1 (13) ABCB1 (13)

Multi-step Multi-step Multi-step Multi-step Multi-step Multi-step Multi-step Multi-step Multi-step Multi-step Multi-step Multi-step Multi-step Multi-step Multi-step Multi-step Multi-step Multi-step Multi-step Multi-step Multi-step

OVCAR-8 MCF-7 MES-SA MCF-7 MDA-MB-231 MCF-7 MCF-7 NCI-H69 COR-L23 MOR GLC4 MCF-7 NCI-H69 MCF-7 MCF-7 MCF-7 S1 IGROV-1 IGROV-1 SF295 GLC4

Doxorubicin Doxorubicin Doxorubicin Docetaxel Docetaxel Doxorubicin Paclitaxel Doxorubicin Doxorubicin Doxorubicin Doxorubicin Etoposide Etoposide Flavopiridol Doxorubicin and verapamil Mitoxantrone Mitoxantrone Topotecan Mitoxantrone Mitoxantrone Mitoxantrone

ABCB1 (14, 15) ABCB1 (17) ABCB1 (19) ABCB1 (20) ABCB1 (20) ABCB1 and ABCG2 (21, 22) ABCB1 and ABCG2 (21, 22) ABCC1 (23) ABCC1 (24, 25) ABCC1 (24, 25) ABCC1 (26) ABCC1 (27) ABCB1 and ABCC1 (28) ABCG2 (29) ABCG2 (30) ABCG2 (31) ABCG2 (32) ABCG2 (33) ABCG2 (33) ABCG2 (34) ABCA2 (35)

Single-step Single-step Single-step Single-step Single-step Single-step Single-step Single-step

MCF-7 MCF-7 IGROV-1 S1 MES-SA MES-SA NCI-H82 GLC4

Doxorubicin Etoposide Doxorubicin Doxorubicin Doxorubicin Paclitaxel Epirubicin Doxorubicin

ABCC4 and ABCG2a (10) ABCG2 (10) ABCG2 (10) ABCG2 (10) ABCB1 (57) ABCB1 (59) ABCC1 (62) ABCC1 (63)

In all cases the overexpression of an ABC transporter was demonstrated at the functional level and the references are given in the parenthesis a ABCG2 is the ABC transporter responsible for MDR and ABCC4 does not confer resistance to doxorubicin

80

Calcagno and Ambudkar

2. Multi-Step Selected Cell Lines To study MDR in vitro, investigators have utilized drug selections to generate resistant cell lines for over 30 years. Selections can be performed on individual clones or on mass populations of cells (11). To establish individual clones, the cells must be cloned so that an individual cell is the source for the entire population, which will then be selected with the drug of choice. This technique boosts one major advantage in that a single gene will be responsible for the MDR. Alternatively, multifactorial MDR will result when an entire cell population is selected (11). The selection of a cell population, on the other hand, more closely mimics the clinical situation and can stem from a spectrum of mechanisms. One of the original multi-step selected cell lines was established by the selection of an individual clone of KB epidermoid carcinoma cells with colchicine (12). The subsequent resistant sublines were generated with increasing single-step selections beginning with KB-3-1, an individual clone from a population of the KB cells. The MDR1 (ABCB1) gene was first identified from these cells. Using this same methodology, four additional KB sublines were created with selections in high concentrations of colchicine, vinblastine, or doxorubicin (13). These sublines were selected with a stepwise selection, and all express high levels of ABCB1. Since their establishment, the various resistant sublines of KB cells have been widely used to investigate MDR mediated by ABCB1. In contrast to the KB cells, MCF-7 breast cancer cells were selected with doxorubicin using the mass population method, yet also expressed ABCB1 (14). This original selection was performed with increasing concentrations of doxorubicin beginning with 10 nM. The final resistant subline, MCF-7 AdrR, was capable of surviving in 10 µM doxorubicin. Later these original doxorubicin-resistant MCF-7 cells were determined to actually be OVCAR-8 ovarian cancer cells, which were resistant to doxorubicin (15). Their nomenclature has changed accordingly to NCI/ ADR-Res or OVCAR-8/ADR (16). Other laboratories independently generated doxorubicin-resistant MCF-7 cells, and one such subline was established by culturing MCF-7 cells in 0.025 µg/ ml doxorubicin and increasing the selection pressure by twofold until the cells grew in the presence of 2 µg/ml doxorubicin (17). Interestingly, these resistant cells also overexpressed ABCB1 and were karyotyped to match MCF-7 cells from ATCC (18). Doxorubicin-resistant sarcoma cells (MES-SA/Dx5) were also one of the early MDR models expressing ABCB1; these cells were selected with increasing concentrations of doxorubicin (19). Lastly, MCF-7 and MDA-MB-231 breast cancer cells exposed to increasing concentrations of docetaxel (up to 30 µM), known as

Molecular Mechanisms of Drug Resistance in Single-Step and Multi-Step

81

MCF-7 TAX30 and MDA-MB-231 TAX30, were also found to overexpress ABCB1 (20). Moreover, ABCB1 was involved in MDR in highly resistant cell lines derived from both the individual clone and population selection techniques. Yet in other more recent studies with MCF-7 cells using multi-step selections with lower concentrations of doxorubicin, both ABCG2 and ABCB1 were expressed (21, 22). In these studies concentrations ranging from 9 to 300 nM doxorubicin were employed. Similarly, MCF-7 cells selected with a multi-step selection with paclitaxel, 0.56 nM to a final concentration of 6.6 nM, also express both transporters, but at a lower level than the doxorubicin-selected MCF-7 cells (21, 22). Remarkably, paclitaxel is not a substrate of ABCG2, yet its selection pressure caused the overexpression of ABCG2. Furthermore, the investigators demonstrated that the clones were isogenic and that MDR was a consequence of adaptation and not a selection of a clone within the population. Involvement of other ABC transporters at lower selection concentrations suggests that ABCB1 may be dominant only at the later stages of MDR in particular cell types. ABCC1 was first reported in a resistant cell line produced with a stepwise doxorubicin selection in a small cell lung cancer cell line, NCI-H69; this was called H69AR and did not express ABCB1 (23). Large-cell (COR-L23), small-cell (H69), and adenocarcinoma (MOR) lung tumor lines continuously selected with increasing concentrations of doxorubicin were also found to express ABCC1 (24, 25). Another small cell lung carcinoma cell line, GLC4, was utilized in resistance studies, and ABCC1 was overexpressed when selected with doxorubicin concentrations augmented from 18 to 1,152 nM (26). Surprisingly, etoposideselected MCF-7 cells also showed overexpression of ABCC1 instead of ABCB1. These cells were generated with a stepwise selection in increasing concentrations of etoposide starting with 200 nM up to 10 µM and revertants were prevented by occasional reselection in 4 µM etoposide (27). Investigators reported that etoposide-selected H69 small cell lung cancer cells expressed both ABCB1 and ABCC1 at the mRNA and protein levels (28). During this selection process, ABCC1 expression preceded that of ABCB1. At a moderate level of ABCC1 expression, rather than continue increasing the expression of ABCC1, the cells turned on the ABCB1 gene (28). It also appears that cell type dictates the particular ABC transporter that is induced and that cells can activate more than one ABC transporter. Investigators have also prepared a variety of resistant cell lines overexpressing ABCG2. For instance, MCF-7 cells selected with increasing concentrations of flavopiridol, MCF-7/FLV1000, expressed wild-type ABCG2 (29). MCF-7 cells selected in the presence of both doxorubicin and verapamil also overexpressed ABCG2; these cells are known as MCF-7 AdVp3000 (30). Unlike

82

Calcagno and Ambudkar

the MCF-7/FLV1000, the MCF-7 AdVp3000 expressed the mutant ABCG2 R482T. MCF-7 cells were also exposed to mitoxantrone and were found to overexpress wild-type ABCG2 (31). When S1 human colon carcinoma cells were selected with mitoxantrone, ABCG2 was also overexpressed (29). Additional sublines were generated when these original S1-M1-3.2 (32) were exposed to higher concentrations of drug, up to a final concentration of 80 µM. These S1 sublines also expressed a mutant R482G ABCG2 protein. Another example of MDR mediated by ABCG2 overexpression occurred in IGROV-1 human ovarian carcinoma cells selected with either topotecan or mitoxantrone (33). Similarly, SF295 human glioblastoma cells showed ABCG2 overexpression when selected in increasing concentrations of mitoxantrone (50–500  nM) (34). Unexpectedly, when a mitoxantrone-resistant small cell lung cancer cell line, GLC4-MITO, was established, ABCA2 upregulation was found (35). In the case of ABCG2overexpressing cell lines in vitro, two gain of function mutations have been identified (R482T and R482G). To our knowledge, no such mutations have been reported in clinical samples positive for ABCG2 to date.

3. Mechanisms of Overexpression in Multi-Step Selected Cell Lines

Gene amplification is the most common method for drug-resistant cells to overexpress a particular ABC transporter. In a comprehensive analysis of 23 drug-resistant cancer cell lines derived from ten different human cancers, it was revealed that changes in gene copy number were implicated in acquired drug resistance (36). Comparative genomic hybridization (CGH) was executed on drug-sensitive and their corresponding drug-resistant sublines, and the regions of gain within the drug-resistant cell lines were consistent with regions encoding ABC transporters in 19 of the 23 cell lines. These changes were further confirmed by fluorescence in situ hybridization (FISH) analysis in these cells. Of particular interest were ABCA3, ABCB1, and ABCC9, which had a greater than twofold increase in copy numbers. Furthermore, gene amplification was in line with gene expression changes present in these resistant cells. Amplified genes are either present in homogeneously staining regions or on extrachromosomal elements such as double-minutes. Investigations of resistant KB cells also showed gene amplification of the ABCB1 gene (37, 38). Double-minute chromosomes were identified in these KB-resistant cell lines. Investigators also determined that KB cells could easily lose their resistance when no selection pressure was present because the gene amplification was only found in the form of double-minutes. Furthermore, in

Molecular Mechanisms of Drug Resistance in Single-Step and Multi-Step

83

this KB resistant series, it was uncovered that in the less-resistant sublines the ABCB1 was activated while in the more highly resistant sublines gene amplification occurred (39). For instance, later studies showed that with the higher colchicine selection pressure, double-minutes were stably maintained even after several months of continuous passaging in culture. The exact formation of the double-minutes in the sequential series of resistant sublines was closely examined using gel electrophoresis. Submicroscopic extrachromosomal circular DNA was revealed in the less-resistant sublines. Consequently, the double-minutes uncovered in the more-resistant sublines were formed by multimerization of these submicroscopic circular DNAs (39). Gene amplification has also been reported for some of the other multi-step selected cell lines previously described. MES-SA/ Dx5, doxorubicin-resistant sarcoma cells, displayed 7q21 alterations and gene amplification (40). However, these cells, unlike the KB-resistant cells, did not possess double-minutes as seen by FISH analysis. In the two breast cancer cell lines selected with docetaxel, MCF-7 TAX30 and MDA-MB-231 TAX30, gene amplification of chromosome 7q in the region that encodes for ABCB1 was discovered using CGH (20). ABCB1 overexpression can also be caused at the transcription level. Investigators have found that in drug-sensitive cells only one transcription start site is used for ABCB1; nonetheless, drug-resistant cells that do not exhibit gene amplification often exploited more than one downstream transcription start site for ABCB1 (41). This substitution of other downstream transcription start sites for ABCB1 within the same cell line was a distinct mechanism that led to the identification of the MED-1 (multiple start site element downstream) in many of the genes with a TATA-less promoter, which have multiple start sites such as ABCB1 (42, 43). This MED-1, GCTCCC/G, was crucial for ABCB1 expression in drug-resistant cells in the cell lines examined. Likewise MEF1, MDR1 promoter-enhancing factor 1, also activated transcription but through an upstream promoter element, −118 to −111 (44). Often drug selections can also cause alterations in genes that appear as gene rearrangements through nonhomologous recombinations. For example, investigators first reported a hybrid ABCB1 mRNA resulting from such a chromosomal rearrangement in a doxorubicin-selected S48-3s human colon adenocarcinoma cell line (45). In these cells, there was a 4;7 translocation resulting with the 3¢ end of ABCB1 adjacent to a transcriptionally active chromosome 4 gene, thus triggering the activation of ABCB1 by the promoter sequences on the adjacent chromosome 4. For this particular gene rearrangement, the subsequent ABCB1 protein structure remained unaltered due to the rearrangement occurring within the 5¢ region of ABCB1. Follow-up studies illustrated that eight other selected cells and two clinical samples had

84

Calcagno and Ambudkar

gene rearrangements (45, 46). Unexpectedly, these gene hybrids differed, suggesting that ABCB1 merely required a sufficiently active promoter to activate it and that the specific promoter was irrelevant. Other ABCB1 mRNA hybrids have also been reported (47). These hybrids were regulated by nearby genomic sequences with similarity to a retroviral element. Nevertheless, no chromosomal rearrangements were discovered in these hybrids. Epigenetic changes have also been reported to activate ABCB1. In the repressed state, the ABCB1 promoter is methylated and enriched in methyl-CpG binding protein 2 (MeCP2). In MCF-7 cells exposed to a stepwise selection with doxorubicin, ABCB1 is overexpressed and in the resistant cells the ABCB1 promoter is completely unmethylated (48). The promoter methylation status of ABCB1 is inversely correlated to the expression of the ABCB1 gene. The loss of methylation at the promoter facilitates the activation of ABCB1 in these resistant cells. In the resistant cell lines that displayed ABCC1 overexpression, gene amplification was also the most common mechanism. The original ABCC1-overexpressing cell line H69AR demonstrated gene amplification with Southern blot analysis (23). Many doubleminutes of chromosome 16 were discovered in the COR-L23/R cells, while the MOR cells exhibited an enlarged copy of chromosome 16 with homogeneously staining regions (24). As with the cells overexpressing ABCB1, highly resistant cell lines such as GLC4/ADR (26), COR-L23/R, and MOR/R predominantly displayed gene amplification of ABCC1 as the mode of gene overexpression. On the contrary, transcriptional activation of ABCC1 was solely responsible for gene overexpression in the weakly resistant SW-1573 30.3  M subline, which had been selected with low concentrations of doxorubicin (49). Of the highly resistant cell lines, only MOR/R presented a combination of gene amplification and gene activation, whereas gene amplification was the main mechanism for gene overexpression in the GLC4/ADR and COR-L23/R selected cell lines. For resistant cell lines overexpressing ABCG2, Southern analysis of MCF-7 AdVp3000 and S1-M1-80 sublines uncovered that only MCF-7 AdVp3000 had gene amplification (50). In later studies, these cell lines as well as MCF-7/MX were further characterized by CGH, FISH, spectral karyotyping, and Southern blotting (51). No amplification was confirmed in the S1-M1-80 subline, while the other two cell lines showed amplification with multiple translocations of chromosome 4. Other investigators evaluated the MCF-7/MX subline and also showed gene amplification by Southern blot analysis (52). In the SF295 MX selected cells, ABCG2 was found amplified by Southern analysis. The sublines selected with the lower concentrations displayed double-minutes containing the ABCG2 gene when examined with FISH and spectral karyotyping. At 500  nM mitoxantrone selection pressure, a

Molecular Mechanisms of Drug Resistance in Single-Step and Multi-Step

85

homogeneously staining region was found integrated into the chromosome causing ABCG2 overexpression (34). Furthermore, Boonstra et  al. also found gene amplification with CGH in the GLC4-MITO cells for ABCA2 on chromosome 9 (35). Analogous to results from ABCB1 promoter studies in drugresistant cells overexpressing ABCB1, ABCG2 has also been shown to have multiple transcription start sites in drug-selected cells (53). Investigators have also reported the expression of novel 5¢UTR variants of transcripts that possess different translation efficiencies. Thus, the ABCG2 protein expression is directed at the posttranscriptional level as a consequence of these 5¢UTR variants. However, no gene rearrangements in the 5¢UTR region were seen. Various epigenetic changes have also been found in multistep selected cell lines that overexpress ABCG2. Chromatin immunoprecipitation (ChIP) has been used to identify histone modifications in these multi-step selected cells. In the S1-M1-80 cells, which show no gene amplification, the ABCG2 proximal promoter displayed histone H3 acetylation (54). Further epigenetic changes were present in this subline as HDAC1 and HDAC3 bound less to the proximal promoter of ABCG2. More importantly, Pol II binding, an indicator of ABCG2 promoter activity, was enhanced in these resistant cells. MicroRNAs (miRNA) have also been implicated in the regulation of genes. Recent reports investigated the effects of miRNA on ABCG2 expression. S1 colon cells possessed a longer 3¢UTR in the ABCG2 mRNA where a putative hsa-miR-519c binding site exists (55). This miRNA binds to this site, causing translational repression and mRNA degradation in the sensitive parental cells. Conversely, the S1-M1-80 cells utilize a noncanonical AUUAAA poly (A) signal to yield a much shorter 3¢UTR lacking this miRNA binding site, thus eluding degradation mediated by hsa-miR-519c. It appears that a combination of epigenetic and miRNA-mediated changes are responsible for the overexpression of ABCG2 in the highly resistant S1-M1-80 cells.

4. Single-Step Selected Cell Lines These and many other multidrug resistant cancer cell lines have been established in  vitro through continual drug selection. Although they offer a sufficient means for investigating the regulation of ABC transporter expression and function, rarely do these continual drug selections emulate what is found in the clinical setting. Thus, insight into the development of MDR by ABC transporters at clinically relevant concentrations and ascertaining which factors induce upregulation of ABC transporters during the initial steps of MDR will afford more advantageous measures

86

Calcagno and Ambudkar

to circumvent MDR. Our goal was to evaluate the expression of ABC transporters following a short low dose selection, which would simulate the conditions encountered in vivo and to compare the gene expression levels of MDR-linked ABC transporters in these sublines selected by a low-dose single step to an established high-dose doxorubicin selected subline (17) and to determine if overexpression of the same ABC transporters occurs. We have recently found that ABC transporter mRNA expression patterns vary with single- vs. multi-step treatment with doxorubicin in MCF-7 breast cancer cells. We established single-step doxorubicin-selected MCF-7 sublines using very low concentrations (14 or 21 nM) (10). Individual clones were selected from a population of 10,000 cells in a 100× 20 mm tissue culture dish exposed to drug for 10  days. Clones were then maintained in drug-free medium following the initial drug selection. We compared these single-step sublines to a previously established multistep doxorubicin-selected MCF-7 subline (17) known to overexpress only ABCB1 at the mRNA and protein levels (Fig.  5.1) due to gene amplification. We evaluated a number of ABC transporters and found that ABCC2, ABCC4, and ABCG2 were overexpressed at the mRNA level in these single-step selected sublines (Fig. 5.2). Yet, only ABCC4 and ABCG2 were overexpressed at the protein level. Both 14 and 21 nM single-step doxorubicin-selected sublines exhibited nearly fivefold resistance to doxorubicin compared with parental MCF-7 cells. However, ABCC4 did not confer resistance to this drug, suggesting that ABCG2 was the major transporter responsible for the development of doxorubicin resistance. Sequencing of ABCG2 in the single-step selected sublines revealed that our in vitro selection resulted in the overexpression of the wild-type ABCG2 and not the gain of function mutations either G or T at amino acid 482. SiRNA studies further confirmed that mainly ABCG2 confered drug resistance in these clones. We also observed that the upregulation of ABCG2 was facilitated by histone hyperacetylation of H3 at the proximal promoter of ABCG2. Similar to what was found in the S1-M1-80 cells, Pol II binding was increased while HDAC1 binding was decreased in the single-step selected sublines. This was the first report of ABCG2 overexpression in MCF-7 cells following a short-term low-dose selection with doxorubicin. To further evaluate if this ABCG2 overexpression was drug and cell line independent, we generated additional sublines of MCF-7 cells with a single-step selection using 300 nM etoposide and two different cancer cell lines, IGROV-1 ovarian cancer cells and S-1 colon tumor cells, with 14 and 21  nM doxorubicin, respectively. To ensure that we were not selecting for a resistant clone, several lower etoposide concentrations, 50, 100, and 200  nM, were first evaluated. These lower selections indicated that all MCF-7 parental cells were able to survive. For IGROV-1

Molecular Mechanisms of Drug Resistance in Single-Step and Multi-Step

87

Fig. 5.1. Analysis of ABC transporter expression and function in the multi-step doxorubicin-selected MCF7/ADR cell line. (a) The delta-delta CT method was used to determine the fold change of ABC transporter gene expression in the multistep doxorubicin-selected cells, MCF7/ADR (17), compared to their parental line, MCF-7. The values represent the mean and standard deviation (n = 2). The overexpression of ABCB1 is highlighted by the black circle. (b) Using C219, the ABCB1-specific monoclonal antibody, the relative quantities of ABCB1 were determined for MCF7/ADR and MCF-7 in whole-cell lysates. Lane 1, MCF-7 control and lane 2, MCF7/ADR, (100,000 cells for all samples). (c) Calcein-AM efflux assays were performed using flow cytometry. Assays compared MCF-7 and MCF7/ADR. Charcoal gray histogram, MCF7/ADR; dark gray histogram, MCF7/ADR cells in the presence of 10 µM cyclosporin A (CSA); black histogram, MCF-7, and light gray histogram, MCF-7 in the presence of cyclosporine A. The schematic on the far left side depicts the multi-step selection with doxorubicin in 0.025 µg/ml doxorubicin and increasing the selection pressure by twofold until the cells grew in the presence of 2 µg/ml doxorubicin.

cells, only the 14 nM doxorubicin selection yielded resistant cells. Five sublines derived from IGROV-1 cells obtained using 14 nM doxorubicin and S-1-resistant clones obtained at a 21 nM doxorubicin concentration were further characterized. Furthermore, we were also able to replicate the upregulation of the same ABC

88

Calcagno and Ambudkar

Fig. 5.2. Single-step doxorubicin-selected clones overexpress ABCG2. (a) The schematic of the single-step selection for the MCF-7 cells with doxorubicin. (b) Characterization of selected ABC transporter gene expression levels in several single-step clones. Doxorubicin-resistant MCF-7 clones were established employing a single-step selection with either 14 or 21 nM treatment for 10 days followed by culturing continuously in drug-free medium. The average fold change compared to parental MCF-7 cells ± SD (n = 4) was calculated using delta-delta Ct method from real-time RT-PCR data. Reference gene is PMCA4 (64). The key for selected five ABC transporters is given in figure. (c) Western blotting analysis of ABCG2 protein using BXP-21 antibody following no treatment (lane 1), 50 nM negative siRNA treatment (lane 2), and 50 nM G2-2 siRNA treatment (lane 3). (d) Cytotoxicity assays with mitoxantrone evaluating the effect of silencing ABCG2 in the 21  nM single-step clone. Dose–response curves were derived from three independent experiments using the CCK-8 assay. White box, 21 nM cells with 12.5 nM G2-2 siRNA and black triangle, 21 nM cells with 12.5 nM negative siRNA. Error bars indicate standard deviation (n= 3).

transporters in the MCF-7 cells using this single-step selection with 300 nM etoposide. ABCG2 was also the dominant overexpressed ABC transporter for these additional sublines. This suggests

Molecular Mechanisms of Drug Resistance in Single-Step and Multi-Step

89

that even a low-dose selection can bring about MDR and that ABCG2 overexpression mediates the early stages of MDR development in certain cell lines. ABCG2 may be protecting against the cytotoxic effects of drugs in our single-step-selected clones, as it does in stem cells (56). Analysis of other mammary stem cell markers in our single-step sublines demonstrated that we did not enrich for cancer stem cells during the single-step selections of these clones. Taken together with the epigenetic alterations that were discovered in these resistant sublines, adaptation as opposed to selection appears to be the mitigating factor for this selection process. Single-step selections have also been performed with the MES-SA, human sarcoma cell line. This protocol used the mass population selection technique, where cells were first plated in a 96-well plate, treated for 2 weeks with 40 nM doxorubicin, grown drug free for two additional weeks, and then individually harvested from each well (57). As with the MES-SA/Dx5, all clones examined expressed ABCB1. The authors used fluctuation analysis to determine that the doxorubicin-resistant clones were derived due to spontaneous mutations. Additionally, no chromosomal alteration or gene amplification was discovered in these singlestep mutants (40). When either etoposide or paclitaxel was used in single-step selections with MES-SA cells, authors found that either no ABCB1 overexpression occurred (58) or that only 44% of the clones expressed ABCB1 (59), respectively. Etoposideselected MES-SA cells showed a reduction in topoisomerase II but no ABC transporter increases. This suggests that ABCB1 substrates have different effects when selecting for ABCB1expressing clones. Furthermore, a single-step selection with 40  nM doxorubicin in the presence of an ABCB1 inhibitor, PSC833, also produced no detectable levels of ABCB1 but rather decreased levels of topoisomerase II a (60). In recent follow-up studies, the authors found that an increase in acetylated H3 modified the chromatin structure of ABCB1 far upstream, 968-bp proximal to the upstream promoter, and initiated upstream transcripts for these single-step selections (61). Equally important, the authors confirmed that these upstream ABCB1 transcripts were spontaneous in nature given that a clonal variant expanded to several million cells without any drug selection also produced these ABCB1 upstream variants. Other single-step selections have generated sublines that overexpress ABCC1. One such example was the H82, a variant of small cell lung cancer, which was selected for 18 h with 69 nM epirubicin. This initial selection yielded a drug-resistant cell line, which was then subsequently selected for 18 h with 14 nM epirubicin, producing an even more resistant line known as H82/E8 (62). These sublines displayed two- to ninefold resistance. Remarkably, the H82/E8 subline remained stably resistant for

90

Calcagno and Ambudkar

over 2  years without further drug treatment. Investigators also selected H69 cells with eight treatments of 14  nM epirubicin followed by maintenance in drug-free medium. This subline is referred to as H69/E8. Only the H82/E8 increased ABCC1 expression (62) while neither cell line expressed ABCB1. Other investigators using a 50  nM single-step doxorubicin-selection with GLC4 small cell lung cancer cells attributed an increase in ABCC1 expression to the activation of the JNK pathway (63).

5. Conclusion Drug selections with both clonal and cell populations have aided in the study of MDR mediated by ABC transporters. For instance, these in vitro techniques have led to the identification of at least three of the most influential ABC drug transporters for MDR. The overexpression of a particular ABC transporter during drug selection appears to depend on a multitude of factors, which include but are not limited to the cell type, the selection regimen, the drug used for selection pressure, as well as the concentrations utilized. These factors suggest that a number of ABC transporters should be evaluated following drug selection in addition to ABCB1. The single-step selection is capable of generating sublines with the MDR phenotype at clinically relevant concentrations while eliminating pleiotropic effects due to long-term drug exposure. With advancements in techniques for analysis at the molecular level and better understanding of gene regulation in the presence of drug, future studies should focus on translational research to improve the success rate of cancer therapies.

Acknowledgments We thank Mr. George Leiman for editorial assistance. This research was supported by the Intramural Research Program of the National Institutes of Health, National Cancer Institute, Center for Cancer Research. References 1. Jemal A, Siegel R, Ward E et al (2008) Cancer statistics. CA Cancer J Clin 58:71–96 2. Gottesman MM (2002) Mechanisms of cancer drug resistance. Annu Rev Med 53:615–627

3. Dean M, Annilo T (2005) Evolution of the ATP-Binding cassette (ABC) transporter superfamily in vertebrates. Annu Rev Genomics Hum Genet 6:123–142

Molecular Mechanisms of Drug Resistance in Single-Step and Multi-Step 4. Dean M, Fojo T, Bates S (2005) Tumour stem cells and drug resistance. Nat Rev Cancer 5:275–284 5. Gottesman M, Ambudkar S (2001) Overview: ABC transporters and human disease. J Bioenerg Biomembr 33:453–458 6. Dean M, Rzhetsky A, Allikmets R (2001) The human ATP-Binding Cassette (ABC) transporter superfamily. Genome Res 11:1156–1166 7. Haimeur A, Conseil G, Deeley R, Cole S (2004) The MRP-related and BCRP/ABCG2 multidrug resistance proteins: biology, substrate specificity and regulation. Curr Drug Metab 5:21–53 8. Tan B, Piwnica-Worms D, Ratner L (2000) Multidrug resistance transporters and modulation. Curr Opin Oncol 12:450–458 9. Szakacs G, Paterson JK, Ludwig JA, BoothGenthe C, Gottesman MM (2006) Targeting multidrug resistance in cancer. Nat Rev Drug Discov 5:219–234 10. Calcagno AM, Fostel JM, To KK et al (2008) Single-step doxorubicin-selected cancer cells overexpress the ABCG2 drug transporter through epigenetic changes. Br J Cancer 98:1515–1524 11. Gottesman MM, Cardarelli C, Goldenberg S, Licht T, Pastan I (1998) Selection and maintenance of multidrug-resistant cells. Methods Enzymol 292:248–258 12. Akiyama S, Fojo A, Hanover JA, Pastan I, Gottesman MM (1985) Isolation and genetic characterization of human KB cell lines resistant to multiple drugs. Somat Cell Mol Genet 11:117–126 13. Shen D, Cardarelli C, Hwang J et al (1986) Multiple drug-resistant human KB carcinoma cells independently selected for high-level resistance to colchicine, adriamycin, or vinblastine show changes in expression of specific proteins. J Biol Chem 261:7762–7770 14. Batist G, Tulpule A, Sinha B et  al (1986) Overexpression of a novel anionic glutathione transferase in multidrug- resistant human breast cancer cells. J Biol Chem 261:15544–15549 15. Scudiero DA, Monks A, Sausville EA (1998) Cell line designation change: multidrug-resistant cell line in the NCI anticancer screen. J Natl Cancer Inst 90:862 16. Liscovitch M, Ravid D (2007) A case study in misidentification of cancer cell lines: MCF-7/ AdrR cells (re-designated NCI/ADR-RES) are derived from OVCAR-8 human ovarian carcinoma cells. Cancer Lett 245:350–352 17. Mehta K (1994) High levels of transglutaminase expression in doxorubicin-resistant human breast carcinoma cells. Int J Cancer 58:400–406

91

18. Devarajan E, Chen J, Multani AS et al (2002) Human breast cancer MCF-7 cell line contains inherently drug-resistant subclones with distinct genotypic and phenotypic features. Int J Oncol 20:913–920 19. Harker WG, Sikic BI (1985) Multidrug (pleiotropic) resistance in doxorubicinselected variants of the human sarcoma cell line MES-SA. Cancer Res 45:4091–4096 20. McDonald SL, Stevenson DA, Moir SE et al (2005) Genomic changes identified by comparative genomic hybridisation in docetaxelresistant breast cancer cell lines. Eur J Cancer 41:1086–1094 21. Guo B, Villeneuve DJ, Hembruff SL et  al (2004) Cross-resistance studies of isogenic drug-resistant breast tumor cell lines support recent clinical evidence suggesting that sensitivity to paclitaxel may be strongly compromised by prior doxorubicin exposure. Breast Cancer Res Treat 85:31–51 22. Villeneuve DJ, Hembruff SL, Veitch Z et  al (2006) cDNA microarray analysis of isogenic paclitaxel- and doxorubicin-resistant breast tumor cell lines reveals distinct drug-specific genetic signatures of resistance. Breast Cancer Res Treat 96:17–39 23. Cole SP, Bhardwaj G, Gerlach JH et al (1992) Overexpression of a transporter gene in a multidrug-resistant human lung cancer cell line. Science 258:1650–1654 24. Barrand MA, Heppell-Parton AC, Wright KA, Rabbitts PH, Twentyman PR (1994) A 190-kilodalton protein overexpressed in nonP-glycoprotein-containing multidrug-resistant cells and its relationship to the MRP gene. J Natl Cancer Inst 86:110–117 25. Twentyman PR, Fox NE, Wright KA, Bleehen NM (1986) Derivation and preliminary characterisation of adriamycin resistant lines of human lung cancer cells. Br J Cancer 53:529–537 26. Zijlstra JG, de Vries EGE, Mulder NH (1987) Multifactorial drug resistance in an adriamycin-resistant human small cell lung carcinoma cell line. Cancer Res 47:1780–1784 27. Schneider E, Horton JK, Yang CH, Nakagawa M, Cowan KH (1994) Multidrug resistanceassociated protein gene overexpression and reduced drug sensitivity of topoisomerase II in a human breast carcinoma MCF7 cell line selected for etoposide resistance. Cancer Res 54:152–158 28. Brock I, Hipfner DR, Nielsen BS et al (1995) Sequential coexpression of the multidrug resistance genes MRP and mdr1 and their products in VP-16 (etoposide)-selected H69 small cell lung cancer cells. Cancer Res 55:459–462

92

Calcagno and Ambudkar

29. Robey RW, Medina-Perez WY, Nishiyama K et  al (2001) Overexpression of the ATPbinding cassette half-transporter, ABCG2 (MXR/BCRP/ABCP1), in flavopiridol-resistant human breast cancer cells. Clin Cancer Res 7:145–152 30. Lee JS, Scala S, Matsumoto Y et  al (1997) Reduced drug accumulation and multidrug resistance in human breast cancer cells without associated P-glycoprotein or MRP overexpression. J Cell Biochem 65:513–526 31. Volk EL, Rohde K, Rhee M et  al (2000) Methotrexate cross-resistance in a mitoxantrone-selected multidrug-resistant MCF7 breast cancer cell line is attributable to enhanced energy-dependent drug efflux. Cancer Res 60:3514–3521 32. Rabindran SK, He H, Singh M et  al (1998) Reversal of a novel multidrug resistance mechanism in human colon carcinoma cells by fumitremorgin C. Cancer Res 58:5850–5858 33. Maliepaard M, van Gastelen MA, de Jong LA et  al (1999) Overexpression of the BCRP/ MXR/ABCP gene in a topotecan-selected ovarian tumor cell line. Cancer Res 59:4559–4563 34. Rao VK, Wangsa D, Robey RW et al (2005) Characterization of ABCG2 gene amplification manifesting as extrachromosomal DNA in mitoxantrone-selected SF295 human glioblastoma cells. Cancer Genet Cytogenet 160:126–133 35. Boonstra R, Timmer-Bosscha H, van EchtenArends J et al (2004) Mitoxantrone resistance in a small cell lung cancer cell line is associated with ABCA2 upregulation. Br J Cancer 90:2411–2417 36. Yasui K, Mihara S, Zhao C et  al (2004) Alteration in copy numbers of genes as a mechanism for acquired drug resistance. Cancer Res 64:1403–1410 37. Shen D, Fojo A, Chin J et al (1986) Human multidrug-resistant cell lines: increased mdr1 expression can precede gene amplification. Science 232:643–645 38. Fojo AT, Whang-Peng J, Gottesman MM, Pastan I (1985) Amplification of DNA sequences in human multidrug-resistant KB carcinoma cells. Proc Natl Acad Sci USA 82:7661–7665 39. Schoenlein PV, Shen DW, Barrett JT, Pastan I, Gottesman MM (1992) Double minute chromosomes carrying the human multidrug resistance 1 and 2 genes are generated from the dimerization of submicroscopic circular DNAs in colchicine-selected KB carcinoma cells. Mol Biol Cell 3:507–520 40. Chen GK, Lacayo NJ, Duran GE et al (2002) Preferential expression of a mutant allele of

the amplified MDR1 (ABCB1) gene in drugresistant variants of a human sarcoma. Genes Chromosomes Cancer 34:372–383 41. Ince TA, Scotto KW (1995) Differential utilization of multiple transcription start points accompanies the overexpression of the P-glycoprotein-encoding gene in Chinese hamster lung cells. Gene 156:287–290 42. Ince TA, Scotto KW (1995) A conserved downstream element defines a new class of RNA polymerase II promoters. J Biol Chem 270:30249–30252 43. Ince TA, Scotto KW (1996) Stable transfection of the P-glycoprotein promoter reproduces the endogenous overexpression phenotype: the role of MED-1. Cancer Res 56:2021–2024 44. Scotto K (2003) Transcriptional regulation of ABC drug transporters. Oncogene 22: 7496–7511 45. Mickley LA, Spengler BA, Knutsen TA, Biedler JL, Fojo T (1997) Gene rearrangement: a novel mechanism for MDR-1 gene activation. J Clin Invest 99:1947–1957 46. Huff LM, Lee J-S, Robey RW, Fojo T (2006) Characterization of gene rearrangements leading to activation of MDR-1. J Biol Chem 281:36501–36509 47. Mickley Huff L, Wang Z et al (2005) Aberrant transcription from an unrelated promoter can result in MDR-1 expression following drug selection in  vitro and in relapsed lymphoma samples. Cancer Res 65:11694–11703 48. Chekhun VF, Kulik GI, Yurchenko OV et al (2006) Role of DNA hypomethylation in the development of the resistance to doxorubicin in human MCF-7 breast adenocarcinoma cells. Cancer Lett 231:87–93 49. Eijdems EW, De Haas M, Coco-Martin JM et al (1995) Mechanisms of MRP over-expression in four human lung-cancer cell lines and analysis of the MRP amplicon. Int J Cancer 60:676–684 50. Miyake K, Mickley L, Litman T et al (1999) Molecular cloning of cDNAs which are highly overexpressed in mitoxantrone-resistant cells: demonstration of homology to ABC transport genes. Cancer Res 59:8–13 51. Knutsen T, Rao VK, Ried T et  al (2000) Amplification of 4q21–q22 and the MXR gene in independently derived mitoxantroneresistant cell lines. Genes Chromosomes Cancer 27:110–116 52. Volk EL, Farley KM, Wu Y et  al (2002) Overexpression of wild-type breast cancer resistance protein mediates methotrexate resistance. Cancer Res 62:5035–5040 53. Nakanishi T, Bailey-Dell KJ, Hassel BA et  al (2006) Novel 5¢ untranslated region variants of

Molecular Mechanisms of Drug Resistance in Single-Step and Multi-Step

54.

55.

56.

57.

58.

BCRP mRNA are differentially expressed in drug-selected cancer cells and in normal human tissues: implications for drug resistance, tissuespecific expression, and alternative promoter usage. Cancer Res 66:5007–5011 To KKW, Polgar O, Huff LM, Morisaki K, Bates SE (2008) Histone modifications at the ABCG2 promoter following treatment with histone deacetylase inhibitor mirror those in multidrugresistant cells. Mol Cancer Res 6:151–164 To KKW, Zhan Z, Litman T, Bates SE (2008) Regulation of ABCG2 expression at the 3¢ untranslated region of its mRNA through modulation of transcript stability and protein translation by a putative MicroRNA in the S1 colon cancer cell line. Mol Cell Biol 28:5147–5161 Zhou S, Schuetz JD, Bunting KD et al (2001) The ABC transporter Bcrp1/ABCG2 is expressed in a wide variety of stem cells and is a molecular determinant of the side-population phenotype. Nat Med 7:1028–1034 Chen G, Jaffrezou JP, Fleming WH, Duran GE, Sikic BI (1994) Prevalence of multidrug resistance related to activation of the mdr1 gene in human sarcoma mutants derived by single-step doxorubicin selection. Cancer Res 54:4980–4987 Jaffrezou JP, Chen G, Duran GE, Kuhl JS, Sikic BI (1994) Mutation rates and mechanisms of resistance to etoposide determined from fluctuation analysis. J Natl Cancer Inst 86:1152–1158

93

59. Dumontet C, Duran GE, Steger KA, BeketicOreskovic L, Sikic BI (1996) Resistance mechanisms in human sarcoma mutants derived by single-step exposure to paclitaxel (Taxol). Cancer Res 56:1091–1097 60. Beketic-Oreskovic L, Duran GE, Chen G, Dumontet C, Sikic BI (1995) Decreased mutation rate for cellular resistance to doxorubicin and suppression of mdr1 gene activation by the cyclosporin PSC 833. J Natl Cancer Inst 87:1593–1602 61. Chen KG, Wang YC, Schaner ME et  al (2005) Genetic and epigenetic modeling of the origins of multidrug-resistant cells in a human sarcoma cell line. Cancer Res 65: 9388–9397 62. Su GM, Davey MW, Davey RA (1998) Induction of broad drug resistance in small cell lung cancer cells and its reversal by paclitaxel. Int J Cancer 76:702–708 63. Shinoda C, Maruyama M, Fujishita T et  al (2005) Doxorubicin induces expression of multidrug resistance-associated protein 1 in human small cell lung cancer cell lines by the c-jun N-terminal kinase pathway. Int J Cancer 117: 21–31 64. Calcagno AM, Chewning KJ, Wu CP, Ambudkar SV (2006) Plasma membrane calcium ATPase (PMCA4): a housekeeper for RT-PCR relative quantification of polytopic membrane proteins. BMC Mol Biol 7:29

sdfsdf

Chapter 6 Pharmacogenetics of ATP-Binding Cassette Transporters and Clinical Implications Ingolf Cascorbi and Sierk Haenisch Abstract Drug resistance is a severe limitation of chemotherapy of various malignancies. In particular efflux transporters of the ATP-binding cassette family such as ABCB1 (P-glycoprotein), the ABCC (multidrug resistance-associated protein) family, and ABCG2 (breast cancer resistance protein) have been identified as major determinants of chemoresistance in tumor cells. Bioavailability depends not only on the activity of drug metabolizing enzymes but also to a major extent on the activity of drug transport across biomembranes. They are expressed in the apical membranes of many barrier tissues such as the intestine, liver, blood–brain barrier, kidney, placenta, testis, and in lymphocytes, thus contributing to plasma, liquor, but also intracellular drug disposition. Since expression and function exhibit a broad variability, it was hypothesized that hereditary variances in the genes of membrane transporters could explain at least in part interindividual differences of pharmacokinetics of a variety of anticancer drugs and many others contributing to the clinical outcome of certain leukemias and further malignancies. Key words:  ATP-binding cassette, Multidrug resistance, Bioavailability, Efflux transporter, Single nucleotide polymorphisms

1. Introduction A number of drugs are actively transported through intestinal enterocytes into portal blood vessels or back into the gut lumen. Moreover, drug bioavailability at the site of action is influenced by active transport processes, a fact that is well known from drugs having a low bioavailability in the central-nervous system due to active export processes at the blood–brain barrier. Most efflux transporters belong to the ABC (ATP-binding cassette) superfamily of membrane proteins, which may influence the intracellular concentration of numerous compounds in a variety of cells and tissues. These transporters play a major role as defense J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_6, © Humana Press, a part of Springer Science + Business Media, LLC 2010

95

96

Cascorbi and Haenisch

mechanism against penetration of xenobiotics such as cytostatics. The energy necessary for the substrate translocation across biomembranes is generated from the hydrolysis of ATP and intermediate phosphorylation of the transporter, enabling active transport of substrates against steep concentration gradients. The ABC transporter P-glycoprotein (Pgp, also known as ABCB1) is one of the best characterized human efflux transporters. There is increasing understanding of its function, regulation, and impact of genetic variants. Aside ABCB1, further members of the ABC transporters related to the phenomenon of multidrug resistance were identified, such as ABCC1 and ABCC2 (multidrug resistance-associated proteins, MRPs) and ABCG2 (breast cancer-related protein, BCRP). To date there are at least 49 members of the ABC transporter family, subdivided into seven subfamilies (http://nutrigene.4t.com/humanabc.htm). The protein size spans 325 amino acids in ABCC13 and up to 5,058 amino acids in ABCA13. In general, most transporters have a size of 1,500 AA. Most ABC transporters are composed of two equal or unequal halves containing a membrane-spanning domain and a nucleotide-binding fold (1).

2. ABCB1 (Pgp) 2.1. Function and Expression

The gene encoding Pgp was multidrug resistance 1 (MDR1, now termed ABCB1), due to the observation that ABC transporters like Pgp were overexpressed in tumor cells conferring to the commonly known phenomenon of multidrug resistance against certain antineoplastic agents (2). Mice have two closely related homologues of ABCB1 (Abcb1a, Abcb1b). Absence of the gene, as being the case in double-knockout mice, is conformable with life. Double-knock-out mice are viable and fertile but are highly sensitive to certain neurotoxins such as ivermectin (3), indicating the important role of ABCB1 in transport across the blood–brain barrier. Interestingly, such knock-out mice develop an inflammatory bowel disease, similar to Crohn’s disease (4). Pgp mediates the apical transport of various hydrophobic substrates including cytostatics such as etoposide, adriamycin, vinblastine as well as lipids, steroids, xenobiotics, and peptides (Table 6.1). It is expressed at the apical side of the brush-border membranes in the intestine, at the canalicular site of hepatocytes, at the apical site of renal tubular cells and of epithelial cells at the blood–brain barrier, protecting against drug penetration into the CNS (5). In general Pgp serves as a functional barrier against drug entry but contributes also to the excretion (Table 6.2) (6–8).

Table 6.1 Typical substrates of the ABC transporters discussed in this chapter ((5, 128–133) and references herein) ABCB1 Drug class MDR1, Pgp

ABCC1

ABCC2

ABCG2

MRP1

MRP2

BCRP

Arsenite, cisplatin, Camptothecin daunoAnticancer Docetacel, doxorubicin, Doxorubicin, rubicin, doxorubicin, doxorubicin, etoposide, drugs etoposide, imatinib, etoposide, gefitinib, etoposide, methotrexate, paclitaxel, teniposide, irinotecan, methoirinotecan, vinblastine, vinblastine, trexate, mitoxanmethotrexate, vincristine vincristine trone, topotecan, vinblastine, vincristine vincristine

Table 6.2 Function and expression of ABCB1, ABCC1, ABCC2, and ABCG2 proteins in human normal tissues (according to (134)) Name

ABCB1

ABCC1

ABCC2

ABCG2

Synonym

MDR1

MRP1

MRP2

BCRP

Locus

Apical

Basolateral

Apical

Apical

Intestinal enterocytes

Vs. gut lumen

Vs. portal blood

Vs. gut lumen

Vs. gut lumen

Duodenum

X

(X)

XX

X

Jejunum

X

(X)

XX

X

Ileum

XX

(X)

(X)

?

Colon

XX

(X)

–

XX

Rectum

X

(X)

?

X

Liver (hepatocytes)

Vs. biliary canaliculi

Vs. blood

Vs. biliary canaliculi

Vs. biliary canaliculi

Blood brain barrier (capillary endothelium)

Vs. blood

Vs. blood

(Vs. blood)

(Vs. blood)

Blood CSF barrier (chorioid plexus epithelium)

Vs. liquor

Vs. blood

Kidney (tubular epithelium)

XX(vs. urinary lumen)

(X)(vs. blood)

Heart (capillary endothelium)

Vs. blood

(X)

Placenta (syncytiotrophoblasts)

Vs. maternal blood

(X)

Testis

Capillary endothelium

(Sertoli cells)

Lymphocytes

X

(X)

Ubiquitous

X

X(vs. urinary lumen) Vs. blood Vs. maternal blood

Vs. maternal blood Capillary endothelium

98

Cascorbi and Haenisch

ABCB1 is also expressed in lymphocytes, but interestingly to a high extent in hematopoietic stem cells, where it may serve to protect these cells from toxins (3, 9, 10). Moreover, ABCB1 may be involved in the migration of dendritic cells (11). In humans, expression of ABCB1 discloses a broad interindividual variability. In the liver, Pgp concentrations reportedly differ up to 50-fold (12); moreover, Pgp expression and activity are subject to markedly drug–drug interactions (13). For example, the antituberculous agent rifampicin is an effective inducer of ABCB1 due to a responsive element to the nuclear receptor PXR in the ABCB1 promoter region (14). Similarly, the multidrug transporter ABCC2 (MRP2) is upregulated by rifampicin (15, 16). ABCB1 induction was also observed after coadministration of carbamazepine (17) or St. John’s worth (18) also well known as PXR trans-acting ligands (19, 20). On the other hand, competition with verapamil may lead to an inhibition of the transporter activity (21). Although ABCB1 appears to be coregulated in many aspects with cytochrome P450 3A4, there is not always a direct relationship to CYP3A4 expression (22). 2.2. Genetic Variability of ABCB1

Less than 100 SNPs have been identified in the coding region; the absolute number including intronic and the 5¢ and 3¢-regions is several hundreds; a systematic determination of the frequency was, however, performed in a limited number (Table 6.3). Figure  6.1 shows a two-dimensional structure of ABCB1 with locations of amino acid replacements and two frequent synonymous SNPs. The first systematic investigation on ABCB1 SNPs revealed a significant correlation of a silent polymorphism in exon 26 (3435C>T; rs1045642) with intestinal Pgp expression levels and oral bioavailability of digoxin, showing significantly decreased intestinal Pgp expression and increased digoxin plasma levels after oral administration among homozygote 3435T carriers (23). In codon 893 in exon 21, up to six different genotypes exist as a result of combinations of the three alleles 2677G>T/A (rs2032582). Moreover, 2677G>T/A and 3435C>T are in linkage disequilibrium (24–26). The frequency of the putatively most interesting 3435C>T SNP differs significantly between ethnicities. The variant 3435T allele has a prevalence of 0.17–0.27 in African Blacks, 0.41–0.47 in Oriental populations, and 0.52–0.57 among Caucasians (25, 27–30). Such genotypic differences may contribute to interethnic differences of drug responses in certain populations, as far the variants have functional relevance.

2.3. Functional Significance of ABCB1 SNPs

The functional significance of the ABCB1 variants is still discussed controversially. A number of studies indicated a loss of function for the 3435T variant, whereas others did not. Interestingly, there are no confirmations that the 3435C variant would be associated

Table 6.3 Frequency of ABCB1 genetic variants in Caucasians, position on DNA, putative effect, and frequencies (134) Position

Amino acid or effect

Frequency of the variant allele

5¢-Flanking −2903 T>C

0.02a

5¢-Flanking −2410 T>C

0.10a

5¢-Flanking −2352 G>A

0.28a

5¢-Flanking −1910 T>C

0.10a

5¢-Flanking −1717 T>C

0.02a

5¢-Flanking −1325 A>G

0.02a

5¢-Flanking −934 A>G

0.10a

5¢-Flanking −692 T>C

0.10a

5¢-Flanking −41 A>G

0.09b

IVS 1a −145 C>G

0.02b

IVS 1b −129 T>C

0.06b

IVS 1b 12 T>C

0.06c

IVS 2 −1 G>A

0.09d

c. 61 A>G

N21D

0.11d

IVS 5 −35 G>C

Intronic

0.006c

IVS 5 −25 G>T

Intronic

0.16c

IVS 6 +139 C>T

Intronic

0.37d

c. 548 A>G

N183S

0.01e

c. 1199 G>A

S400N

0.05d

c. 1236 C>T

Synonymous

0.41d

IVS 12 +44 C>T

Intronic

0.05d

c. 1474 C>T

R492C

0.01e

IVS 17 −76 T>A

Intronic

0.46d

IVS 17 +137 A>G

Intronic

0.006c

c. 2650 C>T

Synonymous

0.03e

c. 2677 G>T/A

A893S/T

0.42d/0.02d

c. 2956 A>G

M986V

0.005b

c. 3320 A>C

Q1107P

0.002d

c. 3396 C>T

Synonymous

0.03c

c. 3421 T>A

S1141T

0.00c

c. 3435 C>T

Synonymous

0.54d

c. 4030

Synonymous

0.005b

c. 4036

Synonymous

0.30b

References: a[42], b[26], c[25], d[28], e[23]

99

100

Cascorbi and Haenisch

Fig. 6.1. Two-dimensional structure of ABCB1 with locations of amino acid replacements and two frequent synonymous SNPs. NBD nucleotide-binding domain (according to (134) ).

with lower activity or expression in Caucasians. The molecular mechanism of how 3435C>T influences Pgp expression is not well understood. Some in  vitro studies indicated a lower rhodamine-123 efflux from CD56+ natural killer cells (31), whereas others could not confirm this finding (32, 33). Other methodologies pointed in the same direction. Determination of allelic abundance in genomic DNA and mRNA in human liver samples showed a significantly higher expression of the 3435C allele than that of the 3435T allele. Moreover, increasing 3435C/3435T ratios after cessation of transcription indicated decreased mRNA stability (34). A recent paper discussed with altered drug and inhibitor interactions as the result of altered conformations of the 3435C and T-variant during translation (35). The missense variant 2677G>T/A coding for the three different amino acids A893S/T exhibits altered transport properties in membrane vesicles from Sf9 insect cells, overexpressing human ABCB1. 893T had a higher vmax for the anticancer drug vincristine than 893S, and vmax of 893S was higher than that of the wildtype 893A, whereas KM was higher for 893S than for 893T or A (36). This study corroborated findings from a Japanese group, thoroughly investigating the ATPase activity from ten nonsynonymous SNP in Sf9 insect cells (37). A further rare missense SNP 1199 G>T (frequency 0.05 in Caucasians, (28)) leading to a Ser400Asn amino acid replacement is associated with lower activity and accordingly higher sensitivity against anticancer drugs. In contrast, a later detected 1199G>A variant caused an amino acid exchange to isoleucin at position 400. In vitro transfected HEK cells carrying this variant, however, exhibited an elevated chemoresistance indicating an elevated transport rate of the modified P-glycoprotein (38). A large study on ABCB1 mRNA (n=32) and protein (n=37) expression was performed in German healthy volunteers (39), showing a broad variability but lack of any association to 2677G>T/A or 3435C>T. In contrast, among Japanese, 3435T

Pharmacogenetics of ATP-Binding Cassette Transporters and Clinical Implications

101

Table 6.4 Functional significance of ABCB1 genetic variants and relevance for clinical outcome Position

Amino acid exchange

Effect of variant

5¢-Flanking −2410 T>C

Decreased mRNAa

5¢-Flanking −692 T>C

Decreased mRNAa

c. 571 G>A

G191R

Reduced chemotherapy resistanceb

c. 1199 G>A

S400N

Elevated activityc

c. 1236 C>T

Synonymous

Increased imatinib disposition and therapy responsed

c. 2677 G>T/A

A893S/T

In vitro increased vmax,q no effect on vincristine,e increased imatinib response in CMLd

c. 3435 C>T

Synonymous

Decreased mRNA and protein expression,f, g decreased in vitro transport,h no effect on expression and bioavailability of talinolol,i no effect on in vitro transport,j, k decreased digoxin bioavailability,l increased etoposid disposition,m no effect on AML or ALL outcome,k better prognosis of multiple myeloma,n better chemotherapy response in breast cancer,o no effect in colon cancerp

References: a[42], b[69], c[38], d[53], e[51], f[23], g[64], h[31], i[39], j[135], k[65–67], l[40, 41], m[52], n[68], o [74], p[70, 71], q[36]

was associated with significantly higher mRNA levels than 3435C, a finding that is in line with the observation that the digoxin bioavailability is lower among Oriental 3435T carriers (24, 40, 41) (Table 6.4). Further genetic variants identified in the 5¢-UTR, particularly −2410T>C, −190T>C, and −692T>C, being in linkage disequilibrium showed a significant association to mRNA levels obtained from Japanese colon cancer specimens (42). Among Japanese liver-transplant recipients, however, there was no correlation to common ABCB1 SNPs, but interestingly there was a correlation to intestinal CYP3A4 expression (43). No association for 3435C>T, but for 2677G>T/A and −129T>C was reported from a Japanese group, who evaluated whether ABCB1 correlates with placenta trophoblast Pgp expression in 100 placentas obtained from Japanese women (26). In contrast, Hitzl et  al. showed that the placenta P-glycoprotein expression was lower when both mother and child were carriers of 3435T compared to levels obtained for pairs of 3435C. However, no influence on the mRNA level was observed (31).

102

Cascorbi and Haenisch

2.4. Association to Drug Bioavailability

As mentioned earlier, the first systematic study on ABCB1 genetic variability and its association to expression and bioavailability was the first one, showing an association of 3435C>T with digoxin plasma levels. TT carriers had higher digoxin levels than CC carriers, a fact that was in line with the findings on Pgp expression levels (23). Numerous studies on the genotype-dependent bioavailability of various drugs, known to underlie Pgp-mediated transport, followed this initial study (for excellent reviews see (44, 45)). Support was given, though less pronounced in subjects being in steady state with 0.25-mg digoxin/day. TT-carriers had a 20% elevated area under the curve (AUC) within the first four hours. Interestingly, the effect was stronger in 2677GG/3435TT carriers (46). A further publication from Europe, summarizing different studies including some African-Blacks, supported these observations with similar differences between 3435TT and CC (47). However, the functional impact of 3435C>T is not consistent in a number of further studies., e.g., the pharmacokinetics of 1 mg orally administered digoxin (the same dose as in the initial study (23)) were not influenced by 3435C>T or 2677G>T/A (48) and a in a Japanese study, the digoxin AUC in the first 4 h was significantly lower in the CC group than in subjects homozygous for TT (41). This opposite tendency was also observed in two further independent Japanese investigations (24, 40, 49). Further thorough investigation on the Pgp substrate talinolol, a b1-blocker, revealed that in Germans carriers of the 2677 TT/TA variants had slightly but significantly elevated drug levels than carriers of at least one wild-type allele (p < 0.05). However, multiple comparisons with combinations of putative functional SNPs did not confirm a significant influence of the ABCB1 genotype on talinolol disposition (39). Conflicting results are reported for the antihistaminic drug fexofenadine (25) and again no effects were observed among Germans (50) (Table 6.4). Few studies on the bioavailability of anticancer drugs also show diverging results. For example, the pharmacokinetics of vincristine were only marginally influenced by the ABCB1 SNPs 2677G>T/A and 3435C>T (51), whereas the disposition of etoposide was apparently influenced by 3435C>T, demonstrating elevated clearance among CC carriers (52). The kinetics of the BCR-ABL inhibitor imatinib was affected by the sense SNP 1236C>T in a study among chronic myelomic leukemia patients (53). The reasons for the discrepancies observed particularly for digoxin are currently unclear. As far as differences of bioavailability for 3435C>T have been found, the association was counterpart in Asians compared with that in Caucasians, suggesting different haplotype constellations among these ethnicities. These effects were more pronounced concerning haplotypes considering 2677G>T/A. However, the majority of studies revealed no differences, and it was concluded that it is unlikely that digoxin

Pharmacogenetics of ATP-Binding Cassette Transporters and Clinical Implications

103

bioavailability is modulated by ABCB1 polymorphisms (54) due to various hormonal and immunological influences on ABCB1 activity (55–57) and possibly a circadian rhythm as could be demonstrated in a mouse model (58). Moreover, saturation of the intestinal ABCB1 transport capacity may surpass any genetic effects (45). Finally other uptake and efflux transporters may contribute to the disposition of digoxin in humans. 2.5. Association to Treatment Outcome

Drug transporters are expressed in many tissues and tissue barriers. Therefore, the question arises whether changes of transporter activity may contribute to different drug concentrations not only in the plasma but particularly at the site of drug action, e.g., beyond the blood–brain barrier in the treatment of epilepsy, inside lymphocytes in the treatment of HIV infections or lymphatic malignancies, or with tumor masses. Although there is increasing evidence of the importance of P-glycoprotein in the etiology of pharmacotherapy-resistant epilepsy, the data concerning the role of ABCB1 variants are controversial (for review see (59)).

2.5.1. Leukemia

Similar as in HIV therapy, it is obviously clear that the expression of ABCB1 in lymphocyte membrane might be of major importance in the treatment outcome of lymphatic leukemia. Accordingly, a number of studies were performed regarding the contribution of ABCB1 genotypes and indeed, there is an increasing number of studies suggesting an association between ABCB1 genotypes and clinical outcome (60–63). Some functional evidence for a role of ABCB1 for lymphocytic P-glycoprotein expression was given by a study from Seedhouse et  al. (64). Investigating lymphocytes from British acute myeloic leukemia (AML) patients, the Pgp expression was significantly higher among the upper percentiles in ABCB1 3435C carriers than T carriers. In a study in 101 Korean AML patients 3435CC was significantly correlated with lower functional ABCB1 function in a daunorubicin intracellular accumulation assay (62). Surprisingly, 3435CC and 2677GG carriers were strongly associated with a higher probability of complete remission and 3-year event-free survival. However, no differences were noted in overall survival according to the ABCB1 MDR1 SNPs. This lack of association was confirmed in a clinical study among elderly AML patients; ABCB1 variants failed to show any association to the treatment outcome or ABCB1 expression and function, as evidenced by rhodamine efflux experiments controlled with the Pgp inhibitor PSC833 (65). In contrast, the molecular response to the BRC-ABL-inhibitor imatinib in French chronic myeloic leukemia (CML) patients was dependent on ABCB1 SNPs (53). Although 3435C>T failed to show a significant influence, patients with the 1236C>T had higher imatinib plasma concentrations and showed also an improved therapy response, whereas presence of the wild-type

104

Cascorbi and Haenisch

2677G variant worsened the clinical response. In another study in a small sample of CML patients and gastrointestinal stromal tumors (GIST), all “classic” ABCB1 SNPs such as 1236C>T, 2677G>T/A, and 3435C>T led to decreased imatinib clearance among variant carriers. In a Korean study on acute lymphoblastic leukemia (ALL), however, there was lack of association of clinical endpoints such as complete remission rates, or relapse-free and event-free survival rates to ABCB1 variants (66). Only rare ABCB1 haplotypes of 2677G>T/A and 3435C>T differed in a large Hungarian ALL study, but overall the genotype distribution was not statistically different (67). In multiple myelomas treated with dexamethasone, doxorubicin and vincristine, ABCB1 3435CT or TT carriers had a better prognosis than 3435CC carriers (p = 0.02) (68). In a recent study, a novel ABCB1 571G>A missense variant detected in 6.4% of leukemia patients was reported, causing a Gly191Arg amino acid change (69). In a stable recombinant cell model, the anticancer drugs doxorubicin HCl, daunorubicin HCl, vinblastine sulfate, vincristine sulfate, taxanes (paclitaxel), and epipodophyllotoxin (etoposide, VP-16) exhibited selectively reduced degree of Pgp-mediated resistance in 561A carriers. In particular, the ABCB1-dependent resistance on vinblastine, vincristine, paclitaxel, and etoposide was fivefold reduced, indicating lower transport capacity of the 191Arg-variant. This could be proven by determination of intracellular drug concentrations. It was suggested that individuals with the ABCB1 571A genotype may be more sensitive to the specific anticancer drugs that are Pgp substrates but may also exhibit a higher risk of side effects (Table 6.4). 2.5.2. Solid Tumors

3. ABCC1 (Multidrug Resistanceassociated Protein 1)

For colorectal cancer, there are only weak data supporting any evidence of an impact of ABCB1 variants to the risk of cancer and to our knowledge no data on the treatment outcome. Overall, there was lack of association to the risk (70, 71) or only differences in small subgroups requiring confirmation (72, 73). Moreover, in the treatment of solid tumors such as breast cancer, 3435C>T was shown to be associated with the clinical response to preoperative chemotherapy (74) (Table 6.4).

The human ABCC1 protein was first identified in the doxorubicinresistant small cell lung cancer cell line H69AR that did not overexpress Pgp (75). ABCC1 serves as a multispecific organic

Pharmacogenetics of ATP-Binding Cassette Transporters and Clinical Implications

105

anion transporter for certain drugs such as folate-based antimetabolites, anthracyclines, plant alkaloids, and antiandrogens. Moreover, it is involved in the transport of GSH- and glutathione conjugates (76) (Table  6.1). ABCC1 is expressed ubiquitously in the human body. In polarized epithelial cells, it is localized to basolateral membranes (Table 6.2). A large number of hereditary polymorphisms have meanwhile been identified (77–80). A thorough investigation on the functional significance of ten nonsynonymous SNPs, leading to amino acid changes C43S, T73I, S92F, T117; R230Q, R633Q, R723Q, A989T, C1047S. R1058Q, and S1512L was performed by Létourneau et al. in transfected HEK293T cells (81) (Table 6.5). None of them influenced significantly the expression level, indicating that the amino acid exchanges do not substantially affect the synthesis or stability of the protein. The overall influence on transport capacity of three different substrates such as leukotriene glutathione conjugate LTC4 was moderate. Lowest capacity was found for A989T, caused by a 2965G>A variant. Kinetic analysis revealed a weakened substrate affinity as indicated by an elevated Km value of the 989T-expressing variant (81). In another study, the silent variants 816G>A, 825T>C, 1684T>C, and 4002G>A were investigated for their impact on mRNA expression in peripheral CD4+ cells of German healthy volunteers without obtaining any significant differences (33). In an approach to scan for genetic signatures within the ABCC1 locus in different ethnicities, a haplotype containing a −260G>C SNP in the 5¢-flanking region was identified, associated with diminished activity in a reporter gene assay (82). This SNP has a frequency of 0.23 in European Americans, 0.55 in African Americans, but 0.00–0.05 in Orientals (Table 6.5). 3.1. Susceptibility to Cancer

The risk of lung cancer with respect to variant in ABCB1 and ABCC1 was investigated in a case-control study of 500 patients with incident lung cancer and 517 controls in a Chinese population. Out of three SNPs in the 3¢ untranslated region of ABCC1 (rs3743527, rs212090, and rs212091) the variant rs212090 genotype was more frequent in a recessive model (OR, 1.37; 95% CI, 1.03–1.83) (83) (Table 6.5).

3.2. Clinical Outcome

In a study on genetic determinants of anthracycline-induced cardiomyopathy in non-Hodgkin lymphoma patients, the ABCC1 Gly671Val variant as well as a haplotype of ABCC2 turned out to be significant risk factors. Data on a significant impact of ABCC1 polymorphism on drug bioavailability or further treatment outcome, however, are lacking (84). No effects on the clinical outcome on a platinum- and taxane-based chemotherapy were observed in the Scottish Randomized Trial in Ovarian Cancer (85).

106

Cascorbi and Haenisch

Table 6.5 Frequency of ABCC1 genetic variants in different populations, position on DNA, putative effect, and frequencies (according to (33, 77–80, 136)) Position

Amino acid or effect

Orientals

Caucasians

Function

c.128G>C

C43S

0.01

–

Elevateda

c. 218C>T

T73I

0.00–0.04

–

c. 257C>T

S92F

0.00

0.00

Decreaseda

c. 350C>T

T117M

–

0.02

(Decreased)a

c. 689G>A

R230N

0.00

0.00

(Decreased)a

c. 816G>A

Synonymous

–

0.04

c. 825T>C

Synonymous

–

0.30

c. 1057G>A

V353M

0.00

0.005

Elevateda

c. 1299G>T

R433S

–

0.01

Elevated vmax of doxorubicin, decreased transport of LTC4a,b

c. 1684T>C

Synonymous

–

0.80

c. 1898G>A

R633Q

–

0.01

(Decreased)a

c. 2012G>T

G671V

–

0.03

Doxorubicine-induced cardiomyopathyc

c. 2168G>A

R723Q

0.01–0.07

–

Decreaseda

c. 2965G>A

A989T

0.00

0.005

(Decreased)a

c. 3140G>C

C1047S

0.00

0.00

c. 3173G>A

R1058Q

0.01

–

c. 4002G>A

Synonymous

–

0.28

c. 4535C>T

S1512L

–

0.03

Decreaseda

References: a[81], b[77], c[84]

4. ABCC2 (Multidrug ResistanceAssociated Protein 2)

The ABCC2 gene encodes for a transmembrane transport pump exhibiting two cytosolic nucleotide binding domains (NBD) and two transmembrane domains (TMD1 and 2) each consisting of six helices. An additional transmembrane domain (TMD0) comprising five helices is located at the N-terminus. The ABCC2 protein is a glycolized protein located in the apical luminal membrane of tissues with function of barriers such as liver, kidney, intestine, placenta, but to newer knowledge weak or even not at the blood–brain barrier (Fig. 6.2). Tumor cells often show

Pharmacogenetics of ATP-Binding Cassette Transporters and Clinical Implications

107

Fig. 6.2. Apical expression of ABCC2 in the intestine and kidney, as well as canalicular expression in the liver and its role as export pump.

an inducible expression of ABCC2, which contributes also to the phenomenon of drug resistance. The ABCC2 gene is located on chromosome 10q24 and spans a total length of 69 kb. It exhibits 32 exons. After translation, a protein of 1,545 amino acids with a molecular weight of 190 kDa is expressed. As typical for an ABC transporter the spectrum of substrates is wide and partly overlapping with that of other ABC transporters. The substrates are mainly organic anionic sulfate, glutathione or glucuronide conjugates of endogenous compounds such as hormones, leukotrienes, bile acids, and bilirubin. But also conjugated and unconjugated exogenous substances of organic but also inorganic origin are pumped out of the tissue into the lumen (Table 6.1). It is discussed that some substrates require cotransportation with glutathione because of their missing anionic character. Because of the transport of bile acids and glutathione from the hepatocytes into the bile duct, ABCC2 plays physiologically an important role in forming bile flow and potentially in detoxification by delivering glutathione for conjugation of xenobiotics. ABCC2 was for the first time described in man in 1996. Besides other ABC transporters ABCC2 was also found to be overexpressed in cancer cells, which exhibit resistance against antineoplastic drugs such as doxorubicin, cisplatin, etoposid, vincristine, SN38, and

108

Cascorbi and Haenisch

methotrexate. Some reagents such as probenecid and diclofenac are shown to inhibit the transporter whereas substances such as rifampicin, dexamethasone, phenobarbital, bile acids, and oltipraz induce the expression via nuclear receptors namely PXR (pregnan X receptor), CAR (constitutive androstane receptor), FXR (farnesoid X receptor), or Nrf2 (nuclear factor-erythroid p45 related factor2) (86-89). 4.1. Genetic Variability

Individuals exhibiting the Dubin–Johnson syndrome show different mutations in the ABCC2 gene leading to impairment or complete loss of function of the transporter. The clinical consequence is a benign conjugated hyperbilirubinemia and pigmentation of liver (86–89). A number of polymorphisms have been also found in the normal population by systematically sequencing revealing substantial differences between ethnicities (Table 6.6). For some SNPs evidence for the functional impact is available from in-vitro experiments but also from pharmacokinetic studies. A −24C>T polymorphism in the 5¢-UTR was firstly described by Ito et al. in a Japanese sample (78). It is in a strong linkage disequilibrium with 3972C>T (90, 91) and was reportedly influencing the gene expression but also drug bioavailability, clinical outcome, and toxicity of xenobiotics (92).

4.2. Functional Significance

Renal allograft transplant recipients harboring the −24T allele show a decreased oral clearance for the immunosuppressant mycophenolic acid (MPA), the active metabolite of mycophenolate mofetil. In consequence, these patients are more protected from a decrease in MPA exposure but with a higher association to mild liver dysfunction (93). A similar observation was made for the oral clearance of MPA, suggesting a decreased biliary excretion in −24T allele carriers as far they exhibited the homozygote 334T>G variant of the uptake transporter SLCO1B3 (94). Interestingly, one study showed a significant higher allele frequency of −24T in patients who had undergone renal organ transplantation suggesting a higher risk to develop renal failure probably due to impaired renal excretion and a higher exposure of renal cells to toxic substances derived from nutrition, drugs, and environment (95). In noncancerous renal cortex tissue, the −24C>T showed a decreased mRNA expression. This functional influence could be confirmed in a luciferase assay after transfection of HepG2 cells with reporter gene vector constructs containing the wild type or variant allele (90). Similar results were received in in  vitro experiments investigating the −24C>T polymorphism and the polymorphism −1774G/del in the 5¢ flanking region of the ABCC2 gene. Both SNPs that were not linked showed a decreased activity in reporter gene assays. Clinically individuals harboring the deletion variant at −1774 showed a significant

Pharmacogenetics of ATP-Binding Cassette Transporters and Clinical Implications

109

Table 6.6 Frequency of ABCC2 genetic variants in Japanese (78), Germans (in parentheses) (92), and 72 cell lines* (91), position on DNA, putative effect, frequencies, and function Position

Amino acid or effect

Frequency

−751 T>A

Transcription?

0.01*

−717 C>T

Transcription?

0.01*

−24 C>T

Translation?

0.19 (0.18)

−23 G>A

Translation?

0.01* (0.00)

c. 56 C>T

P19L

0.01*

c. 234 A>G

Synonymous

0.01*

c. 299 G>A

R100Q

0.01*

c. 842 G>A

S281N

0.01*

c. 1249 G>A

V417I

0.13 (0.21)

c. 1446 C>G

(0.01)

c. 1457 C>T

T486I

0.03* (0.00)

c. 2302 C>T

R768W

0.01 (0.00)

c. 2366 C>T

S789F

0.01 (0.00)

c. 2647 G>A

D883N

0.01*

c. 2882 A>G

K961R

0.01*

c. 2934 G>A

Synonymous

0.05*

c. 3039 C>T

Synonymous

0.01*

c. 3057 G>T

Q1019H

0.01*

c. 3321 G>T

Synonymous

0.01*

c. 3521 G>A

R1174H

0.01*

c. 3542 G>T

(0.001)

c. 3561 G>A

(0.00)

c. 3563 T>A

V1188E

0.01* (0.05)

c. 3732 C>T

N1244K

0.01*

c. 3972 C>T

Synonymous

0.22* (0.34)

c. 4100 C>G

S1367C

0.01*

c. 4290 G>T

Synonymous

0.01*

c. 4348 G>A

A1450T

0.01 (0.00)

c. 4488 C>T

Synonymous

0.01*

c. 4544 G>A

C1515Y

0.01* (0.04)

110

Cascorbi and Haenisch

association to cholestatic or mixed type hepatitis whereas −24T carriers exhibited more often hepatocellular-type hepatitis after intake of drugs or herbal remedies (96). Inline with these reports −24T allele carriers revealed a higher risk to develop a hepatotoxic reaction after intake of diclofenac (97). A missense SNP 1249G>A (Val417Ile) is located in substrate-binding region of the first transmembrane domain and is associated with lower oral bioavailability and increased residual clearance after intravenous administration of the beta-blocker talinolol, indicating a higher activity of the intestinal transporter (92). It was also found to be correlated with renal proximal tubulopathy after treatment with the HIV protease inhibitor tenovir disoproxil fumarate (98), probably due to a toxic concentration of the drug in renal tubular cells after being actively secreted by ABCC2 from the blood into the tubule. It is suggested that the elevated transport capacity toward the substrates 17b-glucuronosyl estradiol, leukotriene C4, and S-glutathionyl 2,4-dinitrobenzene (DNP-SG) is due to elevated protein expression rather than changes of functional properties (99). In methotrexate-treated African American rheumatoid arthritis patients, the 1249A variant allele was associated with higher gastrointestinal toxicity (100). The silent polymorphism 1446C>G is associated with higher ABCC2 mRNA expression in the liver and with a decreased AUC and cmax of the cholesterol lowering drug pravastatin (101) due to an elevated hepatic first pass effect. The SNPs c.3563T>A and c.4544G>A are in a strong linkage and correlated with a higher ABCC2 protein expression in liver (102). 3972C>T was reportedly associated with a fourfold higher risk for occurrence of intrahepatic cholestasis in pregnancy (103). 4.3. Clinical Outcome of Cancer

ABCC2 was shown to be significantly associated with acute doxorubicin toxicity (84). High-dose methotrexate treatment in pediatric ALL induced a two fold higher area under the curve and a ninefold higher risk of intensification of folinate rescue in female patients carrying the −24 variant allele (104). In non-small cell lung cancer patients, ABCC2 −24TT and 3972TT genotypes were associated with higher response rates (p = 0.031 and 0.046, respectively) and longer progression-free survival (p = 0.035 and 0.038, respectively), which was sustained in haplotype analysis, suggesting a more effective exposure to irinotecane (105). However, for haplotype carriers containing the −24C-allele, a less frequent irinotecan-related diarrhea was observed, which was suggested to be the consequence of a less hepatobiliar excretion of the drug (106). In summary there is increasing evidence that polymorphisms of the ABCC2 transporter can influence the first pass effect of the

Pharmacogenetics of ATP-Binding Cassette Transporters and Clinical Implications

111

Table 6.7 ABCC2 polymorphisms currently described to exhibit a clinical influence ABCC2 polymorphism Effect

Clinical impact on

Function

Reference

−1774 G /del

5¢-flanking

Hepatotoxicity of herbal and conventional drugs

Decreased

[96]

−24C>T

5¢-UTR

Hepatotoxicity of drugs e.g. diclofenac and herbal remedies

Decreased

[96, 97]

Oral clearance of Mycophenolic acid

Decreased

[93, 94]

Risk of renal failure, renal expression

Decreased

[90, 95]

Bioavailability and side effects of methotrexate

Decreased

[104]

Tumor response and side effects of irinotecan

Decreased

[105, 106]

Intestinal activity, bioavailability of talinolol

Increased

[92]

Gastrointestinal toxicity of methotrexate

Increased?

[100]

Proximal tubulopathy of tenovir disproxil fumarate

Increased?

[98]

c.1249G>A

V417I

c.1446C>G

T482T

Bioavailability of pravastatin

Increased

[101]

c.3563T>A

V1188E

Higher protein expression in liver

Increased

[102]

c.3972C>T

I1324I

Intrahepatic cholestasis in pregnancy

Decreased

[103]

c.4544G>A

C1515Y

Higher protein expression in liver

Increased

[102]

liver but also the excretion in the intestine and kidney (Fig. 6.2). These genetically differing transport activities caused by either different expressions and/or affinities to the substrates can be observed as differences in bioavailability, drug response, or toxic side effects. Table  6.7 summarizes the effects of the ABCC2 polymorphisms yet associated with clinical impact on the transporter.

5. ABCG2 (Breast Cancer Resistance Protein) 5.1. Function and Substrate Specificity

ABCG2, also termed breast cancer resistance protein (BCRP) or mitoxantrone-resistant protein, is the second member of the G-family of ABC transporters (ABCG2) (107, 108). The ABCG2 gene is located at 4q22 and encodes a 72-kDa membrane protein

112

Cascorbi and Haenisch

composed of 655 amino acids (109). The protein consists of only one ATP-binding region and one transmembrane domain and is referred to as a half-transporter, and its homodimerization may be necessary to transport substrates (110). In normal human tissues, ABCG2 is highly expressed in the placenta, colon, small intestine, and liver (111, 112). ABCB1 and ABCG2 as well as the major vault protein (MVP) are colocalized in microvascular endothelium of epileptogenic human brain tissue (113) (Table  6.2). Interestingly, ABCG2 expression is upregulated in tissues with a low-oxygen environment (114) and also expressed in endothelial cells of human heart vessels. Strikingly, ventricular samples from cardiomyopathic hearts exhibited significantly increased levels of ABCG2 mRNA (115). Its expression is modulated by EGF by activation of MAPK cascade via phosphorylation of ERK1/2 and JNK/SAPK (112). On the basis of its tissue distribution and findings in knockout mice, ABCG2 is speculated to have a major influence on the pharmacokinetic and pharmacodynamic profiles of certain xenobiotics and endogenous substrates and is believed to contribute to multidrug resistance, since typical substrates are cytostatic drugs such as cisplatin, camptothecin, daunorubicin, doxorubicin, etoposide, methotrexate, mitoxantrone, SN-38, topotecan, and vincristine (111, 116) (Table 6.1). ABCG2 interacts with heme and other porphyrins and protects cells and/or tissues from protoporphyrin accumulation under hypoxic conditions (for review see (114)). 5.2. Genetic Variability of ABCG2

5.3. Association to Activity and Drug Bioavailability

In 2001, Honjo et  al. (117) and 1 year later Imai et  al. (118) examined cDNA from cancer cell lines for genetic variants. Both identified 34G>A (V12M) and 421C>A (Q141K). Further a deletion at position 944–949 leading to the lack of amino acid residues A315 and T316 was detected. 421C>A ABCG2-transfected PA317 cells showed markedly decreased protein expression and low-level drug resistance compared with wild-type, whereas the other variants showed similar protein expression and drug resistance compared with wild-type ABCG2-transfected cells. Among 124 healthy Japanese volunteers, the frequency of the variant 421A allele, coding for low ABCG2 expression was 27%. This nonsynonymous polymorphism is located near the ATP-binding site of the half transporter. The 376C>T in exon 4, leading to a premature stop codon, was found in only 1.2% (118) (Table 6.8). Currently, more than 80 variants have been detected. HEK-293 cells, transfected with wild-type or V12N, ABCG2 showed apical staining with an ABCG2 antibody, but high intracellular staining in case of Q141K, suggesting impaired membrane trafficking or incorrect membrane insertion. Moreover, decreased transport rates were found in Sf9 insect cells, transfected with the V12M variant (119). In contrast, in a study by Mizuarai

Pharmacogenetics of ATP-Binding Cassette Transporters and Clinical Implications

113

Table 6.8 Frequency of ABCG2 genetic variants in Japanese, position on DNA, putative effect, and frequencies (according to (137) ) SNP

Amino acid or effect

Frequency of the variant allele

5¢ −20445 C>T

?

0.010

5¢ −20296 A>G

?

0.110

5¢ −19781 A>G

?

0.005

5¢ −19572–69 DCTCA

?

0.235

c. 34 G>A

V12M

0.17 0.04a 0.06b

IVS 2 +16 A>G

?

0.180

IVS 3 +10 A>G

?

0.080

IVS 3 +10 C>T

?

0.005

c. 376 C>T

Q126stop

0.01 0.00a 0.00b

Lack of function

c. 421 C>A

Q141K

0.35 0.11a 0.02b

Reduced activity [120, 124, 137]

IVS 5 −16 A>G

?

0.005

c. 1098 G>A

Synonymous

0.010

IVS 10 +95 T>A

?

0.015

c. 1322 G>A

S441N

0.005

c. 1367 A>G

?

0.165

c. 1465 A>G

F489L

0.005

c.1492 G>C

?

0.335

c. 1515DC

Frame shift

0.005

IVS 13 −42 A>T

?

0.005

IVS 13 −21 C>T

?

0.165

IVS 14 −46 A>G

?

0.500

3¢ 2332 A>TinsA

?

0.070

3¢ 2364 A>C

?

0.005

3¢ 2512 C>T

?

0.005

Caucasians African Americans

a

b

Function

Lack of function

114

Cascorbi and Haenisch

et  al. ABCG2 V12N protein was found to be accumulated intracellulary (120) and always apical expression was observed by Kondo et  al. (121). Correlation of the mRNA expression of intestinal samples of 42 patients revealed no significant association to any polymorphism (122). In 15 ventricular (including 10 cardiomyopathic) and 51 auricular heart tissue samples there was also lack of association of 34G>A or 421C>A to ABCG2 mRNA expression (115). The functional impact of various nonsynonymous polymorphisms was also investigated for methotrexate capacity, showing highly varying transport capacities. Interestingly, V12M was associated with elevated activity compared to the wild-type, whereas ABCG2 with premature stop-codon lacked any activity as expected (37). Best evidence was detected for the functional impact of 421C>A, located in the ATP-binding region. The 421A variant showed 1.3-fold decreased ATPase activity than the wild type (120), and the bioavailability of diflomotecan and topotecan was significantly elevated (123, 124) (Table  6.8). The pharmacokinetics of the anticancer drugs 9-nitrocamptothecin (9NC) appear also to be influenced by ABCG2 variants. The ABCG2 421C>A genotype significantly affected the AUC/does ratio of the 9-aminocamptothecin (9AC) metabolite in metastatic colon cancer patients, being 3.6-fold higher in 421CA carriers than in 421CC wild-type carriers (p = 0.032) (125).

6. Conclusion In summary, the physiological consequences of ABC transporter genetic variants are still only partly understood and the current figure of all findings is puzzling. The overall bioavailability of drugs seems to be only moderately influenced by the currently known ABCB1 SNPs, at least as compared to variants of the cytochrome P450 system (126, 127). It is interesting to note that among bioavailability studies performed in Caucasians often 3435T carriers presented higher plasma concentrations, whereas among Orientals this was the case for 3435C subjects, indicating possible different haplotype clusters in these ethnicities. Consequently, due to the large interindividual variability of Pgp expression in the intestine and in the liver that can still not be explained in full, consideration of genetic variants of this transporter does not appear to be a suitable parameter for individualized drug therapy (45). On the other hand, the association of ABCB1 3435T to improved treatment outcome particularly in the field of lymphatic leukemia gives rise to further investigations

Pharmacogenetics of ATP-Binding Cassette Transporters and Clinical Implications

115

on the genetic control of P-glycoprotein concerning intracellular drug disposition. ABCC1 (MRP1) was also shown to be highly polymorphic. Although in vitro data show some variability of substrate specificity dependent on genetic variants, there is lack of evidence for a functional significance for drug bioavailability. Similarly, the current data on ABCC2 (MRP2) suggest that genetic variants not linked to Dubin-Johnson syndrome seem to play a minor part for drug bioavailability. In contrast, in ABCG2 in  vitro as well as in vivo findings indicate significant differences of expression and substrate transport capacities. However, more studies have to be performed to clarify the partly divergent results. In conclusion, the current knowledge of the functional significance of genetic variants of ABC membrane transporters does not allow selection of a particular SNP to predict an individual’s pharmacokinetics. However, the large number of studies, achieving associations particularly of ABCB1 variants to clinical outcome, strongly support the necessity of further investigation of the role of these fascinating transporters for intracellular drug bioavailability and clinical outcome, particularly of lymphatic and chronic leukemia. References 1. Borst P, Elferink RO (2002) Mammalian ABC transporters in health and disease. Annu Rev Biochem 71:537–592 2. Juranka PF, Zastawny RL, Ling V (1989) P-glycoprotein: multidrug-resistance and a superfamily of membrane-associated transport proteins. FASEB J 3:2583–2592 3. Schinkel AH, Mayer U, Wagenaar E et  al (1997) Normal viability and altered pharmacokinetics in mice lacking mdr1-type (drugtransporting) P-glycoproteins. Proc Natl Acad Sci USA 94:4028–4033 4. Panwala CM, Jones JC, Viney JL (1998) A novel model of inflammatory bowel disease: mice deficient for the multiple drug resistance gene, mdr1a, spontaneously develop colitis. J Immunol 161:5733–5744 5. Fromm MF (2000) P-glycoprotein: a defense mechanism limiting oral bioavailability and CNS accumulation of drugs. Int J Clin Pharmacol Ther 38:69–74 6. Schinkel AH, Wagenaar E, Mol CA, van Deemter L (1996) P-glycoprotein in the blood-brain barrier of mice influences the brain penetration and pharmacological activity of many drugs. J Clin Invest 97:2517–2524 7. Terao T, Hisanaga E, Sai Y, Tamai I, Tsuji A (1996) Active secretion of drugs from

the small intestinal epithelium in rats by P-glycoprotein functioning as an absorption barrier. J Pharm Pharmacol 48:1083–1089 8. Zhang Y, Benet LZ (2001) The gut as a barrier to drug absorption: combined role of cytochrome P450 3A and P-glycoprotein. Clin Pharmacokinet 40:159–168 9. Chaudhary PM, Roninson IB (1991) Expression and activity of P-glycoprotein, a multidrug efflux pump, in human hematopoietic stem cells. Cell 66:85–94 10. Chaudhary PM, Mechetner EB, Roninson IB (1992) Expression and activity of the multidrug resistance P-glycoprotein in human peripheral blood lymphocytes. Blood 80: 2735–2739 11. Randolph GJ (2001) Dendritic cell migration to lymph nodes: cytokines, chemokines, and lipid mediators. Semin Immunol 13:267–274 12. Schuetz EG, Furuya KN, Schuetz JD (1995) Interindividual variation in expression of P-glycoprotein in normal human liver and secondary hepatic neoplasms. J Pharmacol Exp Ther 275:1011–1018 13. Kullak-Ublick GA, Becker MB (2003) Regulation of drug and bile salt transporters in liver and intestine. Drug Metab Rev 35:305–317

116

Cascorbi and Haenisch

14. Geick A, Eichelbaum M, Burk O (2001) Nuclear receptor response elements mediate induction of intestinal MDR1 by rifampin. J Biol Chem 276:14581–14587 15. Fromm MF, Kauffmann HM, Fritz P et  al (2000) The effect of rifampin treatment on intestinal expression of human MRP transporters. Am J Pathol 157:1575–1580 16. Greiner B, Eichelbaum M, Fritz P et al (1999) The role of intestinal P-glycoprotein in the interaction of digoxin and rifampin. J Clin Invest 104:147–153 1 7. Giessmann T, May K, Modess C et  al (2004) Carbamazepine regulates intestinal P-glycoprotein and multidrug resistance protein MRP2 and influences disposition of talinolol in humans. Clin Pharmacol Ther 76:192–200 18. Johne A, Brockmoller J, Bauer S et al (1999) Pharmacokinetic interaction of digoxin with an herbal extract from St John’s wort (Hypericum perforatum). Clin Pharmacol Ther 66:338–345 19. Mannel M (2004) Drug interactions with St John’s wort: mechanisms and clinical implications. Drug Saf 27:773–797 20. Watkins RE, Maglich JM, Moore LB et  al (2003) 2.1 A crystal structure of human PXR in complex with the St. John’s wort compound hyperforin. Biochemistry 42:1430–1438 21. Pauli-Magnus C, von Richter O, Burk O et al (2000) Characterization of the major metabolites of verapamil as substrates and inhibitors of P-glycoprotein. J Pharmacol Exp Ther 293: 376–382 22. Ufer M, Dilger K, Leschhorn L et al (2008) Influence of CYP3A4, CYP3A5, and ABCB1 genotype and expression on budesonide pharmacokinetics: a possible role of intestinal CYP3A4 expression. Clin Pharmacol Ther 84:43–46 23. Hoffmeyer S, Burk O, von Richter O et  al (2000) Functional polymorphisms of the human multidrug-resistance gene: multiple sequence variations and correlation of one allele with P-glycoprotein expression and activity in  vivo. Proc Natl Acad Sci USA 97:3473–3478 24. Horinouchi M, Sakaeda T, Nakamura T et al (2002) Significant genetic linkage of MDR1 polymorphisms at positions 3435 and 2677: functional relevance to pharmacokinetics of digoxin. Pharm Res 19:1581–1585 25. Kim RB, Leake BF, Choo EF et  al (2001) Identification of functionally variant MDR1 alleles among European Americans and African Americans. Clin Pharmacol Ther 70:189–199 26. Tanabe M, Ieiri I, Nagata N et  al (2001) Expression of P-glycoprotein in human

27.

28.

29.

30.

31.

32. 33.

34.

35.

36.

37.

38.

placenta: relation to genetic polymorphism of the multidrug resistance (MDR)-1 gene. J Pharmacol Exp Ther 297:1137–1143 Ameyaw MM, Regateiro F, Li T et al (2001) MDR1 pharmacogenetics: frequency of the C3435T mutation in exon 26 is significantly influenced by ethnicity. Pharmacogenetics 11:217–221 Cascorbi I, Gerloff T, Johne A et  al (2001) Frequency of single nucleotide polymorphisms in the P-glycoprotein drug transporter MDR1 gene in white subjects. Clin Pharmacol Ther 69:169–174 Lee SS, Kim SY, Kim WY et al (2005) MDR1 genetic polymorphisms and comparison of MDR1 haplotype profiles in Korean and Vietnamese populations. Ther Drug Monit 27:531–535 Schaeffeler E, Eichelbaum M, Brinkmann U et al (2001) Frequency of C3435T polymorphism of MDR1 gene in African people. Lancet 358:383–384 Hitzl M, Drescher S, Van der KH et al (2001) The C3435T mutation in the human MDR1 gene is associated with altered efflux of the P-glycoprotein substrate rhodamine 123 from CD56+ natural killer cells. Pharmacogenetics 11:293–298 Brenner SS, Klotz U (2004) P-glycoprotein function in the elderly. Eur J Clin Pharmacol 60:97–102 Oselin K, Mrozikiewicz PM, Gaikovitch E, Pahkla R, Roots I (2003) Frequency of MRP1 genetic polymorphisms and their functional significance in Caucasians: detection of a novel mutation G816A in the human MRP1 gene. Eur J Clin Pharmacol 59:347–350 Wang D, Johnson AD, Papp AC, Kroetz DL, Sadee W (2005) Multidrug resistance polypeptide 1 (MDR1, ABCB1) variant 3435C>T affects mRNA stability. Pharmacogenet Genomics 15:693–704 Kimchi-Sarfaty C, Oh JM, Kim IW et  al (2007) A “silent” polymorphism in the MDR1 gene changes substrate specificity. Science 315: 525–528 Schaefer M, Roots I, Gerloff T (2005) In vitro transport characteristics discriminate wildtype MDR1 (ABCB1) from Ala893Ser and Ala893Thr polymorphisms. Eur J Clin Pharmacol 61:718 Ishikawa T, Sakurai A, Kanamori Y et al (2005) High-speed screening of human ATP-binding cassette transporter function and genetic polymorphisms: new strategies in pharmacogenomics. Methods Enzymol 400:485–510 Crouthamel MH, Wu D, Yang Z, Ho RJ (2006) A novel MDR1 G1199T variant alters

Pharmacogenetics of ATP-Binding Cassette Transporters and Clinical Implications

39.

40.

41.

42.

43.

44.

45. 46.

47.

48.

49.

50.

drug resistance and efflux transport activity of P-glycoprotein in recombinant Hek cells. J Pharm Sci 95:2767–2777 Siegmund W, Ludwig K, Giessmann T, Dazert P, Schroeder E, Sperker B, Warzok R, Kroemer HK, Cascorbi I (2002) The effects of the human MDR1 genotype on the expression of duodenal P-glycoprotein and disposition of the probe drug talinolol. Clin Pharmacol Ther 72:572–583 Kurata Y, Ieiri I, Kimura M et al (2002) Role of human MDR1 gene polymorphism in bioavailability and interaction of digoxin, a substrate of P-glycoprotein. Clin Pharmacol Ther 72:209–219 Sakaeda T, Nakamura T, Horinouchi M et al (2001) MDR1 genotype-related pharmacokinetics of digoxin after single oral administration in healthy Japanese subjects. Pharm Res 18:1400–1404 Taniguchi S, Mochida Y, Uchiumi T et  al (2003) Genetic polymorphism at the 5¢ regulatory region of multidrug resistance 1 (MDR1) and its association with interindividual variation of expression level in the colon. Mol Cancer Ther 2:1351–1359 Goto M, Masuda S, Saito H et  al (2002) C3435T polymorphism in the MDR1 gene affects the enterocyte expression level of CYP3A4 rather than Pgp in recipients of livingdonor liver transplantation. Pharmacogenetics 12:451–457 Eichelbaum M, Fromm MF, Schwab M (2004) Clinical aspects of the MDR1 (ABCB1) gene polymorphism. Ther Drug Monit 26:180–185 Sakaeda T (2005) MDR1 genotype-related pharmacokinetics: fact or fiction? Drug Metab Pharmacokinet 20:391–414 Johne A, Kopke K, Gerloff T et  al (2002) Modulation of steady-state kinetics of digoxin by haplotypes of the P-glycoprotein MDR1 gene. Clin Pharmacol Ther 72:584–594 Verstuyft C, Schwab M, Schaeffeler E et  al (2003) Digoxin pharmacokinetics and MDR1 genetic polymorphisms. Eur J Clin Pharmacol 58:809–812 Gerloff T, Schaefer M, Johne A et al (2002) MDR1 genotypes do not influence the absorption of a single oral dose of 1 mg digoxin in healthy white males. Br J Clin Pharmacol 54:610–616 Morita Y, Sakaeda T, Horinouchi M et  al (2003) MDR1 genotype-related duodenal absorption rate of digoxin in healthy Japanese subjects. Pharm Res 20:552–556 Drescher S, Schaeffeler E, Hitzl M et al (2002) MDR1 gene polymorphisms and disposition

51.

52.

53.

54.

55.

56.

57.

58. 59. 60.

61.

62.

117

of the P-glycoprotein substrate fexofenadine. Br J Clin Pharmacol 53:526–534 Plasschaert SL, Groninger E, Boezen M et al (2004) Influence of functional polymorphisms of the MDR1 gene on vincristine pharmacokinetics in childhood acute lymphoblastic leukemia. Clin Pharmacol Ther 76:220–229 Kishi S, Yang W, Boureau B, Morand S, Das S, Chen P, Cook EH, Rosner GL, Schuetz E, Pui CH, Relling MV (2004) Effects of prednisone and genetic polymorphisms on etoposide disposition in children with acute lymphoblastic leukemia. Blood 103:67–72 Dulucq S, Bouchet S, Turcq B et  al (2008) Multidrug resistance gene (MDR1) polymorphisms are associated with major molecular responses to standard-dose imatinib in chronic myeloid leukemia. Blood 112:2024–2027 Chowbay B, Li H, David M, Cheung YB, Lee EJ (2005) Meta-analysis of the influence of MDR1 C3435T polymorphism on digoxin pharmacokinetics and MDR1 gene expression. Br J Clin Pharmacol 60:159–171 Castagne V, Bonhomme-Faivre L, Urien S et al (2004) Effect of recombinant interleukin-2 pretreatment on oral and intravenous digoxin pharmacokinetics and P-glycoprotein activity in mice. Drug Metab Dispos 32:168–171 Sakaeda T, Nakamura T, Hirai M et al (2002) MDR1 up-regulated by apoptotic stimuli suppresses apoptotic signaling. Pharm Res 19:1323–1329 Siegmund W, Altmannsberger S, Paneitz A et al (2002) Effect of levothyroxine administration on intestinal P-glycoprotein expression: consequences for drug disposition. Clin Pharmacol Ther 72:256–264 Ando H, Yanagihara H, Sugimoto K et al (2005) Daily rhythms of P-glycoprotein expression in mice. Chronobiol Int 22:655–665 Loscher W, Klotz U, Zimprich F, Schmidt D (2009) The clinical impact of pharmacogenetics on the treatment of epilepsy. Epilepsia 50:1–23 Illmer T, Schuler US, Thiede C et al (2002) MDR1 gene polymorphisms affect therapy outcome in acute myeloid leukemia patients. Cancer Res 62:4955–4962 Jamroziak K, Mlynarski W, Balcerczak E et al (2004) Functional C3435T polymorphism of MDR1 gene: an impact on genetic susceptibility and clinical outcome of childhood acute lymphoblastic leukemia. Eur J Haematol 72:314–321 Kim DH, Park JY, Sohn SK et  al (2006) Multidrug resistance-1 gene polymorphisms associated with treatment outcomes in de novo acute myeloid leukemia. Int J Cancer 118: 2195–2201

118

Cascorbi and Haenisch

63. Stanulla M, Schaffeler E, Arens S et al (2005) GSTP1 and MDR1 genotypes and central nervous system relapse in childhood acute lymphoblastic leukemia. Int J Hematol 81: 39–44 64. Seedhouse CH, Grundy M, White P et  al (2007) Sequential influences of leukemiaspecific and genetic factors on p-glycoprotein expression in blasts from 817 patients entered into the National Cancer Research Network acute myeloid leukemia 14 and 15 trials. Clin Cancer Res 13:7059–7066 65. van der Holt B, Van den Heuvel-Eibrink MM, Van Schaik RH et al (2006) ABCB1 gene polymorphisms are not associated with treatment outcome in elderly acute myeloid leukemia patients. Clin Pharmacol Ther 80:427–439 66. Hur EH, Lee JH, Lee MJ et  al (2008) C3435T polymorphism of the MDR1 gene is not associated with P-glycoprotein function of leukemic blasts and clinical outcome in patients with acute myeloid leukemia. Leuk Res 32:1601–1604 67. Semsei AF, Erdelyi DJ, Ungvari I et al (2008) Association of some rare haplotypes and genotype combinations in the MDR1 gene with childhood acute lymphoblastic leukaemia. Leuk Res 32:1214–1220 68. Buda G, Maggini V, Galimberti S et al (2007) MDR1 polymorphism influences the outcome of multiple myeloma patients. Br J Haematol 137:454–456 69. Yang Z, Wu D, Bui T, Ho RJ (2008) A novel human multidrug resistance gene MDR1 variant G571A (G191R) modulates cancer drug resistance and efflux transport. J Pharmacol Exp Ther 327:474–481 70. Petrova DT, Nedeva P, Maslyankov S et  al (2008) No association between MDR1 (ABCB1) 2677G>T and 3435C>T polymorphism and sporadic colorectal cancer among Bulgarian patients. J Cancer Res Clin Oncol 134:317–322 71. Ho GT, Soranzo N, Nimmo ER et al (2006) ABCB1/MDR1 gene determines susceptibility and phenotype in ulcerative colitis: discrimination of critical variants using a gene-wide haplotype tagging approach. Hum Mol Genet 15:797–805 72. Kurzawski M, Drozdzik M, Suchy J et  al (2005) Polymorphism in the P-glycoprotein drug transporter MDR1 gene in colon cancer patients. Eur J Clin Pharmacol 61:389–394 73. Osswald E, Johne A, Laschinski G et  al (2007) Association of MDR1 genotypes with susceptibility to colorectal cancer in older non-smokers. Eur J Clin Pharmacol 63:9–16

74. Kafka A, Sauer G, Jaeger C et  al (2003) Polymorphism C3435T of the MDR-1 gene predicts response to preoperative chemotherapy in locally advanced breast cancer. Int J Oncol 22:1117–1121 75. Cole SP, Bhardwaj G, Gerlach JH et al (1992) Overexpression of a transporter gene in a multidrug-resistant human lung cancer cell line. Science 258:1650–1654 76. Conseil G, Deeley RG, Cole SP (2005) Polymorphisms of MRP1 (ABCC1) and related ATP-dependent drug transporters. Pharmacogenet Genomics 15:523–533 77. Conrad S, Kauffmann HM, Ito K et al (2002) A naturally occurring mutation in MRP1 results in a selective decrease in organic anion transport and in increased doxorubicin resistance. Pharmacogenetics 12:321–330 78. Ito S, Ieiri I, Tanabe M et al (2001) Polymorphism of the ABC transporter genes, MDR1, MRP1 and MRP2/cMOAT, in healthy Japanese subjects. Pharmacogenetics 11:175–184 79. Le Saux O, Urban Z, Tschuch C et al (2000) Mutations in a gene encoding an ABC transporter cause pseudoxanthoma elasticum. Nat Genet 25:223–227 80. Wang H, Hao B, Zhou K et al (2004) Linkage disequilibrium and haplotype architecture for two ABC transporter genes (ABCC1 and ABCG2) in Chinese population: implications for pharmacogenomic association studies. Ann Hum Genet 68:563–573 81. Letourneau IJ, Deeley RG, Cole SP (2005) Functional characterization of non-synonymous single nucleotide polymorphisms in the gene encoding human multidrug resistance protein 1 (MRP1/ABCC1). Pharmacogenet Genomics 15:647–657 82. Wang Z, Wang B, Tang K et al (2005) A functional polymorphism within the MRP1 gene locus identified through its genomic signature of positive selection. Hum Mol Genet 14:2075–2087 83. Wang H, Jin G, Wang H et al (2009) Genetic susceptibility of lung cancer associated with common variants in the 3¢ untranslated regions of the adenosine triphosphate-binding cassette B1 (ABCB1) and ABCC1 candidate transporter genes for carcinogen export. Cancer 115:595–607 84. Wojnowski L, Kulle B, Schirmer M et  al (2005) NAD(P)H oxidase and multidrug resistance protein genetic polymorphisms are associated with doxorubicin-induced cardiotoxicity. Circulation 112:3754–3762 85. Marsh S, Paul J, King CR et  al (2007) Pharmacogenetic assessment of toxicity and

Pharmacogenetics of ATP-Binding Cassette Transporters and Clinical Implications

86.

87.

88. 89.

90.

91.

92.

93.

94.

95.

outcome after platinum plus taxane chemotherapy in ovarian cancer: the Scottish randomised trial in ovarian cancer. J Clin Oncol 25:4528–4535 Konig J, Nies AT, Cui Y, Leier I, Keppler D (1999) Conjugate export pumps of the multidrug resistance protein (MRP) family: localization, substrate specificity, and MRP2mediated drug resistance. Biochim Biophys Acta 1461:377–394 Payen L, Sparfel L, Courtois A et  al (2002) The drug efflux pump MRP2: regulation of expression in physiopathological situations and by endogenous and exogenous compounds. Cell Biol Toxicol 18:221–233 Trauner M, Boyer JL (2003) Bile salt transporters: molecular characterization, function, and regulation. Physiol Rev 83:633–671 Vollrath V, Wielandt AM, Iruretagoyena M, Chianale J (2006) Role of Nrf2 in the regulation of the Mrp2 (ABCC2) gene. Biochem J 395:599–609 Haenisch S, Zimmermann U, Dazert E et al (2007) Influence of polymorphisms of ABCB1 and ABCC2 on mRNA and protein expression in normal and cancerous kidney cortex. Pharmacogenomics J 7:56–65 Itoda M, Saito Y, Soyama A et  al (2002) Polymorphisms in the ABCC2 (cMOAT/ MRP2) gene found in 72 established cell lines derived from Japanese individuals: an association between single nucleotide polymorphisms in the 5¢-untranslated region and exon 28. Drug Metab Dispos 30:363–364 Haenisch S, May K, Wegner D et al (2008) Influence of genetic polymorphisms on intestinal expression and rifampicin-type induction of ABCC2 and on bioavailability of talinolol. Pharmacogenet Genomics 18: 357–365 Naesens M, Kuypers DR, Verbeke K, Vanrenterghem Y (2006) Multidrug resistance protein 2 genetic polymorphisms influence mycophenolic acid exposure in renal allograft recipients. Transplantation 82: 1074–1084 Miura M, Satoh S, Inoue K et  al (2007) Influence of SLCO1B1, 1B3, 2B1 and ABCC2 genetic polymorphisms on mycophenolic acid pharmacokinetics in Japanese renal transplant recipients. Eur J Clin Pharmacol 63:1161–1169 Renders L, Frisman M, Ufer M et al (2007) CYP3A5 genotype markedly influences the pharmacokinetics of tacrolimus and sirolimus in kidney transplant recipients. Clin Pharmacol Ther 81:228–234

119

96. Choi JH, Ahn BM, Yi J et al (2007) MRP2 haplotypes confer differential susceptibility to toxic liver injury. Pharmacogenet Genomics 17:403–415 97. Daly AK, Aithal GP, Leathart JB et al (2007) Genetic susceptibility to diclofenac-induced hepatotoxicity: contribution of UGT2B7, CYP2C8, and ABCC2 genotypes. Gastro­ enterology 132:272–281 98. Izzedine H, Hulot JS, Villard E et al (2006) Association between ABCC2 gene haplotypes and tenofovir-induced proximal tubulopathy. J Infect Dis 194:1481–1491 99. Hirouchi M, Suzuki H, Itoda M et al (2004) Characterization of the cellular localization, expression level, and function of SNP variants of MRP2/ABCC2. Pharm Res 21:742–748 100. Ranganathan P, Culverhouse R, Marsh S et al (2008) Methotrexate (MTX) pathway gene polymorphisms and their effects on MTX toxicity in Caucasian and African American patients with rheumatoid arthritis. J Rheumatol 35:572–579 101. Niemi M, Arnold KA, Backman JT et al (2006) Association of genetic polymorphism in ABCC2 with hepatic multidrug resistanceassociated protein 2 expression and pravastatin pharmacokinetics. Pharmacogenet Genomics 16:801–808 102. Meier Y, Pauli-Magnus C, Zanger UM et al (2006) Interindividual variability of canalicular ATP-binding-cassette (ABC)-transporter expression in human liver. Hepatology 44: 62–74 103. Sookoian S, Castano G, Burgueno A, Gianotti TF, Pirola CJ (2008) Association of the multidrug-resistance-associated protein gene (ABCC2) variants with intrahepatic cholestasis of pregnancy. J Hepatol 48:125–132 104. Rau T, Erney B, Gores R et  al (2006) High-dose methotrexate in pediatric acute lymphoblastic leukemia: impact of ABCC2 polymorphisms on plasma concentrations. Clin Pharmacol Ther 80:468–476 105. Han JY, Lim HS, Yoo YK et  al (2007) Associations of ABCB1, ABCC2, and ABCG2 polymorphisms with irinotecanpharmacokinetics and clinical outcome in patients with advanced non-small cell lung cancer. Cancer 110:138–147 106. de Jong FA, Scott-Horton TJ, Kroetz DL et  al (2007) Irinotecan-induced diarrhea: functional significance of the polymorphic ABCC2 transporter protein. Clin Pharmacol Ther 81:42–49 107. Allikmets R, Schriml LM, Hutchinson A, Romano-Spica V, Dean M (1998) A human

120

Cascorbi and Haenisch

placenta-specific ATP-binding cassette gene (ABCP) on chromosome 4q22 that is invol­ ved in multidrug resistance. Cancer Res 58: 5337–5339 108. Doyle LA, Yang W, Abruzzo LV et al (1998) A multidrug resistance transporter from human MCF-7 breast cancer cells. Proc Natl Acad Sci USA 95:15665–15670 109. Bailey-Dell KJ, Hassel B, Doyle LA, Ross DD (2001) Promoter characterization and genomic organization of the human breast cancer resistance protein (ATP-binding cassette transporter G2) gene. Biochim Biophys Acta 1520:234–241 110. Kage K, Tsukahara S, Sugiyama T et al (2002) Dominant-negative inhibition of breast cancer resistance protein as drug efflux pump through the inhibition of S–S dependent homodimerization. Int J Cancer 97: 626–630 111. Ito K, Suzuki H, Horie T, Sugiyama Y (2005) Apical/basolateral surface expression of drug transporters and its role in vectorial drug transport. Pharm Res 22:1559–1577 112. Meyer Zu Schwabedissen HE, Grube M, Dreisbach A et al (2006) Epidermal growth factor (EGF) mediated activation of the MAP kinase cascade results in altered expression and function of ABCG2 (BCRP). Drug Metab Dispos 34:524–533 113. Sisodiya SM, Martinian L, Scheffer GL et  al (2006) Vascular colocalization of P-glycoprotein, multidrug-resistance associated protein 1, breast cancer resistance protein and major vault protein in human epileptogenic pathologies. Neuropathol Appl Neurobiol 32:51–63 114. Krishnamurthy P, Schuetz JD (2006) Role of ABCG2/BCRP in biology and medicine. Annu Rev Pharmacol Toxicol 46:381–410 115. Meissner K, Heydrich B, Jedlitschky G et al (2006) The ATP-binding cassette transporter ABCG2 (BCRP), a marker for side population stem cells, is expressed in human heart. J Histochem Cytochem 54:215–221 116. Jonker JW, Buitelaar M, Wagenaar E et  al (2002) The breast cancer resistance protein protects against a major chlorophyll-derived dietary phototoxin and protoporphyria. Proc Natl Acad Sci USA 99:15649–15654 117. Honjo Y, Hrycyna CA, Yan QW et al (2001) Acquired mutations in the MXR/BCRP/ ABCP gene alter substrate specificity in MXR/BCRP/ABCP-overexpressing cells. Cancer Res 61:6635–6639 118. Imai Y, Nakane M, Kage K et  al (2002) C421A polymorphism in the human breast cancer resistance protein gene is associated

with low expression of Q141K protein and low-level drug resistance. Mol Cancer Ther 1:611–616 119. Morisaki K, Robey RW, Ozvegy-Laczka C et  al (2005) Single nucleotide polymorphisms modify the transporter activity of ABCG2. Cancer Chemother Pharmacol 56: 161–172 120. Mizuarai S, Aozasa N, Kotani H (2004) Single nucleotide polymorphisms result in impaired membrane localization and reduced atpase activity in multidrug transporter ABCG2. Int J Cancer 109:238–246 121. Kondo C, Suzuki H, Itoda M et  al (2004) Functional analysis of SNPs variants of BCRP/ABCG2. Pharm Res 21:1895–1903 122. Zamber CP, Lamba JK, Yasuda K et al (2003) Natural allelic variants of breast cancer resistance protein (BCRP) and their relationship to BCRP expression in human intestine. Pharmacogenetics 13:19–28 123. Sparreboom A, Gelderblom H, Marsh S et al (2004) Diflomotecan pharmacokinetics in relation to ABCG2 421C>A genotype. Clin Pharmacol Ther 76:38–44 124. Sparreboom A, Loos WJ, Burger H et  al (2005) Effect of ABCG2 genotype on the oral bioavailability of topotecan. Cancer Biol Ther 4:650–658 125. Zamboni WC, Ramanathan RK, McLeod HL et al (2006) Disposition of 9-nitrocamptothecin and its 9-aminocamptothecin metabolite in relation to ABC transporter genotypes. Invest New Drugs 24:393–401 126. Cascorbi I (2006) Genetic basis of toxic reactions to drugs and chemicals. Toxicol Lett 162:16–28 127. Ingelman-Sundberg M (2004) Pharmaco­ genetics of cytochrome P450 and its applications in drug therapy: the past, present and future. Trends Pharmacol Sci 25:193–200 128. Ambudkar SV, Dey S, Hrycyna CA et  al (1999) Biochemical, cellular, and pharmacological aspects of the multidrug transporter. Annu Rev Pharmacol Toxicol 39:361–398 129. Dietrich CG, Geier A, Oude Elferink RP (2003) ABC of oral bioavailability: transporters as gatekeepers in the gut. Gut 52: 1788–1795 1 30. Ho RH, Kim RB (2005) Transporters and drug therapy: implications for drug disposition and disease. Clin Pharmacol Ther 78:260–277 131. Sakurai A, Tamura A, Onishi Y, Ishikawa T (2005) Genetic polymorphisms of ATPbinding cassette transporters ABCB1 and ABCG2: therapeutic implications. Expert Opin Pharmacother 6:2455–2473

Pharmacogenetics of ATP-Binding Cassette Transporters and Clinical Implications 132. Sarkadi B, Ozvegy-Laczka C, Nemet K, Varadi A (2004) ABCG2 – a transporter for all seasons. FEBS Lett 567:116–120 133. Suzuki H, Sugiyama Y (2002) Single nucleotide polymorphisms in multidrug resistance associated protein 2 (MRP2/ABCC2): its impact on drug disposition. Adv Drug Deliv Rev 54:1311–1331 134. Cascorbi I (2006) Role of pharmacogenetics of ATP-binding cassette transporters in the pharmacokinetics of drugs. Pharmacol Ther 112:457–473 135. Kimchi-Sarfaty C, Gribar JJ, Gottesman MM (2002) Functional characterization of coding

121

polymorphisms in the human MDR1 gene using a vaccinia virus expression system. Mol Pharmacol 62:1–6 136. Moriya Y, Nakamura T, Horinouchi M et al (2002) Effects of polymorphisms of MDR1, MRP1, and MRP2 genes on their mRNA expression levels in duodenal enterocytes of healthy Japanese subjects. Biol Pharm Bull 25:1356–1359 137. Kobayashi D, Ieiri I, Hirota T et  al (2005) Functional assessment of ABCG2 (BCRP) gene polymorphisms to protein expression in human placenta. Drug Metab Dispos 33:94–101

sdfsdf

Chapter 7 Flow Cytometric Evaluation of Multidrug Resistance Proteins Adorjan Aszalos and Barbara J. Taylor Abstract There are several ways to detect proteins on cells. One quite frequently used method is flow cytometry. This method needs fluorescently labeled antibodies that can attach selectively to the protein to be investigated for flow cytometric detection. Flow cytometry scans individual cells, virtually without their surrounding liquid, and can scan many cells in a very short time. Because of this advantage of flow cytometry, it was adapted to investigate transport proteins on normal and cancerous human cells and cell lines. These transport proteins play important roles in human metabolism. Absorption in the intestine, excretion at the kidney, protection of the CNS compartment and the fetus from xenobiotics, and other vital functions depend on these transporters. However, several transporters are overexpressed in cancer cells. These overexpressed transporters pump out anticancer drugs from the cells and prevent their curative effects. The detection and quantitation of these types of transporters in cancer cells is important for this reason. Here, we review literature on flow cytometric detection of the three most studied transporters: P-glycoprotein, multidrug resistance-associated proteins, and breast cancer resistance protein. Key words: Transport proteins, P-glycoprotein, Multidrug resistance protein, Breast cancer resistance protein, Flow cytometer

1. Introduction Flow cytometric detection and evaluation of three types of transport proteins will be discussed in this chapter: P-glycoprotein (Pgp, ABCB1), multidrug resistance-associated proteins (MRPs, ABCCs), and breast cancer resistance protein (BCRP, ABCG2). These transport proteins play normal physiological roles in the human body but also cause resistance to cancer chemotherapy. The physiological roles of these transporters are many: they include absorption of molecules in the intestine, regulating passage to the CNS compartment at the blood–brain J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_7, © Humana Press, a part of Springer Science + Business Media, LLC 2010

123

124

Aszalos and Taylor

barrier, passage of metabolites to the urine through the kidney, protecting the placenta from passing toxic agents to the fetus, and other functions at the liver, testis, and the lung. For further details on the expression of the transporters in humans, see refs. (1, 2). Any abnormality in the level of expression of these transporters can result in illness. One reason for resistance to cancer chemotherapy is that cancer cells overexpress some of these transporters and prevent the entry of the chemotherapeutic agents into the cancer cells to exert their curing effects. For these reasons, the analysis of these transporters has become very important in the last two decades. While there are other methods to analyze the presence and amount of these transporters in cells, one of the best methods to analyze, and especially quantitate transporters on the plasma membrane is flow cytometry (3). To distinguish among these transporters by flow cytometric methods, specific antibodies have been developed and specific inhibitors, or modifiers of their function, have been evaluated. Figure 7.1 shows the primary structure of Pgp and the known attachments for three surface antibodies to their epitopes. Antibodies, substrates, and modifiers are listed with the cited methodologies in the text. To distinguish among these three types of transporters, in general, the following molecules could be mentioned. Calcein AM, a fluorescent substrate, is a characteristic substrate for MRP1. Fumitremorgin C (FTC) is a specific inhibitor of BCRP. Rhodamine 123 is not specific but in

Fig. 7.1. Primary structure of P-glycoprotein as positioned in cell membrane, with three antibodies (MRK16, UIC-2, and 17F9) positioned at the known binding epitopes (Adapted from Ambudkar et al. (2003) P-glycoprotein: from genomics to mechanism. Oncogene 22:7468–7485, with permission of the author).

Flow Cytometric Evaluation of Multidrug Resistance Proteins

125

combination with verapamil, a Pgp substrate, can detect Pgp. Expression of the different transporters on cells can be distinguished with the combination of these agents, as detailed in some of the methodologies described later.

2. Evaluation of Pgp by Flow Cytometry

Several research groups have established protocols for detecting Pgp by flow cytometry for reasons of interest in their laboratories. We will now detail general guidelines provided by each of the main laboratories with established protocols. Next, we describe flow cytometric investigations as applied to some particular purposes. Broxterman et  al. analyzed the conditions for determining the amount of Pgp and its functionality on acute myeloid leukemia (AML) cells by flow cytometry (4). Their general recommendation for this evaluation can be summarized as follows. One can estimate the number of Pgp molecules on AML cells by using the antibody (Ab) MRK16, which binds to a surface epitope on the Pgp molecule. Other Abs binding to surface epitopes, such as UIC-2 can also be used. Phycoerythrin (PE)-labeled Abs yield higher sensitivity than fluorescein (FITC)-labeled Abs. The amount of Pgp expressed on AML cells can be determined using cell lines with a known amount of Pgp expressed under the same experimental conditions used for the AML cells. These authors used KB cell lines (KB-3-1, KB8, and KB-8-5), which each express different amounts of Pgp. These reference cell lines should be used fresh from frozen stocks after each 3 or 4 months of culturing. These cells tend to express higher levels of Pgp after longer culturing in the presence of the selecting agents. See Table 7.1 for flow cytometers used. Broxterman et  al. provide some general considerations concerning the functional analysis of Pgp expressed on cells (5). Pgp functionality is measured in order to find the transport capacity of the transporters. It also helps to determine the proper modulators for a particular patient. Fluorescent probes for accumulation and efflux studies include rhodamine 123, DiOC2(3), calceinAM, Hoechst 33342, BCECF-AM, Furo-2-AM, Fluo-3-AM, daunorubicin, and doxorubicin. The first three are particularly suitable to study Pgp modulator molecules. All these fluorescent probes have high ratios of active to passive transport and they equilibrate between cells and the medium relatively quickly. Equilibration of some of these probes, however, depends on certain characteristics of the cells studied, such as their membrane

126

Aszalos and Taylor

Table 7.1 Flow cytometers and characteristic substrates used for detection of P-glycoprotein Detectors and filters Cytometer

Laser

FL1

FL2

FL3

FL4

References

FACSCan

488

530

585

650

NA

(5, 7)

FACSCalibur

488

530

585

670

NA

(3, 14)

Epics-XL

488

525

575

620

670

(10)

FL1

FL2

FL3

FL4

Accumulation and efflux

Rhodamine 123 DiOC2(3) Calcein-AM

Doxorubicin Daunorubicin

–

–

Fluorochromes for antibodies

FITC

PE

PE-TRa

–

Viability

–

PI

7-AAD

7-AADc

b

FITC fluorescein, PE phycoerythrin, PE-TR PE-Texas red tandem, PI propidium iodide, 7-AAD 7-aminoactinomycin D a PE-Texas Red (PE-TR) emission is detected suboptimally in the 650 and 670 filters. More suitable fluorochromes may be PerCP and the PE tandem dyes PE-Cy5, PE-Cy5.5, or PE-Cy7 b FACScan and FACSCalibur c Epics-XL

potential, intracellular pH, intracellular Ca2+, and DNA content. It is important that the Pgp modulator does not interfere in any way, such as through fluorescence, with the fluorescence and transport of the fluorescent probe. In this regard, valspodar was found superior to cyclosporin A and verapamil. Because transport is ATP dependent, sufficient glucose should be present in the medium and the cells should be viable during the study. A group in England led by M. Pallis adapted the Dutch protocol with a slight modification that is based on the work of the group of Broxterman et  al. (5, 6). The adapted protocol was tested for expression and functionality of Pgp in AML and myelodysplastic syndrome cells in a multicenter trial in England. Leith et al. described a method for functional assay of Pgp in AML and cells of bone marrow origin (7). Differences in the efflux of the fluorescent substrate DiOC2(3) in the presence and absence of the modulator molecule can be measured over a time course according to Krishan (8, 9). In these experiments besides the forward vs. side scatter and forward scatter vs. fluorescence, time vs. fluorescence histograms also could be generated. A multicolor analysis for peripheral blood mononuclear cells (PBMCs) was established by Ford et  al. (10). The assay was established for detection of Pgp expression in subsets of blood cells of healthy people.

Flow Cytometric Evaluation of Multidrug Resistance Proteins

127

Feller at al. described a method to analyze Pgp in solid tumor cells (11). Depending on the origin of the cells isolated from solid tumors, the expression of Pgp on the cells as measured by flow cytometry may or may not correspond to the measured Pgp RNA. One reason for this possible discrepancy could be the heterogeneous expression of Pgp in the isolated cells. Aszalos and Weaver described a flow cytometric test for expression of Pgp on cell lines (12). Pgp on cells can be detected with the MRK16 mAb either labeled directly with FITC according to the instructions provided with the labeling kit or by using a secondary FITC Ab. Histograms showing fluorescence intensities when MRK16-FITC, MRK16 plus a secondary FITC Ab and isotype-matched Ab plus anti-isotype FITC Ab were used are shown in Fig. 7.2. Differences in fluorescence shifts between two histograms can be evaluated by the Kolmogorov–Smirnov statistics included with the flow cytometer software. See Table 7.1 for flow cytometers used. Numbers of Pgp molecules on clinical and in  vitro drugselected cells can be determined according to Aleman et al. (3). To determine number of Pgp molecules on cells, a series of beads with increasing numbers of fluorescein molecules is used. Standard fluorescent beads with defined numbers of fluorescein molecules were used as follows: 6,318, 15,877, 53,989, 82,914, 123,338, 170,473, 353,992, and 437,815 fluorochromes. Fluorescence intensities of the beads are obtained with a flow cytometer equipped with a 488-nm laser and a 530 emission filter. Beads are shaken well and mixed two intensities per tube for a total of four tubes. A graph is plotted from the means of individual histograms

Fig. 7.2. Differences in histogram intensity after binding anti-mouse FITC, MRK16-FITC, or MRK-16 + anti-mouse FITC to P-glycoprotein-expressing NIH3T3MDR cells (From Aszalos and Weaver (1998) Estimation of drug resistance by flow cytometry. In: Jaroszeski and Heller (eds) Flow cytometry protocols. Humana, Totowa, NJ, pp 117–122, Figure 2, p. 121, with kind permission of Springer Science).

128

Aszalos and Taylor

of the beads. Means of histograms obtained from 104 cells labeled with MRK16 plus antimouse IgG-FITC or another FITC Ab against a surface epitope of Pgp are matched against the graph. Both the beads and the cells are suspended in the same PBS buffer. The graph for the means of the histograms of the beads is plotted on semilogarithmic paper. Alternatively, QuickCal data analysis software (Bangs Laboratories) can be used. Means of histograms of fluorescence intensities obtained from the tested cells are matched against the plot obtained with the beads. See Table 7.1 for flow cytometers used. Beck et  al. described a consensus recommendation for detection of Pgp in patient’s tumors (13). A multinational workshop was organized for the detection of Pgp in clinical samples. The aim of this workshop was to standardize factors such as reagents, preparation of samples, detection of end-points, and methodology of analysis, in the determination of the role of Pgp in drug resistance in clinical evaluation of a patient’s treatment. The following recommendations were made: Hematological samples or disaggregated solid tumor cells are best analyzed by flow cytometry when the preparation is fresh. Samples can be cryopreserved at −135°C in 20–90% fetal bovine serum with 10% DMSO. Otherwise samples can be kept on ice for 24  h before analysis. Recommended Abs, such as MRK16, UIC-2, and 4E3 are best to use because they recognize external epitopes on the Pgp molecule. The advantage of using Abs that recognize an external epitope is that in flow cytometric analysis, correlation can be made by multicolor analysis with other surface antigens and with functional measurement of Pgp using dye accumulation/ efflux measurements. For flow cytometric detection of fluorochromes attached to primary or secondary Abs, PE is preferred over FITC. The reason for this recommendation is that PE has a higher quantum efficiency and therefore detection of low levels of Pgp expression is more accurate. Also, autofluorescence is less in the PE detector (585 nm) than the FITC (530 nm), so the signal-to-noise ratio is higher with PE. An isotypically matched Ab with the same fluorochrome should be used as a baseline control. Well-characterized cell lines with a known amount of Pgp expressed, as determined by mRNA and flow cytometric methods, should be used to validate flow cytometric assays in a particular laboratory. A cell line developed by Beck et al. (13), CEM/VLB, or the aforementioned KB cell line by Broxterman et al. could be used. For clinically relevant low-level expression of Pgp, the 8226/Dox6 cell line could be used. Normal cells expressing a low level of Pgp should be electronically gated out from the malignant cells based on surface marker expression. Reporting results can be done by evaluation of the mean channel shift between control and sample by the

Flow Cytometric Evaluation of Multidrug Resistance Proteins

129

Kolmogorov–Smirnov statistical method provided with most flow cytometers or by continuous variable data without a cut-off point for positivity. Drug efflux measurement is preferred to a drug accumulation test, using cyclosporin A, verapamil, or valspodar as modulators of Pgp function. For drug efflux measurements, the fluorescent substrates DiOC2(3), rhodamine 123, Hoechst 33342 or the drugs daunorubicin and doxorubicin could be used. The dyes have more favorable uptake and efflux kinetics. The flow cytometric efflux studies should be correlated with Ab-mediated Pgp expression determination. The reason for this necessary correlation is that transporters other than Pgp, such as MRPs, may be responsible for the efflux of the substrate. The consensus report indicates that a tumor’s resistance to chemotherapy does not necessarily correlate with Pgp expression. After considering the aforementioned recommendations, one can use materials and techniques found in several of the earlier described detailed analytical methods. Wang et al. described an assay for quantitative determination of modulation of the function of Pgp by compounds (14). The aim of this study is to quantitatively assess the ability of a compound to modulate the function of Pgp. For this purpose, increasing doses of the tested compound are used in the efflux assay and the inhibition of the function of Pgp is determined in relative % to that of ortho vanadate. While Wang et al. used the CR1R12 cell line, other cell lines exclusively expressing the transporter Pgp could be used for this study. Wang et al. found that 2 µM daunorubicin was the optimal concentration of substrate for the cell line they used. Among the tested compounds, cyclosporin A and progesterone gave the most inhibition of Pgp function, 75 and 60%, respectively. Verapamil and terfenidine inhibition were 40–50%, relative to orthovanadate. Note that second- and thirdgeneration Pgp modulators can achieve greater inhibition of Pgp function than the compounds tested by Wang et al. See Table 7.1 for flow cytometers used.

3. Evaluation of MRPs by Flow Cytometry

Several MRPs have been characterized and described in the literature. Later we detail the known flow cytometric evaluation of some of these transport proteins from laboratories involved in their particular research. Janneh et al. evaluated the expression of Pgp and MRPs in peripheral blood mononuclear cells (PBMCs) for the purpose of determining the interaction of various protease inhibitors at the

130

Aszalos and Taylor

level of Pgp and several MRPs (15). In their study, they determined the uptake of (14C) lopinavir into PBMCs and some specific cell lines expressing various transporter proteins, in combination with specific inhibitors of Pgp, MRP1, MRP2, and organic anion transporter protein. Following this determination, they evaluated the modulating effect of several other protease inhibitors at the level of the transport proteins. Flow cytometry served to determine the presence of MRP1 in PBMCs, obtained from patient buffy coats. The specific modulators used by Janneh et al. (15) in their study are worth mentioning despite the fact that they did not use them in the flow cytometric experiment: tariquidar (Pgp specific), MK571 (MRP specific), frusemid (MRP1/2 specific), dipyridamole (MRP1/Pgp specific), and probenecid (MRP2/OATP specific). These specific blockers can be used to differentiate among the transport proteins for the efflux of drugs and compounds from cells. See Table 7.2 for flow cytometers used. Feller at al. evaluated the expression of MRP in several cell lines (11). Their aim was to find the best combination of fluorescent substrate and modifier of the function of MRP. They came to the conclusion that the best probe to detect the specific function of MRP1 by flow cytometry is to use daunorubicin as substrate and genistein as modulator of the function of MRP1. They also concluded that genistein decreases the fluorescence of rhodamine 123 and calcein-AM in sensitive cells, and therefore these fluorescent substrates cannot be used together with the specific modulator of MRP1, genistein. They also found that valspodin and vincristine are not suitable substrates of MRP1. Meaden et al. compared the expression of MRP in PBMCs between HIV-infected and noninfected patients (16). They concluded that the expression of MRP in PBMCs is the same in HIV-infected and noninfected patients. (Pgp expression is less in HIV infected than noninfected patients.) See Table 7.2 for flow cytometers used. Braga et  al. compared the expression of MRP1 and Pgp in cells expressing both transporters (17). For detection of MRP1 they used carboxy fluorescein diacetate (CFDA), and for Pgp, rhodamine 123. The CFDA is nonfluorescent as is, but is hydrolyzed in cells by esterases to the fluorescent derivative. The two probes can distinguish between Pgp and MRP1 (18), as rhodamine 123 is a substrate of Pgp and CFDA a substrate of MRP1. Braga et al. analyzed the effect of oleanolic acid on the transport properties of the two transporters. They found that oleanolic acid inhibits the function of MRP1 but not that of Pgp. McAleer et al. studied the characteristics of MRP5 (MOAT-C, ABCC11) and determined the substrate specificity of this transporter by flow cytometry (19). A comparison was made between stably-transfected and nontransfected HEKc10 cells using probes,

Flow Cytometric Evaluation of Multidrug Resistance Proteins

131

Table 7.2 Flow cytometers and characteristic substrates used for detection of multidrug resistance-associated protein Detectors and filters Cytometer

Laser

FL1

FL2

FL3

FL4

References

FACSCan

488

530

585

650

NA

(11, 19)

FACSCalibur

488

530

585

670

NA

(17)

Epics-XL

488

525

575

620

670

(15)

FL1

FL2

FL3

FL4

Accumulation and efflux

CMFDA CFDA FDA BCECF-AM Calcein-AM

Doxorubicin Daunorubicin TMR

–

–

Fluorochromes for antibodies

FITC

PE

–

–

FITC fluorescein, PE phycoerythrin, CMFDA 5-chloromethyl fluorescein diacetate, CFDA carboxy fluorescein diacetate, FDA fluorescein diacetate, BCECF-AM 2¢,7¢-bis-(2carboxyethyl)-5(and 6-)carboxy fluorescein acetoxymethyl ester, TMR tetra methyl rosamine

5-chloromethyl fluorescein diacetate (CMFDA), fluorescein diacetate (FDA), 2¢,7¢-bis-(2carboxyethyl)-5(and 6-)carboxy fluorescein acetoxymethyl ester (BCECF-AM), daunorubicin, tetra methyl rosamine (TMR), and calcein-AM. Flow cytometric studies indicated that CMFDA is a substrate of MRP5, but daunorubicin, calcein-AM, and TMR are not. They also found by a non-flow cytometric method, fluorometry, that FDA and BCFCF-AM are also substrates of MRP5. See Table 7.2 for flow cytometers used. Leidert et  al. analyzed the influence of MRP2 (cMOAT, ABCC2) expression in melanoma cells on platinum and DNA adduct formation (20). They found that an inverse correlation exists between expression of MRP2 and adduct formation. In connection with this study, they performed cell cycle analysis on cisplatin-treated cells. This analysis indicated a cisplatin-triggered G2 arrest in both sensitive and resistant cells. MRP expression was done by Northern blot and RT-PCR analyses and not by flow cytometry. One should mention in connection with this study that flow cytometry would have analyzed MPR1 expression on the cell membrane, while the Northern blot plus RT-PCR analyzed all MRP1s in the membrane plus in the cytoplasm. Other MRPs (MRP3, MRP4, MRP7, and MRP8) have been characterized in membrane vesicles and not by flow cytometry. Antibodies have been developed against two of them, MRP3 and

132

Aszalos and Taylor

MRP6. Provided these antibodies are available, references for MRP3 (21) and MRP6 (22) can be found in the reference section for potential use in flow cytometry.

4. Evaluation of BCRP by Flow Cytometry

BCRP is found in the human placenta, bile canaliculi, the colon, small bowel, and in brain microvessel endothelium. It is overexpressed in breast and leukemia cancer tissues. In normal tissues, this transporter protects the organs from potentially toxic xenobiotics. A complete treatment of this transport molecule, including genetics, chemistry, modulators of BCRP, transported molecules, antibodies to BCRP, mutation variants of BCRP, and physiological function was published by Doyle and Ross (23). BCRP effluxes the substrates mitoxantrone, daunorubicin, bisantrene, prozasin, rhodamine123 (only in BCRP mutation variants), topotecan, and LysoTracker. The Pgp substrates verapamil (low dose), vinblastine, paclitaxel, and the MRP1 substrate calcein are not transported by BCRP. Based on these different transport properties, these transporters can be distinguished from one another by flow cytometry. Any standard flow cytometer with a 488-nm laser can be used for detection of BCRP. For these studies, either a FACSCan or a one- or two-laser FACSCalibur was used. The antibodies available (5D3, BXP-21, BXP-34) can be used with FITC (FL1, 530 nm) or PE (FL2, 575 or 585 nm). Substrates transported by BCRP include topotecan, BODIPY-prazosin, pheophorbide a, and BBR 3390 (all detected in FL1, 530  nm); topotecan, daunorubicin, and doxorubicin (all detected in FL2, 575 or 585  nm); and mitoxantrone (detected in FL3, 650 or 670 nm). For increased sensitivity with mitoxantrone, a 2-laser cytometer with a 633–639 laser can be used for excitation with detection in a 660/20 filter. Flow cytometers equipped with a 488-nm laser and a 355-nm laser may be used to detect Hoechst 33342 with SP (side population) cells detected at two emission wavelengths; blue at 424/44 nm and red at 675 nm LP with the signal split by a 640nm LP dichroic mirror. In the study cited here, a FACS Vantage was used. See Tables 7.3 and 7.4. There is no unified flow cytometric experiment to detect BCRP. Therefore, experiments used to measure BCRP functions in cells for different purposes are detailed later in connection with the aims of the individual investigators. Detailed descriptions of the flow cytometers are given earlier. A group led by a scientist at Roswell Park Cancer Institute did a basic study. Minderman et al. (24) studied the sensitivity of

Flow Cytometric Evaluation of Multidrug Resistance Proteins

133

Table 7.3 Flow cytometers and characteristic substrates used for detection of breast cancer resistance protein Detectors and filters Cytometer

Laser

FL1

FL2

FL3

FL4

References

FACSCan

488

530

585

650

NA

(24, 25)

FACSCalibur

488, 635

530

585

670

660

(26, 27, 30)

Epics-XL

488

525

575

620

670

(10)

FL1

FL2

FL3

FL4

Accumulation and efflux

Topotecan BODIPYprazosin Pheophorbide a BBR3390 Rhodamine 123c DiOC2(3)c Calcein-AMc

Topotecan Doxorubicin Daunorubicin

Mitoxantronea

Mitoxantroneb

Fluorochromes for antibodies

FITC GFP

PE

–

–

Viability

–

PI

–

–

FITC fluorescein, PE phycoerythrin, PI propidium iodide, GFP green fluorescent protein a FACSCalibur and Epics-XL, 488 laser, emission at 670 (Calibur) or 620 (Epics) b FACSCalibur, 635-nm laser, emission at 660 c These substrates are used to exclude Pgp and MRP1

Table 7.4 Side population cells for detection of breast cancer resistance protein Detectors, emission filters, and dichroic mirrors Cytometer

Lasers

UV–blue

UV–red

Dichroic mirror

PIa

7-AADa

References

FACS Vantage

488, 355

424/44

675LP

640LP

585

650

(28)

Mo-Flo LSRII

488, 355 488, 355

For these cytometers, combinations of parameters as given below in footnote b can be usedb

The viability dyes propidium iodide or 7-aminoactinomycin D may be used; both are excited at 488 nm A number of filter combinations have been used for the measurement of the blue and red emissions of the Hoechst 33342 dye, which is excited at 351–364 nm. These include 440/40, 450/50, and 450/20 bandpass filters for the blue, 670LP for the red, and 600LP, 610LP, or 635LP for the dichroic mirror a

b

two antibodies (Abs) to BCRP in MCF-MX8, MCF-AdVp3000, and 8226MR20 mitoxantrone-resistant cells and in their parental cell lines. Wild type HL60, the Pgp-expressing cell line HL60/ Adr, and the MRP1-expressing cell line A2780/Dx5b were

134

Aszalos and Taylor

included in the study. The BCRPT482 cell line (containing a mutated BCRP in which arginine 482 was replaced with threonine) was also studied. Their aim was to determine the sensitivity of two Abs, BXP-21 and BXP-34, to two epitopes on BCRP. The detection sensitivity was assessed by flow cytometry and immunohistochemistry. The study included different fluorescent molecules as possible substrates and modulators of BCRP function. Among the different fluorescent molecules, only mitoxantrone was found to be a substrate of BCRP, and therefore its uptake and efflux were studied in the presence and absence of the BCRP modulator. The fluorescence histograms obtained with the fluorescently labeled goat anti-mouse secondary Ab were evaluated with the D parameter of the Kolmogorov–Smirnov statistical method that is included in flow cytometry software. Appropriate isotype control Abs (IgG1 and IgG2a) were used to obtain baseline fluorescence. Excitation was at 488-nm wavelengths with an argon laser and detection at 530 nm for the FITC-conjugated Abs. To evaluate the mitoxantrone concentration in cells, two different excitation wavelengths were used: a 635-nm red diode laser with a 661-nm emission filter and a 488-nm laser with a 670-nm emission filter. The two Abs bind only at internal epitopes. For this reason, cells had to be fixed in formaldehyde for 10 min at RT followed by 90% methanol treatment for 10 min to permeabilize the plasma membrane. Cells were then incubated with the primary Abs or the isotype Abs at 4°C for 60 min. After washing with PBS with 0.01% Tween, cells were incubated with the secondary fluorescent Ab for 20 min at 4°C. Results of the flow cytometry indicated that the two Abs bind only to wild type or mutated BCRP but not to Pgp or MRP1 or lung resistant protein (LRP) in cells. The presence of BCRP was verified by statistical evaluation when the D value was higher than 0.2. This cut off point was in agreement with similar flow cytometric analysis for the positive expression of MRP1 and LRP when cells must be permeabilized for Ab binding to internal epitopes (7). Experiments with both primary Abs resulted in qualitatively the same results, although the D values varied somewhat. Interestingly, the mutant BCRP in 8226/MR20 cells could be detected with flow cytometry with the two Abs but not with immunohistochemistry. Uptake and efflux of mitoxantrone, a substrate of BCRP, was studied with or without modulator molecules. Using excitation at 488 nm and emission at >670 nm was sensitive enough to detect mitoxantrone in cells with low expression of BCRP. A mitoxantrone concentration of 0.01 µM is required for the detection with excitation of the 488-nm laser and emission at 635  nm. Efflux studies were done with the substrates mitoxantrone, DiOC2 (3), rhodamine 123, and doxorubicin in Pgp, MRP1, and wild-type and mutant BCRP-expressing cells. The efflux difference between

Flow Cytometric Evaluation of Multidrug Resistance Proteins

135

measurements in the presence or absence of transporter modulators was recorded. The modulator for Pgp was valspodar 2.5 µM, for MRP1 probenicid 1 mM, and for BCRP fumitremorgin C (FTC) 10 µM. D values were calculated by the Kolmogorov–Smirnov method from histograms obtained with or without the appropriate modulator. The D values were indicative of the extent of modulation of efflux of substrates by the applied modulator of the efflux pump. For example, the D value with mitoxantrone was above 0.29 for all transporter molecules in each cell (on a scale of 0–1), indicating that the selected modulators blocked the function of all the transporters and that mitoxantrone is transported by all these transporters. Contrary to this, D values indicated that efflux of rhodamine123 was substantially blocked in Pgp- and MRP1-expressing cells but only slightly inhibited in BCRP-expressing cells with the appropriate modulators. The method developed by Minderman et al. is well suited to analyzing clinical samples for expression of BCRP because a flow cytometer equipped with the standard 488-nm argon laser could be used (24). Detection of BCRP with the two available Abs would indicate the presence of BCRP even in cells with low levels of expression. Measuring efflux with mitoxantrone can indicate the extent of expression when FTC, a selective modulator of BCRP, is used. Another flow cytometric method was worked out by Kawabata et al. for detecting BCRP in clinical tissues with various levels of the BCRP mRNA (25). The aim of their investigation was to assess how much BCRP expression constitutes drug resistance in lung cancer. They measured BCRP mRNA levels in cell lines known to express various levels of BCRP. They then measured topotecan retention in the same cell lines by flow cytometry and correlated the results of the two methods. The cell lines they used were PC-6/SN2-5, PC-6, NCI-H460, NCI-H441, NCI-H358, and NCI-H69. PC-6 and NCI-H69 did not efflux much topotecan, indicating low levels of BCRP expression. PL-6/SN1-5 effluxed much more topotecan, indicating high levels of expression of BCRP. The other cell lines demonstrated intermediate levels of efflux. Flow cytometry measurements by Kawabata et al. used topotecan as a fluorescent substrate of BCRP because topotecan is a good substrate of even mutant BCRPs. Kawabata et  al. obtained excellent correlation between BCRP mRNA expression, as determined by real-time RT-PCR analysis, and fluorescence intensity of cells after efflux of topotecan as measured by flow cytometry. They selected NCI-H441 from among the cell lines as borderline BCRP-expressing cells, as the amount of BCRP in these cells is the amount necessary to confer drug resistance. After establishing this correlation and establishing this base line expression of BCRP, 23 nontreated non-small cell lung carcinoma tissues were examined by the two methods for BCRP expression.

136

Aszalos and Taylor

Their results indicated that 22% of the tumor tissues had higher efflux-related resistance than the NCI-H441 cells, conferring transport protein-based drug resistance. Rachel Ee and colleagues used flow cytometry to assess the results of small interfering RNA (siRNA) treatment of BCRPexpressing cells to suppress the expression of this transport protein (26). They used essentially the same system as Kawabata et  al., detailed earlier. They used topotecan as a fluorescent substrate, and its accumulation in cells was assessed by a FACSCalibur instrument using 488-nm excitation wavelength. BeWo choriocarcinoma cells, treated or nontreated with siRNA for 24 h, were incubated with 30 µM/L topotecan for 15 min at 37°C and flow cytometric measurement followed. Rabindran et al. examined the effect of fumitremorgin C (FTC) on the retention of the dye BBR 3390, 5 µM and daunorubicin, 1 µM, in BCRP-expressing MCF-7 cells (27). An interesting study was done by Scharenberg et  al. who analyzed the extent of expression of BCRP, Pgp, and MRP1 in lung carcinoma A549, human embryonic kidney HEK293, and several human leukemia cell lines (28). The aim of their study was to characterize the efflux protein expression in a “side population” (SP) of the cell lines as indicated by Hoechst 33342 dye retention and efflux patterns. Hoechst dye excluding cells sorted out from bone marrow and other tissues contain immature stem cells, isolated as CD34+/CD38− or CD34+/KDR+ cells. They wanted to determine how many of these stem cells were among the studied cell lines and what type of transport proteins are expressed in them. To answer these questions, flow cytometric efflux studies were done with the fluorescent Hoechst dye and efflux protein modulators probenicid, verapamil, and FTC to determine the possible existence of the three efflux pumps. They found that FTC blocks most efflux of Hoechst dye, indicating that BCRP is the predominant transporter in A549 cells. RT-PCR experiments supported these findings. They also found that the two other transporters, Pgp and MRP1 are also expressed, but in minor quantities. A characteristic UV–red vs. UV–blue chart indicates the presence of stem cells expressing BCRP (Fig. 7.3). BCRP has several variants. The V12M and Q141K variants were described by Zamber et  al. (29). Three other variants, I206L, N590Y, and D620N, were studied by Vethanayagam et al. for their expression levels and functionalities (30). Expression levels were determined by immunoblotting and functionality by efflux measurements with flow cytometry. All three variants transported mitoxantrone, pheophorbide a, and BODIPY-prazosin. Doyle et al. showed that the resistance of BCRP-expressing cells could be partially reversed by the antibiotic novobiocin (31). In their studies, they used the “comparative growth assay” system developed by Hausner et al. (32).

Flow Cytometric Evaluation of Multidrug Resistance Proteins

137

Fig. 7.3. Influence of inhibitors on the side population of A549 cells stained with Hoechst 33342 dye (From Scharenberg et al. (2002) The ABCG2 transporter is an efficient Hoechst 33342 efflux pump and is preferentially expressed by immature human hematopoietic progenitors. Blood 99:507–512, used by permission).

Acknowledgments This work was funded by the Intramural Research Program of the National Institutes of Health, National Cancer Institute. We would like to thank Dr. Michael Gottesman for his hospitality and encouragement and George Leiman for editorial assistance.

References 1. Schinkel AH, Jonker JW (2003) Mammalian drug efflux transporters of the ATP binding cassette (ABC) family: an overview. Adv Drug Deliv Rev 55:3–29 2. Szakacs G, Paterson JK, Ludwig JA, BoothGenthe C, Gottesman MM (2006) Targeting multidrug resistance in cancer. Nat Rev Drug Discov 5:219–234

3. Aleman C, Annereau JP, Liang XJ et al (2003) P-glycoprotein, expressed in multidrug resistant cells, is not responsible for alterations in membrane fluidity or membrane potential. Cancer Res 63:3084–3091 4. Broxterman HJ, Lankelma J, Pinedo HM et al (1997) Theoretical and practical considerations for the measurement of P-glycoprotein

138

Aszalos and Taylor

function in acute myeloid leukemia. Leukemia 11:1110–1118 5. Broxterman HJ, Lankelma J, Pinedo HM (1996) How to probe clinical tumour samples for P-glycoprotein and multidrug resistance-associated protein. Eur J Cancer 32A: 1024–1033 6. Pallis M, Das-Gupta E (2005) Flow cytometric measurement of functional and phenotypic P-glycoprotein. Methods Mol Med 111:167–181 7. Leith CP, Kopecky KJ, Chen IM et al (1999) Frequency and clinical significance of the expression of the multidrug resistance proteins MDR1/P-glycoprotein, MRP1, and LRP in acute myeloid leukemia: a Southwest Oncology Group Study. Blood 94:1086–1099 8. Krishan A (2000) Monitoring of cellular resistance to cancer chemotherapy: drug retention and efflux. In: Darzynkiewicz Z, Crissman HA, Robinson JP (eds) Cytometry, part B, 3rd edn. Academic, San Diego, CA, pp 193–209 9. Krishan A (2002) Flow cytometric monitoring of drug resistance in human tumor cells. Methods Cell Sci 24:55–60 10. Ford J, Hoggard PG, Owen A, Khoo SH, Back DJ (2003) A simplified approach to determining P-glycoprotein expression in peripheral blood mononuclear cell subsets. J Immunol Methods 274:129–137 11. Feller N, Kuiper CM, Lankelma J et al (1995) Functional detection of MDR1/P170 and MRP/P190-mediated multidrug resistance in tumour cells by flow cytometry. Br J Cancer 72:543–549 12. Aszalos A, Weaver JL (1998) Estimation of drug resistance by flow cytometry. In: Jaroszeski MJ, Heller R (eds) Flow cytometry protocols. Humana, Totowa, NJ, pp 117–122 13. Beck WT, Grogan TM, Willman CL et  al (1996) Methods to detect P-glycoproteinassociated multidrug resistance in patients’ tumors: consensus recommendations. Cancer Res 56:3010–3020 14. Wang EJ, Casciano CN, Clement RP, Johnson WW (2000) In vitro flow cytometry method to quantitatively assess inhibitors of P-glycoprotein. Drug Metab Dispos 28: 522–528 15. Janneh O, Jones E, Chandler B, Owen A, Khoo SH (2007) Inhibition of P-glycoprotein and multidrug resistance-associated proteins modulates the intracellular concentration of lopinavir in cultured CD4 T cells and primary human lymphocytes. J Antimicrob Chemother 60:987–993

1 6. Meaden ER, Hoggard PG, Maher B, Khoo SH, Back DJ (2001) Expression of P-glycoprotein and multidrug resistanceassociated protein in healthy volunteers and HIV-infected patients. AIDS Res Hum Retroviruses 17:1329–1332 17. Braga F, Ayres-Saraiva D, Gattass CR, Capella MA (2007) Oleanolic acid inhibits the activity of the multidrug resistance protein ABCC1 (MRP1) but not of the ABCB1 (P-glycoprotein): possible use in cancer chemotherapy. Cancer Lett 248:147–152 18. Echevarria-Lima J, Kyle-Cezar F, DF PL, Capella L, Capella MA, Rumjanek VM (2005) Expression and activity of multidrug resistance protein 1 in a murine thymoma cell line. Immunology 114:468–475 19. McAleer MA, Breen MA, White NL, Matthews N (1999) pABC11 (also known as MOAT-C and MRP5), a member of the ABC family of proteins, has anion transporter activity but does not confer multidrug resistance when overexpressed in human embryonic kidney 293 cells. J Biol Chem 274:23541–23548 20. Liedert B, Materna V, Schadendorf D, Thomale J, Lage H (2003) Overexpression of cMOAT (MRP2/ABCC2) is associated with decreased formation of platinum-DNA adducts and decreased G2-arrest in melanoma cells resistant to cisplatin. J Invest Dermatol 121:172–176 21. Young LC, Campling BG, Cole SP, Deeley RG, Gerlach JH (2001) Multidrug resistance proteins MRP3, MRP1, and MRP2 in lung cancer: correlation of protein levels with drug response and messenger RNA levels. Clin Cancer Res 7:1798–1804 22. Belinsky MG, Guo P, Lee K et  al (2007) Multidrug resistance protein 4 protects bone marrow, thymus, spleen, and intestine from nucleotide analogue-induced damage. Cancer Res 67:262–268 23. Doyle LA, Ross DD (2003) Multidrug resistance mediated by the breast cancer resistance protein BCRP (ABCG2). Oncogene 22:7340–7358 24. Minderman H, Suvannasankha A, O’Loughlin KL et  al (2002) Flow cytometric analysis of breast cancer resistance protein expression and function. Cytometry 48:59–65 25. Kawabata S, Oka M, Soda H et  al (2003) Expression and functional analyses of breast cancer resistance protein in lung cancer. Clin Cancer Res 9:3052–3057 26. Ee PL, He X, Ross DD, Beck WT (2004) Modulation of breast cancer resistance protein (BCRP/ABCG2) gene expression using RNA interference. Mol Cancer Ther 3:1577–1583

Flow Cytometric Evaluation of Multidrug Resistance Proteins 27. Rabindran SK, Ross DD, Doyle LA, Yang W, Greenberger LM (2000) Fumitremorgin C reverses multidrug resistance in cells transfected with the breast cancer resistance protein. Cancer Res 60:47–50 28. Scharenberg CW, Harkey MA, Torok-Storb B (2002) The ABCG2 transporter is an efficient Hoechst 33342 efflux pump and is preferentially expressed by immature human hematopoietic progenitors. Blood 99:507–512 29. Zamber CP, Lamba JK, Yasuda K et al (2003) Natural allelic variants of breast cancer resistance protein (BCRP) and their relationship to BCRP expression in human intestine. Pharmacogenetics 13:19–28

139

30. Vethanayagam RR, Wang H, Gupta A et  al (2005) Functional analysis of the human variants of breast cancer resistance protein: I206L, N590Y, and D620N. Drug Metab Dispos 33:697–705 31. Doyle LA, Yang W, Gao Y, Ordonez JV, Ross DD (1996) Novobiocin increases the accumulation of daunorubicin in an atypical multidrug-resistant breast cancer subline. Proc Am Soc Clin Oncol 15:398 32. Hausner P, Venzon DJ, Grogan L, Kirsch IR (1999) The “comparative growth assay”: examining the interplay of anti-cancer agents with cells carrying single gene alterations. Neoplasia 1:356–367

sdfsdf

Chapter 8 Targeted Chemotherapy in Drug-Resistant Tumors, Noninvasive Imaging of P-Glycoprotein-Mediated Functional Transport in Cancer, and Emerging Role of Pgp in Neurodagenerative Diseases Jothilingam Sivapackiam, Seth T. Gammon, Scott E. Harpstrite, and Vijay Sharma Abstract Multidrug resistance (MDR) mediated by overexpression of P-glycoprotein (Pgp) is one of the best characterized transporter-mediated barriers to successful chemotherapy in cancer patients and is also a rapidly emerging target in the progression of neurodegenerative disorders such as Alzheimer’s and Parkinson’s diseases. Therefore, strategies capable of delivering chemotherapeutic agents into drug-resistant tumors and targeted radiopharmaceuticals acting as ultrasensitive molecular imaging probes for detecting functional Pgp expression in vivo could be expected to play a vital role in systemic biology as personalized medicine gains momentum in the twenty-first century. While targeted therapy could be expected to deliver optimal doses of chemotherapeutic drugs into the desired targets, the interrogation of Pgp-mediated transport activity in vivo via noninvasive imaging techniques (SPECT and PET) would be beneficial in stratification of patient populations likely to benefit from a given therapeutic treatment, thereby assisting management of drug resistance in cancer and treatment of neurodegenerative diseases. Both strategies could play a vital role in advancement of personalized treatments in cancer and neurodegenerative diseases. Via this tutorial, authors make an attempt in outlining these strategies and discuss their strengths and weaknesses. Key words: Multidrug resistance, P-glycoprotein, SPECT, PET, Metal complexes, Gallium-67/68, Technetium-99m/94m, Chemotherapeutics, Drug transport, Blood–brain barrier, Cancer, Neurodegenerative diseases, Alzheimer’s disease, Parkinson’s disease

1. Introduction Resistance to chemotherapy represents a major obstacle in the treatment of cancer. Many tumors are intrinsically resistant to chemotherapy, whereas others initially respond to treatment, but J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_8, © Humana Press, a part of Springer Science + Business Media, LLC 2010

141

142

Sivapackiam et al.

acquire resistance to selected cytotoxic drugs during chemotherapy. Multidrug resistance (MDR) can be broadly defined as a phenomenon by which tumor cells in vivo and cultured cells in vitro show resistance simultaneously to a variety of structurally and functionally dissimilar cytotoxic and xenobiotic compounds (1–6). While several different genes have been shown to be associated with a MDR phenotype (7–9), MDR mediated by overexpression of the MDR1 gene product, P-glycoprotein (Pgp), represents one of the best characterized barriers to chemotherapeutic treatment in cancer (6, 10). Pgp, a 170-KDa plasma membrane protein, is predicted by sequence analysis to comprise two symmetrical halves that share both homology with a family of ATP-binding cassette (ABC) membrane transport proteins and a common ancestral origin with bacterial transport systems (5, 11). Characterized by 12 transmembrane domains (TMDs) and two nucleotide-binding folds (Fig. 8.1) (3, 5), the protein is thought to hydrolyze ATP to affect outward transport of substrates across or off the cell surface membrane (5, 12). Although the specific protein domains and amino acids involved in substrate recognition continue to be characterized, genetic and biochemical evidence has conventionally been interpreted to show putative membraneassociated domains interacting directly with selected cytotoxic agents to affect transport (10, 13–15). Recently, Pgp from Chinese hamster was purified using dodecyl maltoside, crystals were grown using standard methods, and three-dimensional structure was determined with a resolution limit of 8 Å using cryoelectron microscopy (16). The structure demonstrated that five of the a-helices from each of the TMDs are related by a pseudo-twofold symmetry and two a-helices positioned closest to the axis of symmetry are slightly twisted. The deviation from a true twofold symmetry in the TMD region is likely a consequence of conformational changes induced by the nucleotide binding and results are consistent with the hypothesis that conformational change opens outward a central cavity in TMD region of Pgp and likely mediates transport of its recognized substrates (17).

Fig. 8.1. A cartoon showing a predicted membrane topology of the MDR1 P-glycoprotein wherein protein consists of 12 transmembrane domains, each half contains a nucleotide binding domain (NBD), and both N- and C- terminus are in cytoplasm.

Targeted Chemotherapy in Drug-Resistant Tumors

2. Probable Mechanism(s) for Pgp-Mediated Drug Transport

143

Commonly, Pgp has been postulated to act as an ATP-dependent efflux transporter of xenobiotics and chemotherapeutic drugs (6). In addition, several studies provide evidence supporting alternative mechanisms for the diminished Pgp-mediated drug accumulation in MDR cells. For example, a flippase model has been proposed for Pgp (Fig. 8.2) (18). This model suggests that Pgp flips hydrophobic cytotoxic compounds from the inner to outer leaflet of the lipid bilayer wherein the agents can diffuse away. Alternatively, Pgp acts as a phospholipid flippase (Fig. 8.2) altering membrane permeability, thereby accounting for diminished intracellular concentration of drug and accounting for the broad specificity of Pgp toward its recognized substrates. In addition, cells expressing MDR1 may have intracellular compartments that are more acidic than non-Pgp expressing cells (19, 20), altered intracellular distribution of drug (21), or altered membrane permeability resulting in decreased drug influx (22). For example, Piwnica-Worms and coworkers have shown sustained expression of Pgp without use of chemotherapeutic drugs in a line of stable transfectants, MCF-7/MDR1, using a bicistronic vector for selection of cells (23). Electrical current and drug transport experiments demonstrated insignificant variations in membrane potential or membrane conductance between parental MCF-7 and MCF-7/MDR1 cells, but reduced unidirectional influx and steady-state cellular content of Pgp substrates. There was no change in unidirectional efflux of substrates from cells. These authors concluded that the dominant effect of Pgp in this system was reduction of drug influx, possibly through an increase in intramembranous dipole potential (23). Pgp may also interfere with or alter pathways of apoptosis (programmed cell death) (24, 25), therefore offering protection from cytotoxic compounds.

Fig. 8.2. A putative model showing P-glycoprotein as a transporter of its recognized substrates: (a) pumps substrates from one side of the membrane to the other via hydrophilic channel of the transmembrane domains (classic pump), (b) drugs partitioning into the lipid bilayer are excluded extracellularly (vacuum cleaner), and (c) recognized substrates partition into the bilayer, interact with the hydrophobic binding site in the cytoplasmic domain and translocated or flipped to the outer membrane for exclusion into the extracellular space (flippase model).

144

Sivapackiam et al.

Thus, although the exact mechanism remains to be biochemically elucidated, the observed combined net effect is a decreased intracellular concentration of cytotoxic drugs that results from overexpression of Pgp, thereby rendering chemotherapeutic treatment ineffective in cancer.

3. Role of Pgp in Pharmacokinetics and Its Transcriptional Regulation

In addition to its overexpression in tumors, Pgp is normally located in several tissues involved in excretory functions, including the brush border of proximal tubule cells in the kidney, the biliary surface of hepatocytes, and the apical surface of mucosal cells in the small intestine and colon (26, 27). Pgp also is located on the luminal surface of endothelial cells lining capillaries in the brain and in the testis (28–30) as well as on the apical surface of choroid plexus epithelial cells (29). However, despite its widely disseminated expression, the function of Pgp in normal physiology has not been clearly defined, although Pgp could have a role in protection from xenobiotics (31). Piwnica-Worms and coworkers have also suggested a role for Pgp in intracellular cholesterol trafficking (32). Furthermore, inhibition of Pgp with an MDR modulator could provide an effective means for increasing oral absorption of drugs and reducing drug excretion, resulting in decreased dosing requirements for treatment of cancer and infectious diseases. For example, Pgp modulation had been under evaluation as a means to improve oral absorption of chemotherapeutics and HIV-1 protease inhibitors such as indinavir, nelfinavir, saquinavir, and rotonavir (33, 34). Similarly, use of Pgp inhibitors (commonly known as MDR modulators or reversal agents) could allow drug penetration into Pgp protected sites in the body, such as the brain and selected hematopoietic cells, as has been shown for penetration of protease inhibitors into the central nervous system (34). In addition, transgenic expression of the MDR1 gene has been explored for hematopoietic cell protection in the context of cancer chemotherapy (35–37), wherein Pgp could protect hematopoietic progenitor cells from chemotherapy-induced myelotoxicity. Hematopoietic cells transduced via retroviral-mediated transfer of the MDR1 gene have shown preferential survival in  vivo after treatment with MDR drugs (37) and data from pilot clinical studies had supported this approach (38). Various pathways and their transcription factors that could potentially regulate the activity of MDR1 promoter have also been studied (39). Among these pathways, the PKC pathway is

Targeted Chemotherapy in Drug-Resistant Tumors

145

often activated by stress and chemotherapeutics and has been shown to result in increased levels of MDR1 mRNA. Thus, inhibition of PKC by a variety of inhibitors prevents not only increases in mRNA but also the drug-resistant phenotype. This paradigm fits well with the observed induction of MDR1 in tumors after chemotherapy treatment. The MAPK pathway can also induce upregulation of MDR1. This story is complicated by the fact that some small molecules will activate one arm ERK1/2, but others activate the JNK arm of the pathway. Interestingly, both still result in upregulation of the MDR1 gene. The NFkB pathway is also often misregulated in cancer and chronic inflammation (40). Unfortunately, activators of this pathway also are reported to increase expression of MDR1 in many cell lines. The NFkB pathway under specific contexts such as breast cancer cell line will decrease expression of MDR1 (41). The release of NFkB from the promoter of MDR1 actually decreased expression of the MDR1 gene in this case. The cause of this anomaly has yet to be determined. Recent studies have also shown that activation of the Wnt/b-catenin, yet another prosurvival oncogenic pathway, also upregulates MDR1. In both human and rat brain epithelial cells, b-catenin activators and GSK-3 inhibtors upregulate MDR1 expression and activity (42). Critically, b-catenin pathway activation also upregulated other known drug resistance genes such as MRP2, MRP4, and BCRP. This combination of downstream effects makes the b-catenin pathway an attractive target for therapeutic intervention. Additionally, the sequence-specific binding sites of many of the transcriptional effectors at the MDR1 locus have also been determined. The MDR1 locus highlights one of the more confounding issues when studying gene regulation, long-range enhancers. Several of the enhancer binding sites are thousands of basepairs upstream of the start of MDR1 mRNA transcription. For an example, Pregnane X receptor (PXR) nuclear receptor binds in a region between −7817 and −7864 from the start of the transcript. PXR binding enhances transcription of the MDR1 gene as well as transcriptional reporters (43). Though original study found that Rifampin could enhance MDR1 expression through PXR, subsequent studies have found that small molecules found in some foods and herbs such as hyperforin, tangeretin, ginkgolide A, and ginkolide B could induce expression of MDR1 and activate luciferase reporters of MDR1 promoter via activation of PXR (44, 45). Even further upstream, the vitamin D receptor/ RXRa heterodimer binds to MDR1 promoter region and enhances MDR1 expression in the presence of vitamin D (46). The presence of these enhancer elements lends credence to the normal role of MDR1 preventing the absorption and enhancing the excretion of environmental toxins during and after meals. Finally, one of the most studied transcription factor families, the p53

146

Sivapackiam et al.

family also plays a role in regulating the transcription of the MDR1 gene. The mutant p53 strongly upregulates MDR1 expression, reporter assay activity, and enzyme activity. The region of enhancement was found to be between −58 and −74 from the start of transcription (47). In addition a truncated, oncogenic version of p73 is found to inhibit wild-type (WT) p53, an MDR1 expression inhibitor, and thus also upregulates MDR1 and its associated activity (48). Thus, mutations in p53 and oncogenes targeting p53, one of the oldest known tumor suppressors, likely play a critical role in the upregulation of MDR1 in many tumor samples. Regardless of mode of interaction, the net effect is that Pgp reduces the intracellular concentration of substrates in Pgp-expressing multidrug-resistant cells compared with nonPgp-expressing cells. Therefore, various strategies for delivering of chemotherapeutic agents in drug-resistant targets and radiopharmaceuticals as molecular imaging probes for detection of functional Pgp expression in vivo could be expected to play a vital role in systemic biology as personalized medicine gains momentum in the twenty-first century. Herein, we first describe attempts from various laboratories worldwide to devise ways for targeted delivery of drugs.

4. Strategies for Targeted Chemotherapy of Drug-Resistant Tumors 4.1. Transduction Sequences for Delivery of Drugs

Transduction sequences (when covalently linked to the organic scaffold of interest) could be employed as efficient drug delivery systems for enabling or enhancing uptake of candidate molecules into desired cells or tissues for enhanced therapeutic effects, radiotherapy, and molecular imaging applications. It has been shown that the nuclear transcription activator protein (Tat) crosses the plasma membrane of cells and its ability to penetrate cell membranes could be associated with the highly basic sequence of amino acid residues 49–57 (Tat49–57). Significantly, Tat49–57 is also a highly water-soluble polycation yet exhibits the ability to cross the nonpolar membrane of cells. After the discovery of this membrane-permeant motif, several modifications of this transduction sequence have emerged showing better uptake profiles and decreased cytotoxic effects. Covalent attachment of these selected group of amino acids into peptoids, peptides, spaced oligocarbamates, dendrimers, radioisotopes for imaging, and drug delivery enables their benefits as efficient cargo motifs for targeted drug delivery and biomedical imaging applications (49). The FDA-approved drugs currently used for the treatment of HIV are nucleoside reverse transcriptase inhibitors, nonnucleoside

Targeted Chemotherapy in Drug-Resistant Tumors

147

reverse transcriptase inhibitors, and protease inhibitors (PIs). Among these potent drugs, PIs represent a class of antiretroviral drugs blocking the protease enzyme activity needed for replication of virus. Like any other class of drugs, PIs also encounter pharmacokinetic limitations such as intestinal absorption and insufficient permeability across the blood–brain barrier (BBB) probably due to their recognition as transport substrates of Pgp (50). Thus, strategies to overcome these barriers would be needed and continue to be an area of intense investigations. Specifically, exploiting the peptide sequence comprising 11 amino acids derived from HIV-1 trans-activating transcriptor (TAT) peptide capable of crossing the cell membranes via pathways independent of either existing transporters or receptor-mediated endocytosis, peptide conjugated nanoparticles (NPs) appended to HIV drugs show higher permeability in validated biochemical systems developed for evaluation of the drug transport (51). Using transwell configurations of MDCK (Madine Darby canine kidney) cell monolayers, conjugation of nanoparticles comprising Ritonavir, an HIV drug, to the TAT peptide has been shown to undergo transwell transepithelial transport (apical to basolateral) using MDCK-MDR1 and MDCK WT cell monolayer counterparts. MDCK cells are capable of polarizing growth with formation of tight junctions, which upon proper validation (leak-proof tight monolayers as well as stable expression of functional Pgp) provide efficient and robust biochemical assay for evaluating transporter-mediated efflux of drug candidates. The observed profiles of transepithelial transport suggested the ability of conjugated-NPs containing Ritonavir to undergo 4.4-fold higher transepithelial transwell transport in MDCK-MDR1 (high Pgpexpressing cells) than their WT counterparts, thereby showing the applicability of the strategy for enhanced transport of attached molecules and overcoming resistance pathways. Importantly, the approach has demonstrated more provocative data via the delivery of Tat-conjugated Ritonavir containing NPs across the BBB in FVB male mice (Functional Pgp). Following intravenous injection of free drug, unconjugated nanoparticles, and TAT conjugated NPs containing Ritonavir, the peptide conjugated NPs show highest permeability across the BBB in these mice models, over prolonged period. It has been shown that clearance of Ritonavir over time was consistent with the profile expected for Pgp recognized substrates. However, the uptake of TAT-conjugated NPs containing Ritonavir in FVB mice brains (14th day) was found to be 11-fold higher than that of unconjugated NPs as well as 22-fold higher than that of Ritonavir alone. These data demonstrate versatility of transduction sequence for delivery of PI across the BBB probably via transcytosis across the endothelium of the brain vasculature (51). However, some transduction sequences are known to cause toxicity, thus these TAT peptide conjugates would need to be investigated for other side effects via safety pharmacology studies for

148

Sivapackiam et al.

further validation of the concept. Nevertheless, the strategy (once adequately validated) could allow maintenance of Ritonavir at thereapeutic levels for extended time in reducing the viral load in the CNS, which act as a reservoir for replication of HIV-1 virus. Furthermore, doxorubicin (DOX), a chemotherapeutic agent has also been appended to a TAT peptide fragment (CGGGYGRKKRRQRRR) via a succinimidyl-4-(N-maleimidomethyl) cyclohexane-1-carboxylate (SMCC) linker 1 and evaluated for therapeutic efficacy in drug-resistant MCF7/ADR cells compared with parental therapeutic drug (52). On incorporation of the chemotherapeutic agent to the peptide, the TAT-conjugated DOX was found to be approximately eightfold more active in drug-resistant cells compared with its parental unconjugated counterpart. Additionally, TAT-conjugated DOX was shown to be translocated in the perinuclear region or cytoplasm compared with parental DOX localization in the nuclei of these cells. Although results have been provocative enough to warrant further investigations of this strategy to target drug-resistant tumors, but cytotoxicity data for parental TAT sequence under same conditions would be needed for further rigorous evaluation of the strategy. Finally, PEG-coated liposomes have been designed for formulation of drugs for improved pharmacokinetics and protection from reticO

OH

O OH OH

O

O

OH

S-CGGGYGRKKRRQRRR

O O

O

OH

O N C H

N

O

1

uloendothelium for better delivery to the target sites. Despite these modifications, sterically stabilized liposomal DOX has not shown a marked improvement in efficacy against sarcoma and small cell lung cancer. Therefore, strategy of transduction sequence for efficient delivery has also been employed. As an example, DOX encapsulated in liposome and appended to cell penetrating peptide sequence (TAT, or PEN) has been shown to accumulate in A431 cells 12-fold higher compared with nonconjugated liposomal DOX (53). Although, higher delivery of the drug has been obtained in the targeted cells via this transduction sequence, the potency of formulated DOX was not improved indicating the obstacles in release of the drug within the intracellular compartments. Overall, these strategies demonstrate the ability of transduction sequences to carry the covalently bonded drugs into intracellular

Targeted Chemotherapy in Drug-Resistant Tumors

149

compartments for enhanced delivery of therapeutics, which in turn could lead to enhanced potency of existing chemotherapeutic regimens. However, simple enhanced delivery of drugs within the cells represents a first important component of these strategies. Challenges could arise in terms of selective delivery of drugs in the desired targeted tissues or organs, as systemic biology gains momentum and additional strategies are conceived such as enzyme-cleavable domains within these molecules to allow selective release of drugs to treat malignant cells/tissues compared with their normal counterparts. 4.2. Prodrug Strategy

The advancement and confluence of the disciplines of chemistry and biology have recently opened an even broader range of opportunities at the interface of interdisciplinary sciences wherein scientists could shift emphasis from the total structure of a biologically active molecule to the subset of its functionality that either influences or determines its activity. The central theme behind this strategy involves a slight modification of a chemical entity, which allows an active component of the molecule to reach its target site with comparable or superior function and simultaneously evade interactions with putative transporters and biochemical pathways devised to reduce intracellular concentrations of available drugs thereby impacting efficacy in the targeted tissues. Acting as a microtubule stabilizing agent paclitaxel is a wonderful anticancer drug that has been successfully used for treatment of a variety of tumors, including breast, ovarian, and lung. Unfortunately paclitaxel has been a well-characterized Pgp substrate. Therefore, strategies for modifying the drug to overcome drug resistance pathways have been sought. Several prodrug derivatives of paclitaxel have been developed based on the chemical modification of the hydroxyl groups at position 7 of the baccatin core and position 2¢ of the Taxane side chain (54, 55). Of these prodrugs, 2¢-ethylcarbonate-conjugated TAX has been shown to be accumulated rapidly into the cells. Additionally, chemical modifications in the baccatin portion of TAX also decrease affinity for the Pgp transporter system. Based upon observation that carboxyesterase (CES) isolated from rabbit liver converts the 2¢-ethylester into a parental compound, a gene-directed enzyme-based prodrug therapeutic model was developed (56). The prodrug paclitaxel-2¢-ethylcarbonate 2 was evaluated for its efficacy in taxol-sensitive, SKOV3 cells and Tax-resistant KOC-7c transfected with a plasmid encoding Ra-CES. The uptake levels of the prodrug in SKOV3/TAX60 cells were found to be comparable to that in Pgp-negative SKOV3 cells (56). Additionally, Pgp inhibitor (verapamil)-induced effects were not observed in SKOV3/TAX60 cells indicating ability of the prodrug to overcome efflux pathways of Pgp (56). These results indicated the potential application of GDEPT (GENE Directed Enzyme Prodrug Therapy) strategy for delivery of taxol into cells or tissues

150

Sivapackiam et al.

expressing high levels of Pgp. Although promising data has been generated, these results would need to be rigorously investigated in the presence of Pgp-specific, third-generation inhibitors to validate unambiguously the benefits of the strategy.

C 6 H5 CO NH Ph 3' C 2H 5OCOO

CH 3COO 18

O 1'

11

O

O

OH

19 7

14

1

3

H OH C6 H5 COO CH 3COO

4

O 20

2

Another concept gaining recognition rapidly is an extension of a prodrug strategy, wherein a chemotherapeutic agent of interest is appended via linker to the cargo. Choice of the linker and cargo is of paramount importance. While the linker could allow a selective release of agent in the targeted compartments, the cargo could assist in differentiation of intracellular vs. extracellular targets. Additionally, if the cargo is based upon an encapsulation strategy, it could simultaneously allow an increase in a biological half-life and decreased side effects. These concepts have been validated in an example of mitomycins, antitumor antibiotics produced by Strepromyces caespitosus (57). Mitomycin C (MMC) was linked through a urethane linker to the lipid promoiety (Fig. 8.3). The design of the strategy was based upon hypothesis that most tumors have enriched thiolytic environment. For example, redox enzymes such as glutaredoxin, thiredoxin, and other corresponding reductases are expressed in variety of the tumors, wherein cleavable linker will allow a delivery of the cytotoxic agent. Further, attachment of lipid will enhance the biological half-life by protecting the agent via encapsulation, thereby decreasing systemic toxicity. Thus, the antitumor activity of a cleavable lipid-based prodrug of MMC delivered by STEALTH liposomes (SL) was studied in drug-resistant human ovarian carcinoma A2780/AD model and compared with free MMC including both free as well as SL forms of DOX, an established anticancer drug. It has been shown that SL-prodrug (SL-pMMC) possessed enhanced antitumor activity when compared with the parent MMC, free DOX, and SL-DOX. Therefore, the resultant high antitumor potency of SL-pMMC could arise from its preferential accumulation in the tumor by the enhanced permeability and retention (EPR) effect and suppression of Pgp efflux pathways. Resultantly, the prodrug strategy could also be extended to include other enzyme (caspase 3 or cathepsin D)-induced cleavable domains and more potent cytotoxic agents susceptible to drug-resistant pathways in chemotherapy. Following a similar prodrug strategy (Fig.  8.4), DOX was coupled to a monoclonal antibody directed to the insulin-like

Targeted Chemotherapy in Drug-Resistant Tumors

151

Fig. 8.3. Schematic of a lipid-based prodrug strategy, wherein thiolytic activation of the prodrug results in a release of mitomycin C (MMC).

Fig. 8.4. Schematic showing the advantages of the receptor-targeted conjugate of a chemotherapeutic drug (doxorubicin) vs. nonconjugated drug. Compared with free drug that gets excluded into the extracellular space resulting in decreased efficacy, the conjugated motif gets selectively internalized, released within lysosomal compartments, and demonstrated enhanced potency.

152

Sivapackiam et al.

growth factor-1 receptor (IGF 1R) (58). The IGF 1R has been shown to be highly overexpressed in most tumors and is known to be a well-validated tumor target (59, 60). For accomplishing the prodrug synthesis, the derivatized MAb a-IR3 antibody possessing reactive aldehyde functionality was reacted with the amino group of the DOX and in turn was reduced in situ to obtain the prodrug conjugate. The prodrug conjugate was shown to be bounded to tumor cells selectively and accumulated efficiently only in receptor-expressing cells. The conjugate was postulated to be processed within the lysosomes to release free DOX inside target cells leading to selective activity. Overall, the strategy lead to approximately >200-fold improved therapeutic index and in vivo reduced tumor load with no systemic toxicity. For further rigorous evaluation of the strategy, the labeled compounds would be expected to offer better visualization of pharmacokinetics, as agents coupled to antibodies normally suffer from poor pharmacokinetics. Nevertheless, the prodrug conjugate was not a Pgp substrate and provocative approach represented a critical step toward development of improved and more selective anticancer agents. Overall, this prodrug strategy could also be extended to include enzyme-cleavable domains to allow delivery of more potent cytotoxic agents into cells or tissues susceptible to drug-resistant pathways in chemotherapy.

5. Chemotherapeutic Drugs Recognized by Pgp (Transport Substrates)

Compounds recognized by Pgp are typically characterized as modestly hydrophobic (octanol/water partitioning coefficient, logP > 1), often contain titratable protons with a net cationic charge under physiological conditions, and are predominately “natural products” with an aromatic moiety (61, 62). In addition, incorporation of methoxy functionalities has been shown to enhance Pgp recognition. O

OH R OH

Ph O

O

OH

O

O NH2 OH

R1

N H

OR 2

O

O

OH

O OH

HO C 6H 5 COO H CH3 COO

O

3 R = COCH 2OH

5 R 1 = COC 6H 5 , R 2 = COCH3 (Paclitaxel)

4 R = COCH 3

6 R 1 = COOC(CH 3) 3, R2 = H (Docetaxel)

Targeted Chemotherapy in Drug-Resistant Tumors

153

Among an extensive list of conventional cytotoxic compounds, anthracyclines (doxorubicin 3, daunorubicin 4), taxanes (paclitaxel 5, docetaxel 6), Vinca alkaloids (vincristine 7, vinblastine 8, vindesine 9), and etoposides 10 (VP-16) are examples of clinically important chemotherapeutic drugs recognized by Pgp (6, 7, 61, 63).

OH N N

N H

H N R3 H R1 HO R2

7 R 1 = CHO, R 2 = COOCH 3, R3 = OCOCH3 (Vincristine) 8 R 1 = CH3, R 2 = COOCH3 , R 3 = OCOCH 3 (Vinblastine) 9 R 1 = CH3, R 2 = CONH 2, R3 = OH

The broad diversity in the scaffolds of these agents emphasizes the key characteristic feature of MDR, i.e., the apparent capacity of Pgp to recognize a large group of cytotoxic compounds sharing little or no structural or functional similarities. Furthermore, even targeted drugs have encountered MDR in clinics (64). H O O HO

O OH

O

O

O

O

NH

O O

NH N

O

O

N

N

N

N

OH 10

11

For example, Gleevec 11, a 2-phenylaminopyrimidine derivative (65, 66), while a highly potent inhibitor of receptor tyrosine kinases, such as BCR-Abl and PDGF-R, is also recognized and transported by Pgp (6, 67). Thus, it would appear that even novel targeted therapeutics will remain susceptible to broad specificity of transporter-mediated resistance mechanisms.

154

Sivapackiam et al.

5.1. Pgp Inhibitors

Because clinical studies have documented the poor outcomes associated with Pgp expression in tumors (63, 68), reversal of MDR by nontoxic agents that block the transport activity of Pgp has been an important target for pharmaceutical development. When coadministered with a cytotoxic agent, these nontoxic compounds (MDR reversal agents), enhance net accumulation of relevant cytotoxic drugs within the tumor cells.

O

O

O

O

CN N 12

O

R1

O N

N

N

O O

O

O

O

N R2

O

H N

N

O

H N

O N

N H

N

N H

O

13 R1 = OH, R 2 = CH 2CH3 (Cyclosporin A) 16 R1 = =O, R 2 = CH(CH3 )2 (PSC 833)

H

H

O N 14

N

N

HO

N

N CF3

S 15

Many compounds known to have other pharmacological sites of action initially were used to reverse MDR in cancer cells grown in culture and several underwent pilot clinical trials (61). These compounds included verapamil 12, cyclosporin A 13, quinidine 14, trifluperazine 15, and their derivatives (61). However, these agents had limited clinical utility because of unacceptable toxicities at serum levels of drug needed to modulate Pgp (63).

155

Targeted Chemotherapy in Drug-Resistant Tumors O O N

N O

H

O

O

N H 17

F F H

H

N N

N O

N OH

N O

O

O

O O

O O

N 18

19

Ph O

N

O O

N

O

N OH

N H NH

O

Ph

O

O

O

N

N 20

21

Second-generation modulators, such as dexverapamil, an optically pure verapamil (69), and PSC 833 (Valspodar) 16, a cyclic undecapeptide analogue of cyclosporin A (70) were soon developed with improved efficacy. These were followed by third-generation modulators, such as GF120918 17, a substituted isoquinolinyl acridonecarboxamide (71), LY335979 18, a difluorocyclopropyl dibenzosuberane (72), VX710 (biricodar) 19, an amido-keto-pipecolinate (73), XR9576 (tariquidar) 20, an analog of anthranilamide pharmacophore (74–76), and MS209 21, a quinoline derivatized analog (77, 78), that have been developed more recently.

156

Sivapackiam et al.

N N N

N

O O

O

22

Br

Cl

N

Br N

HN

O N

O

N H

NH 2 HN

O 23

24

Finally, R101933 (Lanquidar) 22 (79, 80), SCH66336 23, earlier developed as farnesyl protein transferase inhibitor (Lonafamib) (81), and ONT 093 24 (82, 83) have also been developed as potent Pgp inhibitors. Upon evaluating chemical structures of these Pgp inhibitors, it is obvious that most of the potent molecules have been designed based upon screening of natural products, pharmacophores of the drug-like molecules, and modification of organic scaffolds known to possess Pgp antagonist activity. Some of common features among these molecules include high hydrophobicity, presence of one or more protonable nitrogen under physiological conditions, two or more aromatic rings, and methyl or methoxy substituents on the aromatic rings. With the availability of structural information of Pgp, the rational drug design would likely emerge to make more potent and specific molecules. Given the two-fold pseudo symmetry in the structure of hamster Pgp, it is quite likely that molecules possessing flexible linker flanked by aromatic rings with methoxy- and/ or methyl-substituents as well as basic nitrogen would start emerging as lead molecules. It is quite possible that these molecules due to flexible spacer would have high probability to find high-affinity recognition sites within the transporter. The loss in the binding energy due to high entropy could be compensated by the gain in enthalpy derived from the fact that molecules would have high degree of freedom to associate with most favorable binding sites. Nevertheless, the MDR phenotype may be modulated more effectively with these more selective reversal agents to improve the efficacy of chemotherapeutic agents by delivering an appropriate

Targeted Chemotherapy in Drug-Resistant Tumors

157

dose at the targeted site. In terms of biomedical imaging applications using Pgp substrates, a noninvasive method of determining Pgp-mediated drug transport activity could be important in the use of new modulators, applications of gene therapy to chemotherapeutic protocols as well as predicting oral absorption, pharmacokinetics, and penetration of MDR drugs into target tissues and brain. Some of the promising radiopharmaceuticals are described in the following section.

6. SPECT and PET Radiopharma­ ceuticals for Noninvasive Imaging

Increasingly, the choice of systemic therapy for cancer is based on a priori analysis of tumor markers to assess the presence or absence of a molecular pathway or target (such as a key receptor or enzyme activity) for a given therapeutic agent. Identification of tumor markers with diagnostic agents assists in the proper selection of patients most likely to benefit from targeted therapy. Measurement of MDR is one potentially important marker in planning systemic therapy. However, expression of Pgp, as detected at the level of messenger RNA or protein, does not always correlate with the functional assessment of Pgp-mediated transport activity. Because Pgp transport activity is affected by specific mutations as well as the phosphorylation state of the protein (5, 84, 85), altered or less active forms of Pgp may be detected by polymerase chain reaction (PCR) or immunohistochemistry, which do not accurately reflect the status of tumor cell resistance. Thus, methods for functionally interrogating Pgp transport activity have been sought (86). Imaging with a radiopharmaceutical that is transported by Pgp may identify noninvasively those tumors in which the transporter is not only expressed, but functional. Thus, significant effort has been directed toward the noninvasive detection of transporter-mediated resistance using planar scintigraphy or single-photon emission computed tomography (SPECT) employing radiolabeled metal complexes as well as positron emission tomography (PET) radionuclide incorporated organic molecules, characterized as transport substrates for Pgp.

+

R R

R

N N

C C

N C Tc C N R 25

C C

N N

R = H2 C

O

R 25a O R O

R = H2 C 25b

158

Sivapackiam et al.

Hexakis(2-methoxyisobutylisonitrile)technetium-99m (commonly known as (99mTc)Sestamibi) 25a, although originally developed as a radiopharmaceutical for myocardial perfusion imaging (87, 88), subsequently was the first metal complex shown to be a Pgp transport substrate (89). Characterized by octahedral geometry around the central technetium(I) core (87, 90), the radiopharmaceutical possesses a cationic charge and modest hydrophobicity similar to many chemotherapeutic agents in the MDR phenotype. In the absence of Pgp expression, this 99mTcisonitrile complex accumulates within the interior of cells in response to the physiologically negative mitochondrial and plasma membrane potentials maintained within cells (91, 92). However, in Pgp-expressing multidrug-resistant tumor cells, net cellular accumulation levels of (99mTc)Sestamibi are inversely proportional to the level of Pgp expression (89, 93–96). Furthermore, complete reversal of the Pgp-mediated exclusion of (99mTc)Sestamibi has been affected by treatment with conventional Pgp inhibitors such as verapamil 12, cyclosporin A 13, and quinidine 14 or newer more potent reversal agents such as PSC 833 16, GF120918 17, LY335979 18, or VX710 19 (32, 62, 89, 93, 94, 97–102), and more recently XR-9576 20 (6, 74). Furthermore, to optimize the transport and Pgp targeting characteristics of 99mTc-isonitrile complexes, several studies investigating structure–activity relationship (SAR) have been performed. In one study, the alkyl chains in (99mTc)Sestamibi were replaced with longer chain ether functionalities. The hexakis(2-ethoxy-isobutylisonitrile)-Tc-99m complex (99mTc-EIBI) was shown to be a transport substrate recognized by Pgp, but with slightly greater nonspecific cell binding than (99mTc)Sestamibi (103) that would result in inferior target/background ratios compared with (99mTc)Sestamibi. PiwnicaWorms and coworkers have also explored substituted aromatic isonitriles (104). A series of substituted arylisonitrile analogs were obtained from their corresponding amines through a reaction with dichlorocarbene under phase transfer catalyzed conditions, and noncarrier-added hexakis(arylisonitrile)-Tc-99m complexes were produced by reaction with pertechnetate in the presence of sodium dithionite (104). SAR studies resulted in a lead compound, 25b, which demonstrated an overall encouraging transport profile in Pgp-expressing cells, but significant nonspecific adsorption to hydrophobic compartments was identified. Nevertheless, results suggested that methoxy substituents, compared with other functionalities, preferentially contributed to enhanced Pgp recognition for this class of compounds. However, none of these radiolabeled complexes exceeded (99mTc)Sestamibi in their Pgptargeting properties. In addition, several entirely different classes of technetium complexes have been identified as Pgp transport substrates. Using a planar Schiff-base moiety and hydrophobic phosphines, nonre-

Targeted Chemotherapy in Drug-Resistant Tumors

159

ducible Tc(III) monocationic compounds known as “Q-complexes” were developed for applications in myocardial perfusion imaging (105, 106). The lead complex for clinical development was trans((1,2-bis(dihydro-2,2,5,5-tetramethyl-3(2H)furanone4-methyleneimino)ethane)bis(tris(3-methoxy-1-propyl) phosphine))Tc(III), known as (99mTc)Furifosmin 26 (107).

O

O

O

O

P N N Tc O O P

O

O

O

O

26

O

O

P N O

Tc

N O

P

O

O

O

O

O

O 27

Because the hydrophobicity and Pgp-targeting properties of these complexes could be readily adjusted by varying functionalities on the Schiff base or phosphine moieties independently, a variety of novel 99mTc-Q-complexes with subtle structural variations were synthesized (97, 108). This approach allowed the coordination environment of the Tc(III) metal core to be maintained while altering the overall electronic environment, thereby enabling evaluation of structural features conferring Pgp-mediated transport properties. Ether functionalities can be incorporated into the equatorial Schiff base ligand by condensation of ethylenediamines with ether-containing b-dicarbonyl compounds (109). The presence of gem-dimethyl groups sterically hinders the attack of diamine at the adjacent carbonyl and the strategy results in regioselective condensation at the less hindered carbonyl. Preparation of the tertiary phosphines was accomplished in a two-step, one-pot reaction involving treatment of 1-chloro-3-methoxy-propane with magnesium in tetrahydrofuran and subsequent reaction of the reagent with dimethylchlorophosphines or dichloro-methylphosphines to provide the necessary substituted phosphines with overall yield of 50–70% (108). The desired 99mTc-Q-complexes were then obtained by a two-step synthetic approach using the phosphines as both reductants and ligands (110). From MDR transport assays in  vitro, the trans(2,2¢-(1,2ethanediyldiimino)bis(1,5-methoxy-5-methyl-4-oxo-hexenyl)) bis(methyl-bis(3-methoxy-1-propyl)phosphine)Tc(III) complex 27 and the trans(5,5¢-(1,2-ethanediyldiimino)bis(2-ethoxy-2-

160

Sivapackiam et al.

methyl-3-oxo-4-pentenyl))-bis(dimethyl(3-methoxy-1-propyl) phosphine)Tc(III) complex 28 were discovered. O O

P N O O

Tc

O

N O

P

O

P O P Tc P O P

O

O

O

O

O O

O 29

28

These complexes mimic (99mTc)Sestamibi for their Pgp recognition properties in vitro (97, 108). In addition, a Tc(V) complex known as (99mTc)Tetrofosmin (111), bis(1,2-bis{bis(2-ethoxyethyl)phosphino}ethane)Tc(V) 29, has been identified as another 99m Tc-complex with highly favorable Pgp-mediated transport properties (108, 112). While these metal complexes do not share any obvious structural homology, they do share the common features of a delocalized cationic charge and modest hydrophobicity. Overall, which of these selected 99mTc-complexes would be most clinically useful in evaluation of the Pgp status of tumors by SPECT imaging continues to be rigorously investigated.

O

O

OH 2 O H 2O C Tc H 2O C O C O 30

N O

N

C C

N C Tc C O

C C

O O

31

Another class of technetium-based radiopharmaceuticals has emerged on the basis of pioneering work done on the development of an air- and water-stable organometallic aqua complex (99mTc(OH2)3(CO)3)+ 30 obtained from reaction of pertechnetate in saline under 1 atm of CO (113). Because it was shown that the coordinated water molecules were labile, thus exchangeable with other ligands, complexes with heterogeneous ligands could be generated. Based upon these observations, these water molecules were substituted with methoxy-isobutylisonitrile ligands to

Targeted Chemotherapy in Drug-Resistant Tumors

161

obtain another novel class of Pgp-targeted radiopharmaceuticals, tris(carbonyl)tris(2-methoxy-isobutylisonitrile)technetium(I); (99mTc(CO)3(MIBI)3)+ 31 (114). Cellular accumulation of 31 in human epidermal carcinoma KB 3-1 (−Pgp) and KB 8-5 (+Pgp) cells demonstrated uptake profiles inversely proportional to Pgp expression indicating it to be recognized as a transport substrate (115).

CH 3 OCH2 CH 2CH 2

N

CH 2 CH 2CH 2OCH 3 N CH2 CH 2 OCH 2 CH 3 S P Tc N S P CH2 CH 2 OCH 2 CH 3 CH 2CH2 CH 2 OCH3 CH 2 CH 2CH 2OCH 3

R 32 R = CH 2CH2 OCH3 33 R = CH 2CH2 OCH2 CH3

Recently, a new class of nitrido technetium-99m-agents of the type (99mTc(N)(DTC)(PNP)), wherein DTC is a dithiocarbamate ligand and PNP an aminodiphosphine ligand, as potential myocardial imaging agents has been developed (116). Within this novel class, two compounds, (99mTc(N)(DBODC)(PNP5))+ 32 and (99mTc(N)(DBODC)(PNP3))+ 33 (DBODC = bis(N-ethoxyethyl) dithiocarbamato; PNP5 = bis (dimethoxypropyl-phosphinoethyl) ethoxyethylamine, PNP3 = bis-(dimethoxypropylphosphinoethyl)methoxyethylamine) have shown interesting pharmacokinetic profiles that could enable these molecules as myocardial perfusion imaging agents. Importantly, following administration of cyclosporine A (13), a first-generation Pgp inhibitor, intravenous injection of (99mTc(N)(DBODC)(PNP5))+ 32 showed a delayed clearance in lungs, liver, kidney, very significant increase in activity in the intestinal tissue, and a simultaneous decrease in endoluminal contents of the tracer. These results are consistent with their postulation of being Pgp-recognized substrates (117). However, cyclosporin A (13) is not an extremely specific Pgp inhibitor, thus results would need to be validated in the presence of more specific inhibitors such as PSC 833 16 or LY 335979 18 in same rat models. Having demonstrated specificity of the agent in presence of more specific inhibitors, the agent could potentially provide an imaging marker for probing functional Pgp expression in cancer. Availability of such a wide selection of chemical structures of technetium complexes has offered a powerful tool for molecular modeling in selecting critical structural characteristics needed in a given chemical entity for their recognition as Pgp substrate. Taking the advantage of this existing armamentarium, genetic algorithms

162

Sivapackiam et al.

(GA) have been used to develop specific technetium metal–ligand force field parameters for the MM3 force field. These parameters have been developed using automated procedures within the program FFGenerAtor from a combination of crystal structures and ab initio calculations (118). These models have been allowed to predict uptake of technetium complexes in the liver and kidney, two critical organs expressing Pgp. Therefore, the development of new metal ligand parameters to model the biological properties of radiolabeled organometallic complexes provides a versatile tool to scientists in the field of biomedical imaging applications. Organic scaffolds capable of coordinating other metals have also been explored. Multidentate ligands with an N4O2 donor core have the ability to form stable monomeric, monocationic, hydrophobic complexes with a variety of main group (119, 120) and transition metals (121–123). Schiff-base Ga(III) complexes were previously reported as potential PET radiopharmaceuticals with utility as myocardial perfusion imaging agents (124, 125). These complexes, exemplified by the lead compound 34, demonstrated pharmacological profiles consistent with their potential utility as PET probes of Pgp activity in tumors (126–128). Substituted salicylaldehydes were obtained by ortho-formylation of phenols and substituted linear tetramine was obtained through a reaction of dibromoethane and 2,2-dimethylpropane-1,3diamine at room temperature. The triaryl precursors containing a central imidazolidine ring were synthesized by condensation of an appropriate substituted linear tetramine with substituted salicylaldehydes. The desired metal(III) complexes were obtained by cleavage of the imidazolidine ring via transmetallation reactions using appropriate metal(III) acetylacetonates. O O

O

O HN

68 Ga

N

O

NH HN

N O

O

67 Ga

N

NH N

O

O 34

O 35

For evaluation of Pgp-targeting properties, these compounds were screened for their cytotoxicity in Pgp-expressing tumor cells and their appropriate control (−Pgp) cells (Fig.  8.5). Selected metal(III) complexes were able to differentiate between human epidermal carcinoma KB 3-1 (−Pgp) and KB 8-5 (+Pgp) cells

Targeted Chemotherapy in Drug-Resistant Tumors

163

Fig. 8.5. Cell survival studies in parental KB 3-1 (Pgp−) and multidrug-resistant KB 8-5 (Pgp+) cells at increasing concentrations of R-ENBPI gallium(III) (R = 4,6-dimethoxy (a); 3-methoxy (b)) complexes, colchicine (25 mM , positive control). Each point represents the mean of triplicate determinations; bars represent ± SEM when larger than symbol; solid lines are a spline presentation of the data.

through differential cytotoxicity profiles indicating Pgp-mediated transport (127). The active metal complexes containing 4,6-dimethoxy-substituted aromatic rings (Fig. 8.5a) were more potent than their corresponding 3-methoxy analogs (Fig. 8.5b). SAR studies led to discovery of a lead compound, ({bis(3-ethoxy2-hydroxy-benzylidene)-N,N¢-bis(2,2-dimethyl-3-aminopropyl) ethylenediamine}-gallium(III)), (Ga-ENBDMPI)+ 35. The crystal structure provided direct evidence that gallium is coordinated symmetrically and simultaneously to the N4O2 donor core of the ligand. The structure revealed that the central gallium is hexacoordinated, involving two phenoxy oxygens (O1 and O2), two secondary amine nitrogens (N2 and N3), and two imine nitrogens (N1 and N4) with overall trans-pseudooctahedral geometry (Fig.  8.6). For biochemical studies, the ligand was labeled with radioactive gallium-67/68 through a simple transmetallation reaction of 67/68Ga(acetylacetonate)3 and ligand dissolved in ethanol. The radiolabeled compound was purified through a C-18 sep-pack. For analysis of metabolites, the lead SPECT radiotracer, 67Ga-ENBDMPI

164

Sivapackiam et al.

Fig. 8.6. ORTEP drawing of ((3-ethoxy ENBDMPI)Ga)+(ClO4)− showing the crystallographic numbering scheme. Atoms are represented by thermal ellipsoids (20% probability).

was incubated in buffer and calf serum at 37°C for 2 h. In addition, the radiopharmaceutical was extracted from mouse tissues (liver and kidney) 15 min after tail-vein injection. Radio-TLC analysis of tissue fractions indicated primarily a nonmetabolized agent (Fig.  8.7), because net cell contents of hydrophobic and cationic radiopharmaceuticals transported by Pgp are a function of both passive potentialdependent influx and transporter-mediated efflux. Therefore, favorable cationic 67Ga-complexes would likely penetrate KB 3-1 cells as result of the inwardly directed electrochemical driving forces. Furthermore, membrane potential-dependent influx in KB 3-1 cells in 120 mm K+/20 mM Cl− buffer containing the potassium ionophore valinomycin (1 mg/ml) was shown to collapse mitochondrial and plasma membrane potentials toward zero and reduce net tracer uptake of membrane potential-responsive hydrophobic cations (129). Resultantly, net accumulation of the 67Ga-complexes would be reduced in high K+/valinomycin buffer, and furthermore, tracer levels greater than that expected from equilibrium distribution into the water spaces under these conditions would provide one measure of nonspecific adsorption to intracellular hydrophobic compartments. Conversely, Pgp-mediated outward transport of 67Ga-complexes

Targeted Chemotherapy in Drug-Resistant Tumors

165

Fig. 8.7. Radio-TLC analysis of 67Ga-ENBDMPI 35 in buffer, calf-serum, mouse liver, and mouse kidney.

would be expected to decrease net cellular accumulation in KB 8-5 cells compared with KB 3-1 cells. Following these criteria, 67 Ga-ENBDMPI accumulated in cells showing profiles inversely proportional to Pgp expression (Fig. 8.8). While residual uptake in control KB 3-1 cells in the presence of high K+/valinomycin buffer indicated modest nonspecific interaction, the inhibitor GF120918induced uptake in MDR KB 8-5 cells (Fig. 8.8) demonstrated target specificity (126, 129). Additionally, a gallium(III) complex 36 of the naphthol–Schiff-base ligand has been developed and evaluated in human epidermal carcinoma cells (130). The compound showed selective cytotoxicity against KB 3-1 cells (Pgp−) compared with KB 8-5 cells (Pgp+) indicating its recognition as a transport substrate and thereby exclusion from drug-resistant (Pgp+) cells (Fig.  8.9). These results suggested that radiolabeled analogues of these Ga(III)complexes could also provide templates for 68Ga-PET radiopharmaceuticals to probe Pgp transport activity in tumors (128).

OMe O HN

Ga

N

NH N

O MeO 36

166

Sivapackiam et al.

Fig. 8.8. Accumulation of 67Ga-ENBDMPI 35 in KB 3-1 cells (Pgp−) and MDR KB 8-5 cells (Pgp+) as indicated. Shown is net uptake (fmol/mg protein/nMo) at 90 min. Each bar represents the mean of four determinations; line above the bar denotes +SEM.

Fig. 8.9. Cell survival studies in parental KB 3-1 (Pgp−) and MDR KB 8-5 (Pgp+) cells at increasing concentrations of gallium(III) complex of naphthol Schiff-base ligand and a colchicine (25 mM; positive control).

While gallium(III) radiopharmaceuticals are undergoing preclinical evaluation, technetium-99m-based radiotracers have been clinically evaluated for assessment of the Pgp-mediated transport activity. Toward this objective, validation of successful inhibition of the transport function is necessary to evaluate the effects of Pgp inhibitors on patient outcomes. Clinical data indicate that (99mTc)Sestamibi can be used to detect inhibition of Pgp-mediated transport function in patients (Fig. 8.10). Overall, pharmacokinetic studies demonstrated retention of activity in the liver and kidney (Pgp positive tissues) of patients treated with PSC 833, (16), 2  h postinjection of the radiopharmaceutical (Fig.  8.10) (99, 100). In addition, the heart a Pgp negative tissue, acting as an internal control tissue, displayed no difference in retention of activity in either pre-PSC 833 (left panel, Fig. 8.10) or post-PSC 833 (right panel, Fig. 8.10). Overall, results indicated the potential

Targeted Chemotherapy in Drug-Resistant Tumors

167

Fig. 8.10. Effect of the Pgp inhibitor, PSC 833 (16) on pharmacokinetics of (99mTc)Sestamibi in vivo: Whole body posterior planar image obtained at 2  h postinjection of radiopharmaceutical. Left Panel: The pharmacokinetics demonstrated decreased retention of the radiotracer in the liver and kidney, 2 h postinjection of the radiotracer in the absence of PSC 833. Right Panel: Following 24-h treatment with PSC 833, image was obtained 2 h postinjection of radiotracer. Because PSC 833 blocks the Pgp, an efflux transporter, considerable amount of radioactivity was retained in the liver and kidney (right panel). Note that heart, being a Pgp (−ve) organ, retains the same amount of activity in both cases (Pre-PSC 833 and Post-PSC 833; left and right panels).

of (99mTc)Sestamibi to serve as an efficient diagnostic probe of Pgp-mediated transport activity in vivo. PET radiopharmaceuticals offer enhanced spatial resolution and quantification capabilities compared with SPECT agents. It is also noteworthy that gap in terms of resolution and quantification between SPECT and PET detectors has been drastically decreasing with discovery of advanced SPECT cameras. To probe Pgp transport activity, PET-based radiopharmaceuticals have been actively investigated on three fronts: (a) employing SPECT organic ligands capable of accommodating PET radionuclides, (b) bioinorganic radiolabeled complexes, and (c) conventional PET organic medicinals. Among organic scaffolds that coordinate both SPECT and PET radionuclides, two validated examples make use of the PET radionuclides Tc-94m and Ga-68. Thus, the radiosynthesis and biochemical validation of 94mTc-Sestamibi and 68Ga-ENBPI complexes have been reported (126, 131). As expected chemically, the highly favorable Pgp-targeting properties of these metal complexes were retained upon transformation from SPECT agents to PET imaging agents.

168

Sivapackiam et al.

N

N O CuII H N N O

37

On another front, organic scaffolds capable of accommodating PET radionuclides that generate novel metallopharmaceuticals through short synthetic routes have been reported. Thus, based upon rigorous prior contributions (132, 133), a stable, monocationic radiolabeled complex of copper(II) 37 was obtained as a potential 64Cu-radiopharmaceutical (PET) for targeting Pgp (134). The desired diiminedioxime ligand was synthesized from 2,3dimethyl-propane-1,2-diamine and heptane-2,3-dione-3-oxime. Cellular accumulation in MES-SA (−Pgp) and MES-SA (+Pgp) cells demonstrated uptake profiles inversely proportional to Pgp expression. Furthermore, an inhibitor-induced accumulation was observed in Pgp (+) cells. Ph P Cu P Ph Ph Ph

Ph Ph P P Ph Ph

38

Bidentate tertiary phosphine ligands have the ability to generate stable copper(I) complexes through a one-step synthesis in quantative yields (135, 136) and represent another class of potential 64 Cu-radiopharmaceuticals targeting Pgp. These complexes previously demonstrated potent antitumor properties compared with their free ligands alone (137, 138). As with 99mTc-Q-complexes, herein phosphines were exploited as both ligands and reducing agents to generate cationic, hydrophobic, and tetrahedral copper(I) complexes 38 with 1,2-bis(diphenylphosphino)ethane. These potential PET radiopharmaceuticals show evidence of Pgptargeting properties (139, 140). Although several leads exist for a 64 Cu-radiopharmaceutical for interrogation of Pgp by PET, these radiopharmaceuticals show only modest Pgp-targeting properties compared with 94mTc-Sestamibi (25A) or 68Ga-ENBDMPI (35). To demonstrate the potential use of radiolabeled probes as imaging markers of Pgp-mediated transport in vivo, quantitative pharmacokinetic analysis in mice was performed following

Targeted Chemotherapy in Drug-Resistant Tumors

169

intravenous injection of tracers. Mice have two isoforms of Pgp (mdr1a and mdr1b), which confer MDR (141). Drugtransporting mdr1a Pgp isoform is expressed in capillary endothelial cells of the brain wherein the protein is a major component of the BBB (142). Here, Pgp limits entry of a variety of amphipathic compounds into the central nervous system (141). Drug-transporting Pgp isoforms are also expressed along the biliary cannalicular surface of hepatocytes wherein the transporters function to secrete substrates into the bile (141). In this regard, mdr1a/1b(−/−) gene disrupted (Knockout; KO) mice, which have no drug-transporting Pgp, are a robust model for evaluation of candidate MDR agents by interrogating net tracer accumulation into brain and liver tissues (97). Therefore, analysis of initial tissue uptake and retention of 67Ga-ENBDMPI 35 in mdr1a/1b(−/−) mice in comparison to wild-type (WT) FVB mice was performed (Fig. 8.11). Relative to WT, the KO mice showed tenfold more radiotracer 35 in brain parenchyma 5  min after

Fig. 8.11. Pharmacokinetics of 67Ga-ENBDMPI 35 in brain (a) and liver (b) of FVB mice. Wild-type (WT) and mdr1a/1b (−/−) (183) mice were administered 35 by bolus injection into a lateral tail vein and organs harvested at the indicated times for analysis. Data are expressed as percent of injected dose of radioactivity per gram tissue at each respective time point. Data points represent the mean of 2–4 determinations each; bars represent ±SEM when larger than the symbol.

170

Sivapackiam et al.

injection of the complex (Fig. 8.11a). Furthermore, the AUC5–120 of 35 in mdr1a/1b(−/−) brain was 62.5 ± 16.1 mCi (injected mCi)−1 (g tissue)−1 × 100 × min (n = 8), a value 17-fold greater than that in wild-type mice (p < 0.005). While comparing to FDA-approved radiopharmaceuticals for noninvasive imaging, 67Ga-ENBDMPI showed three- to fivefold greater differences in brain AUC5–120 between WT and KO mice than 99mTc-Sestamibi (143) and 99mTcTetrofosmin (144), further documenting its potential superiority for imaging Pgp activity in  vivo. In addition, the liver showed twice the retained activity of 67Ga-ENBDMPI in KO mice compared with WT mice (Fig. 8.11b). Thus, the data further validated 35 as a substrate for Pgp in vivo and showed its potential as a molecular imaging probe of transporter function and inhibition in the living organism. A third focus has been directed toward incorporation of conventional PET radionuclides C-11 or F-18 into small organic medicinals characterized as substrates or inhibitors known to interact with Pgp (145–147). Employing this strategy, various labeled PET agents, including 11C-colchicine, 11C-verapamil, 11 C-daunomycin, and 11C/18F-paclitaxel, have been reported (147–155). While promising, some of these PET agents suffer from modest radiochemical yields and others from complex pharmacokinetics in  vivo mediated, at least in part, by rapid metabolism of the radiolabeled compounds. In addition, these 11 C/18F-radiopharmaceuticals are obtained in custom-designed, dedicated facilities located near cyclotrons that produce these isotopes, thereby imposing potential restrictions in their widespread distribution.

7. Emerging Role of Pgp in Progression of Neurodegenerative Diseases

Because of its location in the brain endothelial cells, Pgp plays an important role in brain detoxification processes. Thus, in principle an optimal function of this transporter at the BBB could ensure a healthy brain function. The downregulation of Pgp expression at the BBB has also been postulated to an enhanced influx of toxic compounds in brains of patients leading to symptoms of neuropathological disorders such as Parkinsonian syndrome (156, 157). Importantly and interestingly, patients with Parkinson’s disease who could have been exposed to toxins earlier in life have been shown to have C3435T polymorphisms of MDR1 gene with a related decreased Pgp function (156). Using 11C-Verapamil as a marker of Pgp function, patients with late stages of Parkinson’s disease have shown a higher uptake of the tracer in the brain compared with their normal counterparts (158). However, results in patients with earlier stages of the disease were not conclusive just

Targeted Chemotherapy in Drug-Resistant Tumors

171

displaying different regional localization of the tracer in diseased subjects compared with their normal controls. It must be noted that verapamil is a modest Pgp substrate that could result in lower overall signal in the brain leading to complications in interpretation of the data at earlier stages of the disease. Furthermore, Pgp has also been known to inhibit the activity of caspases regulating neuronal apoptosis in neurodegenerative diseases such as Alzheimer’s disease (AD) and Creutzfeldt–Jakob disease (25, 159). Pgp has been shown to be present in astrocytes (to a lesser extent) (160) as well as microglia (161) and has been shown to be induced in neurons by epileptic seizures (162, 163). Finally, Pgp has been shown to be downregulated in patients with AD. The AD brain is associated with loss of neurons in regions of the brain responsible for learning and memory. AD involves the appearance of two distinct abnormal proteinaceous deposits: extracellular amyloid plaques, which are characteristic of AD, and intracellular neurofibrillary tangles that also are found in other neurodegenerative disorders (164–166). AD amyloid plaques comprise primarily a family of proteins known collectively as Ab proteins (167), derived from the ubiquitously expressed cell surface amyloid precursor protein (APP). Indeed, several lines of investigation suggest that Ab accumulation is an initiating event in the pathogenic cascade of AD (168–170). Conversion of Ab within the extracellular spaces of the brain into toxic forms is accelerated at higher concentrations, implying that pathways influencing Ab production or elimination could modulate disease progression. Ab is secreted from neurons into brain interstitial fluid, where it is eliminated by proteolytic degradation (171, 172), passive bulk flow (173), and active transport across the BBB (174, 175). The latter, representing efflux across the BBB into the peripheral circulation, appears to be a substantial pathway for elimination of CNS-derived Ab (175). Recent studies demonstrated that low-density lipoprotein receptor-related protein (LRP1) is a major Ab efflux transporter at the BBB (176) but other Ab transporters have been proposed (177, 178). Region-specific levels of Ab deposition in postmortem AD brains are correlated inversely to the local level of Pgp in brain capillaries as assessed by immunohistochemistry (179). Using mdr1a/1b (−/−) double knockout mice, Ab removal from the brain was reported to be mediated at least partially by Pgp activity at the BBB (180). The study showed that acute inhibition of Pgp activity using a selective Pgp modulator increases Ab levels in brain interstitial fluid within hours of treatment. Importantly, mdr1a/1b (−/−) double knockout mice show enhanced Ab plaque formation in a mouse model of AD (180). These data strongly suggested that Pgp normally transports Ab out of the brain and that perturbation of Ab efflux directly affects Ab accumulation within the brain. Thus, there could be a correlation between Pgp and Ab metabolism in vivo such that Pgp activity at the BBB could impact risk for

172

Sivapackiam et al.

developing AD. Recently, these observations have been reinforced by evaluating transepithelial transwell transport of fluorescent Ab1–40/42 using MDR1 transfected LLC (a porcine-derived proximal renal tube epithelial) cells (181). O OH

OH OH

O

N

HO O

O O

O

O

O

OH

39

Fortunately, there are several commonly used drugs that could potentially alter Pgp function. For example, dexamethasone, morphine, rifampin, and St. John’s Wort are reported to induce Pgp activity, while verapamil, cyclosporine A, erythromycin 39, progesterone 40, HIV protease inhibitors, and several statins inhibit Pgp activity (6, 182). Typically, these medicinals only show modest impact on Pgp activity in short-term assays at clinically relevant concentrations, but with chronic use or in combination, these drugs may have a greater effect on Pgp function, and, consequently, in Parkinson’s disease or on brain Ab levels in AD. Recently, it has been shown that treatment of AD patients for 3 months with rifampin 41, an inducer of Pgp expression lead to noticeable improvement in their cognitive function.

HO O

O 40

O

O

H H

CH3 COO

H

N

O O

O

N

OH 41

8. Conclusions In summary, Pgp recognizes and outwardly transports a wide variety of bioinorganic complexes with a broad diversity of scaffolds in their chelation cores as well as conventional anticancer drugs.

Targeted Chemotherapy in Drug-Resistant Tumors

173

With the discovery of transduction sequences and prodrug approaches, chemotherapeutic drugs could possibly be delivered into drug-resistant cells/tumors. In addition to described strategies, future applications could involve enzyme-induced cleavage of the linkers attached to chemotherapeutic drugs for specific treatments. Both small organic medicinals as well as coordination compounds are amenable to incorporation of PET or SPECT radionuclides that may enable diagnostic imaging for functional interrogation of the MDR phenotype in cancer patients. In particular, clinically approved 99mTc-complexes have already shown to be beneficial in the evaluation of functional Pgp transport activity in human tumors in vivo. Therefore, contributions of bioinorganic radiochemistry and conventional organic radiochemistry are rapidly growing to enhance the existing armamentarium of targeted molecular imaging agents that would be capable of functionally analyzing the presence of Pgp in  vivo. With the advent of systemic biology and personalized medicine, both targeted chemotherapy as well as biomedical imaging could play a vital role in the near future. Finally, Pgp offers an interesting example of a protein that may unravel unparalleled discoveries at the interface of oncology and neurosciences, wherein a better understanding of the transporter in one discipline could assist the management (both therapeutic and diagnostic components) of diseases in the other.

Acknowledgments The authors are grateful to Prof. David Piwnica-Worms for inspiring and helpful discussions including coworkers (both past and present) for their contributions that led to successful execution of this tutorial. The financial assistance for the educational component of this contribution is also acknowledged in parts from NIH P50 CA94056, AG030498, and American Health Assistance Foundation, A2007-383 grants. References 1. Ling V, Thompson LH (1974) Reduced permeability in CHO cells as a mechanism of resistance to colchicine. J Cell Physiol 83: 103–111 2. Gerlach JH, Endicott JA, Juranka PF et  al (1986) Homology between P-glycoprotein and a bacterial haemolysin transport protein suggests a model for multidrug resistance. Nature 324:485–489

3. Gros P, Ben Neriah Y, Croop JM, Housman DE (1986) Isolation and expression of a complementary DNA that confers multidrug resistance. Nature 323:728–731 4. Endicott JA, Ling V (1989) The bioche­mistry of P-glycoprotein-mediated multidrug resistance. Annu Rev Biochem 58:137–171 5. Gottesman MM, Pastan I (1993) Biochemistry of multidrug resistance mediated by the

174

6. 7. 8.

9. 10.

11. 12.

13.

14.

15.

16. 17. 18. 19.

20.

Sivapackiam et al. multidrug transporter. Annu Rev Biochem 62:385–427 Gottesman M, Fojo T, Bates S (2002) Multidrug resistance in cancer: role of ATP-dependent transporters. Nat Rev Cancer 2:48–58 Bosch I, Croop J (1996) P-glycoprotein multidrug resistance and cancer. Biochim Biophys Acta 1288:F37–F54 Borst P, Evers R, Kool M, Wijnholds J (2000) A family of drug transporters: the multidrug resistance-associated proteins. J Natl Cancer Inst 92:1295–1302 Kruh G, Belinsky M (2003) The MRP family of drug efflux pumps. Oncogene 22:7537–7552 Ambudkar S, Dey S, Hrycyna C, Ramachandra M, Pastan I, Gottesman M (1999) Biochemical, cellular, and pharmacological aspects of the multidrug transporter. Annu Rev Pharmacol Toxicol 39:361–398 Ames G (1986) Bacterial periplasmic transport systems: structure, mechanism and evolution. Annu Rev Biochem 55:397–425 Gros P, Dhir R, Croop J, Talbot F (1991) A single amino acid substitution strongly modulates the activity and substrate specificity of the mouse mdr1 and mdr3 drug efflux pumps. Proc Natl Acad Sci USA 88:7289–7293 Bruggemann EP, Germann UA, Gottesman MM, Pastan I (1989) Two different regions of phosphoglycoprotein are photoaffinity-labeled by azidopine. J Biol Chem 264:15483–15488 Yoshimura A, Kuwazuru Y, Sumizawa T et al (1989) Cytoplasmic orientation and twodomain structure of the multidrug transporter, P-glycoprotein, demonstrated with sequence-specific antibodies. J Biol Chem 264:16282–16291 Raviv Y, Pollard H, Bruggemann E, Pastan I, Gottesman M (1990) Photosensitized labeling of a functional multidrug transporter in living drug-resistant tumor cells. J Biol Chem 265:3975–3980 Rosenberg M, Callaghan R, Modok S, Higgins C (2005) Three dimensional structure of P-glycoprotein. J Biol Chem 280:2857–2862 Higgins C, Linton K (2004) The ATP switch model for ABC transporters. Nat Struct Mol Biol 11:918–926 Higgins CF, Gottesman MM (1992) Is the multidrug transporter a flippase? Trends Biochem Sci 17:18–21 Altan N, Chen Y, Schindler M, Simon S (1998) Defective acidification in human breast tumor cells and implications for chemotherapy. J Exp Med 187:1583–1598 Hurwitz S, Terashima M, Mizunuma N, Slapak C (1997) Vesicular anthracycline accumulation

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

in doxorubicin-selected U-937 cells: participation of lysosomes. Blood 89:3745–3754 Lautier D, Canitrot Y, Deeley R, Cole S (1996) Multidrug resistance mediated by the multidrug resistance protein (MRP) gene. Biochem Pharmacol 52:967–977 Stein WD, Cardarelli C, Pastan I, Gottesman MM (1994) Kinetic evidence suggesting that the multidrug transporter differentially handles influx and efflux of its substrates. Mol Pharmacol 45:763–772 Luker GD, Flagg TP, Sha Q et  al (2001) MDR1 P-glycoprotein reduces influx of substrates without affecting membrane potential. J Biol Chem 276:49053–49060 Robinson L, Roberts W, Ling T et al (1997) Human MDR1 protein overexpression delays the apoptotic cascade in Chinese hamster ovary fibroblasts. Biochemistry 36:11169–11178 Smyth MJ, Krasovskis E, Sutton VR, Johnstone RW (1998) The drug efflux protein P-glycoprotein additionally protects drug-resistant tumor cells from multiple forms of caspase-dependent apoptosis. Proc Natl Acad Sci USA 95:7024–7029 Thiebaut F, Tsuruo T, Hamada H, Gottesman MM, Pastan I, Willingham MC (1987) Cellular localization of the multidrug-resistance gene product P-glycoprotein in normal human tissues. Proc Natl Acad Sci USA 84:7735–7738 Hitchins RN, Harman DH, Davey RA, Bell DR (1988) Identification of a multidrug resistance associated antigen (P-glycoprotein) in normal human tissues. Eur J Cancer Clin Oncol 24:449–454 Cordon-Cardo C, OBrien J, Casals D et al (1989) Multidrug-resistance gene (P-glycoprotein) is expressed by endothelial cells at bloodbrain barrier sites. Proc Natl Acad Sci USA 86:695–698 Rao V, Dahlheimer J, Bardgett M et al (1999) Choroid plexus epithelial expression of MDR1 P-glycoprotein and multidrug resistance-associated protein contribute to the blood-cerebrospinal fluid drug-permeability barrier. Proc Natl Acad Sci USA 96:3900–3905 Croop JM, Raymond M, Haber D et al (1989) The three mouse multidrug resistance (mdr) genes are expressed in a tissue-specific manner in normal mouse tissues. Mol Cell Biol 9:1346–1350 Schinkel A, Mayer U, Wagenaar E et al (1997) Normal viability and altered pharmacokinetics in mice lacking mdr1-type (drug-transporting) P-glycoproteins. Proc Natl Acad Sci USA 94:4028–4033 Luker G, Nilsson K, Covey D, Piwnica-Worms D (1999) MDR1 P-glycoprotein enhances

Targeted Chemotherapy in Drug-Resistant Tumors esterification of plasma membrane cholesterol. J Biol Chem 274:6979–6991 33. van Asperen J, van Tellingen O, van der Valk M, Rozenhart M, Beijnen J (1998) Enhanced oral absorption and decreased elimination of paclitaxel in mice cotreated with cyclosporin A. Clin Cancer Res 4:2293–2297 34. Kim RB, Fromm MF, Wandel C et al (1998) The drug transporter P-glycoprotein limits oral adsorption and brain entry of HIV-1 protease inhibitors. J Clin Invest 101:289–294 35. Sorrentino B, Brandt S, Bodine D et al (1992) Selection of drug-resistant bone marrow cells in  vivo after retroviral transfer of human MDR1. Science 257:99–103 36. Podda S, Ward M, Himelstein A et al (1992) Transfer and expression of the human multiple drug resistance gene into live mice. Proc Natl Acad Sci USA 89:9676–9680 37. Hanania E, Fu S, Roninson I, Zu Z, Gottesman M, Deisseroth A (1995) Resistance to taxol chemotherapy produced in mouse marrow cells by safety-modified retroviruses containing a human MDR-1 transcription unit. Gene Ther 2:279–284 3 8. Moscow J, Huang H, Carter C et  al (1999) Engraftment of MDR1 and NeoR gene-transduced hematopoietic cells after breast cancer chemotherapy. Blood 94:52–61 39. Shtil AA, Azare J (2005) Redundancy of biological regulation as the basis of emergence of multidrug resistance. Int Rev Cytol 246:1–29 40. Courtois G, Gilmore TD (2006) Mutations in the NF-kappaB signaling pathway: implications for human disease. Oncogene 25:6831–6843 41. Ogretmen B, Safa A (1999) Negative regulation of MDR1 promoter activity in MCF-7, but not in multidrug resistant MCF-7/Adr, cells by cross-coupled NF-kB/p65 and c-Fos transcription factors and their interaction with the CAAT region. Biochemistry 38:2189–2199 42. Lim JC, Kania KD, Wijesuriya H et al (2008) Activation of beta-catenin signalling by GSK-3 inhibition increases p-glycoprotein expression in brain endothelial cells. J Neurochem 106:1855–1865 43. Geick A, Eichelbaum M, Burk O (2001) Nuclear receptor response elements mediate induction of intestinal MDR1 by rifampin. J Biol Chem 276:14581–14587 44. Moore LB, Goodwin B, Jones SA et al (2000) St. John’s wort induces hepatic drug metabolism through activation of the pregnane X receptor. Proc Natl Acad Sci USA 97:7500–7502 45. Satsu H, Hiura Y, Mochizuki K, Hamada M, Shimizu M (2008) Activation of pregnane X

175

receptor and induction of MDR1 by dietary phytochemicals. J Agric Food Chem 56:5366–5373 46. Saeki M, Kurose K, Tohkin M, Hasegawa R (2008) Identification of the functional vitamin D response elements in the human MDR1 gene. Biochem Pharmacol 76:531–542 47. Sampath J, Sun D, Kidd VJ et  al (2001) Mutant p53 cooperates with ETS and selectively up-regulates human MDR1 not MRP1. J Biol Chem 276:39359–39367 48. Vilgelm A, Wei JX, Piazuelo MB et al (2008) DeltaNp73alpha regulates MDR1 expression by inhibiting p53 function. Oncogene 27:2170–2176 49. Polyakov V, Sharma V, Dahlheimer J, Pica C, Luker G, Piwnica-Worms D (2000) Novel Tat-peptide chelates for direct transduction of technetium-99  m and rhenium into human cells for imaging and radiotherapy. Bioconjug Chem 11:762–771 50. Lee C, Gottesman MM (1998) HIV-1 protease inhibitors and the MDR1 multidrug transporters. J Clin Invest 101:287–288 51. Rao K, Reddy M, Horning J, Labhasetwar V (2008) TAT-conjugated nanoparticles for the CNS delivery of anti-HIV drugs. Biomaterials 29:4429–4438 5 2. Liang J, Yang V (2005) Synthesis of doxorubicin-peptide conjugate with multidrug resistant tumor cell killing activity. Biorg Med Chem Lett 15:5071–5075 53. Tseng Y, Liu J, Hong R (2002) Translocation of liposomes into cancer cells by cell-penetrating peptides penetratin and tat: a kinetic and efficacy study. Mol Pharmacol 62:864–872 54. Magri N, Kingston D (1986) Modified taxols: 2. Oxidation products of taxol. J Org Chem 51:797–802 55. Chordia M, Yuan H, Jagtap P et  al (2001) Synthesis and bioactivity of 2, 4-diacyl analogues of paclitaxel. Biorg Med Chem 9:171–178 56. Tanino T, Nawa A, Kondo E et  al (2007) Paclitaxel-2¢-ethylcarbonate prodrug can circumvent P-glycoprotein mediated cellular efflux to increase drug cytotoxicity. Pharm Res 24:555–565 57. Zalipsky S, Saad M, KIwan R, Ber E, Yu N, Minko T (2007) Antitumor activity of new liposomal prodrug of mitomycin C in multidrug resistant solid tumor: insight of the mechanism of action. J Drug Target 15:518–530 58. Guillemard V, Saragovi H (2004) Prodrug chemotherapeutics bypass p-glycoprotein resistance and kill tumors with high efficacy and target-dependent selectivity. Oncogene 23:3613–3621

176

Sivapackiam et al.

59. Adams T, Epa V, Garrett T, Ward C (2000) Structure and function of the type 1 insulinlike growth factor receptor. Cell Mol Life Sci 57:1050–1093 60. Baserga R (1999) The IGF-I receptor in cancer research. Exp Cell Res 253:1–6 61. Ford JM, Hait WN (1990) Pharmacology of drugs that alter multidrug resistance in cancer. Pharmacol Rev 42:155–199 62. Pichler A, Prior J, Piwnica-Worms D (2004) Imaging reversal of multidrug resistance in living mice with bioluminescence: MDR1 P-glycoprotein transports coelenterazine. Proc Natl Acad Sci USA 101:1702–1707 63. Sandor V, Fojo T, Bates S (1998) Future perspectives for the development of P-glycoprotein modulators. Drug Resist Updat 1:190–200 64. Luo J, Manning B, Cantley L (2003) Targeting the PI3K-Akt pathway in human cancer: rationale and promise. Cancer Cell 4:257–262 65. Buchdunger E, Zimmermann J, Mett H et  al (1996) Inhibition of the Abl protein-tyrosine kinase in vitro and in vivo by a 2-phenylaminopyrimidine derivative. Cancer Res 56:100–104 66. Schindler T, Bornmann W, Pellicena P, Miller WT, Clarkson B, Kuriyan J (2000) Structural mechanism for STI-571 inhibition of Abelson tyrosine kinase. Science 289:1938–1942 67. Mahon FX, Deininger MWN, Schultheis B et al (2000) Selection and characterization of BCR-ABL positive cell lines with differential sensitivity to the tyrosine kinase inhibitor STI571: diverse mechanisms of resistance. Blood 96:1070–1079 68. Trock B, Leonessa F, Clarke R (1997) Multidrug resistance in breast cancer: a metaanalysis of MDR1/gp170 expression and its possible functional significance. J Natl Cancer Inst 89:917–931 69. Wilson W, Jamis-Dow C, Bryant et al (1995) Phase I and pharmacokinetic study of the multidrug resistance modulator dexverapamil with EPOCH chemotherapy. J Clin Oncol 13:1985–1994 70. Gaveriaux C, Boesch D, Jachez B (1991) PSC 833 a non-immunosuppressive cyclosporin analog, is a very potent multidrug-resistance modifier. J Cell Pharmacol 2:225–234 71. Hyafil F, Vergely C, Du Vignaud P, GrandPerret T (1993) In vitro and in vivo reversal of multidrug resistance by GF120918, an acridonecarboxamide derivative. Cancer Res 53:4595–4602 72. Dantzig A, Shepard R, Cao J et  al (1996) Reversal of P-glycoprotein-mediated multidrug resistance by a potent cyclopropyldibenzosuberane modulator, LY335979. Cancer Res 56:4171–4179

73. Germann U, Ford P, Schlakhter D, Mason V, Harding M (1997) Chemosensitization and drug accumulation effects of VX-710, verapamil, cyclosporin A, MS-209, and GF120918 in multidrug resistant HL60/ADR cells expressing the multidrug resistance-associated protein MRP. Anticancer Drugs 8:141–155 74. Agrawal M, Abraham J, Balis F et  al (2003) Increased Tc-99m-Sestamibi accumulation in normal liver and drug-resistant tumors after the administration of the glycoprotein inhibitor, XR9576. Clin Cancer Res 9:650–656 75. Mistry P, Stewart A, Dangerfield W et  al (2001) In vitro and in  vivo reversal of P-glycoprotein-mediated multidrug resistance by a novel potent modulator, XR9576. Cancer Res 61:749–758 76. Martin C, Berridge G, Mistry P, Higgins C, Charlton P, Callaghan R (1999) The molecular interaction of the high affinity reversal agent XR9576 with P-glycoprotein. Br J Pharmacol 128:403–411 77. Suzuki T, Fukazawa N, San-nohe K (1997) Structure activity relationship of newly synthesized quinoline derivatives for reversal of multidrug resistance in cancer. J Med Chem 40:2047–2052 78. Suzuki T, Mabuchi K, Fukazawa N (1999) Synthesis of glucuronides of multidrug resistance reversing drug MS209. Biorg Med Chem Lett 9:659–662 79. van Zuylen L, Sparrboom A, van der Gaast A et al (2000) The orally administered p-glycoprotein inhibitor R101933 does not alter the plasma pharmacokinetics of docetaxel. Clin Cancer Res 6:1365–1371 80. van Zuylen L, Sparrboom A, van der Gaast A et  al (2002) Disposition of docetaxel in the presence of P-glycoprotein inhibition by intravenous administration of R101933. Eur J Cancer 38:1090–1099 81. Wong E, Casciano C, Clement R, Johnson W (2001) The farnesyl protein transferase inhibitor SCH66336 is a potent inhibitor of MDR1 product p-glycoprotein. Cancer Res 61:7525–7529 82. Sarshar S, Zhang C, Moran EK et al (2000) 2, 4, 5-Trisubstituted imidazoles: novel nontoxic modulators of P-glycoprotein mediated multidrug resistance. Part 1. Biorg Med Chem Lett 10:2599–2601 83. Zhang C, Sarshar S, Moran EK et al (2000) 2, 4, 5-Trisubstituted imidazoles: novel nontoxic modulators of P-glycoprotein mediated multidrug resistance. Part 2. Biorg Med Chem Lett 10:2603–2605 84. Germann UA, Willingham MC, Pastan I, Gottesman MM (1990) Expression of the

Targeted Chemotherapy in Drug-Resistant Tumors human multidrug transporter in insect cells by a recombinant baculovirus. Biochemistry 29:2295–2303 85. Ahmad S, Safa AR, Glazer RI (1994) Modulation of P-glycoprotein by protein kinase Ca in a baculovirus expression system. Biochemistry 33:10313–10318 86. Homolya L, Hollo Z, Germann UA, Pastan I, Gottesman MM, Sarkadi B (1993) Fluorescent cellular indicators are extruded by the multidrug resistance protein. J Biol Chem 268:21493–21496 87. Abrams MA, Davison A, Jones AG, Costello CE, Pang H (1983) Synthesis and characterization hexakis(alkyl isocyanide) and hexakis(arylisocyanide) complexes of technetium(I). Inorg Chem 22:2798–2800 88. Wackers FJ, Berman D, Maddahi J et  al (1989) Tc-99m-hexakis 2-methoxy isobutylisonitrile: human biodistribution, dosimetry, safety and preliminary comparison to thallium-201 for myocardial perfusion imaging. J Nucl Med 30:301–309 89. Piwnica-Worms D, Chiu M, Budding M, Kronauge J, Kramer R, Croop J (1993) Functional imaging of multidrug-resistant P-glycoprotein with an organotechnetium complex. Cancer Res 53:977–984 90. Kronauge JF, Davison A, Roseberry AM, Costello CE, Maleknia S, Jones AG (1991) Synthesis and identification of monocation Tc(CPl)6+ in Tc(CNC(CH3)2COOCH3)6Cl and its hydrolysis products. Inorg Chem 30:4265–4271 91. Piwnica-Worms D, Kronauge J, Chiu M (1990) Uptake and retention of hexakis (2-methoxy isobutyl isonitrile) technetium(I) in cultured chick myocardial cells: mitochondrial and plasma membrane potential dependence. Circulation 82:1826–1838 92. Backus M, Piwnica-Worms D, Hockett D et al (1993) Microprobe analysis of Tc-MIBI in heart cells: calculation of mitochondrial potential. Am J Physiol (Cell) 265:C178–C187 93. Piwnica-Worms D, Rao V, Kronauge J, Croop J (1995) Characterization of multidrugresistance P-glycoprotein transport function with an organotechnetium cation. Biochemistry 34:12210–12220 94. Ballinger J, Hua H, Berry B, Firby P, Boxen I (1995) 99mTc-Sestamibi as an agent for imaging P-glycoprotein-mediated multi-drug resistance: in vitro and in vivo studies in a rat breast tumour cell line and its doxorubicin-resistant variant. Nucl Med Comm 16:253–257 95. Ballinger JR, Sheldon KM, Boxen I, Erlichman C, Ling V (1995) Differences between accumulation of Tc-99m-MIBI and Tl-201thallous chloride in tumor cells: role of P-glycoprotein. Q J Nucl Med 39:122–128

177

96. Cordobes M, Starzec A, Delmon-Moingeon L et  al (1996) Technetium-99m-Sestamibi uptake by human benign and malignant breast tumor cells: correlation with mdr gene expression. J Nucl Med 37:286–289 97. Luker G, Rao V, Crankshaw C, Dahlheimer J, Piwnica-Worms D (1997) Characterization of phosphine complexes of technetium (III) as transport substrates of the multidrug resistance (MDR1) P-glycoprotein and functional markers of P-glycoprotein at the blood-brain barrier. Biochemistry 36:14218–14227 98. Bosch I, Crankshaw C, Piwnica-Worms D, Croop J (1997) Characterization of functional assays of P-glycoprotein transport activity. Leukemia 11:1131–1137 99. Luker GD, Fracasso PM, Dobkin J, PiwnicaWorms D (1997) Modulation of the multidrug resistance P-glycoprotein: detection with Tc-99m-Sestamibi in vivo. J Nucl Med 38:369–372 100. Chen C, Meadows B, Regis J et  al (1997) Detection of in  vivo P-glycoprotein inhibition by PSC 833 using Tc-99m-Sestamibi. Clin Cancer Res 3:545–552 101. Barbarics E, Kronauge J, Cohen D, Davison A, Jones A, Croop J (1998) Characterization of P-glycoprotein transport and inhibition in vivo. Cancer Res 58:276–282 102. Peck R, Hewett J, Harding M et al (2001) Phase I and pharmacokinetic study of the novel MDR1 and MRP1 inhibitor biricodar administered alone and in combination with doxorubicin. J Clin Oncol 19:3130–3141 103. Piwnica-Worms D, Kronauge J (1994) Tc-99m-EIBI is a transport substrate for the multidrug resistance (MDR) P-glycoprotein. J Nucl Med 35:230P 104. Herman LW, Sharma V, Kronauge JF, Barbarics E, Herman LA, Piwnica-Worms D (1995) Novel hexakis(areneisonitrile) technetium(I) complexes as radioligands targeted to the multidrug resistance P-glycoprotein. J Med Chem 38:2955–2963 105. Jurisson S, Berning D, Jia W, Ma D-S (1993) Coordination compounds in nuclear medicine. Chem Rev 93:1137–1156 106. Deutsch E, Vanderheyden J-L, Gerundini P et  al (1987) Development of nonreducible technetium-99m(III) cations as myocardial perfusion imaging agents: initial experience in humans. J Nucl Med 28:1870–1880 107. Rossetti C, Vanoli G, Paganelli G et al (1994) Human biodistribution, dosimetry and clinical use of technetium(III)-99m–Q12. J Nucl Med 35:1571–1580 108. Crankshaw C, Marmion M, Luker G et al (1998) Novel Tc(III)-Q-complexes for functional imaging of the multidrug resistance (MDR1) P-glycoprotein. J Nucl Med 39:77–86

178

Sivapackiam et al.

109. Woulfe S, Dunn T, Marmion M, MacDonaold J, Rogic M, Deutsch E (1991) The synthesis of analogous N2O2, N2OS and diamide dimercapto Schiff base ligands for use in radiopharmaceutical inaging agents. J Nucl Med 32:1101 110. Jurisson SS, Dancey K, McPartlin M, Tasker PA, Deutsch E (1984) Synthesis, characterization, and electrochemical properties of technetium complexes containing both tetradentate Schiff base and monodentate tertiary phosphine ligands: single-crystal structure of trans-(N, N¢ethylenebis(acetylacetone iminato)) bis(triphe-ylphosphine)technetium(III) hexafluorophosphate. Inorg Chem 23:4743–4749 111. Kelly J, Forster A, Higley B et  al (1993) Technetium-99m-tetrofosmin as a new radiopharmaceutical for myocardial perfusion imaging. J Nucl Med 34:222–227 112. Ballinger JR, Bannerman J, Boxen I, Firby P, Hartman NG, Moore MJ (1996) Technetium99m-Tetrofosmin as a substrate for P-glycoprotein: in vitro studies in multidrugresistant breast tumor cells. J Nucl Med 37:1578–1582 113. Alberto R, Schibli R, Egli A, Schubiger A (1998) A novel organometallic aqua complex of technetium for the labeling of biomolecules: synthesis of (99mTc(OH2)3(CO)3)+ from (99mTcO4)- in aqueous solution and its reaction with a bifunctional ligand. J Am Chem Soc 120:7987–7988 114. Marmion M, MacDonald J (2000) Preparation and biodistribution of 99mTc(I)tricarbonyl complexes containing isonitrile and phosphine ligands. J Nucl Med 41:40P 115. Dyszlewski M, Blake H, Dahlheimer J, Pica C, Piwnica-Worms D (2002) Characterization of a novel Tc-99m-carbonyl complex as a functional probe of MDR1 P-glycoprotein transport activity. Mol Imaging 1:24–35 116. Hatada K, Riou L, Ruiz M et al (2004) 99mTcN-DBODC5, a new myocardial perfusion imaging agent with rapid liver clearance: comparison with 99mTc-sestamibi and 99mTc-tetrofosmin in rats. J Nucl Med 45:2095–2101 117. Bolzati C, Cavazza-Ceccato M, Agostini S, Tokunaga S, Casara D, Bandoli G (2008) Subcellular distribution and metabolism studies of the potential myocardial imaging agent (99mTc(N)(DBODC)(PNP5))+. J Nucl Med 49:1336–1344 118. Wolohan P, Reichert D (2007) Molecular modeling of hexakis(areneisonitrile) technetium(I), tricarbonyl-cyclopentadienyl technetium and technetium(V)-oxo complexes: MM3 parameter development and

prediction of biological properties. J Mol Graph Model 25:616–632 119. Wong E, Liu S, Lugger T, Hahn FE, Orvig C (1995) Hexadentate N4O2 amine phenol complexes of gallium and indium. Inorg Chem 34:93–101 120. Yang LW, Liu S, Wong E, Rettig SJ, Orvig C (1995) Complexes of trivalent metal ions with potentially heptadentate N4O3 Schiff base and amine phenol ligands of variable rigidity. Inorg Chem 34:2164–2178 121. Sarma BD, Bailar JC (1955) The stereochemistry of metal complexes with polydentate ligands. Part I. J Am Chem Soc 77:5476–5480 122. Sarma BD, Ray KR, Sievers RE, Bailar JC (1964) The stereochemistry of metal chelates with multidentate ligands. Part II. J Am Chem Soc 86:14–16 123. Tweedle MF, Wilson LJ (1976) Variable spin iron(III) chelates with hexadentate ligands derived from triethylenetetramine and varius salicyaldehydes. J Am Chem Soc 98:4824–4834 124. Tsang B, Mathias C, Green M (1993) A gallium-68 radiopharmaceutical that is retained in myocardium: 68  Ga-((4, 6-MeO)2sal)2BA PEN)+. J Nucl Med 34:1127–1131 125. Tsang B, Mathias C, Fanwick P, Green M (1994) Structure-distribution relationship for metal-labelled myocardial imaging agents: comparison of a series of cationic gallium(III) complexeswithhexadentatebis(salicylaldimine) ligands. J Med Chem 37:4400–4406 126. Sharma V, Beatty A, Wey S-P et  al (2000) Novel gallium(III) complexes transported by MDR1 P-glycoprotein: potential PET imaging agents for probing P-glycoprotein-mediated transport activity in  vivo. Chem Biol 7: 335–343 127. Sharma V, Crankshaw C, Piwnica-Worms D (1996) Effects of multidrug resistance (MDR1) P-glycoprotein expression levels and coordination metal on the cytotoxic potency of multidentate (N4O2) ethylenediamine-bis(propyl(R-benzylimino))metal(III) cations. J Med Chem 39:3483–3490 128. Sharma V, Wey SP, Bass L et  al (1996) Monocationic N4O2 Schiff-Base phenolate complexes of gallium(III): novel PET imaging agents of the human multidrug resistance (MDR1) P-glycoprotein. J Nucl Med 37:51P 129. Sharma V, Prior J, Belinsky M, Kruh G, Piwnica-Worms D (2005) Characterization of a gallium-67/gallium-68 radiopharmaceutical for SPECT and PET imaging of MDR1 P-glycoprotein transport activity in  vivo: validation in multidrug resistant tumors and at the blood-brain barrier. J Nucl Med 46:354–364

Targeted Chemotherapy in Drug-Resistant Tumors 130. Harpstrite SE, Prior JL, Rath NP, Sharma V (2007) Metalloprobes: synthesis, characterization, and potency of a novel gallium(III) complex in human epidermal carcinoma cells. J Inorg Biochem 101:1347–1353 131. Bigott H, McCarthy D, Wust F, Dahlheimer J, Piwnica-Worms D, Welch M (2001) Production, processing and uses of 94m-Tc. J Labeled Cpd Radiopharm 44:S119–S121 132. Anderson O, Packard A (1980) Structure variations in macrocyclic copper(II) complexes:crystal and molecular structure of cyano(difluoro-3, 3-(trimethylenedinitrilo)bis(2-butanone oxi mato)borato)copper II-methanol(Cu(cyclops) CN).CH3OH. Inorg Chem 19:2941–2945 133. Anderson O, Packard A (1979) Structural variations in macrocyclic copper(II) complexes: crystal and molecular structures of (Cu(cyclops) H2O)(ClO4) and (Cu(PreH)H2O)(ClO4). H2O. Inorg Chem 18:1940–1947 134. Packard A, Kronauge J, Barbarics E, Kiani S, Treves S (2002) Synthesis and biodistibution of a lipophilic 64Cu-labeled monocationic copper(II) complex. Nucl Med Biol 29: 289–294 135. Lewis J, Zweit J, Dearling J, Rooney B, Blower P (1996) Copper(I) bis diphosphine complexes as basis for radiopharmaceuticals for positron emission tomography and targeted radiotherapy. Chem Commun 1093–1094 136. Blower P (1998) Small coordination compounds as radiopharmaceuticals for cancer targeting. Transition Met Chem 23:109–112 137. Camus A, Marsich N, Nardin G (1976) Transition Met Chem 1:205 138. Berners-Price S, Johnson R, Mirabelli C, Faucette L, McCabe F, Sadler P (1987) Copper(I) complexes with bidentate tertiary phosphine ligands: solution chemistry and antitumor activity. Inorg Chem 26:3383–3387 139. Zweit J, Lewis J, Dearling J, Rooney B, Coley H, Kelland L (1997) Copper-64diphosphine complexes: potential PET tracers for the assessment of multidrug resistance in tumors (abstract). J Nucl Med 38:133P 140. Lewis J, Dearling J, Sosabowski J et al (2000) Copper bis(diphosphine)complexes: radiopharmaceuticals for detection of multidrug resistance in tumors by PET. Eur J Nucl Med 27:638–646 141. Schinkel AH, Mol CAAM, Wagenaar E, van Deemter L, Smit JJM, Borst P (1995) Multidrug resistance and the role of P-glycoprotein knockout mice. Eur J Cancer 31A:1295–1298 142. Schinkel A, Smit J, van Tellingen O et  al (1994) Disruption of the mouse mdr1a P-glycoprotein gene leads to a deficiency in

179

the blood-brain barrier and to increased sensitivity to drugs. Cell 77:491–502 143. Slapak C, Dahlheimer J, Piwnica-Worms D (2001) Reversal of multidrug resistance with LY335979: functional analysis of P-glycoprotein-mediated transport activity and its modulation in vivo. J Clin Pharmacol 41:29S–38S 144. Chen W, Luker K, Dahlheimer J, Pica C, Luker G, Piwnica-Worms D (2000) Effects of MDR1 and MDR3 P-glycoproteins, MRP1 and BCRP/MXR/ABCP on transport of Tc-99mTetrofosmin. Biochem Pharmacol 60:413–426 145. Mehta BM, Rosa E, Fissekis JD, Bading JR, Biedler JL, Larson SM (1992) In-vivo identification of tumor multidrug resistance with tritum-3-colchicine. J Nucl Med 33:1373–1377 146. Mehta B, Rosa E, Biedler J, Larson S (1994) In vivo uptake of carbon-14-colchicine for identification of tumor multidrug resistance. J Nucl Med 35:1179–1184 147. Elsinga PH, Franssen JF, Hendrikse NH et al (1996) Carbon-11-labeled daunorubicin and verapamil for probing P-glycoprotein in tumors with PET. J Nucl Med 37:1571–1575 148. Hendrikse N, Franssen E, van der Graaf W, Vaalburg W, de Vries E (1999) Visualization of multidrug resistance in  vivo. Eur J Nucl Med 26:283–293 149. Hendrikse N, de Vries EG, Eriks-Fluks L, et al (1999) A new in  vivo method to study P-glycoprotein transport in tumors and the blood–brain barrier. Cancer Res 59:2411–2416 150. Levchenko A, Mehta B, Lee J-B et al (2000) Evaluation of 11C-colchicine for PET imaging of multiple drug resistance. J Nucl Med 41:493–501 151. Luurtsema G, Molthoff C, Windhorst A et al (2003) (R)- and (S)-(11C) Verapamil as PET-tracers for measuring P-glycoprotein function: in vitro and in vivo evaluation. Nucl Med Biol 30:747–751 152. Hendrikse N, Vaalburg W (2002) Imaging of P-glycoprotein function in vivo with PET. Novartis Found Symp 243:137–145 153. Ravert H, Klecker R, Collins J et al (2002) Radiosynthesis of (C-11)paclitaxel. J Labelled Cpd Radiopharm 45:471–477 154. Kiesewetter D, Eckelman W (2001) Radiochemical synthesis of (18F)fluoropaclitaxel (abstract). J Labelled Cpd Radiopharm 44:S903–S905 155. Kiesewetter D, Jagoda E, Kao C et al (2003) Fluoro-, bromo-, and iodopaclitaxel derivatives: synthesis and biological evaluation. Nucl Med Biol 30:11–24 156. Drozdzik M, Bialecka M, Mysliwiec K, Honczarenko K, Stankiewicz J, Sych Z (2003)

180

Sivapackiam et al.

Polymorphism in the P-glycoprotein drug transporter MDR1 gene: a possible link between environmental and genetic factors in Parkinson’s disease. Pharmacogenetics 13: 259–263 157. Furuno T, Landi M, Ceroni M et al (2002) Expression polymorphism of the blood-brain barrier component p-glycoprotein (MDR1) in relation to Parkinson’s disease. Pharmacogenetics 12:529–534 158. Bartels A, Willemsen A, Kortekaas R et  al (2008) Decreased blood-brain barrier p-glycoprotein function in the progression of Parkinson’s disease, PSP, and MSA. J Neural Transm 115:1001–1009 159. Tainton K, Smyth M, Jackson J et al (2004) Mutational analysis of P-glycoprotein: suppression of caspase activation in the absence of ATP dependent drug efflux. Cell Death Differ 11:1028–1037 160. Schlachestzki F, Pardridge W (2003) P-glycoprotein and cavolin-1-alpha in endothelium and astrocytes of primate brain. NeuroReport 14:2041–2046 161. Lee G, Schlichter L, Bendayan M, Bendayan R (2001) Functional expression of P-glycoprotein in rat brain microglia. J Pharmacol Exp Ther 299:204–212 162. Volk H, Burkhardt K, Potschka H, Chen J, Becker A, Loscher W (2004) Neuronal expression of the drug efflux transporter P-glycoprotein in rat hippocampus after limbic seizures. Neuroscience 123:751–759 163. Volk H, Potschka H, Loscher W (2005) Immunohistochemical localization of P-glycoprotein in rat brain and detection of its increased expression by seizures are sensitive to fixation and staining variables. J Histochem Cytochem 53:517–531 164. McKhann G, Drachman D, Folstein M (1984) Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA work group under the auspices of the department of health and human services task force on Alzheimer’s disease. Neurology 34:939–944 165. Weiner M (1997) Alzheimer’s disease: diagnosis and treatment. Harvard Rev Psychiatry 4:306–316 166. Yanker B (1996) Mechanisms of neuronal degeneration in Alzheimer’s disease. Neuron 16:921–932 167. Lansbury PJ (1996) A reductionist view of Alzheimer’s disease. Acc Chem Res 29:317–321 168. Teller J (1996) Presence of soluble amyloid ß-peptide precedes amyloid plaque formation in Down’s syndrome. Nat Med 2:93–95

169. Games D, Adams D, Alessandrini R et  al (1995) Alzheimer-type neuropathology in transgenic mice overexpressing V717F ß-amyloid precursor protein. Nature 373:523–527 170. Hsiao K, Chapman P, Nilsen S et al (1996) Correlative memory deficits, Aß elevation, and amyloid plaques in transgenic mice. Science 274:99–102 171. Farris W, Mansourian S, Chang Y et al (2003) Insulin-degrading enzyme regulates the levels of insulin, amyloid b, and b-amyloid precursor protein intyracellular domain in vivo. Proc Natl Acad Sci USA 100:4162–4167 172. Iwata N, Mizukami H, Shirotani K et al (2004) Presynaptic localization of neprilysin contributes to efficient clearance of amyloid-beta peptide in mouse brain. J Neurosci 24:991–998 173. Weller R, Massey A, Newman T, Hutchings M, Kuo Y, Roher A (1998) Cerebral amyloid angiopathy: amyloid beta accumulates in putative interstitial fluid drainage pathways in Alzheimer’s disease. Am J Pathol 153:725–733 174. DeMattos R, Bales K, Cummins D, Paul S, Holtzman D (2002) Brain to plasma amyloid-b efflux: a measure of brain amyloid burden in a mouse model of Alzheimer’s disease. Science 295:2264–2267 175. Shibata M, Yamada S, Kumar S et al (2000) Clearance of Alzheimer’s amyloid-ss(1–40) peptide from brain by LDL receptor-related protein-1 at the blood-brain barrier. J Clin Invest 106:1489–1499 176. Zlokovic B (2005) Neurovascular mechanisms of Alzheimer’s neurodegeneration. Trends Neurosci 28:202–208 177. Deane R, Wu Z, Sagare A et al (2004) LRP/ amyloid b peptide interaction mediates differential brain efllux of Ab isoforms. Neuron 43:333–344 178. Zlokovic B, Yamada S, Holtzman D, Ghiso J, Frangione B (2000) Clearance of amyloid b pepitde from brain: transport or metabolism? Nat Med 6:718–719 179. Vogelgesang S, Cascorbi I, Schroeder R et  al (2002) Deposition of Alzheimer’s b amyloid is inversely correlated with P-glycoprotein expression in the brains of elderly non-demented humans. Pharmacogenetics 12:535–541 180. Cirrito JR, Deane R, Fagan AM et al (2005) P-glycoprotein deficiency at the blood-brain barrier increases amyloid-beta deposition in an Alzheimer’s disease mouse model. J Clin Invest 115:3285–3290 181. Kuhnke D, Jedlitschky G, Grube M et  al (2007) MDR1-P-Glycoprotein (ABCB1) Mediates Transport of Alzheimer’s amyloid-beta

Targeted Chemotherapy in Drug-Resistant Tumors peptides–implications for the mechanisms of Abeta clearance at the blood-brain barrier. Brain Pathol 17:347–353 182. Marzolini C, Paus E, Buclin T, Kim R (2004) Polymorphisms in human MDR1 (P-glycoprotein): recent advances and clinical relevance. Clin Pharmacol Ther 75:13–33

181

183. Hrycyna CA, Ramachandra M, Ambudkar SV et  al (1998) Mechanism of action of human P-glycoprotein ATPase activity. Photochemical cleavage during a catalytic transition state using orthovandate reveals cross talk between two ATP sites. J Biol Chem 273:16631–16634

sdfsdf

Chapter 9 Epigenetic Regulation of Multidrug Resistance 1 Gene Expression: Profiling CpG Methylation Status Using Bisulphite Sequencing Emma K. Baker and Assam El-Osta Abstract Methylation of CpG dinucleotides is one of the major epigenetic processes involved in the regulation of gene expression. Catalyzed by DNA methyltransferases, hypermethylation of CpG islands in promoter regions is typically associated with gene silencing. DNA methylation plays an important role in normal differentiation, development, and maintenance of genomic stability, with aberrant CpG methylation being linked with a number of disease states. Three CpG islands within a 1.15- kb region characterize the chromatin landscape surrounding the transcriptional start site of the multidrug resistance 1 (MDR1) gene. We and others have demonstrated that hypermethylation of this region is correlated with MDR1 gene silencing and the inability of chemotherapeutic agents to activate MDR1 transcription. The bisulphite sequencing and cloning method allows a precise interpretation of the methylation status of each individual CpG dinucleotide in the MDR1 region. Key words: Epigenetics, Chromatin, MDR1, CpG methylation, CpG island, MeCP2, HDAC, Bisulphite sequencing, Gene silencing

1. Introduction The expression of a gene is inherently determined by chromatin structure and function, and its conformation can be mediated by a number of epigenetic mechanisms. Methylation of the 5-carbon position of CpG dinucleotides catalyzed by DNA methyltransferases is one of the major epigenetic processes involved in controlling chromatin architecture and is biologically important for normal differentiation, development, and maintenance of genomic stability (1). Hypermethylation of CpG clusters, also known as CpG islands, which are associated with the promoter regions of approximately 60% of genes (2), is correlated with J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_9, © Humana Press, a part of Springer Science + Business Media, LLC 2010

183

184

Baker and El-Osta

long-term epigenetic silencing. CpG methylation can induce gene silencing by directly interfering with transcription factor binding (3); however, silencing is more commonly brought about by targeted binding of methyl-CpG-binding-domain (MBD) protein complexes that recruit histone deacetylase (HDAC) activity such as the MeCP2 corepressor complex (4, 5). A number of studies have identified that MDR1 transcriptional silencing is correlated with methylation of the promoter (6–10). We have further demonstrated that the methylation-dependent silencing of MDR1 is mediated by the recruitment of the MBD protein, MeCP2, HDAC1, and Brm, the catalytic subunit of the ATPase-dependent chromatin remodeller, SWI/SNF (7, 11). A CpG island is defined by a region of DNA greater than 200  bp, with a G/C content above 0.5 and an observed or expected presence of CpG dinucleotides above 0.6 (12). According to these specifications, the MDR1 gene has three CpG islands within a 1.15-kb region composed of the promoter, exon 1, and intron 1 sequences (Fig. 9.1). The methylation profile of the 66 CpG dinucleotides in the islands (Fig. 9.1) can be precisely determined using the bisulphite conversion technique (Fig. 9.2). The method is dependent upon the ability of sodium bisulphite to distinguish between unmethylated and methylated cytosine residues in single-stranded DNA, unmethylated residues being converted to uracil (13, 14). Amplifying with primers that recognize the modified single-stranded template results in uracil residues being represented as thymine residues in downstream sequencing analyses, allowing the differentiation between methylated and unmethylated cytosine residues. Direct sequencing of the PCR product provides a population average of the CpG dinucleotide methylation status. However, a CpG methylation pattern is not typically uniformally represented in every cell, especially if the DNA sample is derived from a mixed cell population. A more informative approach is to clone and sequence multiple PCR products to yield a very high resolution map of the methylation

Fig. 9.1. Schematic of the MDR1 CpG islands. The sequence of the MDR1 transcriptional start site (−409 to +719 bp) is characterized by three CpG islands within a region encompassing the promoter (black), exon 1 (light gray), and intron 1 (dark gray) (approximately 1.15 kb). The numbering of nucleotides is relative to the transcription initiation start site. 66 CpG dinucleotides within the 1.15-kb region are represented as circles in the schematic. The black arrows indicate the positioning of the three CpG islands, as predicted using the Methprimer software, freely available at http://www.urogene. org/methprimer/.

Epigenetic Regulation of Multidrug Resistance 1 Gene Expression

185

Fig.  9.2. Schematic of the sodium bisulphite conversion reaction. (a) Single-stranded DNA (induced by denaturation) is modified with sodium bisulphite converting unmethylated cytosines (C) to uracil (U), leaving methylated cytosines (Cm) unmodified. (b) The bisulphite converted DNA is amplified by PCR using primers designed to the converted sequence and lacking CpG dinucleotides in their recognition sequence. PCR amplification results in uracil (U) being replaced with thymine (T).

profile. The protocol we describe here includes the reagents and practical notes that will allow users to perform the bisulphite conversion and PCR amplification, cloning and sequencing steps with high reproducibility.

2. Materials 2.1. Cell Culture and DNA Extraction

1. RPMI 1640 medium (+l-glutamine) (Gibco Invitrogen, Mount Waverly, VIC, AUS) supplemented with 10% (v/v) FBS (JHR Biosciences, Lenexa, KS, USA) and 16 mg/ml gentamicin (Pizer, New York, NY, USA). Store at 4°C with protection from light. 2. Phosphate buffered saline (without magnesium and calcium) (PBS) was generated by the Baker IDI Heart and Diabetes Institute media kitchen staff for use in all experiments. Can be stored at room temperature for up to 36 months. 3. Cell Scrapers (Sarstedt, Newton, NC, USA). 4. DNeasy® Blood and Tissue Kit (Qiagen, Hilden, Germany). 5. Ethanol Absolute Analytical Grade (LabServ Biolab, Clayton, VIC, AUS).

2.2. Bisulphite Conversion

1. EZ DNA Methylation™ Kit (ZYMO Research, Orange, CA, USA). 2. Nuclease-free water (Ambion, Austin, TX, USA). Store at room temperature.

186

Baker and El-Osta

2.3. PCR Amplification

  1. Resuspend synthetic oligonucleotides in TE buffer at a concentration of 100 rmole/ml (Sigma, St Louis, MS, USA). The sequences of the oligonucleotides used to amplify the MDR1 region from bisulphite converted DNA are as follows: MDR1 external for 5¢-gaaattgttaagtatgttgaag-3¢; MDR1 external rev 5¢-aaattctcctcctttactcctc-3¢; MDR1 internal for 5¢-aaggtgttaggaagtagaaagg-3¢; MDR1 internal rev 5¢-tcttcttctttactcctccatt-3¢. Store at −20°C.   2. Taq polymerase supplied with PCR Taq buffer and 25 mM MgCl2 (Promega, Madison, WI, USA). Taq polymerase should be stored on ice when setting up reactions to maintain its enzymatic activity. Limit its exposure to temperatures greater than −20°C.   3. 2¢-Deoxynucleotide-5¢triphosphates (dNTPs) (100  mM) (Promega, Madison, WI, USA). 10 mM dNTP stocks of each of the four 2¢-deoxynucleotide-5¢triphosphates were diluted in nuclease-free water (Ambion). Store at −20°C.   4. Nuclease-free water (Ambion, Austin, TX, USA). Store at room temperature.

2.4. Restriction Enzyme Digestion and Agarose Gel Electrophoresis

  1. BglII and Sau3A restriction enzymes and dedicated buffers (Promega, Madison, WI, USA).   2. Nuclease-free water (Ambion, Austin, TX, USA). Store at room temperature.   3. 3 M NaAc pH5.2 (Amresco, Solon, OH, USA).   4. Ethanol Absolute Analytical Grade (LabServ Biolab, Clayton, VIC, AUS).   5. 70% Ethanol (Absolute Analytical Grade). Diluted in nucleasefree water.   6. Agarose I™(Amresco, Solon, OH, USA).   7. TAE buffer (1×): 40  mM Tris-acetate and 1  mM EDTA. Store at room temperature.   8. 10  mg/ml Ethidium Bromide dissolved in distilled water (Sigma, St Louis, MS, USA). Ethidium bromide is a known mutagen and care should be taken to limit exposure by wearing gloves and disposing of gels containing ethidium bromide in carcinogen biohazard bins marked for incineration.   9. pGEM DNA Marker (Promega, Madison, WI, USA). 10. Amresco® Gel Loading Buffer 4× with Bromophenol Blue (Amresco, Solon, OH, USA).

2.5. PCR Band Excision

  1. TAE buffer (1×): 40  mM Tris-acetate and 1  mM EDTA. Store at room temperature.

Epigenetic Regulation of Multidrug Resistance 1 Gene Expression

187

2. 10  mg/ml Ethidium Bromide dissolved in distilled water (Sigma, St Louis, MS, USA). Ethidium bromide is a known mutagen and care should be taken to limit exposure by wearing gloves and disposing of gels containing ethidium bromide in carcinogen biohazard bins marked for incineration. 3. pGEM DNA Marker (Promega, Madison, WI, USA). 4. Sterile scalpel blades. 5. Agarose I™(Amresco, Solon, OH, USA). 6. Amresco® Gel Loading Buffer 4× with Bromophenol Blue (Amresco, Solon, OH, USA). 7. Wizard® SV Gel and PCR Clean-up System (Promega, Madison, WI, USA). 2.6. Cloning of PCR Fragments

1. pCR®II-TOPO® vector kit with chemically competent Escherichia coli strain, One Shot® TOP10 cells and S.O.C. medium (Invitrogen, Mount Waverly, VIC, AUS). 2. Ampicillin (100 mg/ml)-selective Luria Bertani (LB) agar plates were prepared beforehand by the Baker IDI Heart and Diabetes Institute media kitchen staff for use in all experiments. Plates should be sealed with parafilm and stored inverted at 4°C to prevent condensation occurring on the agar. 3. 5¢ Bromo-4-chloro-3-indolyl-b-d-galactopyranoside (X-gal) dissolved in N,N-dimethylformamide (Sigma, St Louis, MS, USA) at 40 mg/ml. Store at −20°C protected from light. 4. LB broth was supplied by Baker IDI Heart and Diabetes Institute media kitchen staff and was supplemented with 100 mg/mL ampicillin (Sigma, St Louis, MS, USA). 5. Promega Wizard® Plus SV Miniprep kit (Promega, Madison, WI, USA).

2.7. Sequencing PCR Inserts

1. Big Dye Terminator Chemistry Version 3.1 (Applied Biosystems, Scoresby, VIC, AUS). 2. TOPO M13for (-20) and M13rev sequencing primers for pCR®II-TOPO® vector inserts. Suspend synthetic oligonucleotides in TE buffer at 3.2 rmole/ml (Sigma, St Louis, MS, USA). The sequences of the oligonucleotides are: M13for (-20) 5¢-gtaaaacgacggccag-3¢; M13rev 5¢-caggaaacagctatgac-3¢ (Invitrogen, Mount Waverly, VIC, AUS) (see Note 16). 3. Automated high-throughput capillary electrophoresis system on a 3100 Genetic Analyser (Applied Biosystems, Scoresby, VIC, AUS). 4. All sequencing reactions were carried out by the Baker IDI Heart and Diabetes Institute DNA Sequencing Facility.

188

Baker and El-Osta

3. Methods Hypermethylation of the MDR1 promoter is correlated with long-term silencing and an inability to induce transcription in response to a number of stimuli including chemotherapeutic agents and HDAC inhibitors (6, 7). Profiling of the methylation status of the promoter is therefore important when examining MDR1 transcriptional response in cell models and the clinical setting, where potential response to drugs is important. Bisulphite sequencing and cloning enables the precise determination of the methylation status of every individual CpG site within a given region. The method is dependent upon the complete conversion of unmethylated cytosine residues to uracil on single-stranded DNA mediated by sodium bisulphite. Following the conversion, the DNA must be purified (desalted), desulphonated, and neutralized (to remove the bisulfite adduct from the uracil ring, desulphonating the uracil sulphonate to uracil) and finally precipitated for use in downstream analyses. An inefficient conversion can lead to false positives preventing accurate methylation profiling. While methodology for non-kit bisulphite conversion is available (not discussed in this review), the EZ DNA Methylation™ Kit simplifies and streamlines the process in three steps. The in-column desulphonation step eliminates tedious precipitation steps that are necessary in non-kit methodology. The EZ DNA Methylation™ Kit also limits the amount of DNA material required for conversion, some non-kit protocols needing up to 20 mg of DNA due to excessive degradation. When the manufacturer’s (ZYMO) recommended DNA input is adhered to, the bisulphite reaction should be efficient, in excess of 99%. Primer design is a critical component in bisulphite sequencing for an unbiased representation of all CpG sites within a given region (see Note 1) and has a number of constraints, including no CG dinucleotides in the primer sequence. In addition, the modified single-stranded DNA template hampers primer selection and amplicon length is limited (optimal is less than 300 bp) due to sodium bisulphite reaction-mediated DNA degradation (see Note 2). Amplicon lengths greater than 300 bp, however, have been successfully designed as discussed in this review, often by implementing a nested PCR step. While direct sequencing of the PCR fragment would provide an average population profiling of every CpG site, cell to cell and copy number variations in individual CpG methylation can be identified by cloning and sequencing of multiple PCR fragments and is recommended for the most accurate representation of the methylation profile in that DNA sample. The following section details

Epigenetic Regulation of Multidrug Resistance 1 Gene Expression

189

the methodology used to establish the MDR1 methylation profile in HeLa cells. 3.1. Preparation of DNA

The ZYMO EZ DNA Methylation™ Kit indicates that the optimal DNA input is between 200 and 500  ng, with a maximum input of 2 mg of DNA. The DNA input volume for the kit must be less than 45 ml. Because of the considerable length of the MDR1 amplicon and inherent DNA degradation during sodium bisulphite conversion, 1.5 mg of DNA is used in the reaction to maximize the chances of a successful PCR amplification. We have routinely had successful conversions using 1.5-mg DNA with the kit. 1. In order to extract enough DNA at a sufficient concentration, culture HeLa cells in a 60-mm tissue culture dish in 4-ml RPMI 1640 medium until they are confluent. At confluency, the plate should contain approximately 2.5 × 106 HeLa cells. 2. Wash the cells twice with PBS to remove residual RPMI 1640 medium before harvesting. 3. Harvest the cells by scraping them off the tissue culture dish into 5 ml of PBS using a cell scraper and transfer to a 15-ml falcon tube. Pellet the cells by centrifuging at 1,000 g for 4 min at room temperature and discard the supernatant. 4. Prepare genomic DNA from the cell pellet using the DNeasy® Blood and Tissue Kit according to the manufacturer’s instructions (Qiagen). At the elution step, collect the DNA in two separate elutions, each of 100 ml (see Note 3).

3.2. Bisulphite Conversion of DNA

The ZYMO EZ DNA Methylation™ Kit efficiently bisulphite converts genomic DNA using a simplified and streamlined approach compared to more traditional non-kit methods. 1. Prepare 1.5 mg of genomic DNA in 45 ml of nuclease-free water for bisulphite conversion using the EZ DNA Methylation™ Kit. Carry out the conversion according to the manufacturer’s instructions (ZYMO) (see Note 4). 2. According to the instructions of the EZ DNA Methylation™ Kit, the bisulphite converted DNA is eluted in 10 ml of M-Elution buffer (provided in the kit). Increase this volume to 40 ml with nuclease-free water to provide sufficient template for multiple PCR reactions.

3.3. PCR Amplification of the 1.15-kb MDR1 Fragment

A 1.15-kb region of the MDR1 promoter (Fig. 9.1) is amplified in a nested PCR reaction using the primers detailed in subheading 9.2.3. The fragment is initially amplified to confirm that the bisulphite conversion has been successful using a restriction enzyme digest assay (Subheading 9.3.4). Once confirmation of a successful conversion is obtained, the fragment is amplified again to clone and sequence the MDR1 fragment (Subheading 9.3.5).

190

Baker and El-Osta

1. Amplify 3 ml of the diluted bisulphite converted DNA (Subheading  9.3.2) in five identical (see Note 5) 50-ml PCR reactions that contain 2.5 units of Taq polymerase, 2  mM MgCl2, 1× PCR Taq buffer, 0.25  mM of each dNTP, and 25 rmole of each of the primers MDR1 external for and MDR1 external rev. Amplify the primary product using the cycling parameters: 1× 95°C/5  min; 35× 95°C/45  s, 55°C/30  s, 72°C/2 min, with a final extension period of 1× 72°C/10 min. Include a negative water control in the PCR assay, consisting only of the reaction components and water as a template to exclude contamination. 2. Combine the five PCR reactions and make a 1/500 dilution with nuclease-free water. Amplify 5 ml of the diluted products in five individual 50-ml nested PCR reactions (see Note 5), using 25 rmole of each of the primers MDR1 internal for and MDR1 internal rev, 2.5 units of Taq polymerase, 2  mM MgCl2, 1× PCR Taq buffer, and 0.25  mM of each dNTP. Amplify the secondary product using the cycling parameters: 1× 95°C/5  min; 35× 95°C/45  s, 55°C/30  s, 72°C/2  min (see Note 6), with a final extension period of 1× 72°C/30 min (see Note 7). Include a negative water control, consisting only of the reaction components and water as a template in the PCR assay to exclude contamination. 3. Place the PCR products designated for gel-extraction purification (Subheading 9.3.5) and cloning (Subheading 9.3.6) on ice (see Note 8). It is not essential to keep PCR products designated for restriction enzyme analyses on ice and these may be stored at room temperature for several hours, or alternatively at −20°C for longer periods of time. 3.4. Restriction Enzyme Digestion and Agarose Gel Electrophoresis

The restriction enzymes BglII and Sau3A contain cytosine residues in their sequence recognition motifs that are not within a CpG sequence. The BglII and Sau3A sequence recognition motifs are found within the 1.15-kb region of the MDR1 promoter. Bisulphite conversion would remove their sequence recognition motifs from the MDR1 fragment, as the cytosine residues would be converted to uracil in the reaction. Therefore, a successful conversion of the DNA would prevent digestion of the MDR1 PCR product with the restriction enzymes BglII and Sau3A. 1. Pool the five PCR products designated for restriction enzyme digestion analyses and precipitate with 1/10 (v/v) 3 M NaAc pH5.2 and 2.5 volumes of ethanol at −20°C for a minimum of 1 h. Collect the precipitated PCR fragments by centrifugation at 15,000 g for 15 min at 4°C. Wash the pellet with 1 ml 70% ethanol and centrifuge at 13,000 rpm for 5 min at 4°C. Remove the supernatant, air dry the pellet for 4–5 min at room temperature, and dissolve in 30-ml nuclease-free water with pipetting.

Epigenetic Regulation of Multidrug Resistance 1 Gene Expression

191

2. To confirm that the bisulphite conversion reaction is complete, digest 10 ml of the precipitated PCR products with each of the restriction enzymes BglII and Sau3A in 20-ml reactions for 2 h at 37°C according to the manufacturer’s recommendations (Promega). After the incubation, heat the reactions to 70°C for 20 min to cease the enzymatic activity. 3. PCR product restriction enzyme digests are visualized by agarose gel electrophoresis. In a glass 250-ml conical flask, mix the constituents for a 1% agarose gel (100  ml 1× TAE, 1-g Agarose I™, 10 ml 10 mg/ml ethidium bromide) and heat for approximately 90  s on a medium setting of a microwave to totally resuspend the solution, taking care not to overboil the solution. While in a liquid state, pour the gel solution into a gel tray (150 mm × 100 mm) equipped with a comb with loading well dimensions of 4 mm × 1.5 mm. Incubate at room temperature until the agarose has polymerized. Remove the comb and transfer the gel tray with the agarose gel to an electrophoresis chamber and submerge in 1× TAE buffer. 4. Combine the restriction enzyme digest products (20 ml) with gel loading dye to a final concentration of 1 ml and load into individual wells of the agarose gel. In a separate well, load 1 ml of the pGEM DNA Marker (used to confirm the fragment sizes). Electrophorese the gel at 120  V for approximately 30 min to size fractionate the restriction digest products. 5. We visualize the restriction digest bands using a ChemiDoc XRS (Biorad) (Fig.  9.3). Successful bisulphite conversion is observed in the left hand panel of Fig. 9.3 by the lack of digestion of the 1.15-kb fragment with the enzymes BglII and Sau3A. In the right hand panel of Fig.  9.3, digestion of the 1.15-kb fragment with BglII and Sau3A indicates incomplete conversion of the genomic DNA by sodium bisulphite. 3.5. PCR Band Excision

Once the successful bisulphite conversion has been confirmed by restriction enzyme digestion (Subheading  9.3.4), amplify the bisulphite converted DNA by PCR (Subheading  9.3.3) for a second time and purify by agarose gel extraction for cloning (Subheadings 9.3.5 and 3.6). 1. Pool the five PCR reactions designated for cloning and mix gently. Leave the products on ice until size fractionation by agarose gel electrophoresis (see Note 8). 2. PCR products are separated by agarose gel electrophoresis. In a glass 250-ml conical flask, mix the constituents for a 1% agarose gel (100  ml 1× TAE, 1-g Agarose I™, 10 ml 10  mg/ml ethidium bromide) and heat for approximately 90  s on a medium setting of a microwave to totally resuspend the solution taking care not to overboil the solution. While in a liquid state, pour the gel solution into a gel tray (150 mm × 100 mm)

192

Baker and El-Osta

Fig. 9.3. Confirmation of successful sodium bisulphite conversion of genomic DNA by restriction enzyme digestion of PCR products. Restriction enzyme-digested PCR products were size fractionated on a 1% TAE agarose gel. A 1.15-kb band indicates nondigestion of the PCR product by the restriction enzymes BglII and Sau3A. Smaller fragments after digestion with BglII and Sau3A are evidence of incomplete conversion of genomic DNA by sodium bisulphite. pGEM® DNA size marker (M).

equipped with a comb with loading well dimensions of 4 mm × 1.5 mm. Incubate at room temperature until the agarose has polymerized. Remove the comb and transfer the gel tray with the agarose gel to an electrophoresis chamber and submerge in 1× TAE buffer. 3. Combine 20 ml of the pooled PCR products with gel loading dye to a final concentration of 1× and load into individual wells of the gel. In a separate well, load 1 ml of the pGEM DNA Marker (used to confirm the PCR product sizes). Electrophorese the gel at 120 V for approximately 30 min to size fractionate the PCR products. 4. Clean the surface of an ultraviolet (UV) light box by wiping with 70% ethanol, and place the agarose gel on the light source. Ethidium bromide fluoresces under UV light when intercalated with nucleic acid polymers, allowing the visualization of the PCR products. Limit skin and eye exposure to the UV light source by using perspex screens or protective equipment. 5. Excise the correctly sized PCR fragments from the gel using sterile scalpel blades. A single blade should be used for each PCR fragment to prevent cross-contamination of samples. While it is essential to excise the entire fragment, limit the amount of

Epigenetic Regulation of Multidrug Resistance 1 Gene Expression

193

nonessential agarose in the excision. Try to work as efficiently as possible, as continual exposure of the gel to UV light increases the risk of DNA lesions, damaging the fragments. 6. Recover the DNA from the agarose gel piece using the Wizard® SV Gel and PCR Clean-up System (Promega) as specified by the manufacturer. 7. Place the purified fragment on ice until vector ligation (see Note 8). 3.6. Cloning of PCR Fragments

Cloning the PCR fragments enables the sequencing of individual alleles providing an in-depth representation of the methylation profile in that DNA sample, compared to directly sequencing the PCR fragment, which only portrays an average methylation profile. 1. Agarose gel purification dilutes the PCR product. If excising a very faint band, it will be necessary to use the maximal template volume (4 ml) in the vector ligation. Only minimal template addition (1 ml) in the vector ligation should be necessary if strong PCR bands were evident under UV irradiation. 2. Subclone the gel-purified PCR products into the pCR®IITOPO® vector as specified by the manufacturer (Invitrogen). Include a control ligation, where PCR product template is replaced by water. Extend the benchtop room temperature ligation to 30 min to improve the efficiency of subcloning the large MDR1 fragment. 3. During this time, warm the S.O.C. medium (Invitrogen) to room temperature. Equilibrate a waterbath to 42°C, and warm the ampicillin-selective Luria Bertani (LB) agar plates to 37°C. Working near a blue flame of a Bunsen burner and using aseptic techniques to prevent contamination, spread the warm ampicillin-selective Luria Bertani (LB) agar plates with 40 ml of 40  mg/ml X-gal to enable blue/white colony screening of positive PCR product insert clones (white colonies) (see Note 9). Return the agar plates to 37°C until needed. 4. Transform the PCR product-vector ligations into the chemically competent E. coli strain One Shot® TOP10 cells by heat shock as specified by the manufacturer of the pCR®II-TOPO® vector kit (Invitrogen). We find that the transformation efficiency is improved by extending the 42°C heat shock time to 45  s. A water bath is used for the heat shock incubation to ensure equal tube exposure to heat, which cannot be guaranteed when using a standard dry cell heat block. After heat shock, immediately transfer the tubes to ice. 5. Add 250 ml of S.O.C. medium to the tubes, seal the lids, and shake horizontally (200 rpm) at 37°C for 1 h.

194

Baker and El-Osta

6. Working near a blue flame of a Bunsen burner and using aseptic techniques to prevent contamination, spread 50 ml of the transformed cells onto the prewarmed ampicillin-selective Luria Bertani (LB) agar plates with X-gal using a sterile plate spreader (see Note 10). Incubate the plates overnight at 37°C inverted, to prevent condensation collecting on the agar. 7. Successful cloning will be evidenced by a lack of colonies on the agar plate spread with the water control transformed cells and a high ratio of white to blue colonies on plates spread with PCR product transformations. Pick single white colonies with sterile pipette tips and inoculate to tubes containing 3 ml of LB broth supplemented with 100 mg/ml ampicillin. Incubate the culture in a shaking incubator overnight at 37°C (approximately 16–18 h). We routinely pick 5–10 colonies for the initial analysis. More clones can be analyzed at a later date if needed, as viable colonies can still be picked from plates stored at 4°C for up to 4 weeks and inoculated directly to LB broth supplemented with 100  mg/ml ampicillin. Seal plates with parafilm, wrap in aluminium foil, and store inverted at 4°C. 8. Recover plasmid DNA containing MDR1 inserts using the Promega Wizard® Plus SV Minipreps as specified by the manufacturer (Promega). Bacterial cells can be pelleted and stored at −20°C indefinitely if the plasmid DNA cannot be recovered immediately. 3.7. Sequencing PCR Inserts

1. 300 ng of purified plasmid DNA (a minimum of 100 ng in a 13-ml volume can be sequenced) is sequenced by the DNA Sequencing Laboratory at the Baker IDI Heart and Diabetes Institute using an automated, high-throughput, capillary electrophoresis system on a 3100 Genetic Analyser (Applied Biosystems). 2. Sequencing reactions are carried out using Big Dye Terminator Chemistry Version 3.1 (Applied Biosystems) with TOPO M13for (-20) and M13rev sequencing primers for pCR®IITOPO® vector inserts (Invitrogen) (see Note 11). 3. Align the bisulphite converted MDR1 sequences against a template of the unconverted genomic DNA sequence to determine the methylation status of each cytosine residue. Examine the bisulphite converted MDR1 sequencing files for cytosine residues in non-CpG contexts. If evident, discard these sequencing files as this is indicative of incomplete bisulphite conversion of the DNA. Align the sequences. 4. Bisulphite sequencing data are often presented in some form of color-coded schematic. Figure  9.4 demonstrates how we would typically present data for the MDR1 gene. Each row of circles should represent a single cloned allele and each circle

Epigenetic Regulation of Multidrug Resistance 1 Gene Expression

195

Fig. 9.4. The methylation status of the MDR1 promoter, exon 1 and intron 1 region (−409 to +719 bp) in HeLa cells was analyzed by bisulphite sequencing. Each row of circles represents a single cloned allele and each circle represents a single CpG dinucleotide. White circles – unmethylated CpG dinucleotide and black circles – methylated CpG dinucleotide. (a) Schematic of the MDR1 1.15 kb region, including promoter, exon 1, and intron 1. The numbering of nucleotides is relative to the transcription initiation start site. The black arrows indicate the positioning of the three CpG islands. (b) Methylation profile of the MDR1 1.15 kb region in HeLa cells.

should represent a single CpG dinucleotide. Black circles represent a methylated cytosine and white circles represent an unmethylated cytosine.

4. Notes 1. The primer sequences for amplifying a 1.15-kb region of the MDR1 locus encompassing 3 CpG islands are included in this protocol. If the methylation profiles of different MDR1 regions or different genes were of interest, primers specifically for bisulphite-modified DNA analyses can be designed using Methprimer software, freely available at http://www.urogene. org/methprimer/. 2. DNA degradation in bisulphite conversion reactions has been suggested to be as high as 90%. The high rate of degradation often results in poor PCR amplification and can limit the amplicon size for primer design. The ZYMO EZ DNA Methylation™ Kit advertizes that 80% of the DNA template is recovered. Despite the company recommendation of using between 200 and 500 ng of DNA, 1.5 mg is used to improve the total yield of DNA template recovered, aiding in amplification of the large 1.15-kb MDR1 fragment. 3. Elute the bound DNA into a new Eppendorf tube in two separate 100-ml elutions. The concentration of the eluted DNA should be well within the limits (>0.04 mg/ml) of converting 1.5 mg of DNA in the allowable 45-ml volume.

196

Baker and El-Osta

  4. The Ct conversion reagent liquid solution in the EZ DNA Methylation™ Kit is light sensitive, so minimize its exposure to light by containing in aluminum foil. Unused CT reagent liquid solution can be stored for up to 1 month at −20°C. Warm to 37°C and vortex before use. It is essential that the 50°C incubation is carried out in darkness. Exposure to light will cause incomplete conversion. A heat block can be used for the incubation if the block is covered with aluminum foil.   5. Amplify the primary and secondary PCR reactions in five identical reactions in order to minimize the effects of PCR artifacts.   6. The number of rounds in the secondary PCR may need to be reduced if the product is strongly amplified, potentially within the range of 25 cycles. Additionally, if only very weak products are observed, increase the cycles to 40–45.   7. The final extension time in the secondary PCR method is increased to 30 min because it can improve Taq-polymerasemediated addition of a single deoxyadenosine (A) to the 3¢ ends of PCR products, which is necessary for cloning into the pCR®II-TOPO® vector. Taq is least efficient at adenylating next to another adenosine residue and most efficient at adenylating next to a cytosine residue. Therefore, primer designs typically should not contain 5¢ T residues and would be improved by a 5¢ G residue as recommended by the pCR®IITOPO® vector kit manufacturer, Invitrogen. The current primer designs do not follow these recommendations; however, with care taken to handle the PCR products postamplification and precloning, efficient ligation reactions can be performed.   8. The loss of adenylation at the 3¢ ends of PCR products mediated by Taq-polymerase can be increased by temperature fluctuations and delays in processing. The pCR®II-TOPO® vector is supplied linearized with single 3¢-thymidine (T) overhangs for TA cloning® (Invitrogen), which allows the direct insertion of Taq polymerase-amplified PCR products. Taq-polymerase adds a single deoxyadenosine (A) to the 3¢ ends of PCR products during amplification, which enables the ligation of the PCR products with the pCR®II-TOPO® vector. Delays between amplifying the PCR product and vector ligation can lead to inadvertent loss of the 3¢ A overhang resulting in inefficient cloning. Methods are available to add 3¢ A-overhangs to the PCR products postamplification; however, the cloning reaction is generally not as efficient.   9. X-gal is used to indicate whether E. coli One Shot® TOP10 cells express the b-galactosidase enzyme. The pCR®II-TOPO® vector contains the LacZa gene, which encodes for a truncated

Epigenetic Regulation of Multidrug Resistance 1 Gene Expression

197

version of the b-galactosidase enzyme. The multiple cloning site of the pCR®II-TOPO® vector is located within the LacZa gene. When the LacZa gene is intact in these cells, the X-gal is cleaved by b-galactosidase producing an insoluble blue product (5,5¢-dibromo-4,4¢-dichloro-indigo) resulting in a blue colony. When the MDR1 PCR product has been successfully ligated into the multiple cloning site of the pCR®IITOPO® vector, the LacZa gene open reading frame is disrupted, preventing the production of the b-galactosidase resulting in a white colony. 10. A steel plate spreader can be sterilized by dipping it in 100% ethanol and exposing to the blue flame of a Bunsen burner. Make sure the loop is cooled sufficiently before applying to the agar plate, or you risk killing the bacterial cells with excessive heat. Alternatively, single-use sterile plate spreaders can be made by bending a glass Pasteur pipette using the blue flame of the Bunsen burner. 11. Sequencing with the M13rev sequencing primer typically returns good-quality sequencing reads in excess of 700  bp indicating that the Big Dye Terminator Chemistry works well despite typically high C/T contents in bisulphite converted sequences. Sequencing with the M13for (-20) sequencing primer is often compromised by being significantly reduced in read length, suggesting an incompatibility between the Big Dye Terminator Chemistry and high T/G content. The combined read length from the two sequencing reactions is typically long enough to cover the 1.15-kb MDR1 insert. If sequencing coverage is lacking however, internal sequencing primers should be designed to a bisulphite converted MDR1 sequence that lacks CpG dinucleotides. References 1. Bird A (2002) DNA methylation patterns and epigenetic memory. Genes Dev 16:6–21 2. Larsen F, Gundersen G, Lopez R, Prydz H (1992) CpG islands as gene markers in the human genome. Genomics 13:1095–1107 3. Bird AP, Wolffe AP (1999) Methylationinduced repression–belts, braces, and chromatin. Cell 99:451–454 4. Jones PL, Veenstra GJ, Wade PA et al (1998) Methylated DNA and MeCP2 recruit histone deacetylase to repress transcription. Nat Genet 19:187–191 5. Nan X, Ng HH, Johnson CA et  al (1998) Transcriptional repression by the methyl-CpGbinding protein MeCP2 involves a histone deacetylase complex. Nature 393:386–389

6. Baker EK, Johnstone RW, Zalcberg JR, El-Osta A (2005) Epigenetic changes to the MDR1 locus in response to chemotherapeutic drugs. Oncogene 24:8061–8075 7. El-Osta A, Kantharidis P, Zalcberg JR, Wolffe AP (2002) Precipitous release of methyl-CpG binding protein 2 and histone deacetylase 1 from the methylated human multidrug resistance gene (MDR1) on activation. Mol Cell Biol 22:1844–1857 8. Kantharidis P, El-Osta A, deSilva M et al (1997) Altered methylation of the human MDR1 promoter is associated with acquired multidrug resistance. Clin Cancer Res 3:2025–2032 9. Kusaba H, Nakayama M, Harada T et  al (1999) Association of 5’ CpG demethylation

198

Baker and El-Osta

and altered chromatin structure in the promoter region with transcriptional activation of the multidrug resistance 1 gene in human cancer cells. Eur J Biochem 262: 924–932 10. Nakayama M, Wada M, Harada T et al (1998) Hypomethylation status of CpG sites at the promoter region and overexpression of the human MDR1 gene in acute myeloid leukemias. Blood 92:4296–4307 11. Harikrishnan KN, Chow MZ, Baker EK et al (2005) Brahma links the SWI/SNF chromatinremodeling complex with MeCP2-dependent

transcriptional silencing. Nat Genet 37: 254–264 12. Gardiner-Garden M, Frommer M (1987) CpG islands in vertebrate genomes. J Mol Biol 196:261–282 13. Clark SJ, Harrison J, Paul CL, Frommer M (1994) High sensitivity mapping of methylated cytosines. Nucl Acids Res 22:2990–2997 14. Frommer M, McDonald LE, Millar DS et al (1992) A genomic sequencing protocol that yields a positive display of 5-methylcytosine residues in individual DNA strands. Proc Natl Acad Sci USA 89:1827–1831

Chapter 10 Expression and Function of P-Glycoprotein in Normal Tissues: Effect on Pharmacokinetics Frantisek Staud, Martina Ceckova, Stanislav Micuda, and Petr Pavek Abstract ATP-binding cassette (ABC) drug efflux transporters limit intracellular concentration of their substrates by pumping them out of cell through an active, energy dependent mechanism. Several of these proteins have been originally associated with the phenomenon of multidrug resistance; however, later on, they have also been shown to control body disposition of their substrates. P-glycoprotein (Pgp) is the first detected and the best characterized of ABC drug efflux transporters. Apart from tumor cells, its constitutive expression has been reported in a variety of other tissues, such as the intestine, brain, liver, placenta, kidney, and others. Being located on the apical site of the plasma membrane, Pgp can remove a variety of structurally unrelated compounds, including clinically relevant drugs, their metabolites, and conjugates from cells. Driven by energy from ATP, it affects many pharmacokinetic events such as intestinal absorption, brain penetration, transplacental passage, and hepatobiliary excretion of drugs and their metabolites. It is widely believed that Pgp, together with other ABC drug efflux transporters, plays a crucial role in the host detoxication and protection against xenobiotic substances. On the other hand, the presence of these transporters in normal tissues may prevent pharmacotherapeutic agents from reaching their site of action, thus limiting their therapeutic potential. This chapter focuses on P-glycoprotein, its expression, localization, and function in nontumor tissues and the pharmacological consequences hereof. Key words: P-glycoprotein, Expression, Function, Localization, Normal tissues, Pharmacokinetics, Drug absorption, Drug distribution, Drug excretion

1. Introduction Drug efflux transporters of the ATP-binding cassette (ABC) family are membrane-embedded proteins that limit intracellular concentration of substrate agents by pumping them out of cell through an active, energy dependent mechanism. Several of these proteins, mainly P-glycoprotein (Pgp, also known as ABCB1), multidrug resistance-associated protein 1 (MRP1, also known as ABCC1), and breast cancer resistance protein (BCRP, also known J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_10, © Humana Press, a part of Springer Science + Business Media, LLC 2010

199

200

Staud et al.

as ABCG2, MXR, and ABCP), have been associated with the phenomenon of multidrug resistance in cancer therapy. In addition, these transporters have also been localized in normal tissues where they seem to control body disposition of their substrates, such as brain penetration, intestinal absorption, maternal-fetal transport, or hepatobiliary excretion. Therefore, ABC drug efflux transporters are widely believed to play a crucial role in host detoxication and protection against xenobiotic substances. Pgp, the paradigm ABC drug efflux transporter, is the first detected and to date the best characterized of the family of ABC drug efflux transporters. It was first identified in 1972 by Juliano and Ling (1) as a surface phosphoglycoprotein expressed in drug-resistant Chinese hamster ovary cells. It gained worldwide attention about three decades ago for its role in the phenomenon of multidrug resistance in tumor cells (2–4). Subsequently, constitutive expression of Pgp has been described in a variety of other tissues including the liver, intestine, kidney, pancreas, adrenal, capillary endothelium of blood–brain and blood–testis barrier, choroid plexus, placental trophoblast, and others (Figs.  10.1 and 10.3) (5, 6). The polarized, apical membrane localization of Pgp causes that its substrates are preferentially translocated from basolateral to the apical side of the epithelium. Thus, Pgp limits the influx and facilitates the efflux of its substrates, eventually preventing their intracellular accumulation. Many in vitro and in  vivo studies have demonstrated high impact of Pgp on drug pharmacokinetics in these organs (5). It is likely that Pgp and other ABCs have evolved in these “normal” tissues to protect them from potentially damaging effects of toxic compounds. The function of Pgp can be divided, based on its anatomical localization, into three steps: (i) Pgp limits intestinal absorption of drugs; (ii) once the drug is in the systemic circulation, Pgp protects its passage to sensitive organs and tissues; and finally (iii) Pgp also facilitates elimination of drugs and metabolites into bile and urine (7). Therefore, it greatly affects the fate of a drug in the body as well as effectiveness of drug treatment. However, as pointed by Schinkel and Jonker (5), Pgp will only result in noticeable distribution effects if the rate of active transport for a certain compound is substantially relative to passive diffusion rate. If not, the pump activity will be overwhelmed by the passive diffusion of the component.

2. Substrates, Inhibitors, and Inducers of P-glycoprotein

Pgp transports an extremely wide variety of chemically and structurally diverse compounds. Pgp substrates are usually organic molecules ranging in size from about 200 Da to over 1,000 Da.

Expression and Function of P-Glycoprotein in Normal Tissues

201

Fig. 10.1. Localization of P-glycoprotein in human nontumor tissues; see text for detailed description. For P-glycoprotein localization in the placenta see Fig. 10.3.

Although the structure-activity relationship for Pgp substrates has not been fully elucidated to date, it seems that both lipophilicity and number of hydrogen bonds are the most important properties for Pgp affinity (8). Thus, most substrates are uncharged or weakly basic in nature, but some acidic compounds can also be transported. It is of clinical importance that a large number of drugs, or their metabolites, of various pharmacotherapeutic groups have been recognized as Pgp substrates (5, 7). The list of substrates and inhibitors is constantly growing and includes, for example, cytotoxic drugs, HIV protease inhibitors, antibiotics, opioids, antiemetics, as well as diagnostic dyes rhodamine 123 or Hoechst 33342 (see Table  10.1). In addition, considerable overlap in substrates between Pgp and CYP3A4 is often discussed, pointing out synergistic effect of CYP3A4 and Pgp in detoxication processes, mainly in the small intestine (see Subheading 4.1) (9).

202

Staud et al.

Table 10.1 List of Pgp substrates, inhibitors, and inducers. Based on [5, 12, 16, 42, 136] Pgp substrates

Pgp inhibitors

Pgp inducers

Analgesics: morphine, methadone Anthelmintics: ivermectin, abamectin Antibiotics: erythromycin, tetracyclines, fluoroquinolines Anticancer drugs: vinblastine, vincristine, paclitaxel, doxorubicin, daunorubicin, mitoxantrone, etoposide, methotrexate, topotecan Antidiarrheal agents: loperamide Antiemetics: domperidone, ondansetron Antiepileptic drugs: phenytoin, carbamazepine, lamotrigine, phenobarbital, felbamate, gabapentin, topiramate Anti-gout agents: colchicine Calcium channel blockers: verapamil Cardiac glycosides: digoxin Corticoids: dexamethasone, hydrocortisone, corticosterone, cortisol, aldosterone, triamcinolone Diagnostic dyes: rhodamine 123, Hoechst 33342 HIV protease inhibitors: saquinavir, ritonavir, indinavir Immunosuppressive agents: cyclosporine, sirolimus, tacrolimus Psychotropic drugs: chlorpromazine, clozapine, desipramine, domperidone, flupentixol, imipramine, nortryptiline, sertaline, amitryptiline, doxepin, venlafaxine, paroxetine

1st Generation verapamil, cyclosporine, quinidine, quinine, amiodarone, cremophore EL 2nd Generation PSC-833 (valspodar), GF120918 (elacridar), VX-710 (biricodar), dexverapamil 3rd Generation OC 144-093 (ONT-093), LY335979 (zosuquidar), XR9576 (tariquidar), R101933 (laniquidar)

verapamil midazolam rapamycin reserpine rifampicin phenobarbitol St John’s Wort clotrimazole

The presence of Pgp in normal tissues protects the cells against harmful compounds but, on the other hand, may prevent pharmacotherapeutic agents from entering systemic circulation (Pgp in the small intestine) or their site of action (Pgp in the brain). Therefore, Pgp inhibitors have been searched for to improve pharmacotherapy, for example, to overcome multidrug resistance in anticancer treatment (10), and to enable drug penetration behind the blood–tissue barriers such as the intestine (to increase bioavailabilty of orally administered drugs), to the brain (to increase availability of CNS acting compounds) or fetus (for in utero fetal therapy) (11). Apart from low-molecular

Expression and Function of P-Glycoprotein in Normal Tissues

203

inhibitors of three generations, other tools for Pgp inhibition have been exploited including monoclonal antibodies (MRK16) or RNA interference (12). Several compounds have been found to upregulate the expression and increase the amount and transport activity of Pgp in tissues. Substrates, inhibitors, and inducers of Pgp are listed in Table 10.1. In clinical practice, Pgp-mediated drug–drug interactions are likely to occur if Pgp substrates/inhibitors/inducers are administered to a patient at the same time as described, for example, for digoxin-quinidine, digoxin/talinolol-rifampin, or digoxin-St John’s Wort (7). In their comprehensive review, Lin et  al. summarize in detail drug–drug interactions mediated by inhibition and induction of Pgp and their effect on drug absorption, distribution, and elimination (13).

3. ABCB1 Gene Structure, Regulation, and Polymorphisms

Human Pgp is coded by ABCB1 (MDR1) gene located on chromosome 7q21.1. The gene consists of 29 exons, numbered −1 to 28, spanning more than 200 kb of genomic DNA; however, only 27 exons code for Pgp protein. ABCB1 gene possesses two distinct promoter regions, including an upstream promoter at the beginning of exon −1 and the more preferred downstream promoter located within exon 1. The ATG translation initiation codon is located within exon 2. Unlike its murine analogue, the ABCB1 promoter lacks a consensus TATA box (core DNA sequence 5¢-TATAAA-3¢) within the proximal promoter region. However, like many other TATA-less promoters, basal transcription is directed by an initiator (Inr) sequence that encompasses the transcription start site (+1). Sequences between −6 and +11 are sufficient for proper initiation of transcription and are likely acting as the binding sites of the RNA Pol II transcriptional complex (Fig. 10.2a) (14, 15). Controlling the expression of ABCB1 in both tumor and normal tissues is a multifactorial event that has not been fully elucidated to date (16). Studies of the ABCB1 promoter and regulation of ABCB1 transcription have shown that transcription is under control of the complex regulatory network. In addition to transcriptional regulation by transcription factors such as those of the Sp family, NF-Y, AP-1, the tumor suppressor protein p53 and nuclear receptors (Pregnane X receptor, Vitamin D receptor, and Constitutive androstane receptor), ABCB1 transcription is also influenced by epigenetic modification such as DNA methylation and histone acetylation and via different signal transduction and

204

Staud et al.

Fig.  10.2. Structure, genomic organization, transcriptional regulation, and major single nucleotide polymorphisms of human ABCB1 (MDR1) gene. (a) Genomic, mRNA and protein structure. (b) Transcriptional regulation of the MDR1 gene.

mitogen-activate cascades such as p38, JNK, ERK, PKC and NF-kB. In addition, many pathophysiological and stress conditions such as inflammation and hypoxia affect transactivation and gene expression of ABCB1 illustrating the complexity of regulatory mechanisms (16–18) (Fig. 10.2b). Genetic polymorphisms of drug transporters and metabolizing enzymes, mainly CYP450, represent a source of interindividual variability in drug pharmacokinetics, drug effectiveness and/or toxicity. Genetic polymorphisms of ABCG1 gene have been reported both in animals (19) and in human (20). Over 50 single nucleotide polymorphisms (SNPs) have been identified so far (21) including coding and noncoding regions of the gene, and it can be predicted that many more will be found in future. Majority of coding region SNPs are nonsynonymous. However, allele frequency for most of the coding region SNPs is low (<8%) in different ethnic populations, with the exception of 3 SNPs in exon 12 (C1236T), exon 21 (G2677T/A), and exon 26 (C3435T) (22). SNPs C1236T and C3435T are synonymous, whereas G2677T/A causes an amino acid substitution of Ala by Ser/Thr. These 3 SNPs are in linkage disequilibrium with an allele frequency of 45–55% in whites and 5% to 10% in African Americans. Haplotype analysis of the ABCB1 SNPs revealed two major haplotypes, ABCB1*1 and ABCB1*13 (http://www. pharmgkb.org/search/annotatedGene/abcb1). Despite inconsistency in reported data, most studies suggest that C3435T represents a main functional polymorphism, accounting for

Expression and Function of P-Glycoprotein in Normal Tissues

205

1.5- to 2-fold changes in mRNA levels and/or protein expression in some tissues. Previous haplotype analysis has not revealed any functional polymorphisms in the promoter region that are in linkage with C3435T (22, 23). Thus, how can a synonymous SNP C3435T affect ABCB1 mRNA expression? Recent research has indicated that both C1236T and C3435T variants appear to change mRNA structure, while G2677T did not. However, only synonymous SNP C3435T affects mRNA stability resulting in decreased expression of P-glycoprotein transporter (22).

4. Role of P-glycoprotein in Drug Absorption

4.1. P-glycoprotein in the Small Intestine

Absorption is a pharmacokinetic event by which drugs enter systemic (blood) circulation after extravascular administration. Absorption is a primary focus in pharmaceutical development as the drug must be absorbed before any medicinal effects can take place. Drugs may be administered by a number of ways and their absorption will depend on many factors; however, the details are beyond the scope of this chapter. In the following text, we will focus only on intestinal drug absorption, which is the most common in clinical practice and is under the influence of Pgp. Small intestine is the principal site for absorption of ingested compounds – nutritional, therapeutic, or toxic. Intestinal enterocytes form a selective barrier to xenobiotics and drugs consisting of specific membrane transporters and biotransformation enzymes (24). Drug absorption from the gastrointestinal tract (GIT) is affected by many factors such as physical–chemical properties of the molecule (lipophilicity and solubility, pKa, molecular weight) and anatomical/physiological properties of the absorption site (gastric emptying time, gastrointestinal motility, intestinal pH, area size, and blood circulation). Most drugs administered per os enter the blood circulation by passive lipid diffusion; because of its large size reaching up to 200  m2, small intestine is the primary site of drug absorption. Recently, several influx and efflux transporters have been identified along the length of gastrointestinal tract, that may increase or decrease drug absorption, respectively. Since Pgp substrates are generally hydrophobic compounds, it is obvious that Pgp in the GIT will affect absorption of many drugs. Pgp expression has been described in human intestine as well as in animal and cellular models (24). Being localized to the top villi of enterocytes, Pgp is conveniently located to recognize its

206

Staud et al.

substrates and pump them back to the intestinal lumen (Fig. 10.1). However, Pgp levels in the small intestine are not uniform; inconsistent expression of the transporter was described among epithelial villi cells with higher expression in the columnar cells than in the crypt (6). Similarly, Pgp expression along the intestine is not uniform but increases from the stomach to the colon (25). Functional expression of Pgp in the intestine was first demonstrated in the cellular model of Caco-2 cells (cancer cells of human colon widely used experimentally to predict intestinal drug absorption) with various substrates (26–29) as well as in Pgp knock-out mice (30). Oral absorption of paclitaxel was found to be several times higher in Pgp-deficient mice compared with wild type animals confirming that P-pg hinders absorption of drugs by pumping them back to intestinal lumen. Similar effect in human has also been suggested based on several studies using Pgp substrates and inhibitors such as cyclosporine (31) and docetaxel (32). In addition, Pgp has been shown to function as an excretory protein in the intestine; after i.v. injection of digoxin to Pgp expressing mice, considerable portion of the dose was secreted into the intestinal lumen within 90 min post injection. When Pgp was blocked by an inhibitor or Pgp knock-out mice were used, this secretory effect was not observed (33). This effect was also confirmed in human (34). However, not all Pgp substrates are significantly affected in their passage across the wall of the small intestine. Since Pgp is a saturable transporter, it is obvious that it can play a major role preferentially for substrate-drugs given in low doses, not exceeding their Km value, for example, digoxin (35, 36). On the other hand, cyclosporine and paclitaxel are substantially limited by Pgp in their intestinal absorption even when given in relatively high doses. Since ABCB1 is a highly polymorphic gene with several SNPs identified to date, interindividual variability exists in the expression of intestinal Pgp. The study of Lown and colleagues (37) described eightfold differences in a population of 25 patients using immunoblotting; in their study, Pgp accounted for 30% of the variability in cyclosporine pharmacokinetics. Furthermore, intraindividual variability in intestinal Pgp expression has also been described in the literature (38). Variability in Pgp expression in the intestine may, therefore, contribute to variable drug absorption in population as proposed by Hoffmeyer et al. (20). An interesting aspect of Pgp expression in the small intestine is its interaction with metabolizing enzymes, specifically the 3A4 isozyme of the cytochrome P450. As stated above, considerable overlap exists in substrate specificity and tissue distribution between Pgp and CYP3A4. Enterocytes express both Pgp and CYP3A4, leading to a drug efflux-metabolism grouping a functional tandem to defend organism against harmful compounds. According to this hypothesis, first proposed by Benet

Expression and Function of P-Glycoprotein in Normal Tissues

207

et al. (39), interaction of a drug with Pgp in the small intestine leads to repeated cycles of absorption-excretion which increases the access of drug to metabolism by CYP3A4 (9). By this means, cyclosporine, an immunosuppressive agent, is metabolized from 51% in the enterocytes of the gut wall, and only 8% is biotransformed in the hepatocytes (40). In addition, since there is also an overlap in inducers of Pgp and CYP3A4, administration of a common inducer such as dexamethasone will upregulate the expression of both Pgp and CYP3A4 in the intestine, resulting in profound decrease in the bioavailability of common substrate as observed in the case of indinavir (41). Functional expression of Pgp in the small intestine is desirable to protect organism from absorption of detrimental substances. On the other hand, however, it limits intestinal absorption of orally administered compounds, making oral dosing of certain drugs complicated and unpredictable as well as expensive. Search for Pgp inhibitors that would block intestinal Pgp in  vivo was started and indeed, directed inhibition of Pgp in the small intestine was found to improve the bioavailability of orally administered Pgp substrates. Dramatic increase in intestinal absorption of paclitaxel was observed when Pgp inhibitors, PSC833, GF120918, or cyclosporine, were coadministered (5, 11).

5. Role of P-glycoprotein in Drug Distribution

5.1. P-glycoprotein in the Brain

Distribution is a phase of pharmacokinetics by which a drug escapes blood circulation and enters body tissues. The rate of distribution generally depends on the physical-chemical properties of the molecule and on characteristics of the distribution tissues. Sensitive cells and organs are separated from blood circulation by “physiological barriers” that limit drug transport from blood to the tissue/organ. These barriers usually consist of a (i) mechanical component, formed by tissue layers and their cell–cell junctions and (ii) active component, formed by efflux drug transporter(s) and/or metabolizing enzymes. The most important and best characterized are the blood–brain barrier, placental barrier, and blood–testis barrier, all of which are known to express Pgp. The brain is a crucial organ, extremely sensitive to the toxic effects of exogenous compounds. The main interfaces between the CNS and the blood circulation are the blood–brain barrier (BBB) and the blood–cerebrospinal fluid barrier (BCSFB). Due to the much larger surface area of the BBB in comparison to that of BCSFB, the BBB is considered the primary interface for solute exchange

208

Staud et al.

between the CNS and blood. Therefore, only BBB is covered in the following text in detail. The BBB is a mechanical as well as active barrier between the systemic circulation and brain cells that protects the tissue from potentially harmful substances and contributes to brain homeostasis. Passive paracellular transport of hydrophilic molecules is hindered by the presence of tight junctions (zonullae occludentes) between endothelial cells. This barrier is further fortified by astrocyte foot processes and the presence of active efflux transporters (42) as shown in Fig. 10.1. Drug transport from blood to CNS is governed by similar conditions as in other body barriers; that is, physical–chemical properties (lipophilicity, pKa, molecular weight) and anatomical/ physiological properties of the tissue (local pH and blood circulation). Accordingly, only lipid soluble molecules can cross the barrier between blood and brain by passive diffusion and indeed, strong correlation between octanol/water partition coefficient and brain permeability was described (43, 44). However, several drugs that fulfill the requirements for rapid passage across membranes by lipid diffusion have been found to cross the BBB rather slowly. This is the case of, for example, cyclosporine, vincristine or etoposide, all good Pgp substrates. Pgp was first identified in the brain tissue in 1989 by CordonCardo et  al. (45) and Thiebaut et  al. (46) using monoclonal antibodies. In the brain, Pgp is localized in the luminal, bloodfacing, membrane of the endothelial cells (47); other researchers reported its expression also in astrocytes (48, 49) and microglia (50). While Pgp expression in the endothelial cells limits transfer of drugs from blood to brain tissue, presence of Pgp in other brain cells might affect drug disposition within the organ (36). Functional expression of Pgp in the brain was first described in kinetic studies in  vitro investigating transport of vincristine across bovine or mouse brain endothelial cells (51, 52). Ohnishi et  al. (53) have also demonstrated Pgp mediated transport of doxorubicin across BBB in  vivo in rat and in  vitro in primary cultured brain capillary endothelial cells. The most convincing description of Pgp activity in the brain was reported in in  vivo studies investigating pharmacokinetics of various substrates in knock-out mice (54). Recently, intraindividual variability of Pgp expression in the BBB was suggested by Bartels et al. (55) who reported decline in Pgp function in specific brain regions with aging. They hypothesize that loss of Pgp activity in the brain may have impact on progressive development of neurodegenerative disease. Expression of Pgp in the brain is an evolutionary step to protect the sensitive tissue against potential toxins. On the other hand, however, Pgp limits drug transport to brain for CNS acting drugs and, therefore, reduces their pharmacotherapeutic effectiveness.

Expression and Function of P-Glycoprotein in Normal Tissues

209

Several reports have been published over the last few years indicating that overexpression of drug efflux transporters in the brain has been linked with CNS diseases such as pharmaco-resistant epilepsia and schizophrenia. Tishler et al. (56) proposed that Pgp may play a clinically significant role by limiting access of antiepileptic drugs to the brain; therefore, elevated levels of Pgp in the brain may contribute to the refractoriness of seizures in patients with treatment-resistant epilepsy. Subsequently, coadministration of CNS acting drugs with a potent Pgp inhibitor might prove a viable method to improve therapy of patients with drug-resistant forms of epilepsia (57). Also, the brain is considered a shelter for HIV virus; antiretroviral drugs are, therefore, desired to cross the BBB which is a problem for HIV protease inhibitors that are Pgp substrates (e.g. saquinavir). Pgp-mediated drug interactions in the brain have been described in several experimental models as well as in human. For example, Sadeque et al. (58) reported respiratory depression of loperamide when coadministered with quinidine; when given alone, loperamide showed no adverse effects. Enhanced effects of methadone were observed when administered together with quinidine (59); enhanced neurotoxicity of colchicine coadministered with verapamil was described in a case report (60). 5.2. P-glycoprotein in the Placenta

Placenta is the organ that brings into close apposition blood circulations of mother and fetus, while maintaining separation of the two blood systems. One of the placenta’s major roles is to regulate the exchange of nutrients and gases between mother and fetus and to remove fetal waste products. The rate-limiting barrier for transplacental transfer of most substances is formed by polarized layer of syncytiotrophoblast (61, 62), which covers each placental villous tree and consists of microvillous brush-border (maternal blood-facing) membrane and basal (fetal circulation-facing) membrane (Fig. 10.3). Basically, any chemical substance administered to mother is able to permeate across the placenta (63, 64). Disclosure of Pgp and other ABC drug efflux transporters in the placental barrier brought new sights into the problematic of transplacental pharmacokinetics (12). Expression of Pgp has been found in the placental tissue of various mammalian species including humans. Intensive immunoreactivity for Pgp was initially reported in the trophoblast layer, whereas the endothelial cells of placental fetal capillaries were proved to be Pgp negative (45). Expression of placental Pgp during pregnancy is not uniform; Pgp levels in mice and rat seem to reach its maximum during the second half of pregnancy and decrease toward the term (65, 66). In human, on the other hand, decrease in placental Pgp expression with advancing gestation was described (67, 68). In a recent report of Nanovskaya et al. (69), the materno-fetal transfer of methadone appears to be lower in

210

Staud et al.

Fig. 10.3. Schematic structure of human placenta. (a) Cross section of uterus at the term of pregnancy showing fetus connected with the placenta via the umbilical cord. In the detailed schema, structure of cotyledon, placental functional unit, is depicted. Chorion, the fetal part of the placenta, consists of chorionic plate and chorionic villi that are washed by maternal blood entering intervillous space through spiral arteries in decidua basalis. The oxygen and nutrients from maternal blood cross the surface trophoblast layer of chorionic villi, enter the fetal blood, and are brought to the fetus via umbilical vein. The deoxygenated blood is conducted from the fetus through two umbilical arteries. (b) Hematoxylin-eosin stained paraffin sections of terminal villi in human third trimester placenta (microphotograph courtesy of Dr. Nachtigal). (c) Schematic description of the terminal villi section showing localization of Pgp at the apical microvillous membrane of syncytiotrophoblast and presence of other placental drug efflux transporters. Reproduced with permission from [12].

preterm compared with term placentas, corresponding with the lower levels of Pgp at the end of pregnancy. Placental Pgp, therefore, appears to be upregulated in early phase of human pregnancy, which could help protect the fetus from potentially toxic xenobiotics at a time when it is most vulnerable to such toxicity.

Expression and Function of P-Glycoprotein in Normal Tissues

211

The functional activity of Pgp transporter in the placenta was initially confirmed by the uptake of vincristine in the membrane vesicles prepared from human trophoblast (70). Subsequent studies employing alternative in  vitro models demonstrated the functional expression of Pgp in primary cultures of human cytotrophoblasts and in human placental BeWo (clone c30) cell line (71) and revealed an exclusive expression of Pgp on the brush-border (maternal side-facing) membrane of trophoblast directing the basolateral-to-apical efflux of Pgp substrates into the maternal compartment (72). The importance of placental Pgp in protecting the fetus was confirmed by many in vivo experiments using genetically modified mice (73–75) or using in situ perfusion of the rat placenta (76, 77). Using the model of isolated perfused human placenta, no statistically significant effects of Pgp inhibitors quinidine or verapamil on the materno-fetal transfer of digoxin were observed in the initial work of Holcberg et al. (78). However, subsequent studies with indinavir and vinblastine have brought clear evidence for differential transfer of Pgp substrates in the intact human placenta (79, 80). Subsequently, other studies confirmed limited transfer of Pgp substrates from mother to fetus, including protease inhibitor saquinavir (81, 82), methadone (83, 84) and another opioid agonist l-alpha-acetylmethadol (85). These data suggest that the activity of placental Pgp can affect transplacental pharmacokinetics of a variety of drugs, which can have important consequences for pharmacotherapy during pregnancy. Substantial variability in Pgp expression has recently been observed among placentas in rat (66) as well as in humans (68, 86). Correlation between nine single nucleotide polymorphisms (SNPs) and Pgp expression was examined in placentas obtained from 100 Japanese women (87). An association between synonymous SNP in exon 26 (C3435T) and nonsynonymous SNP in exon 21 (G2677A,T, Ala893Ser) was observed. Individuals having the G2677(A,T) and/or T-129C mutant allele in the promoter region further showed significantly lower expression of placental Pgp (87). Interestingly, when different ABCB1 genotype/haplotype of mother and her fetus were examined in 73 Caucasians, Pgp expression appeared to be significantly lower if both mother and infant were homozygous for the 3435T (TT/tt) compared with maternal and fetal homozygotes for the C allele (88). Moreover, placentas from mothers carrying polymorphisms for both 3435T and 2677T also possessed significantly lower Pgp expression compared with placentas of wild type individuals (CC/GG) (88). The recent studies, however, have demonstrated that although the polymorphisms C3435T and G2677A/T altered the expression levels of Pgp in the human placenta, they did not have any consequences in Pgp-mediated transfer of saquinavir in dually

212

Staud et al.

perfused human term placenta (81, 82). Therefore, it still remains unclear to which extent ABCB1 polymorphisms influence expression and/or function of placental Pgp. In pharmacotherapy of pregnant women, usage of drugs that are substrates of Pgp will be favored to reduce fetal drug exposure, toxicity, and risk of malformations. On the other hand, drugs that are not recognized as Pgp substrates could be preferred in cases where fetus is the target of pharmacotherapy. For example, ivermectine was found to be limited in its placental transfer and therefore suggested to be used in the treatment of onchocerciasis in pregnant women living in areas with high risk of this infectious disease (89). Similarly, favoring antracyclines and taxanes in adjuvant treatment of pregnant women has been suggested (90). Placental Pgp presents a risk of drug–drug interactions; for example, digoxin is a drug commonly used to treat fetal tachycardia via maternal drug administration (91, 92). Interaction with other cardiac drugs, such as verapamil, could result in increased fetal plasma levels and toxic effects. On the other hand, inhibition of placental Pgp may be beneficial in the treatment with HIV protease inhibitors to reduce the rate of mother-to-child HIV transmission in HIV positive pregnant women (75, 93, 94). Similarly, for the treatment of fetal tachycardia, it has been suggested that pharmacological inhibition of Pgp would be beneficial to enhance digoxin availability to the fetus, while minimizing drug exposure of the mother (95). 5.3. P-glycoprotein in the Blood–Testis Barrier

Germ cells in the testis are separated from blood stream by the blood–testis barrier (BTB), formed by tight connections between Sertoli cells (Fig.  10.1). In addition to Sertoli–cell barrier, the concept of BTB also encompasses the endothelial lining of the blood and lymphatic vessels (96). Similarly to other blood– tissue barriers, the role of BTB is to limit the passage of toxic agents into the sensitive tissue (germ cells in the seminiferous tubules). The transfer of drugs across the BTB follows the principles common to other membrane-composing barriers: lipid-soluble compounds permeate readily into the tubules, while large hydrophilic molecules or substances that bind to plasma proteins usually penetrate only slowly (97). High levels of Pgp mRNA were initially demonstrated in Chinese hamster testes (98); immunoreactivity for Pgp was further found in testes of rat (46) and in endothelial cells of the human testicular capillaries (45, 99, 100). In addition to endothelial expression, MDR1 mRNA was also detected in male germ cells (99). This expression appears to be present in somatic and haploid cells, but not in the mitotic and meiotic germ cells (101). MDR1 genes mRNA and Pgp were also found in resident macrophages

Expression and Function of P-Glycoprotein in Normal Tissues

213

and Leydig cells of the testicular interstitial compartment in rat, mouse, and guinea pig as well as in human (101). The protective role of testicular Pgp has been initially demonstrated in Pgp-deficient mice in which considerably greater levels of its substrate, ivermectine, were detected in the testes compared with control animals (102). Subsequent study revealed that pharmacological inhibition of Pgp increased the penetration of doxorubicin to rat testis (103). Protection of testes from exposure to HIV protease inhibitors has been suggested in a study employing mice model and a potent Pgp inhibitor LY-335979 (104). Intravenous administration of the inhibitor resulted in increased concentration of nelfinavir in testes (up to fourfold). In a clinically more realistic administration schedule involving repeated oral administration of saquinavir, 3.9-fold increase in testicular concentration of the Pgp substrate was observed due to Pgp inhibition by GF120918 (75). In the recent work of Choo et al. (105), the BTB was, however, suggested to be more resistant to pharmacological inhibition than other tissue sites such as the lymphocytes. Nevertheless, from the above studies, it seems evident that Pgp represents one of the essential molecules conferring functionality to the BTB. Expression and function of Pgp in the BTB seem to have mainly toxicological consequences. The absence of MDR1 expression in mitotic and meiotic germ cells probably explains their particular vulnerability to various anticancer drugs. In contrast, expression of Pgp in the haploid cells most likely reflects the ability of spermatozoa to assume their own antidrug defense (101). High rate of gonad damage leading to long-term or permanent oligo- or azoospermia is one of the serious side effects of anticancer treatment (106, 107). There is also evidence for growing incidence of deleterious effect of various xenobiotics on the testicular function (108, 109). From this point of view, suitable pharmacological or genetic manipulation of Pgp expression and/or function in testes could help decrease the toxicity of xenobiotics or therapeutic agents to the vulnerable male germ cells.

6. Role of P-glycoprotein in Drug Excretion

Drug excretion is the final process of pharmacokinetics that irreversibly removes drug and/or its metabolite(s) from the organism. The main organs of excretion are liver and kidneys clearing the drugs from blood and transferring them into bile and urine, respectively. The excretion process in these organs is a complex interplay of both passive and active events.

214

Staud et al.

6.1. P-glycoprotein in the Liver

Regarding excretion into the bile, the drug must first cross the sinusoidal (basolateral) membrane of the hepatocytes by passive diffusion and/or active transport. Accordingly, basolateral membrane of hepatocytes contains spectrum of uptake transport proteins that transport endo- and xenobiotics intracellularly into the cells. Compounds, thereafter, diffuse within the intracellular space where they may be metabolized, based on their physicalchemical properties, by biotransformation enzymes of phase I (e.g. cytochrome P450) and/or phase II (e.g. conjugation). Biliary excretion of drugs/metabolites takes place at the canalicular (apical) membrane of hepatocytes and is accomplished against concentration gradient, that is, requires energy from ATP hydrolysis (110). In addition to excretory function of hepatocytes, the intact hepato-biliary barrier is necessary for biliary elimination of compounds. This barrier is formed by tight junction interconnections between hepatocytes and creates cell plates strictly sealing the bile canaliculi, therefore, ensuring bile flow without its backward leakage into the blood (111) (see Fig.  10.1 for schematic depiction). Therefore, the whole cascade of these mechanisms has to be taken into consideration when biliary excretion of drugs is evaluated. The role of Pgp in the biliary excretion of drugs was first suggested by combining the knowledge of its capacity to export drugs with the results of tissue distribution studies showing that Pgp is highly expressed at the canalicular membrane of hepatocytes (6). As the exporting transporter capable of drug efflux against concentration gradient, the predicted function for Pgp was extrusion of its substrates from the hepatocyte into the bile. Consequently, with the contribution of intestinal Pgp, substrates ultimately leave the body via the feces, and all this results in plasma clearing and detoxifying function. Indeed, since the first reports of potential involvement of Pgp in the biliary excretion of drugs were published (112, 113), numerous studies have confirmed the importance of hepatic Pgp in pharmacokinetics of its substrates. Generation of mice with the targeted disruption of mdr1a and mdr1a/1b (114) has greatly enhanced the understanding of the in vivo relevance of Pgp to hepatic elimination of drugs. Absence of Pgp in knockout mice results in several fold reduction of biliary clearance of Pgp substrates, such as digoxin or vecuronium (13). In the liver, Pgp contributes to the first-pass elimination of drugs after oral administration and/or forms essential pathway for drug biliary excretion after parenteral administration (5). Nevertheless, the hepatic expression and function of this transporter are highly interindividually and intraindividually variable, which can produce unpredictable changes in drug plasma concentrations and therapeutic effects. Variability in human liver Pgp expression has been recently demonstrated by Meier et  al. (115) who described unimodal distribution of Pgp in Caucasian

Expression and Function of P-Glycoprotein in Normal Tissues

215

population with 30-fold variability in liver content of Pgp. No correlation was found between hepatic Pgp and ABCB1 genotypes. Because the relationship has already been observed between some variant alleles and intestinal Pgp expression (116, 117), other factors such as liver diseases and concomitant drug treatment are suggested to mitigate the genetic effect on hepatic Pgp expression. Many drug–drug interactions related to the inhibition or induction of hepatic Pgp have been reported to date (13). Of broad clinical relevance is the inhibitory interaction between digoxin and other drugs such as verapamil, amiodarone, quinidine (118), clarithromycin (119) or atorvastatin (120) which have been shown to increase the area under the plasma concentrationtime curve of digoxin. Conversely, induction of MDR1 gene associated with rifampin therapy has been shown to lower plasma levels of digoxin (121). List of Pgp inducers and inhibitors is shown in Table 10.1. In addition to drugs, several liver diseases such as cholestasis, primary biliary cirrhosis, hepatocellular carcinomas, chronic hepatitis C virus infection, and submassive cell necrosis are associated with changes in the Pgp expression (122– 124). Subsequent impairment of Pgp substrate pharmacokinetics could be exemplified by significant reduction of doxorubicin biliary clearance during endotoxin-induced down-regulation of hepatic Pgp protein expression (125). As in the intestine, strong overlap in substrate specificity with CYP3A4 highly contributes to Pgp function in the liver. While high expression of CYP3A4 in hepatocytes provides capacity for metabolic inactivation of substrates, the consequent excretion by Pgp further reduces the disposition of drug in organism. Coordinated function of both proteins is also given by mutual regulation of their expression by pregnane X receptor (PXR) and constitutive androstane receptor (CAR) (126), ligand-activated nuclear receptors (see Fig. 10.2). 6.2. P-glycoprotein in the Kidney

Urinary excretion of drugs is a complex interplay of three mechanisms: glomerular filtration, tubular secretion, and tubular reabsorption. The basic prerequisite for excretion into urine is water solubility of drug and/or its metabolites as no transporting carrier is available in this body fluid. Accordingly, the primary mechanism for drug excretion is glomerular filtration, a passive process influenced by molecule size, concentration of unbound fraction of drug in plasma and renal blood perfusion. Tubular secretion and partly also reabsorption are active mechanisms requiring transporters; the function of renal proximal tubular epithelial cells is the most significant factor for active drug renal excretion. Herein, similar to hepatocytes, the drug is taken up at the basolateral membrane by a spectrum of transporters, and limited biotransformation may occur during intracellular diffusion (127, 128).

216

Staud et al.

The transport across luminal brush-border (apical) membrane is realized by several active transporters (128). In the kidney, Pgp is expressed at the apical (luminal) brushborder membrane of proximal tubular cells (6) (Fig. 10.1) with the expected function to pump substrates from intracellular space into urine. This was confirmed by in vitro cellular studies using specific tubular confluent cell lines as well as using in situ isolated perfused rat kidney or in vivo pharmacokinetic studies. As demonstrated with model Pgp substrates, digoxin, and rhodamine-123, administration of either Pgp inhibitor (e.g. verapamil or quinidine) or pretreatment with inducer may produce several fold changes in tubular secretion of Pgp substrates (129, 130). Interestingly, experiments using Pgp knockout mice have given controversial results. While some studies using mdr1a (−/−) or mdr1a/1b (−/−) animals showed expected reduction in renal clearance of Pgp substrates (131, 132), other reports demonstrated increased renal excretion of the substrate drugs (133). The reason for this contradiction is not clear, but upregulation of other transporter system in knockout animals is suggested to be the cause (13). In humans, several well controlled studies revealed contribution of renal Pgp to substrate drug excretion. Most data originate from inhibitory studies with digoxin where reduced renal clearance of the drug was reported after coadministration with, for example, clarithromycin or itraconazol (134, 135).

7. Concluding Remarks Pgp together with other ABC drug efflux transporters was initially thought to have an importance only in the phenomenon of multidrug resistance in cancer treatment. Although their physiological role is still not completely elucidated, extensive research of the last couple of decades has clearly documented the expression and function of these transporters in many nontumor tissues. From the information summarized in this chapter, it is evident that the presence of Pgp in normal tissues has a huge impact on the fate of drugs/toxins inside the organism on several levels, including absorption from the small intestine, body distribution, and elimination. Since Pgp is an important factor determining plasma drug concentrations, it is obvious that any alteration in Pgp expression and function (due to inhibition, induction or genetic polymorphisms) may lead to changed systemic exposure. It is generally accepted that coadministration of drugs that interact with this transporter (as a substrate, inhibitor, or inducer) can result in

Expression and Function of P-Glycoprotein in Normal Tissues

217

drug–drug interactions that affect the pharmacokinetics and pharmacodynamics of the coadministered drugs. Inhibition or induction of this transporter may affect oral bioavailability, biliary and renal clearances, brain uptake, or transplacental penetration of drugs. As a result, variations in drug response, adverse drug reactions or loss of efficacy may be observed. Drug–drug interactions related to transporters are being documented with increasing frequency and are important to consider both by physicians in clinical practice and by researchers in the process of developing a new molecular entity with therapeutic potential. It is essential to evaluate possible drug interactions prior to market approval as well as during the postmarketing period. The FDA has published several documents that provide guidance to industry and agency reviewers regarding the use of various methodologies to address metabolic and transporter drug–drug interaction issues. In 2004, FDA published a concept paper  to facilitate the discussion of study design, data analysis, and implication of drug interactions for dosing and labeling. A new draft guidance including additional discussions on emerging areas, such as drug transporters, was published in September 2006. Also, decision trees to determine whether an investigational drug is a substrate or inhibitor for Pgp and whether an in vivo drug interaction study with a Pgp inhibitor is needed are available on the FDA website (http://www.fda.gov/cder/guidance/6695dft. htm). Finally, thorough knowledge of physiological and pharmacological role of Pgp and other drug efflux transporters of the ABC family should lead to many research as well as clinical benefits such as designing safer and more effective molecules, predicting possible drug interactions and kinetic profiles, and assisting in patient-tailored dosing. References 1. Juliano RL, Ling V (1976) A surface glycoprotein modulating drug permeability in Chinese hamster ovary cell mutants. Biochim Biophys Acta 455:152–162 2. Bosch I, Croop J (1996) P-glycoprotein multidrug resistance and cancer. Biochim Biophys Acta 1288:F37–F54 3. Goldstein LJ, Gottesman MM, Pastan I (1991) Expression of the MDR1 gene in human cancers. Cancer Treat Res 57:101–119 4. Kartner N, Riordan JR, Ling V (1983) Cell surface P-glycoprotein associated with multidrug resistance in mammalian cell lines. Science 221:1285–1288 5. Schinkel AH, Jonker JW (2003) Mammalian drug efflux transporters of the ATP binding

cassette (ABC) family: an overview. Adv Drug Deliv Rev 55:3–29 6. Thiebaut F, Tsuruo T, Hamada H et al (1987) Cellular localization of the multidrug-resistance gene product P-glycoprotein in normal human tissues. Proc Natl Acad Sci USA 84:7735–7738 7. Fromm MF (2004) Importance of P-glycoprotein at blood–tissue barriers. Trends Pharmacol Sci 25:423–429 8. Seelig A, Landwojtowicz E (2000) Structureactivity relationship of P-glycoprotein substrates and modifiers. Eur J Pharm Sci 12:31–40 9. Benet LZ, Cummins CL (2001) The drug efflux-metabolism alliance: biochemical aspects. Adv Drug Deliv Rev 50(Suppl 1):S3–S11

218

Staud et al.

10. Krishna R, Mayer LD (2000) Multidrug resistance (MDR) in cancer. Mechanisms, reversal using modulators of MDR and the role of MDR modulators in influencing the pharmacokinetics of anticancer drugs. Eur J Pharm Sci 11:265–283 11. Breedveld P, Beijnen JH, Schellens JH (2006) Use of P-glycoprotein and BCRP inhibitors to improve oral bioavailability and CNS penetration of anticancer drugs. Trends Pharmacol Sci 27:17–24 12. Ceckova-Novotna M, Pavek P, Staud F (2006) P-glycoprotein in the placenta: expression, localization, regulation and function. Reprod Toxicol 22:400–410 13. Lin JH (2003) Drug–drug interaction mediated by inhibition and induction of P-glycoprotein. Adv Drug Deliv Rev 55:53–81 14. Chen CJ, Clark D, Ueda K et  al (1990) Genomic organization of the human multidrug resistance (MDR1) gene and origin of P-glycoproteins. J Biol Chem 265:506–514 15. Choudhuri S, Klaassen CD (2006) Structure, function, expression, genomic organization, and single nucleotide polymorphisms of human ABCB1 (MDR1), ABCC (MRP), and ABCG2 (BCRP) efflux transporters. Int J Toxicol 25: 231–259 16. Callaghan R, Crowley E, Potter S, Kerr ID (2008) P-glycoprotein: so many ways to turn it on. J Clin Pharmacol 48:365–378 17. Labialle S, Gayet L, Marthinet E, Rigal D, Baggetto LG (2002) Transcriptional regulators of the human multidrug resistance 1 gene: recent views. Biochem Pharmacol 64:943–948 18. Scotto KW (2003) Transcriptional regulation of ABC drug transporters. Oncogene 22: 7496–7511 19. Umbenhauer DR, Lankas GR, Pippert TR et al (1997) Identification of a P-glycoproteindeficient subpopulation in the CF-1 mouse strain using a restriction fragment length polymorphism. Toxicol Appl Pharmacol 146:88–94 20. Hoffmeyer S, Burk O, von Richter O et  al (2000) Functional polymorphisms of the human multidrug-resistance gene: multiple sequence variations and correlation of one allele with P-glycoprotein expression and activity in  vivo. Proc Natl Acad Sci USA 97:3473–3478 21. Kimchi-Sarfaty C, Oh JM, Kim IW et al (2007) A “silent” polymorphism in the MDR1 gene changes substrate specificity. Science 315: 525–528 22. Wang D, Sadee W (2006) Searching for polymorphisms that affect gene expression and mRNA processing: example ABCB1 (MDR1). AAPS J 8:E515–E520

23. Streetman DS (2007) Clinical pharmacogenetics of the major adenosine triphosphate binding cassette and solute carrier drug transporters. J Pharm Pract 20:219–233 24. Takano M, Yumoto R, Murakami T (2006) Expression and function of efflux drug transporters in the intestine. Pharmacol Ther 109: 137–161 25. Fojo AT, Ueda K, Slamon DJ et  al (1987) Expression of a multidrug-resistance gene in human tumors and tissues. Proc Natl Acad Sci USA 84:265–269 26. Augustijns PF, Bradshaw TP, Gan LS, Hendren RW, Thakker DR (1993) Evidence for a polarized efflux system in CACO-2 cells capable of modulating cyclosporin A transport. Biochem Biophys Res Commun 197:360–365 27. Burton PS, Conradi RA, Hilgers AR, Ho NF (1993) Evidence for a polarized efflux system for peptides in the apical membrane of Caco-2 cells. Biochem Biophys Res Commun 190:760–766 28. Hunter J, Hirst BH, Simmons NL (1993) Drug absorption limited by P-glycoproteinmediated secretory drug transport in human intestinal epithelial Caco-2 cell layers. Pharm Res 10:743–749 29. Hunter J, Jepson MA, Tsuruo T, Simmons NL, Hirst BH (1993) Functional expression of P-glycoprotein in apical membranes of human intestinal Caco-2 cells. Kinetics of vinblastine secretion and interaction with modulators. J Biol Chem 268:14991–14997 30. Sparreboom A, van Asperen J, Mayer U et al (1997) Limited oral bioavailability and active epithelial excretion of paclitaxel (Taxol) caused by P-glycoprotein in the intestine. Proc Natl Acad Sci USA 94:2031–2035 31. Meerum Terwogt JM, Malingre MM, Beijnen JH et  al (1999) Coadministration of oral cyclosporin A enables oral therapy with paclitaxel. Clin Cancer Res 5:3379–3384 32. Malingre MM, Richel DJ, Beijnen JH et  al (2001) Coadministration of cyclosporine strongly enhances the oral bioavailability of docetaxel. J Clin Oncol 19:1160–1166 33. Mayer U, Wagenaar E, Beijnen JH et  al (1996) Substantial excretion of digoxin via the intestinal mucosa and prevention of longterm digoxin accumulation in the brain by the mdr 1a P-glycoprotein. Br J Pharmacol 119:1038–1044 34. Drescher S, Glaeser H, Murdter T et al (2003) P-glycoprotein-mediated intestinal and biliary digoxin transport in humans. Clin Pharmacol Ther 73:223–231 35. Chiou WL, Chung SM, Wu TC, Ma C (2001) A comprehensive account on the role of efflux

Expression and Function of P-Glycoprotein in Normal Tissues transporters in the gastrointestinal absorption of 13 commonly used substrate drugs in humans. Int J Clin Pharmacol Ther 39:93–101 36. Lin JH, Yamazaki M (2003) Role of P-glycoprotein in pharmacokinetics: clinical implications. Clin Pharmacokinet 42:59–98 37. Lown KS, Mayo RR, Leichtman AB et  al (1997) Role of intestinal P-glycoprotein (mdr1) in interpatient variation in the oral bioavailability of cyclosporine. Clin Pharmacol Ther 62:248–260 38. Masuda S, Uemoto S, Hashida T et al (2000) Effect of intestinal P-glycoprotein on daily tacrolimus trough level in a living-donor small bowel recipient. Clin Pharmacol Ther 68: 98–103 39. Benet LZ, Izumi T, Zhang Y, Silverman JA, Wacher VJ (1999) Intestinal MDR transport proteins and P-450 enzymes as barriers to oral drug delivery. J Control Release 62:25–31 40. Wacher VJ, Salphati L, Benet LZ (2001) Active secretion and enterocytic drug metabolism barriers to drug absorption. Adv Drug Deliv Rev 46:89–102 41. Lin JH, Chiba M, Chen IW et al (1999) Effect of dexamethasone on the intestinal first-pass metabolism of indinavir in rats: evidence of cytochrome P-450 3A (correction of P-450 A) and p-glycoprotein induction. Drug Metab Dispos 27:1187–1193 42. Loscher W, Potschka H (2005) Blood–brain barrier active efflux transporters: ATP-binding cassette gene family. NeuroRx 2:86–98 43. Levin VA (1980) Relationship of octanol/ water partition coefficient and molecular weight to rat brain capillary permeability. J Med Chem 23:682–684 44. Tsuji A, Tamai I (1997) Blood–brain barrier function of P-glycoprotein. Adv Drug Deliv Rev 25:287–298 45. Cordon-Cardo C, O’Brien JP, Casals D et  al (1989) Multidrug-resistance gene (P-glycoprotein) is expressed by endothelial cells at blood–brain barrier sites. Proc Natl Acad Sci USA 86:695–698 46. Thiebaut F, Tsuruo T, Hamada H et al (1989) Immunohistochemical localization in normal tissues of different epitopes in the multidrug transport protein P170: evidence for localization in brain capillaries and crossreactivity of one antibody with a muscle protein. J Histochem Cytochem 37:159–164 47. Beaulieu E, Demeule M, Ghitescu L, Beliveau R (1997) P-glycoprotein is strongly expressed in the luminal membranes of the endothelium of blood vessels in the brain. Biochem J 326 (Pt 2):539–544

219

48. Decleves X, Regina A, Laplanche JL et  al (2000) Functional expression of P-glycoprotein and multidrug resistance-associated protein (Mrp1) in primary cultures of rat astrocytes. J Neurosci Res 60:594–601 49. Pardridge WM, Golden PL, Kang YS, Bickel U (1997) Brain microvascular and astrocyte localization of P-glycoprotein. J Neurochem 68:1278–1285 50. Lee G, Schlichter L, Bendayan M, Bendayan R (2001) Functional expression of P-glycoprotein in rat brain microglia. J Pharmacol Exp Ther 299:204–212 51. Tatsuta T, Naito M, Oh-hara T, Sugawara I, Tsuruo T (1992) Functional involvement of P-glycoprotein in blood–brain barrier. J Biol Chem 267:20383–20391 52. Tsuji A, Terasaki T, Takabatake Y et al (1992) P-glycoprotein as the drug efflux pump in primary cultured bovine brain capillary endothelial cells. Life Sci 51:1427–1437 53. Ohnishi T, Tamai I, Sakanaka K et al (1995) In vivo and in  vitro evidence for ATPdependency of P-glycoprotein-mediated efflux of doxorubicin at the blood–brain barrier. Biochem Pharmacol 49:1541–1544 54. Schinkel AH (1999) P-Glycoprotein, a gatekeeper in the blood–brain barrier. Adv Drug Deliv Rev 36:179–194 55. Bartels AL, Kortekaas R, Bart J et al (2008) Blood–brain barrier P-glycoprotein function decreases in specific brain regions with aging: a possible role in progressive neurodegeneration. Neurobiol Aging 56. Tishler DM, Weinberg KI, Hinton DR et  al (1995) MDR1 gene expression in brain of patients with medically intractable epilepsy. Epilepsia 36:1–6 5 7. Loscher W (2007) Drug transporters in the epileptic brain. Epilepsia 48(Suppl 1): 8–13 58. Sadeque AJ, Wandel C, He H, Shah S, Wood AJ (2000) Increased drug delivery to the brain by P-glycoprotein inhibition. Clin Pharmacol Ther 68:231–237 59. Kharasch ED, Hoffer C, Whittington D (2004) The effect of quinidine, used as a probe for the involvement of P-glycoprotein, on the intestinal absorption and pharmacodynamics of methadone. Br J Clin Pharmacol 57:600–610 60. Troger U, Lins H, Scherrmann JM, Wallesch CW, Bode-Boger SM (2005) Tetraparesis associated with colchicine is probably due to inhibition by verapamil of the P-glycoprotein efflux pump in the blood–brain barrier. BMJ 331:613

220

Staud et al.

61. Enders AC, Blankenship TN (1999) Comparative placental structure. Adv Drug Deliv Rev 38: 3–15 62. van der Aa EM, Peereboom-Stegeman JH, Noordhoek J, Gribnau FW, Russel FG (1998) Mechanisms of drug transfer across the human placenta. Pharm World Sci 20:139–148 63. Audus KL (1999) Controlling drug delivery across the placenta. Eur J Pharm Sci 8: 161–165 64. Pacifici GM, Nottoli R (1995) Placental transfer of drugs administered to the mother. Clin Pharmacokinet 28:235–269 65. Kalabis GM, Kostaki A, Andrews MH et  al (2005) Multidrug resistance phosphoglycoprotein (ABCB1) in the mouse placenta: fetal protection. Biol Reprod 73:591–597 66. Novotna M, Libra A, Kopecky M et al (2004) P-glycoprotein expression and distribution in the rat placenta during pregnancy. Reprod Toxicol 18:785–792 67. Sun M, Kingdom J, Baczyk D et  al (2006) Expression of the multidrug resistance P-glycoprotein, (ABCB1 glycoprotein) in the human placenta decreases with advancing gestation. Placenta 27:602–609 68. Gil S, Saura R, Forestier F, Farinotti R (2005) P-glycoprotein expression of the human placenta during pregnancy. Placenta 26: 268–270 69. Nanovskaya TN, Nekhayeva IA, Hankins GD, Ahmed MS (2008) Transfer of methadone across the dually perfused preterm human placental lobule. Am J Obstet Gynecol 198 (126):e121–e124 70. Nakamura Y, Ikeda S, Furukawa T et al (1997) Function of P-glycoprotein expressed in placenta and mole. Biochem Biophys Res Commun 235:849–853 71. Utoguchi N, Chandorkar GA, Avery M, Audus KL (2000) Functional expression of P-glycoprotein in primary cultures of human cytotrophoblasts and BeWo cells. Reprod Toxicol 14:217–224 72. Ushigome F, Takanaga H, Matsuo H et  al (2000) Human placental transport of vinblastine, vincristine, digoxin and progesterone: contribution of P-glycoprotein. Eur J Pharmacol 408:1–10 73. Lankas GR, Wise LD, Cartwright ME, Pippert T, Umbenhauer DR (1998) Placental P-glycoprotein deficiency enhances susceptibility to chemically induced birth defects in mice. Reprod Toxicol 12:457–463 74. Smit JW, Huisman MT, van Tellingen O, Wiltshire HR, Schinkel AH (1999) Absence or pharmacological blocking of placental

P-glycoprotein profoundly increases fetal drug exposure. J Clin Invest 104:1441–1447 75. Huisman MT, Smit JW, Wiltshire HR et  al (2001) P-glycoprotein limits oral availability, brain, and fetal penetration of saquinavir even with high doses of ritonavir. Mol Pharmacol 59:806–813 76. Pavek P, Fendrich Z, Staud F et  al (2001) Influence of P-glycoprotein on the transplacental passage of cyclosporine. J Pharm Sci 90:1583–1592 77. Pavek P, Staud F, Fendrich Z et  al (2003) Examination of the functional activity of P-glycoprotein in the rat placental barrier using rhodamine 123. J Pharmacol Exp Ther 305:1239–1250 78. Holcberg G, Sapir O, Tsadkin M et  al (2003) Lack of interaction of digoxin and P-glycoprotein inhibitors, quinidine and verapamil in human placenta in  vitro. Eur J Obstet Gynecol Reprod Biol 109:133–137 79. Sudhakaran S, Ghabrial H, Nation RL et  al (2005) Differential bidirectional transfer of indinavir in the isolated perfused human placenta. Antimicrob Agents Chemother 49: 1023–1028 80. Sudhakaran S, Rayner CR, Li J et  al (2008) Inhibition of placental P-glycoprotein: impact on indinavir transfer to the foetus. Br J Clin Pharmacol 65:667–673 81. Molsa M, Heikkinen T, Hakkola J et  al (2005) Functional role of P-glycoprotein in the human blood–placental barrier. Clin Pharmacol Ther 78:123–131 82. Rahi M, Heikkinen T, Hakkola J et  al (2008) Influence of adenosine triphosphate and ABCB1 (MDR1) genotype on the P-glycoprotein-dependent transfer of saquinavir in the dually perfused human placenta. Hum Exp Toxicol 27:65–71 83. Nekhayeva IA, Nanovskaya TN, Deshmukh SV et  al (2005) Bidirectional transfer of methadone across human placenta. Biochem Pharmacol 69:187–197 84. Nanovskaya T, Nekhayeva I, Karunaratne N et al (2005) Role of P-glycoprotein in transplacental transfer of methadone. Biochem Pharmacol 69:1869–1878 85. Nekhayeva IA, Nanovskaya TN, Hankins GD, Ahmed MS (2006) Role of human placental efflux transporter P-glycoprotein in the transfer of buprenorphine, levo-alpha-acetylmethadol, and paclitaxel. Am J Perinatol 23:423–430 86. Kolwankar D, Glover DD, Ware JA, Tracy TS (2005) Expression and function of ABCB1 and ABCG2 in human placental tissue. Drug Metab Dispos 33:524–529

Expression and Function of P-Glycoprotein in Normal Tissues 87. Tanabe M, Ieiri I, Nagata N et  al (2001) Expression of P-glycoprotein in human placenta: relation to genetic polymorphism of the multidrug resistance (MDR)-1 gene. J Pharmacol Exp Ther 297:1137–1143 88. Hitzl M, Schaeffeler E, Hocher B et al (2004) Variable expression of P-glycoprotein in the human placenta and its association with mutations of the multidrug resistance 1 gene (MDR1, ABCB1). Pharmacogenetics 14:309–318 89. Brown KR (1998) Changes in the use profile of Mectizan: 1987–1997. Ann Trop Med Parasitol 92(Suppl 1):S61–S64 90. Gadducci A, Cosio S, Fanucchi A et al (2003) Chemotherapy with epirubicin and paclitaxel for breast cancer during pregnancy: case report and review of the literature. Anticancer Res 23:5225–5229 91. Kleinman CS, Nehgme RA (2004) Cardiac arrhythmias in the human fetus. Pediatr Cardiol 25:234–251 92. Oudijk MA, Ruskamp JM, Ambachtsheer BE et  al (2002) Drug treatment of fetal tachycardias. Paediatr Drugs 4:49–63 93. Capparelli E, Rakhmanina N, Mirochnick M (2005) Pharmacotherapy of perinatal HIV. Semin Fetal Neonatal Med 10:161–175 94. Clayette P, Jorajuria S, Dormont D (2000) Significance of P-glycoprotein for the pharmacology and clinical use of HIV protease inhibitors. AIDS 14:235–236 95. Ito S (2001) Transplacental treatment of fetal tachycardia: implications of drug transporting proteins in placenta. Semin Perinatol 25:196–201 96. Ploen L, Setchell BP (1992) Blood–testis barriers revisited. A homage to Lennart Nicander. Int J Androl 15:1–4 97. Setchell BP, Main SJ (1978) Drugs and the blood–testis barrier. Environ Health Perspect 24:61–64 98. Baas F, Borst P (1988) The tissue dependent expression of hamster P-glycoprotein genes. FEBS Lett 229:329–332 99. Trezise AE, Romano PR, Gill DR et al (1992) The multidrug resistance and cystic fibrosis genes have complementary patterns of epithelial expression. EMBO J 11:4291–4303 100. Sugawara I, Akiyama S, Scheper RJ, Itoyama S (1997) Lung resistance protein (LRP) expression in human normal tissues in comparison with that of MDR1 and MRP. Cancer Lett 112:23–31 101. Melaine N, Lienard MO, Dorval I et  al (2002) Multidrug resistance genes and p-glycoprotein in the testis of the rat, mouse,

221

Guinea pig, and human. Biol Reprod 67: 1699–1707 102. Lankas GR, Cartwright ME, Umbenhauer D (1997) P-glycoprotein deficiency in a subpopulation of CF-1 mice enhances avermectininduced neurotoxicity. Toxicol Appl Pharmacol 143:357–365 103. Hughes CS, Vaden SL, Manaugh CA, Price GS, Hudson LC (1998) Modulation of doxorubicin concentration by cyclosporin A in brain and testicular barrier tissues expressing P-glycoprotein in rats. J Neurooncol 37:45–54 104. Choo EF, Leake B, Wandel C et  al (2000) Pharmacological inhibition of P-glycoprotein transport enhances the distribution of HIV-1 protease inhibitors into brain and testes. Drug Metab Dispos 28:655–660 105. Choo EF, Kurnik D, Muszkat M et al (2006) Differential in  vivo sensitivity to inhibition of P-glycoprotein located in lymphocytes, testes, and the blood–brain barrier. J Pharmacol Exp Ther 317:1012–1018 106. Kreuser ED, Hetzel WD, Billia DO, Thiel E (1990) Gonadal toxicity following cancer therapy in adults: significance, diagnosis, prevention and treatment. Cancer Treat Rev 17:169–175 107. Viviani S, Santoro A, Ragni G et  al (1985) Gonadal toxicity after combination chemotherapy for Hodgkin’s disease. Comparative results of MOPP vs ABVD. Eur J Cancer Clin Oncol 21:601–605 108. Sharpe RM (1993) Declining sperm counts in men – is there an endocrine cause? J Endocrinol 136:357–360 109. Sharpe RM, Fisher JS, Millar MM, Jobling S, Sumpter JP (1995) Gestational and lactational exposure of rats to xenoestrogens results in reduced testicular size and sperm production. Environ Health Perspect 103:1136–1143 110. Roberts MS, Magnusson BM, Burczynski FJ, Weiss M (2002) Enterohepatic circulation: physiological, pharmacokinetic and clinical implications. Clin Pharmacokinet 41:751–790 111. Kojima T, Yamamoto T, Murata M et  al (2003) Regulation of the blood–biliary barrier: interaction between gap and tight junctions in hepatocytes. Med Electron Microsc 36:157–164 112. Kamimoto Y, Gatmaitan Z, Hsu J, Arias IM (1989) The function of Gp170, the multidrug resistance gene product, in rat liver canalicular membrane vesicles. J Biol Chem 264:11693–11698 113. Watanabe T, Miyauchi S, Sawada Y et  al (1992) Kinetic analysis of hepatobiliary

222

Staud et al.

transport of vincristine in perfused rat liver. Possible roles of P-glycoprotein in biliary excretion of vincristine. J Hepatol 16:77–88 114. Schinkel AH, Mayer U, Wagenaar E et  al (1997) Normal viability and altered pharmacokinetics in mice lacking mdr1-type (drug-transporting) P-glycoproteins. Proc Natl Acad Sci USA 94:4028–4033 115. Meier Y, Pauli-Magnus C, Zanger UM et al (2006) Interindividual variability of canalicular ATP-binding-cassette (ABC)-transporter expression in human liver. Hepatology 44: 62–74 116. Cascorbi I (2006) Role of pharmacogenetics of ATP-binding cassette transporters in the pharmacokinetics of drugs. Pharmacol Ther 112:457–473 117. Marzolini C, Paus E, Buclin T, Kim RB (2004) Polymorphisms in human MDR1 (P-glycoprotein): recent advances and clinical relevance. Clin Pharmacol Ther 75:13–33 118. Kakumoto M, Takara K, Sakaeda T et  al (2002) MDR1-mediated interaction of digoxin with antiarrhythmic or antianginal drugs. Biol Pharm Bull 25:1604–1607 119. Wakasugi H, Yano I, Ito T et al (1998) Effect of clarithromycin on renal excretion of digoxin: interaction with P-glycoprotein. Clin Pharmacol Ther 64:123–128 120. Boyd RA, Stern RH, Stewart BH et al (2000) Atorvastatin coadministration may increase digoxin concentrations by inhibition of intestinal P-glycoprotein-mediated secretion. J Clin Pharmacol 40:91–98 121. Greiner B, Eichelbaum M, Fritz P et  al (1999) The role of intestinal P-glycoprotein in the interaction of digoxin and rifampin. J Clin Invest 104:147–153 122. Micuda S, Brcakova E, Fuksa L et al (2008) P-glycoprotein function and expression during obstructive cholestasis in rats. Eur J Gastroenterol Hepatol 20:404–412 123. Zollner G, Fickert P, Silbert D et  al (2003) Adaptive changes in hepatobiliary transporter expression in primary biliary cirrhosis. J Hepatol 38:717–727 124. Zollner G, Wagner M, Fickert P et al (2005) Hepatobiliary transporter expression in human hepatocellular carcinoma. Liver Int 25:367–379 125. Hidemura K, Zhao YL, Ito K et  al (2003) Shiga-like toxin II impairs hepatobiliary transport of doxorubicin in rats by

down-regulation of hepatic P glycoprotein and multidrug resistance-associated protein Mrp2. Antimicrob Agents Chemother 47: 1636–1642 126. Pavek P, Dvorak Z (2008) Xenobioticinduced transcriptional regulation of xenobiotic metabolizing enzymes of the cytochrome P450 superfamily in human extrahepatic tissues. Curr Drug Metab 9:129–143 127. Lash LH, Putt DA, Cai H (2008) Drug metabolism enzyme expression and activity in primary cultures of human proximal tubular cells. Toxicology 244:56–65 128. Launay-Vacher V, Izzedine H, Karie S et  al (2006) Renal tubular drug transporters. Nephron Physiol 103:p97–p106 129. Hori R, Okamura N, Aiba T, Tanigawara Y (1993) Role of P-glycoprotein in renal tubular secretion of digoxin in the isolated perfused rat kidney. J Pharmacol Exp Ther 266:1620–1625 130. Micuda S, Fuksa L, Mundlova L et al (2007) Morphological and functional changes in p-glycoprotein during dexamethasone-induced hepatomegaly. Clin Exp Pharmacol Physiol 34:296–303 131. Kawahara M, Sakata A, Miyashita T, Tamai I, Tsuji A (1999) Physiologically based pharmacokinetics of digoxin in mdr1a knockout mice. J Pharm Sci 88:1281–1287 132. Sasabe H, Kato Y, Suzuki T et  al (2004) Differential involvement of multidrug resistance-associated protein 1 and P-glycoprotein in tissue distribution and excretion of grepafloxacin in mice. J Pharmacol Exp Ther 310: 648–655 133. Smit JW, Schinkel AH, Weert B, Meijer DK (1998) Hepatobiliary and intestinal clearance of amphiphilic cationic drugs in mice in which both mdr1a and mdr1b genes have been disrupted. Br J Pharmacol 124:416–424 134. Angirasa AK, Koch AZ (2002) P-glycoprotein as the mediator of itraconazole–digoxin interaction. J Am Podiatr Med Assoc 92:471–472 135. Rengelshausen J, Goggelmann C, Burhenne J et  al (2003) Contribution of increased oral bioavailability and reduced nonglomerular renal clearance of digoxin to the digoxin– clarithromycin interaction. Br J Clin Pharmacol 56:32–38 136. Linnet K, Ejsing TB (2008) A review on the impact of P-glycoprotein on the penetration of drugs into the brain. Eur Neuro­ psychopharmacol 18:157–169

Chapter 11 Molecular Mechanism of ATP-Dependent Solute Transport by Multidrug Resistance-Associated Protein 1 Xiu-bao Chang Abstract Millions of new cancer patients are diagnosed each year and over half of these patients die from this devastating disease. Thus, cancer causes a major public health problem worldwide. Chemotherapy remains the principal mode to treat many metastatic cancers. However, occurrence of cellular multidrug resistance (MDR) prevents efficient killing of cancer cells, leading to chemotherapeutic treatment failure. Over-expression of ATP-binding cassette transporters, such as P-glycoprotein, breast cancer resistance protein and/or multidrug resistance-associated protein 1 (MRP1), confers an acquired MDR due to their capabilities of transporting a broad range of chemically diverse anticancer drugs across the cell membrane barrier. In this review, the molecular mechanism of ATP-dependent solute transport by MRP1 will be addressed. Key words: MRP1, ATP, NBD, GSH, Transport, MDR

1. Introduction Biological membrane surrounding a living cell is mainly composed of lipid bilayer which prevents the free flow of solutes across the membrane barrier. Many proteins are inserted into the hydrophobic biological membrane and able to transport hydrophilic solutes across the membrane bilayer. These membraneembedded transporters have been classified into two major families: ATP-binding cassette (ABC) transporter superfamily and solute carrier transporter superfamily (1). There are 49 members of ABC transporters in human being that were divided into seven subfamilies (www.gene.ucl.ac.uk/nomenclature/genefamily/abc.html or www.4t.com/humanabc.htm), i.e., ABCA, including 12 members; ABCB, 11 members; ABCC, 13 members; ABCD, 4 members; ABCE, 1 member; ABCF, 3 members and J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_11, © Humana Press, a part of Springer Science + Business Media, LLC 2010

223

224

Chang

ABCG, 5 members. Although all these ABC transporters have a common structure of nucleotide binding domain (NBD) and trans-membrane domain (TMD), their variant primary sequences determine their specific structures and functions. For example, ABCA1 transports cholesterol into blood stream (onto HDL), whereas ABCB2/ABCB3 (TAP1/TAP2) delivers peptides to the lumen of endoplasmic reticulum. In addition, some of the ABC transporters have functions in tissues important for absorption, such as in lung and gut, for metabolism and elimination, such as in liver and kidney, and for maintaining the barrier functions, such as blood–brain barrier, blood–cerebral spinal fluid barrier, blood– testis barrier and the maternal–fetal barrier. Thus, these ABC transporters play important roles in the absorption, disposition and elimination of the structurally diverse array of the endobiotics and xenobiotics. All these transporting processes require ATP. In this review, I will focus on the molecular mechanism of ATPdependent solute transport by multidrug resistance-associated protein 1 (MRP1 or ABCC1).

2. Brief View of MRP1 Protein MRP1 cDNA was cloned from the multidrug resistant (MDR) small cell lung cancer cell line H69AR (2), which was obtained from a drug-sensitive small cell lung cancer cell line H69 by stepwise selection in media containing doxorubicin (3). The genomic DNA fragments coding for MRP1 were significantly amplified in H69AR cell (2). Human MRP1 gene was mapped to chromosome 16p13.1 (4) and encompasses at least 200,000 base pairs containing 31 exons (5) that encode a membrane-embedded glycoprotein consisting of 1,531 amino acids (2). The amount of MRP1 mRNA in H69AR cell was much higher than those were in the parental H69 cell (2). Consistent with the high level of MRP1 mRNA, these MDR cancer cells express high levels of MRP1 protein (6–8). Over-expression of MRP1 confers H69AR cell resistance to multiple anticancer drugs. Although MRP1 shares less than 15% amino acid identity (2) with the first eukaryotic multidrug transporter P-glycoprotein (Pgp) identified in a Chinese hamster ovary cell (9, 10), many anticancer drugs transported by Pgp are also the substrates of MRP1 (Table 11.1). MRP1 belongs to the ABCC subfamily (11–13) with a typical 13 amino acid deletion between Walker A and Walker B motifs in NBD1 (14). In contrast, these 13 amino acid residues are present in NBD2 of ABCC members, as well as in both NBDs of most other eukaryotic ABC transporters (11). The significance of the

Molecular Mechanism of ATP-Dependent Solute Transport by Multidrug Resistance

225

Table 11.1 Partial list of potential MRP1 substrates Endobiotics Leukotriene C4, leukotriene D4, leukotriene E4 Prostaglandin A2-SG Hydroxynonenal-SG 15-Deoxy-D12,14-PGJ2-SG Phosphatidylcholine Estradiol-17-b-D-glucuronide Glucuronosylbilirubin Estrone 3-sulfate Dehydroepiandrosterone 3-sulfate Sulfatolithocholyl taurine Oxidized glutathione (GSSG) Drugs/Xenobiotics Aflatoxin B1-epoxide-SG Chlorambucil-SG, melphalan-SG 4-(Methylnitrosamino)-1-(3-pyridyl)-1-butanol(NNAL)-O-glucuronide Daunorubicin, doxorubicin, etoposide Vinblastine, vincristine, SN-38 Methotrexate Antiandrogens (flutamide) Saquinavir, ritonavir N-acetyl-Leu-Leu-norleucinal (ALLN) Acetaminophen-SG Ethacrynic acid-SG Cyclophosphamide 4-(Glutathione-S-yl)-quinoline 1-oxide Metolachlor Methoxychlor Arsenic-(SG)3, potassium antimonite Others Calcein-AM, CalceinBCECF 99mTc-sestamibi 99mTc-tetrofosmin Dinitrophenyl-SG Bimane-SG N-ethylmaleimide-SG Sulforaphane-SG Monoglutathionyl curcumin

deletion of these 13 amino acid residues in NBD1 of ABCC members is still not known. However, insertion of these 13 equivalent amino acid residues from Pgp into NBD1 results in the expression of an inactive MRP1 protein (15). Hydrophobicity plot analysis of MRP1 protein indicates that this protein has a similar topological core-structure as Pgp (10, 16–18), except that

226

Chang

Fig. 11.1. Topology model of human MRP1 protein. MRP1 protein is a typical member of ABCC subfamily containing an extra N-terminal fragment composed of transmembrane domain 0 (TMD0 including transmembrane segment TM1–5) and L0 linker region and a similar core structure as P-glycoprotein having TMD1 (including TM6–11), nucleotide binding domain 1 (NBD1), linker region 1 (L1), TMD2 (including TM12–17) and NBD2. The three tree signs indicate the potential N-linked oligosaccharide chains at N19, N23 and N1006.

there is an extra N-terminal trans-membrane domain (TMD0 including TM1-5) and a cytoplasmic L0 linker region (19, 20) as shown in Fig. 11.1. In fact, most of the ABCC subfamily members contain this additional N-terminal TMD0 and L0 linker region, except ABCC4, ABCC5 and ABCC7 (14, 21, 22). L0 linker region is essential for drug transport (23, 24), whereas TMD0 is not (24). However, deletion of the TMD0 significantly reduced the ATP-dependent leukotriene C4 (LTC4) transport activity (24). Figure 11.1 shows that there are three potential N-linked glycosylation sites located at N19, N23 and N1006. Complexglycosylated mature MRP1 protein has an apparent molecular weight of 190 kDa, whereas inhibition of the complex-glycosylation by tunicamycin or treatment with endoglycosidase PNGase F generated an MRP1 protein with an apparent molecular weight of 165–170 kDa (25, 26), consistent with the predicted molecular weight from the deduced amino acid sequence of MRP1 (2).

3. MRP1 Expression, Substrates, and Possible Physiological Function of the Protein

MRP1 is expressed in many different organs and cell types (27), with high levels in lung, testis, kidney, skeletal and cardiac muscles, and in placenta (1, 2, 27, 28). MRP1 is also expressed in stomach, spleen, colon, thyroid, urinary bladder and adrenal (29). The expression of MRP1 in choroid plexus is approximately fivefold higher than in lung (30), which is consistent with the

Molecular Mechanism of ATP-Dependent Solute Transport by Multidrug Resistance

227

finding that Mrp1 mRNA level is very high in the choroid plexus of rats (31). Although MRP1 is expressed almost ubiquitously in many different organs (27), it is nevertheless found primarily in specific cell types, such as in bronchial epithelial cells and hyperplastic type II pneumocytes in lung; in Leydig and Sertoli cells in testis; in proliferating Paneth cells in colon; in placental syncytiotrophoblasts and epithelial cells of the endoplacental yolk sac; and in mast cells, eosinophils, helper T-cells, as well as erythrocytes in circulatory system (32–44). MRP1 is also expressed in blood– brain barrier and the choroid plexus of the blood–cerebrospinal fluid barrier (45–47). Since MRP1 is primarily expressed in these specific cell types, the location of this membrane-embedded glycoprotein in the membrane of these polarized cells should predict its physiological function. MRP1 was observed at the basal side of Sertoli and Leydig cells (48) and proximal tubular cells (49–51). In fact, MRP1 is predominantly localized to the basolateral membrane in the polarized cells (35, 43, 51–54). The basolateral membrane location of MRP1 protein drastically contrasts with the apical membrane localization of other ABC transporters, such as Pgp (55), MRP2 (56), breast cancer resistance protein (57) and cystic fibrosis transmembrane-conductance regulator (58–60). Since MRP1 couples ATP binding/hydrolysis to solute transport, both ATP and solute transported by this protein should be considered MRP1 substrates. However, ATP is specifically considered the substrate of the two NBD ATPases of MRP1. Thus, MRP1 substrates, in this review, mean that solutes transported by this protein. Cells over-expressing MRP1 are resistant to a wide variety of anticancer drugs, such as those drugs listed in Table  11.1, suggesting that they should be the substrates of MRP1. However, experiments performed in vitro with membrane vesicles containing MRP1 demonstrated that MRP1 had no ability to transport the un-modified anticancer drugs (61), such as daunorubicin or vincristine, unless physiological amount of glutathione (GSH) was added to the reaction mixture (62, 63), implying that MRP1 might co-transport the hydrophobic anticancer drugs with GSH. GSH itself is a relatively poor substrate of MRP1, with an apparent Km value of more than 1 mM (64–66). However, in the presence of hydrophobic anticancer drug, the Km value for GSH decreased to approximately the same range (~90  mM) as its oxidized product GSSG (62, 67, 68), further confirming that GSH might be co-transported with the hydrophobic compounds. Indeed, GSH can be transported by MRP1 only certain amount of hydrophobic compound, such as verapamil, present in the solution (69), implying that MRP1 substrates should contain two portions: hydrophobic and hydrophilic. In fact, the endobiotic MRP1

228

Chang

substrate leukotriene C4 (LTC4) listed in Table  11.1 is a glutathione-S-conjugated LTA4. Further studies resulted in the elucidation of MRP1 as a primary active ATP-dependent membrane transporter for many other amphiphilic anionic conjugates, including S-(2,4-dinitrophenyl)-GS, a dipeptide Cys-gly conjugated LTD4, a single amino acid Cys-conjugated LTE4 or an N-acetylcysteine-S-conjugated N-acetyl-LTE4, monochloro-monoglutathionyl melphalan, a glucuronate-conjugated 17b-glucuronosyl estradiol, monoglucuronosyl bilirubin, bisglucuronosyl bilirubin, 6a-glucuronosyl hyodeoxycholate, sulfate-conjugated 3a-sulfatolithocholyltaurine and glucuronosyl etoposide (61, 62, 70–77). LTC4 is synthesized in mast cells, basophils, eosinophils, dendritic cells, macrophages, neutrophils, platelets, kidney, brain, liver microsomes and endothelial cells (78–81). Although whether LTC4 synthesized in these cells is exported by MRP1 is not convincingly proved, leukotriene release from eosinophils and mast cells in response to IgE-mediated inflammation is reduced in Mrp1−/− knockout mice (77). The Mrp1−/− knockout mice also displayed an impaired immune response to contact sensitization (77). Since LTC4 was found to be the best substrate of MRP1 with the highest affinity known to date (75), the LTC4 synthesized in the cells mentioned above might be exported by MRP1. Since MRP1 is a membrane-embedded glycoprotein, many trans-membrane segments of the protein should interact with the membrane phospholipids. In addition, phospholipids, similar to typical MRP1 substrates, in biological membrane contain hydrophobic portion and hydrophilic portion. Thus, it is possible that MRP1 protein interacts with the phospholipids in membrane bilayer and flop the phospholipids from inner leaflet to outer leaflet. Indeed, experiments with red blood cells from Pgp and Mrp1 knockout (Mdr1a/1b−/−/Mdr2−/−/Mrp1−/−) mice revealed that Mrp1 is responsible for the observed flop of the 1-oleoyl-2(6(7-nitrobenz-2-oxa-1,3-diazol-4-yl)amino)caproyl (ONBD) fluorescent labeled lipid analogs (82). However, no indications for outward transport of endogenous phosphatidylserine (PS) have been observed in these experiments (82, 83). Thus, whether MRP1 can transport the endogenous phospholipids was not clear. The following experiments clearly indicate that MRP1 can redistribute the endogenous phospholipids. The redistribution of the ONBD-labeled lipid analogs, at the equilibrium in the absence of MRP1 inhibitor, was very similar to that of endogenous phospholipids (84). Upon treatment with MRP1 inhibitors, such as verapamil or indomethacin, the translocations of the ONBDlabeled phosphatidylcholine (PC) and sphingomyelin and the endogenous natural PC and sphingomyelin were significantly reduced (84). MRP1 also flops the NBD-labeled glucosylceramide in a GSH-dependent manner and is inhibited by the MRP1 inhibitors sulfinpyrazone or indomethacin, but not by the Pgp

Molecular Mechanism of ATP-Dependent Solute Transport by Multidrug Resistance

229

inhibitor PSC833 (85). Flow cytofluorimetry studies with external PS labeling showed that Pgp and MRP1 have strong effects on cell lipid pools (86), implying that both Pgp and MRP1 participate in the redistribution of phospholipids in membrane bilayer. The flopping of ONBD-labeled PS from the inner leaflet to the outer leaflet was highly enhanced when GSH was entrapped inside the resealed ghosts derived from human erythrocytes containing MRP1 (87, 88), implying that GSH is essential for the ATPdependent lipid translocation. This conclusion was further proved by using membrane vesicles derived from MRP1 cDNA transfected baby hamster kidney cell (89). Since cells over-expressing MRP1 become MDR, physiological function of this protein is predicted to transport the toxic endoand/or xeno-biotics across the biological membrane and to protect the cells from these toxic compounds. However, MRP1 is located basolaterally (51) and tends to pump drugs into the body, rather than excreting them into the bile, urine or gut (12, 21). Indeed, no decreased drug accumulation has been observed in Mrp1−/− knockout mice (77, 90, 91). However, tissues that normally express relatively high levels of Mrp1, such as testis, kidney, oropharyngeal mucosa etc., are hypersensitive to etoposide in Mrp1−/− knockout mice (42, 77), indicating that MRP1 has non-redundant protective functions against etoposide in these tissues (42). The basolateral location of Mrp1 in the Sertoli cells of testis also allows this protein to protect the germline cells against toxic drugs (21). A similar situation appears to exist in the choroid plexus, where many substances enter the cerebrospinal fluid from the epithelial cells covering the plexus. It has been found that Mrp1 mRNA level is very high in the choroid plexus of rats (31) and the high levels of Mrp1 in these cells (30, 92) play a crucial role in preventing the entry of drugs into the cerebrospinal fluid (CSF) (45). This is consistent with the reports that Mrp1 transports xenobiotics from CSF to blood (1). Conversely, Mrp1−/− knockout mice displayed increased passage of MRP1 substrates from the blood to the CSF (45). This conclusion is further supported by the fact that Mrp1/ Mdr1a/Mdr1b triple knockout mice accumulate tenfold more etoposide in CSF than in Mdr1a/Mdr1b double knockout mice after intravenous administration of the drug (45). Thus, Mrp1 may function as a barrier to drug penetration into CSF.

4. Substrate Binding Sites in MRP1 Protein For the occurrence of ATP-dependent MRP1 substrate transport, substrate binding and releasing must occur at discrete sites with different affinities, accompanied with variant conformational

230

Chang

changes of the MRP1 protein. As mentioned, MRP1 transports anionic conjugates, such as GSH-S-conjugated aliphatic, prostanoid or heterocyclic compounds, across membrane bilayer (65, 71, 73, 74, 93–97). Thus, these amphiphilic anionic substrates containing hydrophilic and hydrophobic portions must interact with MRP1 at two discrete regions. Neither GSH alone nor vincristine alone is a strong inhibitor of MRP1-mediated transport of LTC4 (65). However, together they (GSH + vincristine) act as relatively potent competitive inhibitors (65), implying that the un-conjugated GSH may bind to a similar region of hydrophilic portion (GSH) of LTC4, whereas vincristine may bind to a similar region of hydrophobic portion (LTA4) of LTC4. In another report the labeling of MRP1 with (125I)11-azidophenyl agosterol A ((125I)azido-AG-A) was shown to be GSH dependent (98), indicating that hydrophilic GSH binding site is different from that of the hydrophobic portion. The N-terminal half (N-half) and C-terminal half (C-half) of MRP1 were labeled with LTC4 (99–102), implying that there are at least two LTC4 binding sites, one in N-half and another in C-half. Expression of either N-half alone or C-half alone did not produce efficient LTC4 labeling (101), indicating the need for a cooperative interaction between the two halves. This conclusion was further confirmed by using radioactive-labeled agosterol A + GSH to label MRP1 protein (98). In addition, labeling of N-half of MRP1 with iodoaryl-azido-LTC4 (IAALTC4) is localized to a region encompassing TMD1 and NBD1 and dependent on the presence of L0 linker region but not the TMD0 (101), whereas labeling of C-half of MRP1 protein is localized to a region encompassing trans-membrane segment 14–17 (TM14– 17) without requirement of NBD2 (101). 125I-labeled azido tricyclic isoxazole LY475776 (hydrophobic portion) binds to a region encompassing the TM16–17 of MRP1 protein in a GSH dependent manner (103, 104). When N-(hydrocinchonidin-8¢yl)-4-azido-2-hydroxybenzamide (IACI) (105) or iodoaryl azidorhodamine 123 (106) was used to label MRP1 protein, two fragments, 111 kDa (N-half) and 85 kDa (C-half), were labeled, consistent with the results derived from the LTC4 labeling (99–101). Further proteolytic digestion of the IACI (105) or iodoaryl azido-rhodamine 123 (106) labeled MRP1 fragments generated a 6 kDa fragment from the 85 kDa C-half and 6 kDa plus 4  kDa fragments from the 111  kDa N-half, which were further narrowed down to MRP1 sequences encoding TM10–11 and 16–17 by using hemaglutinin tagged MRP1 (107), suggesting that the hydrophobic portion of the MRP1 substrates, such as rhodamine 123, mainly binds to the hydrophobic TM10–11 and TM16–17. The TM1–5 (TMD0), 10–11, 12 and 16–17 were labeled by using a photoreactive IAALTC4 which contains both hydrophobic and hydrophilic portions of the substrate (102),

Molecular Mechanism of ATP-Dependent Solute Transport by Multidrug Resistance

231

further proved that there are multiple binding sites in MRP1 molecule. Interestingly, the TM10–11 and TM16–17 of MRP1 protein correspond to the photoaffinity labeled TM5–6 and TM11–12 of Pgp protein (108–111), implying that the hydrophobic portion of the substrates might bind to similar regions in different ABC transporters. The results derived from mutagenesis analysis indicate that the sequences in TMs may be directly or indirectly involved in hydrophobic portion of MRP1 substrate binding. Mutations of C43S in TM1 (112); P343A, K332L and K332D in TM6 (113, 114); W445A and P448A in TM8 (113, 115); T550A, T556A and P557A in TM10 (113, 116); N590A, F594A, P595A, N597A, S604A and S605A in TM11 (113, 117, 118); E1089Q, E1089A, E1089L, E1089N, K1092, S1097 and N1100 in TM14 (119, 120); R1197K in TM16 (121); Y1236F, T1241A, T1242A, T1242C, T1242S, T1242L, Y1243F, N1245A, W1246C, W1246A, W1246F, W1246Y or R1249K in TM17 (121–124) significantly affect MRP1 function. The following example clearly indicates that these amino acid sequences in the TMs mutated in above list contribute to hydrophobic portion of MRP1 substrate binding as well as substrate specificity. Mouse Mrp1 has 88% identical amino acid sequences of human MRP1 (40), but this murine protein does not provide resistance to doxorubicin, daunomycin, epirubicin and colchicine (94), presumably reflecting the structure differences between them. In addition, mouse Mrp1 is a relatively poor transporter of E217bG (125). Replacement of the C-terminal third part of mouse Mrp1 with the corresponding region of human MRP1 encoding amino acids 959–1531 produced the hybrid protein capable of conferring some levels of resistance to doxorubicin and epirubicin (125). In addition, this hybrid protein transported E217bG with an efficiency similar to that of human MRP1 (125), indicating that the third part of the C-terminal half of human MRP1 is important for doxorubicin, epirubicin and E217bG binding. Further mutagenesis analysis indicates that converting the mouse Mrp1 Q1086 to E, which is in corresponding position of E1089 in human MRP1, creates a protein that conferred resistance to doxorubicin (119, 123), suggesting that E1089 in human MRP1 is required for anthracycline resistance. In consistent with these results, converting human E1089 to mouse Q at the corresponding position (human MRP1/ E1089Q) markedly decreased resistance to anthracycline, but without affecting LTC4 and E217bG transport (119), suggesting that anthracycline and hydrophobic portion of LTC4 or E217bG might bind to slightly different regions within MRP1 protein. In contrast to labeling with hydrophobic portion of the MRP1 substrates, such as rhodamine 123, the radioactive labeled IAAGSH, which contains the hydrophilic portion of GSH, labeled not only the two hydrophobic TM10–11 and

232

Chang

TM16–17 regions, but also two hydrophilic cytoplasmic L0 and L1 linker regions (126, 127), implying that the hydrophilic portion of IAAGSH may bind to the two hydrophilic cytoplasmic linker regions. However, it has not made clear why IAAGSH also labeled the two hydrophobic TM10–11 and TM16–17 regions. It could be due to the presence of hydrophobic iodoaryl in IAAGSH or the azido group located on the iodoaryl. In other reports, labeling of MRP1 with 125I-11-azidophenyl agosterol A (125I-azido-AG-A) (98) or with 125I-labeled azido tricyclic isoxazole LY475776 (103, 104) was shown to be GSH dependent, indicating that GSH binding site is separated from the hydrophobic portion binding sites. The C-half of MRP1 was labeled by 125I-azido-AG-A whereas GSH bound to the L0 region (98). Thus, the residues in L0 and L1 regions may be involved in binding of the hydrophilic portion, such as GSH, of the MRP1 substrates. L0 is essential for photoreactive IAALTC4 labeling (101) as well as for GSH and LTC4 binding (52). The functional boundaries of L0 linker region have been defined to C208-N260 region and this linker region is required for basolateral trafficking in polarized MDCK-I cell (52) as well as for ATP-dependent LTC4 transport (52, 128). The fact that only compounds with hydrophilic portion of the MRP1 substrates, such as GSH, MK571 or LTC4, can inhibit the IAAGSH labeling (126, 127) implies that main binding force of IAAGSH come from the hydrophilic portion of GSH. Furthermore, some hydrophobic compounds, such as vincristine or verapamil, stimulate the IAAGSH labeling (126), further supporting above conclusion. However, these results did not answer the question whether the hydrophilic portions of MRP1 substrates, such as GSH, glucuronide or sulfate, bind to the same region or different regions. Glucuronate conjugated estrogen E217bG inhibits the photolabeling of MRP1 with 3H-LTC4 in a concentration dependent manner (101), implying that LTC4 and E217bG may bind to the similar regions within the MRP1 protein. Since both LTC4 and E217bG contain hydrophobic and hydrophilic portions of the substrates, these results cannot distinguish whether the hydrophobic portion or hydrophilic portion or both bind to similar regions within MRP1 protein. Interestingly, tobacco-specific carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone and its metabolite 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) are responsible for the induction of lung cancer and the formation of glucuronide-conjugated NNAL is an important step in detoxification of these compounds. However, NNALO-glucuronide itself is not a good substrate of MRP1, unless millimolar-range of GSH is added (129), implying that GSH might bind to one place that is not occupied by b-O-glucuronide conjugate. Furthermore, estrone-3-sulfate itself is a poor competitor of MRP1-mediated transport of LTC4, however, in the presence of GSH, its inhibitory potency is markedly

Molecular Mechanism of ATP-Dependent Solute Transport by Multidrug Resistance

233

increased (130), indicating that GSH and estrone-3-sulfate bind to two distinct places.

5. ATP-Dependent Substrate Transport by MRP1

For ATP-dependent substrate transport to occur, both amphiphilic substrate and ATP must bind to MRP1 protein at certain stage of the transport process. However, binding of ATP induced conformational changes of the MRP1 protein (131–133) and shifted the substrate binding from high to low affinity site (99–101). Thus, substrate must bind to the high affinity site before it is shifted to low affinity site upon ATP binding. Hydrophobic portion, hydrophilic portion or amphiphilic substrates themselves can bind to MRP1 protein (73, 98–102, 105, 106). In addition, ATP itself can also bind to MRP1 protein (15, 134–136). However, ATP binding or ATPase activity of the purified MRP1 protein were significantly enhanced in the presence of MRP1 substrates, including doxorubicin, etoposide, GSH, doxorubicin + GSH, etoposide + GSH, GSSG, E217bG, LTC4, the dietary flavonoids kaempferol and naringenin, etc. (15, 134–139), presumably reflecting the enhancement effects caused by conformational changes induced by substrate binding (133, 140). Interestingly, MRP1 protein becomes more “compact” or less water accessible upon the hydrophilic portion of MRP1 substrate GSH binding (140). In contrast, the protein becomes more “relaxed” or more water accessible upon the hydrophobic portion of MRP1 substrate doxorubicin binding (140). Interestingly, in the presence of both GSH and doxorubicin, the water accessibility of the MRP1 protein is very similar to the condition where no substrate was added (140), presumably, GSH and doxorubicin bind to different regions of MRP1 and induce different conformational changes. Regardless of whether the conformational changes were induced by hydrophobic portion of MRP1 substrate or hydrophilic portion or both, all these conformational changes enhanced ATP binding/hydrolysis in MRP1 protein (15, 134–139). Structure analyses of some bacterial ABC transporter NBDs indicate that the two NBDs form a dimer in which the two ATP molecules are each sandwiched between the Walker A motif from one NBD and the LSGGQ ABC signature motif from another (141–145), meaning that the two ATP molecules are sandwiched between the two NBDs. Although studies with electron microscopy of negatively stained single particle revealed the structure information of the MRP1 protein to 22 Ǻ resolution (146), essentially no structural data exists for the two non-equivalent

234

Chang

NBDs of MRP1 that form hetero-dimer in the presence of Mg++ and ATP. The extremely low intrinsic ATPase activities of the purified recombinant NBD1 alone or NBD2 alone (147–149) imply that they need to form hetero-dimer to have relatively high ATPase activity. However, the extremely low intrinsic ATPase activity of the purified NBD1 of human MRP1 allowed them to obtain the crystal structure of wild-type NBD1 in a complex with Mg·ATP (150), revealed an unexpected non-productive conformation of the putative catalytic site (150). However, these structural analyses did not provide further information to indicate that the two non-equivalent NBDs of MRP1 form hetero-dimer in the presence of Mg++ and ATP. Although it has been found that the purified recombinant NBD1 and NBD2 transiently interact with each other in the presence of Mg++ and ATP and the residue G771 in the LSGGQ signature sequence of NBD1 is involved in NBD1·ATP·ATP·NBD2 hetero-dimer formation (149), where no increase in the ATPase activity could be detected in the NBD1 and NBD2 mixture (149). These results indicate that this labile interaction is not sufficient to form NBD1·ATP·ATP·NBD2 sandwich structure as shown for E171Q-mutated bacterial ABC transporter MJ0796 (143), suggesting that the other parts of MRP1 may be required for the effective hetero-dimer formation. However, substitution of the conserved glycine residue at the fourth position of LSGGQ motif with an A or a D residue in NBD1 (G771D or G771A) or in NBD2 (G1433D or G1433A) lost their abilities to transport substrate across the membrane (99, 151, 152). Further analyses of these mutants indicated that G771D mutation enhanced the (a-32P)-ATP binding on ice at the mutated NBD1, but completely inhibited the vanadate (Vi)-dependent ADP trapping at the intact NBD2 at 37 ºC (153). Conversely, the G1433D mutation in the ABC signature sequence of NBD2 enhanced the (a-32P)-ATP binding on ice at the mutated NBD2, but completely inhibited the Vi-dependent ADP trapping at the intact NBD1 at 37 ºC (153), implying that the LSGGQ signature motif in NBD1 may be involved in ATP binding at the NBD2, whereas the LSVGQ signature motif in NBD2 may be involved in ATP binding at the NBD1. Experimental data support the notion that the TMDs including linker regions and intracellular loops form substrate binding sites and provide a translocation pathway across the biological membranes (98–101, 105–107, 111, 119, 122–124, 154, 155). All ABC transporters, such as MRP1, transport solute in an ATPdependent manner (61, 62, 71–74, 76). Poorly hydrolysable ATP analog ATPgS or non-hydrolysable ATP analog AMP-PNP or AMP-PCP does not support the ATP-dependent solute transport (61, 62, 71), implying that ATP hydrolysis is required for the transport. However, binding of poorly hydrolysable ATP analog ATPgS to wt MRP1 significantly inhibits the 3H-LTC4 labeling

Molecular Mechanism of ATP-Dependent Solute Transport by Multidrug Resistance

235

(99, 100), implying that ATPgS binding might be sufficient to transport the bound LTC4 from high to low affinity site. This conclusion was further supported by the fact that ATPgS or ATP binding to the incompetent E1455D or E1455Q mutants, which were unable to hydrolyze the bound ATP, significantly inhibited the 3H-LTC4 labeling (100, 156, 157). However, the nonhydrolysable ATP analog AMP-PNP or AMP-PCP are unable to inhibit the 3H-LTC4 labeling (100). Interestingly, the nonhydrolysable ATP analog AMP-PNP mainly bound to NBD1 (158) and this binding at NBD1 cannot enhance the ADP trapping at NBD2 (159). Presumably, replacement of the oxygen atom between the b- and g-phosphate of ATP with nitrogen (in AMP-PNP) or carbon (in AMP-PCP) cannot induce proper conformational changes of the protein so that the bound substrate cannot be shifted from high to low affinity site. This kind of interpretation can also be used to explain why ATPgS (with an oxygen atom between the b- and g-phosphate of ATP) binding at NBD1 induced proper conformational changes and enhanced ADP trapping at NBD2 (159) and the binding at both NBDs was sufficient to transport the bound LTC4 from high to low affinity site (99, 100). Thus, the following two points are important for the ATP-dependent solute transport by MRP1, i.e., nucleotide binding and proper conformational changes of MRP1 induced by the nucleotide binding. If that is the case, blocking binding should inhibit the translocation of the bound substrate from high to low affinity site. Indeed, several mutations, such as K684E, K1333E, K684R, K1333R, D792N, D1454N, G771A and G1433A, significantly diminished ATP binding and lost the ability to shift the bound substrate from high to low affinity site (99). Conversely, mutation of the putative catalytic residue E1455 to a short chain D residue, E1455D, markedly increased the affinity of the mutated NBD2 for ATP while decreased its ability to hydrolyze ATP (100), leading to significantly increase the a-32P-ATP labeling regardless of whether Vi is present or not (100). Binding of ATP, ATP + Vi, or ATPgS to E1455D significantly inhibited the LTC4 labeling (100), further supporting the above hypothesis that occupancies of both NBD1 and NBD2 by nucleotide binding without hydrolysis may be sufficient to transport the bound substrate across membrane bilayer. However, the fact that the poorly hydrolysable ATP analog ATPgS cannot support the nucleotide-dependent solute transport is still a puzzle. Thus, it is still not clear whether nucleotide binding alone is sufficient to transport the solute across the plasma membrane or not. Based on our results, we have proposed that the release of bound nucleotide from the two NBDs brings the MRP1 back to its original conformation so that the protein can start a new cycle of ATP-dependent solute transport (156, 160, 161). Thus, the time required to release 50% of the bound

236

Chang

nucleotide from MRP1 (T1/2) should be related to the rate of ATP-dependent solute transport. T1/2 for the Vi-trapped (a-32P)-8-N3ADP is ~5  min, whereas T1/2 for the occluded (a-32P)-8-N3AMP-PNP (without Vi) is ~15  min (unpublished data). T1/2 of (a-32P)-8-N3ATP for wt MRP1 (without Vi) is too short to be accurately determined. Thus, ATP hydrolysis may be responsible for the facilitated release of the bound nucleotides. If that is the case, even if nucleotide binding alone is sufficient to transport solute across the plasma membrane, the substrate transported by MRP1 in the presence of the poorly hydrolysable ATP analog ATPgS might be too low to be detected. The above conclusion means that ATP hydrolysis is required for the efficient substrate transport by MRP1. However, this conclusion did not indicate whether the ATP hydrolysis at NBD1 or NBD2 or both are required to facilitate the release of the bound nucleotides. In the case of Pgp, the two NBDs have been shown to be functionally equivalent with identical ATP hydrolysis steps occurring alternately at each NBD (162, 163). If one NBD enters the transition-state conformation, the other site is prohibited from doing so (162, 164, 165). Modification of either site completely abolished the function of Pgp (162, 166–170). Based on these results, Senior et al. proposed a model of coupling ATP hydrolysis alternately at each NBD with a solute transported in each cycle (171). Sauna et al. proposed a different model, i.e., the first ATP hydrolysis is associated with transport of a solute, whereas the second ATP hydrolysis brings the Pgp back to its original conformation so that the protein can start a new cycle of ATP dependent solute transport (172, 173), meaning that one solute transported by Pgp requires two ATP hydrolyses. In this model, ATP binding/hydrolysis sites of Pgp are recruited in a random manner and only one site is utilized at any given time because of the conformational changes induced by ATP hydrolysis drastically reduce the affinity of the second site for nucleotide (172), suggesting that the two NBDs of Pgp play equal role during ATP-dependent solute transport. In the case of MRP1, the following points clearly indicate that its two non-equivalent NBDs have distinctive properties and different functions. (1) Substitutions of the Walker A S685 or Walker B D792 in NBD1 with a hydrophobic residue resulted in mis-processing of the protein, whereas substitutions of the corresponding residues in NBD2 did not have any significant effect on the protein folding (136, 157, 174, 175), indicating that the two non-equivalent NBDs have different steric structures. (2) Photoaffinity labeling experiments with 8-azido-ATP also revealed an asymmetry between NBD1 and NBD2, with (g-32P)-8-N3ATP preferentially labeling NBD1 (15, 136), with trapping of the ATP hydrolysis product, ADP, occurring primarily at NBD2 (15, 135, 136), implying that NBD2 has higher ATPase

Molecular Mechanism of ATP-Dependent Solute Transport by Multidrug Resistance

237

activity than that of NBD1. (3) ADP trapping by Vi at NBD2 stabilized ATP binding at NBD1 (136), whereas ATP binding at NBD1 allosterically enhanced ADP trapping or AMP-PNP binding at NBD2 (158, 159). (4) ATP binding sites of MRP1 seem not to be recruited in a random fashion because photolabeling by the non-hydrolysable ATP analog (a-32P)-8-N3AMP-PNP in the absence of other nucleotides occurred predominately at NBD1 and this labeling was significantly enhanced at NBD2 by a low concentration of ATP (158), implying that NBD1 has higher affinity for nucleotide than NBD2. Further studies confirmed this conclusion (176). (5) Mutations of the consensus Walker motifs in the two NBDs resulted in having different effects on solute transport (15, 136, 175). (6) Substitution of the putative catalytic residue E1455 in NBD2 with a non-acidic amino acid almost completely abolished ATP-dependent LTC4 transport (100, 156). Conversely, substitution of the corresponding putative catalytic residue D793 in NBD1 with a non-acidic amino acid increased the rate of ATP-dependent LTC4 transport (156, 175). These results are consistent with our previous conclusion that ATP binding to NBD1 mainly plays a regulatory role, whereas ATP hydrolysis at NBD2 plays a crucial role during ATPdependent solute transport by MRP1 (176). Based on the results mentioned above, we have proposed a model of ATP-dependent solute transport by MRP1 as shown in Fig. 11.2. The model in Fig. 11.2 showed that ATP bound at NBD1 might not be hydrolyzed during ATP-dependent solute transport by MRP1. The experimental data derived from structural and mutational analyses strongly support this conclusion. Structure analysis of a prokaryotic ABC transporter MJ0796 revealed that the acidic amino acid directly adjacent to the aspartic acid in Walker B motif may function as a putative catalytic base (141). However, this putative catalytic base D793 in human MRP1NBD1 points away from the bound ATP (150), resulted in an inefficient ATP hydrolysis. Substitution of this putative catalytic base D793 in NBD1 with a non-acidic amino acid does not abolish the ATP-dependent solute transport (100, 156, 175), implying that ATP hydrolysis at NBD1 of MRP1 is not important for the ATP-dependent solute transport. In contrast, substitution of the corresponding putative catalytic base (E1455) in NBD2 with either an L or a Q has completely abolished the ATP-dependent solute transport (100, 136, 156), implying that ATP hydrolysis at NBD2 plays a crucial role during ATP-dependent solute transport by MRP1. Structure analyses of many prokaryotic ABC transporter NBDs have revealed that histidine residue in H-loop participates Mg·ATP binding by formation of hydrogen bond with the g-phosphate of the bound Mg·ATP (141, 177–180). Additionally, this histidine residue interacts with the putative catalytic base to

238

Chang

Fig.  11.2. Schematic representation of the ATP-dependent substrate transport by MRP1. In the initial unoccupied molecule the squares represent the three transmembrane domains including TMD0, TMD1 and TMD2. The circles represent the two cytoplasmic NBDs. The triangle represents the N-terminal cytoplasmic loops including the L0 linker region. The different shapes of the symbols represent the conformational changes induced by substrate binding, ATP binding/ hydrolysis and release of substrate and nucleotides. Substrate binding to the N-terminal cytoplasmic regions and the TMDs of MRP1 induces conformational changes of the protein and enhances the affinity for ATP at the NBD1 (1). ATP binding at NBD1 induces conformational changes of the protein and enhances the affinity for ATP at NBD2 (2). ATP binding at NBD2 further induces conformational changes of the protein and transports the bound substrate from high to low affinity site (3). The substrate bound at the low affinity site is then automatically dissociated from the protein and released to the outside of the cell. ATP bound at the NBD2 is efficiently hydrolyzed to generate two negatively charged ATPhydrolysis products ADP and inorganic phosphate Pi (4). These two negatively charged molecules repel each other to provide a “power stroke” and, in the meantime, to facilitate the release of ADP and Pi from NBD2 (5 and 6). The release of ADP and Pi from NBD2 facilitates the dissociation of the ATP (regardless of whether it is hydrolyzed or not) from NBD1. The release of ATP from NBD1 brings the MRP1 protein back to its original conformation so that the molecule can start a new cycle of ATP-dependent solute transport (7). Reproduced from Ref. (161) with permission.

form a catalytic dyad (150, 181). Substitution of the E residue directly adjacent to the aspartic acid in Walker B motif with a residue that eliminates the negative charge of the E residue has abolished its ATPase activity and ATP-dependent solute transport (100, 143, 156, 169, 177, 182). On the other hand, substitution of the H residue in H-loop with a residue that avoids the interactions with the putative catalytic base resulted in deficient ATPase activity (181, 183–185). Based on these results, Zaitseva and coworkers suggested that the substrate-assisted catalysis, rather

Molecular Mechanism of ATP-Dependent Solute Transport by Multidrug Resistance

239

Fig. 11.3. Model of histidine-assisted general catalytic base catalysis. The model speculated that the histidine residue H1486 in H-loop of MRP1 interacts with the g-phosphate of the bound Mg·ATP and the carboxyl group of the putative catalytic base E1455 so that the carboxyl group of the putative catalytic base can efficiently activate the water molecule located in that position to attack the g-phosphate of the bound Mg·ATP.

than general catalytic base catalysis, might operate in ABC transporters (181). We have found that substitution of the H residue in H-loop of MRP1-NBD1 with a residue, such as H827L or H827F, that avoids the interactions with the putative catalytic base D793 in NBD1 does not have a significant effect on the ATP-dependent LTC4 transport, whereas substitution of the H1486 residue in H-loop of NBD2 with the residues that potentially form hydrogen bond with the putative catalytic base E1455 has yielded functional MRP1 proteins with variant effects on the ATP-dependent LTC4 transport (186). Based on these results, we suggest that the interactions of the histidine residue in H-loop with the g-phosphate of the bound Mg·ATP and the putative catalytic base hold the g-phosphate and the catalytic base firmly so that the putative catalytic base can efficiently activate the water molecule to attack the g-phosphate of the bound Mg·ATP (termed histidine-assisted general catalytic base catalysis as shown in Fig. 11.3).

6. Conclusion and Future Prospects A tremendous amount of work has been done since the cloning of MRP1 cDNA in 1992. Although only low resolution structure of human MRP1 protein has been obtained to date (146), high resolution structure of recombinant human MRP1-NBD1 has been solved (150). With the progress made in the structure analyses of prokaryotic membrane-embedded proteins, such as

240

Chang

BtuCD (145), Sav1866 (187), Hl1470/1 (188), maltose transporter (189) or MsbA (190), there is no reason to doubt that the high resolution structure of MRP1 protein will be solved in the near future. Many endobiotics and xenobiotics, such as those listed in Table 11.1, have been found to be the substrates of MRP1. Their (high affinity) binding sites have been identified, via cross-linking and site-directed mutagenesis. In addition, site-directed mutagenesis was used to elucidate the mechanism of ATP-dependent solute transport. With the increasing success in determining the structures of ABC transporters, including MRP1, it is reasonable to predict that a more detailed mechanism of ATP-dependent solute transport by MRP1 will be elucidated. The finding that MRP1 transports amphiphilic anionic conjugates across biological membrane clearly indicates that it is different from the traditional drug transporter Pgp. The discovery of the ATP-dependent GSH-S-conjugate or the oxidized glutathione GSSG transport by MRP1 indicates that its function is not only related to detoxification but also to regulate the redox-state in the cells expressing this protein. Thus, characterization of MRP1 protein has dramatically increased appreciation of the potential for energy-dependent drug transporters to contribute to clinical MDR encountered in the treatment of many diseases. From one point of view, many normal tissues or cells require the function of MRP1 to protect them from the toxic endo-biotics and xeno-biotics. In addition, many MRP transporters, including MRP1, MRP2 and other MRPs, are also the major determinants of the distribution and disposition of both physiological and pharmacological substrates in human and animals. From another point of view, over-expression of this protein in cancer cells provides a big challenge to selectively kill these cancer cells. However, depletion of the endogenous GSH by MRP1-mediated transport (69, 191–197) provides a possibility to hyper-sensitize the MRP1-over-expressing cancer cells to anticancer drugs. However, a suitable protocol has not been established to treat the cancer patients. With the increasing success in characterizing MRP1 protein, there is a reason to be optimistic in achieving the goal of combating the MDR caused by MRP1.

Acknowledgments I thank Irene Beauvais who has helped me prepare the manuscript. This work was supported by Grant CA89078 from the National Cancer Institute, National Institutes of Health.

Molecular Mechanism of ATP-Dependent Solute Transport by Multidrug Resistance

241

References 1. Choudhuri S, Klaassen CD (2006) Structure, function, expression, genomic organization, and single nucleotide polymorphisms of human ABCB1 (MDR1), ABCC (MRP), and ABCG2 (BCRP) efflux transporters. Int J Toxicol 25:231–259 2. Cole SP, Bhardwaj G, Gerlach JH et al (1992) Overexpression of a transporter gene in a multidrug-resistant human lung cancer cell line (see comments). Science 258:1650–1654 3. Mirski SE, Gerlach JH, Cole SP (1987) Multidrug resistance in a human small cell lung cancer cell line selected in adriamycin. Cancer Res 47:2594–2598 4. Slovak ML, Ho JP, Bhardwaj G et al (1993) Localization of a novel multidrug resistanceassociated gene in the HT1080/DR4 and H69AR human tumor cell lines. Cancer Res 53:3221–3225 5. Grant CE, Kurz EU, Cole SP, Deeley RG (1997) Analysis of the intron-exon organization of the human multidrug-resistance protein gene (MRP) and alternative splicing of its mRNA. Genomics 45:368–378 6. Grant CE, Valdimarsson G, Hipfner DR et al (1994) Overexpression of multidrug resistance-associated protein (MRP) increases resistance to natural product drugs. Cancer Res 54:357–361 7. Marquardt D, McCrone S, Center MS (1990) Mechanisms of multidrug resistance in HL60 cells: detection of resistance-associated proteins with antibodies against synthetic peptides that correspond to the deduced sequence of P-glycoprotein. Cancer Res 50:1426–1430 8. Krishnamachary N, Center MS (1993) The MRP gene associated with a non-P-glycoprotein multidrug resistance encodes a 190-kDa membrane bound glycoprotein. Cancer Res 53:3658–3661 9. Juliano RL, Ling V (1976) A surface glycoprotein modulating drug permeability in Chinese hamster ovary cell mutants. Biochim Biophys Acta 455:152–162 10. Chen CJ, Chin JE, Ueda K et al (1986) Internal duplication and homology with bacterial transport proteins in the mdr1 (P-glycoprotein) gene from multidrug-resistant human cells. Cell 47:381–389 11. Hipfner DR, Deeley RG, Cole SP (1999) Structural, mechanistic and clinical aspects of MRP1. Biochim Biophys Acta 1461:359–376 12. Borst P, Evers R, Kool M, Wijnholds J (1999) The multidrug resistance protein family. Biochim Biophys Acta 1461:347–357

13. Dean M, Rzhetsky A, Allikmets R (2001) The human ATP-binding cassette (ABC) transporter superfamily. Genome Res 11: 1156–1166 14. Cole SP, Deeley RG (1998) Multidrug resistance mediated by the ATP-binding cassette transporter protein MRP. Bioessays 20:931–940 15. Gao M, Cui HR, Loe DW et  al (2000) Comparison of the functional characteristics of the nucleotide binding domains of multidrug resistance protein 1. J Biol Chem 275: 13098–13108 16. Zhang JT, Ling V (1991) Study of membrane orientation and glycosylated extracellular loops of mouse P-glycoprotein by in  vitro translation. J Biol Chem 266:18224–18232 17. Gros P, Croop J, Housman D (1986) Mammalian multidrug resistance gene: complete cDNA sequence indicates strong homology to bacterial transport proteins. Cell 47:371–380 18. Gerlach JH, Endicott JA, Juranka PF et  al (1986) Homology between P-glycoprotein and a bacterial haemolysin transport protein suggests a model for multidrug resistance. Nature 324:485–489 19. Bakos E, Hegedus T, Hollo Z et  al (1996) Membrane topology and glycosylation of the human multidrug resistance-associated protein. J Biol Chem 271:12322–12326 20. Hipfner DR, Almquist KC, Leslie EM et  al (1997) Membrane topology of the multidrug resistance protein (MRP). A study of glycosylation-site mutants reveals an extracytosolic NH2 terminus. J Biol Chem 272: 23623–23630 21. Borst P, Evers R, Kool M, Wijnholds J (2000) A family of drug transporters: the multidrug resistance-associated proteins. J Natl Cancer Inst 92:1295–1302 22. Riordan JR, Rommens JM, Kerem B et  al (1989) Identification of the cystic fibrosis gene: cloning and characterization of complementary DNA. Science 245:1066–1073 23. Lee SH, Altenberg GA (2003) Transport of leukotriene C4 by a cysteine-less multidrug resistance protein 1 (MRP1). Biochem J 370: 357–360 24. Bakos E, Evers R, Szakacs G et  al (1998) Functional multidrug resistance protein (MRP1) lacking the N-terminal transmembrane domain. J Biol Chem 273:2167–32175 25. Krishnamachary N, Ma L, Zheng L, Safa AR, Center MS (1994) Analysis of MRP gene expression and function in HL60 cells

242

Chang

isolated for resistance to adriamycin. Oncol Res 6:119–127 26. Almquist KC, Loe DW, Hipfner DR et  al (1995) Characterization of the M(r) 190, 000 multidrug resistance protein (MRP) in drugselected and transfected human tumor cell. Cancer Res 55:102–110 27. Flens MJ, Zaman GJ, van der Valk P et al (1996) Tissue distribution of the multidrug resistance protein. Am J Pathol 148:1237–1247 28. St-Pierre MV, Serrano MA, Macias RI et  al (2000) Expression of members of the multidrug resistance protein family in human term placenta. Am J Physiol Regul Integr Comp Physiol 279:R1495–R1503 29. Zaman GJ, Versantvoort CH, Smit JJ et  al (1993) Analysis of the expression of MRP, the gene for a new putative transmembrane drug transporter, in human multidrug resistant lung cancer cell lines. Cancer Res 53:1747–1750 30. Nishino J, Suzuki H, Sugiyama D et al (1999) Transepithelial transport of organic anions across the choroid plexus: possible involvement of organic anion transporter and multidrug resistance-associated protein. J Pharmacol Exp Ther 290:289–294 31. Choudhuri S, Cherrington NJ, Li N, Klaassen CD (2003) Constitutive expression of various xenobiotic and endobiotic transporter mRNAs in the choroid plexus of rats. Drug Metab Dispos 31:1337–1345 32. Atkinson DE, Greenwood SL, Sibley CP, Glazier JD, Fairbairn LJ (2003) Role of MDR1 and MRP1 in trophoblast cells, elucidated using retroviral gene transfer. Am J Physiol Cell Physiol 285:C584–C591 33. Brechot JM, Hurbain I, Fajac A, Daty N, Bernaudin JF (1998) Different pattern of MRP localization in ciliated and basal cells from human bronchial epithelium. J Histochem Cytochem 46:513–517 34. Lohoff M, Prechtl S, Sommer F et al (1998) A multidrug-resistance protein (MRP)-like transmembrane pump is highly expressed by resting murine T helper (Th) 2, but not Th1 cells, and is induced to equal expression levels in Th1 and Th2 cells after antigenic stimulation in vivo. J Clin Invest 101:703–710 35. Leslie EM, Deeley RG, Cole SP (2005) Multidrug resistance proteins: role of P-glycoprotein, MRP1, MRP2, and BCRP (ABCG2) in tissue defense. Toxicol Appl Pharmacol 204:216–237 36. Nagashige M, Ushigome F, Koyabu N et al (2003) Basal membrane localization of MRP1 in human placental trophoblast. Placenta 24: 951–958

37. Pascolo L, Fernetti C, Pirulli D et al (2003) Effects of maturation on RNA transcription and protein expression of four MRP genes in human placenta and in BeWo cells. Biochem Biophys Res Commun 303:259–265 38. Peng KC, Cluzeaud F, Bens M et  al (1999) Tissue and cell distribution of the multidrug resistance-associated protein (MRP) in mouse intestine and kidney. J Histochem Cytochem 47:757–768 39. St-Pierre MV, Stallmach T, Freimoser Grundschober A et  al (2004) Temporal expression profiles of organic anion transport proteins in placenta and fetal liver of the rat. Am J Physiol Regul Integr Comp Physiol 287:R1505–R1516 40. Stride BD, Valdimarsson G, Gerlach JH et al (1996) Structure and expression of the messenger RNA encoding the murine multidrug resistance protein, an ATP-binding cassette transporter. Mol Pharmacol 49:962–971 41. Tribull TE, Bruner RH, Bain LJ (2003) The multidrug resistance-associated protein 1 transports methoxychlor and protects the seminiferous epithelium from injury. Toxicol Lett 142:61–70 42. Wijnholds J, Scheffer GL, van der Valk M et al (1998) Multidrug resistance protein 1 protects the oropharyngeal mucosal layer and the testicular tubules against drug-induced damage. J Exp Med 188:797–808 43. Wright SR, Boag AH, Valdimarsson G et  al (1998) Immunohistochemical detection of multidrug resistance protein in human lung cancer and normal lung. Clin Cancer Res 4:2279–2289 44. Deeley RG, Westlake C, Cole SP (2006) Transmembrane transport of endo- and xenobiotics by mammalian ATP-binding cassette multidrug resistance proteins. Physiol Rev 86:849–899 45. Wijnholds J, deLange EC, Scheffer GL et  al (2000) Multidrug resistance protein 1 protects the choroid plexus epithelium and contributes to the blood-cerebrospinal fluid barrier. J Clin Invest 105:279–285 46. Mercier C, Masseguin C, Roux F, Gabrion J, Scherrmann JM (2004) Expression of P-glycoprotein (ABCB1) and Mrp1 (ABCC1) in adult rat brain: focus on astrocytes. Brain Res 1021:32–40 47. de Lange EC (2004) Potential role of ABC transporters as a detoxification system at the blood-CSF barrier. Adv Drug Deliv Rev 56:1793–1809 48. Bart J, Hollema H, Groen HJ et  al (2004) The distribution of drug-efflux pumps, Pgp,

Molecular Mechanism of ATP-Dependent Solute Transport by Multidrug Resistance

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

BCRP, MRP1 and MRP2, in the normal blood-testis barrier and in primary testicular tumours. Eur J Cancer 40:2064–2070 Cha SH, Sekine T, Fukushima JI et al (2001) Identification and characterization of human organic anion transporter 3 expressing predominantly in the kidney. Mol Pharmacol 59:1277–1286 Schaub TP, Kartenbeck J, Konig J et al (1999) Expression of the MRP2 gene-encoded conjugate export pump in human kidney proximal tubules and in renal cell carcinoma. J Am Soc Nephrol 10:1159–69 Evers R, Zaman GJ, van Deemter L et  al (1996) Basolateral localization and export activity of the human multidrug resistanceassociated protein in polarized pig kidney cells. J Clin Invest 97:1211–1218 Westlake CJ, Qian YM, Gao M et  al (2003) Identification of the structural and functional boundaries of the multidrug resistance protein 1 cytoplasmic loop 3. Biochemistry 42: 14099–14113 Hipfner DR, Gauldie SD, Deeley RG, Cole SP (1994) Detection of the M(r) 190, 000 multidrug resistance protein, MRP, with monoclonal antibodies. Cancer Res 54:5788–5792 Roelofsen H, Vos TA, Schippers IJ et al (1997) Increased levels of the multidrug resistance protein in lateral membranes of proliferating hepatocyte-derived cells. Gastroenterology 112:511–521 Thiebaut F, Tsuruo T, Hamada H et al (1987) Cellular localization of the multidrugresistance gene product P-glycoprotein in normal human tissues. Proc Natl Acad Sci USA 84:7735–7738 Kartenbeck J, Leuschner U, Mayer R, Keppler D (1996) Absence of the canalicular isoform of the MRP gene-encoded conjugate export pump from the hepatocytes in Dubin–Johnson syndrome. Hepatology 23:1061–1066 Maliepaard M, Scheffer GL, Faneyte IF et al (2001) Subcellular localization and distribution of the breast cancer resistance protein transporter in normal human tissues. Cancer Res 61:3458–3464 Hoogeveen AT, Keulemans J, Willemsen R et  al (1991) Immunological localization of cystic fibrosis candidate gene products. Exp Cell Res 193:435–437 Marino CR, Matovcik LM, Gorelick FS, Cohn JA (1991) Localization of the cystic fibrosis transmembrane conductance regulator in pancreas. J Clin Invest 88:712–716 Crawford I, Maloney PC, Zeitlin PL et  al (1991) Immunocytochemical localization of

243

the cystic fibrosis gene product CFTR. Proc Natl Acad Sci USA 88:9262–9266 61. Jedlitschky G, Leier I, Buchholz U et  al (1996) Transport of glutathione, glucuronate, and sulfate conjugates by the MRP geneencoded conjugate export pump. Cancer Res 56:988–994 62. Loe DW, Almquist KC, Deeley RG, Cole SP (1996) Multidrug resistance protein (MRP)mediated transport of leukotriene C4 and chemotherapeutic agents in membrane vesicles. Demonstration of glutathione-dependent vincristine transport. J Biol Chem 271: 9675–9682 63. Renes J, de Vries EG, Nienhuis EF, Jansen PL, Muller M (1999) ATP- and glutathionedependent transport of chemotherapeutic drugs by the multidrug resistance protein MRP1. Br J Pharmacol 126:681–688 64. Leslie EM, Deeley RG, Cole SP (2001) Toxicological relevance of the multidrug resistance protein 1, MRP1 (ABCC1) and related transporters. Toxicology 167:3–23 65. Loe DW, Deeley RG, Cole SP (1998) Characterization of vincristine transport by the M(r) 190, 000 multidrug resistance protein (MRP): evidence for cotransport with reduced glutathione. Cancer Res 58:5130–5136 66. Salerno M, Garnier-Suillerot A (2001) Kinetics of glutathione and daunorubicin efflux from multidrug resistance protein overexpressing small-cell lung cancer cells. Eur J Pharmacol 421:1–9 67. Leslie EM, Deeley RG, Cole SP (2003) Bioflavonoid stimulation of glutathione transport by the 190-kDa multidrug resistance protein 1 (MRP1). Drug Metab Dispos 31: 11–15 68. Leier I, Jedlitschky G, Buchholz U et al (1996) ATP-dependent glutathione disulphide transport mediated by the MRP gene-encoded conjugate export pump. Biochem J 314:433–437 69. Loe DW, Deeley RG, Cole SP (2000) Verapamil stimulates glutathione transport by the 190kDa multidrug resistance protein 1 (MRP1). J Pharmacol Exp Ther 293:530–538 70. Jedlitschky G, Leier I, Buchholz U et  al (1997) ATP-dependent transport of bilirubin glucuronides by the multidrug resistance protein MRP1 and its hepatocyte canalicular isoform MRP2. Biochem J 327(Pt 1):305–310 71. Loe DW, Almquist KC, Cole SP, Deeley RG (1996) ATP-dependent 17 beta-estradiol 17-(beta-D-glucuronide) transport by multidrug resistance protein (MRP). Inhibition by cholestatic steroids. J Biol Chem 271: 9683–9689

244

Chang

72. Jedlitschky G, Leier I, Buchholz U, Center M, Keppler D (1994) ATP-dependent transport of glutathione S-conjugates by the multidrug resistance-associated protein. Cancer Res 54:4833–4836 73. Leier I, Jedlitschky G, Buchholz U et  al (1994) The MRP gene encodes an ATPdependent export pump for leukotriene C4 and structurally related conjugates. J Biol Chem 269:27807–27810 74. Muller M, Meijer C, Zaman GJ et al (1994) Overexpression of the gene encoding the multidrug resistance-associated protein results in increased ATP-dependent glutathione S-conjugate transport. Proc Natl Acad Sci USA 91:13033–13037 75. Keppler D, Leier I, Jedlitschky G (1997) Transport of glutathione conjugates and glucuronides by the multidrug resistance proteins MRP1 and MRP2. Biol Chem 378:787–791 76. Leier I, Jedlitschky G, Buchholz U, Keppler D (1994) Characterization of the ATP-dependent leukotriene C4 export carrier in mastocytoma cells. Eur J Biochem 220:599–606 77. Wijnholds J, Evers R, van Leusden MR et al (1997) Increased sensitivity to anticancer drugs and decreased inflammatory response in mice lacking the multidrug resistanceassociated protein. Nat Med 3:1275–1279 78. Schroder O, Sjostrom M, Qiu H, Jakobsson PJ, Haeggstrom JZ (2005) Microsomal glutathione S-transferases: selective up-regulation of leukotriene C4 synthase during lipopolysaccharide-induced pyresis. Cell Mol Life Sci 62:87–94 79. Shimada K, Navarro J, Goeger DE et al (1998) Expression and regulation of leukotrienesynthesis enzymes in rat liver cells. Hepatology 28:1275–1281 80. Mayatepek E (2000) Leukotriene C4 synthesis deficiency: a member of a probably underdiagnosed new group of neurometabolic diseases. Eur J Pediatr 159:811–818 81. Scoggan KA, Jakobsson PJ, Ford-Hutchinson AW (1997) Production of leukotriene C4 in different human tissues is attributable to distinct membrane bound biosynthetic enzymes. J Biol Chem 272:10182–10187 82. Dekkers DW, Comfurius P, Schroit AJ, Bevers EM, Zwaal RF (1998) Transbilayer movement of NBD-labeled phospholipids in red blood cell membranes: outward-directed transport by the multidrug resistance protein 1 (MRP1). Biochemistry 37:14833–14837 83. Borst P, Zelcer N, van Helvoort A (2000) ABC transporters in lipid transport. Biochim Biophys Acta 1486:128–144

84. Dekkers DW, Comfurius P, van Gool RG, Bevers EM, Zwaal RF (2000) Multidrug resistance protein 1 regulates lipid asymmetry in erythrocyte membranes. Biochem J 350 (Pt 2):531–535 85. Raggers RJ, van Helvoort A, Evers R, van Meer G (1999) The human multidrug resistance protein MRP1 translocates sphingolipid analogs across the plasma membrane. J Cell Sci 112(Pt 3):415–422 86. Mannechez A, Collet B, Payen L et  al (2001) Differentiation of the Pgp and MRP1 multidrug resistance systems by mobile lipid 1H-NMR spectroscopy and phosphatidylserine externalization. Anticancer Res 21: 3915–3919 87. Kamp D, Haest CW (1998) Evidence for a role of the multidrug resistance protein (MRP) in the outward translocation of NBDphospholipids in the erythrocyte membrane. Biochim Biophys Acta 1372:91–101 88. Sohnius A, Kamp D, Haest CW (2003) ATP and GSH dependence of MRP1-mediated outward translocation of phospholipid analogs in the human erythrocyte membrane. Mol Membr Biol 20:299–305 89. Huang Z, Chang X, Riordan JR, Huang Y (2004) Fluorescent modified phosphatidylcholine floppase activity of reconstituted multidrug resistance-associated protein MRP1. Biochim Biophys Acta 1660:155–163 90. Lorico A, Rappa G, Finch RA et  al (1997) Disruption of the murine MRP (multidrug resistance protein) gene leads to increased sensitivity to etoposide (VP-16) and increased levels of glutathione. Cancer Res 57:5238–5242 91. Rappa G, Finch RA, Sartorelli AC, Lorico A (1999) New insights into the biology and pharmacology of the multidrug resistance protein (MRP) from gene knockout models. Biochem Pharmacol 58:557–562 92. Rao VV, Dahlheimer JL, Bardgett ME et  al (1999) Choroid plexus epithelial expression of MDR1 P glycoprotein and multidrug resistance-associated protein contribute to the blood-cerebrospinal-fluid drug-permeability barrier. Proc Natl Acad Sci USA 96: 3900–3905 93. Muller M, de Vries EG, Jansen PL (1996) Role of multidrug resistance protein (MRP) in glutathione S-conjugate transport in mammalian cells. J Hepatol 24:100–108 94. Stride BD, Grant CE, Loe DW et  al (1997) Pharmacological characterization of the murine and human orthologs of multidrug-resistance protein in transfected human embryonic kidney cells. Mol Pharmacol 52:344–353

Molecular Mechanism of ATP-Dependent Solute Transport by Multidrug Resistance 95. Gao M, Loe DW, Grant CE, Cole SPC, Deeley RG (1996) Reconstitution of ATPdependent leukotriene C4 transport by Co-expression of both half-molecules of human multidrug resistance protein in insect cells. J Biol Chem 271:27782–27787 96. Keppler D, Leier I, Jedlitschky G, Mayer R, Buchler M (1996) The function of the multidrug resistance proteins (MRP and cMRP) in drug conjugate transport and hepatobiliary excretion. Adv Enzyme Regul 36:17–29 97. Lautier D, Canitrot Y, Deeley RG, Cole SP (1996) Multidrug resistance mediated by the multidrug resistance protein (MRP) gene. Biochem Pharmacol 52:967–977 98. Ren XQ, Furukawa T, Aoki S et  al (2001) Glutathione-dependent binding of a photoaffinity analog of agosterol A to the C-terminal half of human multidrug resistance protein. J Biol Chem 276:23197–23206 99. Payen L, Gao M, Westlake C et  al (2005) Functional interactions between nucleotide binding domains and leukotriene C4 binding sites of multidrug resistance protein 1 (ABCC1). Mol Pharmacol 67:1944–1953 100. Payen LF, Gao M, Westlake CJ, Cole SP, Deeley RG (2003) Role of carboxylate residues adjacent to the conserved core Walker B motifs in the catalytic cycle of multidrug resistance protein 1 (ABCC1). J Biol Chem 278:38537–38547 101. Qian YM, Qiu W, Gao M et  al (2001) Characterization of binding of leukotriene C4 by human multidrug resistance protein 1: evidence of differential interactions with NH2- and COOH-proximal halves of the protein. J Biol Chem 276:38636–38644 102. Karwatsky J, Leimanis M, Cai J, Gros P, Georges E (2005) The leucotriene C4 binding sites in multidrug resistance protein 1 (ABCC1) include the first membrane multiple spanning domain. Biochemistry 44:340–351 103. Qian YM, Grant CE, Westlake CJ et  al (2002) Photolabeling of human and murine multidrug resistance protein 1 with the high affinity inhibitor (125I)LY475776 and azidophenacyl-(35S)glutathione. J Biol Chem 277:35225–35231 104. Mao Q, Qiu W, Weigl KE et  al (2002) GSH-dependent photolabeling of multidrug resistance protein MRP1 (ABCC1) by (125I) LY475776. Evidence of a major binding site in the COOH-proximal membrane spanning domain. J Biol Chem 277:28690–28699 105. Daoud R, Desneves J, Deady LW et al (2000) The multidrug resistance protein is photoaffinity labeled by a quinoline-based drug at multiple sites. Biochemistry 39:6094–6102

245

106. Daoud R, Kast C, Gros P, Georges E (2000) Rhodamine 123 binds to multiple sites in the multidrug resistance protein (MRP1). Biochemistry 39:15344–15352 107. Daoud R, Julien M, Gros P, Georges E (2001) Major photoaffinity drug binding sites in multidrug resistance protein 1 (MRP1) are within transmembrane domains 10–11 and 16–17. J Biol Chem 276: 12324–12330 108. Greenberger LM (1993) Major photoaffinity drug labeling sites for iodoaryl azidoprazosin in P-glycoprotein are within, or immediately C-terminal to, transmembrane domains 6 and 12. J Biol Chem 268:11417–11425 109. Loo TW, Clarke DM (1997) Identification of residues in the drug-binding site of human P-glycoprotein using a thiol-reactive substrate. J Biol Chem 272:31945–31948 110. Dey S, Ramachandra M, Pastan I, Gottesman MM, Ambudkar SV (1997) Evidence for two nonidentical drug-interaction sites in the human P-glycoprotein. Proc Natl Acad Sci USA 94:10594–10599 111. Shapiro AB, Fox K, Lam P, Ling V (1999) Stimulation of P-glycoprotein-mediated drug transport by prazosin and progesterone. Evidence for a third drug-binding site. Eur J Biochem 259:841–850 112. Leslie EM, Letourneau IJ, Deeley RG, Cole SP (2003) Functional and structural consequences of cysteine substitutions in the NH2 proximal region of the human multidrug resistance protein 1 (MRP1/ABCC1). Biochemistry 42:5214–5224 113. Koike K, Conseil G, Leslie EM, Deeley RG, Cole SP (2004) Identification of proline residues in the core cytoplasmic and transmembrane regions of multidrug resistance protein 1 (MRP1/ABCC1) important for transport function, substrate specificity, and nucleotide interactions. J Biol Chem 279:12325–12336 114. Haimeur A, Deeley RG, Cole SP (2002) Charged amino acids in the sixth transmembrane helix of multidrug resistance protein 1 (MRP1/ABCC1) are critical determinants of transport activity. J Biol Chem 277: 41326–41333 115. Koike K, Oleschuk CJ, Haimeur A et  al (2002) Multiple membrane-associated tryptophan residues contribute to the transport activity and substrate specificity of the human multidrug resistance protein, MRP1. J Biol Chem 277:49495–49503 116. Zhang DW, Nunoya K, Vasa M et al (2006) Mutational analysis of polar amino acid residues within predicted transmembrane helices

246

Chang

10 and 16 of multidrug resistance protein 1 (ABCC1): effect on substrate specificity. Drug Metab Dispos 34:539–546 117. Zhang DW, Nunoya K, Vasa M et al (2004) Transmembrane helix 11 of multidrug resistance protein 1 (MRP1/ABCC1): identification of polar amino acids important for substrate specificity and binding of ATP at nucleotide binding domain 1. Biochemistry 43:9413–9425 118. Campbell JD, Koike K, Moreau C et  al (2004) Molecular modeling correctly predicts the functional importance of Phe594 in transmembrane helix 11 of the multidrug resistance protein, MRP1 (ABCC1). J Biol Chem 279:463–468 119. Zhang DW, Cole SP, Deeley RG (2001) Identification of an amino acid residue in multidrug resistance protein 1 critical for conferring resistance to anthracyclines. J Biol Chem 276:13231–13239 120. Zhang DW, Gu HM, Situ D et  al (2003) Functional importance of polar and charged amino acid residues in transmembrane helix 14 of multidrug resistance protein 1 (MRP1/ ABCC1): identification of an aspartate residue critical for conversion from a high to low affinity substrate binding state. J Biol Chem 278:46052–46063 121. Situ D, Haimeur A, Conseil G et al (2004) Mutational analysis of ionizable residues proximal to the cytoplasmic interface of membrane spanning domain 3 of the multidrug resistance protein, MRP1 (ABCC1): glutamate 1204 is important for both the expression and catalytic activity of the transporter. J Biol Chem 279:38871–38880 122. Ito K, Olsen SL, Qiu W, Deeley RG, Cole SP (2001) Mutation of a single conserved tryptophan in multidrug resistance protein 1 (MRP1/ABCC1) results in loss of drug resistance and selective loss of organic anion transport. J Biol Chem 276:15616–15624 123. Zhang DW, Cole SP, Deeley RG (2001) Identification of a nonconserved amino acid residue in multidrug resistance protein 1 important for determining substrate specificity: evidence for functional interaction between transmembrane helices 14 and 17. J Biol Chem 276:34966–34974 124. Zhang DW, Cole SP, Deeley RG (2002) Determinants of the substrate specificity of multidrug resistance protein 1: role of amino acid residues with hydrogen bonding potential in predicted transmembrane helix 17. J Biol Chem 277:20934–20941 125. Stride BD, Cole SP, Deeley RG (1999) Localization of a substrate specificity domain

in the multidrug resistance protein. J Biol Chem 274:22877–22883 126. Karwatsky J, Daoud R, Cai J, Gros P, Georges E (2003) Binding of a photoaffinity analogue of glutathione to MRP1 (ABCC1) within two cytoplasmic regions (L0 and L1) as well as transmembrane domains 10–11 and 16–17. Biochemistry 42:3286–3294 127. Karwatsky JM, Georges E (2004) Drug binding domains of MRP1 (ABCC1) as revealed by photoaffinity labeling. Curr Med Chem Anticancer Agents 4:19–30 128. Bakos E, Evers R, Calenda G et  al (2000) Characterization of the amino-terminal regions in the human multidrug resistance protein (MRP1). J Cell Sci 113(Pt 24): 4451–4461 129. Leslie EM, Ito K, Upadhyaya P et  al (2001) Transport of the beta -O-glucuronide conjugate of the tobacco-specific carcinogen 4-(methylnitrosamino)-1-(3-pyridyl)-1butanol (NNAL) by the multidrug resistance protein 1 (MRP1). Requirement for glutathione or a non-sulfur-containing analog. J Biol Chem 276:27846–27854 130. Qian YM, Song WC, Cui H, Cole SP, Deeley RG (2001) Glutathione stimulates sulfated estrogen transport by multidrug resistance protein 1. J Biol Chem 276:6404–6411 131. Buyse F, Hou YX, Vigano C et  al (2006) Replacement of the positively charged Walker A lysine residue with a hydrophobic leucine residue and conformational alterations caused by this mutation in MRP1 impair ATP binding and hydrolysis. Biochem J 397:121–130 132. Manciu L, Chang XB, Riordan JR, Ruysschaert JM (2000) Multidrug resistance protein MRP1 reconstituted into lipid vesicles: secondary structure and nucleotide-induced tertiary structure changes. Biochemistry 39: 13026–13033 133. Manciu L, Chang XB, Riordan JR, Buyse F, Ruysschaert JM (2001) Nucleotide-induced conformational changes in the human multidrug resistance protein MRP1 are related to the capacity of chemotherapeutic drugs to accumulate or not in resistant cells. FEBS Lett 493:31–35 134. Taguchi Y, Yoshida A, Takada Y, Komano T, Ueda K (1997) Anti-cancer drugs and glutathione stimulate vanadate-induced trapping of nucleotide in multidrug resistance-associated protein (MRP). FEBS Lett 401:11–14 135. Nagata K, Nishitani M, Matsuo M et  al (2000) Nonequivalent nucleotide trapping in the two nucleotide binding folds of the human multidrug resistance protein MRP1. J Biol Chem 275:17626–17630

Molecular Mechanism of ATP-Dependent Solute Transport by Multidrug Resistance 136. Hou Y, Cui L, Riordan JR, Chang XB (2000) Allosteric interactions between the two non-equivalent nucleotide binding domains of multidrug resistance protein MRP1. J Biol Chem 275:20280–20287 137. Leslie EM, Mao Q, Oleschuk CJ, Deeley RG, Cole SP (2001) Modulation of multidrug resistance protein 1 (MRP1/ABCC1) transport and atpase activities by interaction with dietary flavonoids. Mol Pharmacol 59: 1171–1180 138. Mao Q, Leslie EM, Deeley RG, Cole SP (1999) ATPase activity of purified and reconstituted multidrug resistance protein MRP1 from drug-selected H69AR cells. Biochim Biophys Acta 1461:69–82 139. Chang XB, Hou YX, Riordan JR (1997) ATPase activity of purified multidrug resistance-associated protein (published erratum appears in J Biol Chem 1998 Mar 27;273(13):7782). J Biol Chem 272: 30962–30968 140. Manciu L, Chang XB, Buyse F et al (2003) Intermediate structural states involved in MRP1-mediated drug transport. Role of glutathione. J Biol Chem 278:3347–3356 141. Smith PC, Karpowich N, Millen L et al (2002) ATP binding to the motor domain from an ABC transporter drives formation of a nucleotide sandwich dimer. Mol Cell 10:139–149 142. Chen J, Lu G, Lin J, Davidson AL, Quiocho FA (2003) A tweezers-like motion of the ATP-binding cassette dimer in an ABC transport cycle. Mol Cell 12:651–661 143. Moody JE, Millen L, Binns D, Hunt JF, Thomas PJ (2002) Cooperative, ATPdependent association of the nucleotide binding cassettes during the catalytic cycle of ATP-binding cassette transporters. J Biol Chem 277:21111–21114 144. Verdon G, Albers SV, Dijkstra BW, Driessen AJ, Thunnissen AM (2003) Crystal structures of the ATPase subunit of the glucose ABC transporter from Sulfolobus solfataricus: nucleotide-free and nucleotide-bound conformations. J Mol Biol 330:343–358 145. Locher KP, Lee AT, Rees DC (2002) The E. coli BtuCD structure: a framework for ABC transporter architecture and mechanism. Science 296:1091–1098 146. Rosenberg MF, Mao Q, Holzenburg A et  al (2001) The structure of the multidrug resistance protein 1 (MRP1/ABCC1). crystallization and single-particle analysis. J Biol Chem 276:16076–16082 147. Cool RH, Veenstra MK, van Klompenburg W et al (2002) S-decyl-glutathione nonspecifically stimulates the ATPase activity of the

247

nucleotide-binding domains of the human multidrug resistance-associated protein, MRP1 (ABCC1). Eur J Biochem 269:3470–3478 148. Kern A, Felfoldi F, Sarkadi B, Varadi A (2000) Expression and characterization of the N- and C-terminal ATP-binding domains of MRP1. Biochem Biophys Res Commun 273:913–919 149. Ramaen O, Sizun C, Pamlard O, Jacquet E, Lallemand JY (2005) Attempts to characterize the NBD heterodimer of MRP1: transient complex formation involves Gly771 of the ABC signature sequence but does not enhance the intrinsic ATPase activity. Biochem J 391: 481–490 150. Ramaen O, Leulliot N, Sizun C et al (2006) Structure of the human multidrug resistance protein 1 nucleotide binding domain 1 bound to Mg2+/ATP reveals a non-productive catalytic site. J Mol Biol 359:940–949 151. Szentpetery Z, Sarkadi B, Bakos E, Varadi A (2004) Functional studies on the MRP1 multidrug transporter: characterization of ABC-signature mutant variants. Anticancer Res 24:449–455 152. Szentpetery Z, Kern A, Liliom K et al (2004) The role of the conserved glycines of ATPbinding cassette signature motifs of MRP1 in the communication between the substratebinding site and the catalytic centers. J Biol Chem 279:41670–41678 153. Ren XQ, Furukawa T, Haraguchi M et  al (2004) Function of the ABC signature sequences in the human multidrug resistance protein 1. Mol Pharmacol 65:1536–1542 154. Pascaud C, Garrigos M, Orlowski S (1998) Multidrug resistance transporter P-glycoprotein has distinct but interacting binding sites for cytotoxic drugs and reversing agents. Biochem J 333(Pt 2):351–358 155. Martin C, Berridge G, Higgins CF et  al (2000) Communication between multiple drug binding sites on P-glycoprotein. Mol Pharmacol 58:624–632 156. Yang R, McBride A, Hou YX, Goldberg A, Chang XB (2005) Nucleotide dissociation from NBD1 promotes solute transport by MRP1. Biochim Biophys Acta 1668:248–261 157. Yang R, Scavetta R, Chang XB (2008) Interaction between the bound Mg.ATP and the Walker A serine residue in NBD2 of multidrug resistance-associated protein MRP1 plays a crucial role for the ATP-dependent leukotriene C4 transport. Biochemistry 47: 8456–8464 158. Hou YX, Cui L, Riordan JR, Chang XB (2002) ATP binding to the first nucleotidebinding domain of multidrug resistance

248

Chang

protein MRP1 increases binding and hydrolysis of ATP and trapping of ADP at the second domain. J Biol Chem 277:5110–5119 159. Hou YX, Riordan JR, Chang XB (2003) ATP binding, not hydrolysis, at the first nucleotidebinding domain of multidrug resistanceassociated protein MRP1 enhances ADP.Vi trapping at the second domain. J Biol Chem 278:3599–3605 160. Zhao Q, Chang XB (2004) Mutation of the aromatic amino acid interacting with adenine moiety of ATP to a polar residue alters the properties of multidrug resistance protein 1. J Biol Chem 279:48505–48512 161. Chang XB (2007) A molecular understanding of ATP-dependent solute transport by multidrug resistance-associated protein MRP1. Cancer Metastasis Rev 26:15–37 162. Senior AE (1998) Catalytic mechanism of P-glycoprotein. Acta Physiol Scand Suppl 643:213–218 163. Senior AE, al-Shawi MK, Urbatsch IL (1998) ATPase activity of Chinese hamster P-glycoprotein. Methods Enzymol 292: 514–523 164. Urbatsch IL, Sankaran B, Weber J, Senior AE (1995) P-glycoprotein is stably inhibited by vanadate-induced trapping of nucleotide at a single catalytic site. J Biol Chem 270: 19383–19390 165. Urbatsch IL, Sankaran B, Bhagat S, Senior AE (1995) Both P-glycoprotein nucleotidebinding sites are catalytically active. J Biol Chem 270:26956–26961 166. Senior AE, Bhagat S (1998) P-glycoprotein shows strong catalytic cooperativity between the two nucleotide sites. Biochemistry 37:831–836 167. Carrier I, Julien M, Gros P (2003) Analysis of catalytic carboxylate mutants E552Q and E1197Q suggests asymmetric ATP hydrolysis by the two nucleotide-binding domains of P-glycoprotein. Biochemistry 42:12875–12885 168. Urbatsch IL, Beaudet L, Carrier I, Gros P (1998) Mutations in either nucleotide-binding site of P-glycoprotein (Mdr3) prevent vanadate trapping of nucleotide at both sites. Biochemistry 37:4592–4602 169. Urbatsch IL, Julien M, Carrier I et al (2000) Mutational analysis of conserved carboxylate residues in the nucleotide binding sites of P-glycoprotein. Biochemistry 39:14138–14149 170. Azzaria M, Schurr E, Gros P (1989) Discrete mutations introduced in the predicted nucleotide-binding sites of the mdr1 gene abolish

its ability to confer multidrug resistance. Mol Cell Biol 9:5289–5297 171. Senior AE, al-Shawi MK, Urbatsch IL (1995) The catalytic cycle of P-glycoprotein. FEBS Lett 377:285–289 172. Sauna ZE, Ambudkar SV (2001) Characteri­ zation of the catalytic cycle of ATP hydrolysis by human P- glycoprotein. The two ATP hydrolysis events in a single catalytic cycle are kinetically similar but affect different functional outcomes. J Biol Chem 276: 11653–11661 173. Sauna ZE, Ambudkar SV (2000) Evidence for a requirement for ATP hydrolysis at two distinct steps during a single turnover of the catalytic cycle of human P-glycoprotein. Proc Natl Acad Sci USA 97:2515–2520 174. Yang R, Scavetta R, Chang XB (2008) The hydroxyl group of S685 in Walker A motif and the carboxyl group of D792 in Walker B motif of NBD1 play a crucial role for multidrug resistance protein folding and function. Biochim Biophys Acta 1778:454–465 175. Cui L, Hou YX, Riordan JR, Chang XB (2001) Mutations of the Walker B motif in the first nucleotide binding domain of multidrug resistance protein MRP1 prevent conformational maturation. Arch Biochem Biophys 392:153–161 176. Yang R, Cui L, Hou Y-X, Riordan JR, Chang XB (2003) ATP binding to the first nucleotide binding domain of multidrug resistance-associated protein plays a regulatory role at low nucleotide concentration, whereas ATP hydrolysis at the second plays a dominant role in ATP-dependent leukotriene C4 transport. J Biol Chem 278: 30764–30771 177. Hung LW, Wang IX, Nikaido K et  al (1998) Crystal structure of the ATP-binding subunit of an ABC transporter. Nature 396: 703–707 178. Lu G, Westbrooks JM, Davidson AL, Chen J (2005) ATP hydrolysis is required to reset the ATP-binding cassette dimer into the resting-state conformation. Proc Natl Acad Sci USA 102:17969–17974 179. Hopfner KP, Karcher A, Shin DS et al (2000) Structural biology of Rad50 ATPase: ATP-driven conformational control in DNA double-strand break repair and the ABCATPase superfamily. Cell 101:789–800 180. Diederichs K, Diez J, Greller G et al (2000) Crystal structure of MalK, the ATPase subunit of the trehalose/maltose ABC transporter of the archaeon Thermococcus litoralis. EMBO J 19:5951–5961

Molecular Mechanism of ATP-Dependent Solute Transport by Multidrug Resistance 181. Zaitseva J, Jenewein S, Jumpertz T, Holland IB, Schmitt L (2005) H662 is the linchpin of ATP hydrolysis in the nucleotide-binding domain of the ABC transporter HlyB. EMBO J 24:1901–1910 182. Ernst R, Kueppers P, Klein CM et al (2008) A mutation of the H-loop selectively affects rhodamine transport by the yeast multidrug ABC transporter Pdr5. Proc Natl Acad Sci USA 105:5069–5074 183. Shyamala V, Baichwal V, Beall E, Ames GF (1991) Structure-function analysis of the histidine permease and comparison with cystic fibrosis mutations. J Biol Chem 266: 18714–18719 184. Davidson AL, Sharma S (1997) Mutation of a single MalK subunit severely impairs maltose transport activity in Escherichia coli. J Bacteriol 179:5458–5464 185. Nikaido K, Ames GF (1999) One intact ATP-binding subunit is sufficient to support ATP hydrolysis and translocation in an ABC transporter, the histidine permease. J Biol Chem 274:26727–26735 186. Yang R, Chang XB (2007) Hydrogen-bond formation of the residue in H-loop of the nucleotide binding domain 2 with the ATP in this site and/or other residues of multidrug resistance protein MRP1 plays a crucial role during ATP-dependent solute transport. Biochim Biophys Acta 1768:324–335 187. Dawson RJ, Locher KP (2006) Structure of a bacterial multidrug ABC transporter. Nature 443:180–185 188. Pinkett HW, Lee AT, Lum P, Locher KP, Rees DC (2007) An inward-facing conformation of a putative metal-chelate-type ABC transporter. Science 315:373–377 189. Oldham ML, Khare D, Quiocho FA, Davidson AL, Chen J (2007) Crystal structure of a catalytic intermediate of the maltose transporter. Nature 450:515–521

249

190. Ward A, Reyes CL, Yu J, Roth CB, Chang G (2007) Flexibility in the ABC transporter MsbA: alternating access with a twist. Proc Natl Acad Sci USA 104:19005–19010 191. Trompier D, Chang XB, Barattin R et  al (2004) Verapamil and its derivative trigger apoptosis through glutathione extrusion by multidrug resistance protein MRP1. Cancer Res 64:4950–4956 192. Perrotton T, Trompier D, Chang XB, Di Pietro A, Baubichon-Cortay H (2007) (R)and (S)-verapamil differentially modulate the multidrug-resistant protein MRP1. J Biol Chem 282:31542–31548 193. Salerno M, Loechariyakul P, Saengkhae C, Garnier-Suillerot A (2004) Relation between the ability of some compounds to modulate the MRP1-mediated efflux of glutathione and to inhibit the MRPl-mediated efflux of daunorubicin. Biochem Pharmacol 68: 2159–2165 194. Cole SP, Downes HF, Mirski SE, Clements DJ (1990) Alterations in glutathione and glutathione-related enzymes in a multidrugresistant small cell lung cancer cell line. Mol Pharmacol 37:192–197 195. Campling BG, Baer K, Baker HM, Lam YM, Cole SP (1993) Do glutathione and related enzymes play a role in drug resistance in small cell lung cancer cell lines? Br J Cancer 68:327–335 196. Rappa G, Gamcsik MP, Mitina RL et al (2003) Retroviral transfer of MRP1 and gammaglutamyl cysteine synthetase modulates cell sensitivity to L-buthionine-S, R-sulphoximine (BSO): new rationale for the use of BSO in cancer therapy. Eur J Cancer 39:120–128 197. Zaman GJ, Lankelma J, van Tellingen O et al (1995) Role of glutathione in the export of compounds from cells by the multidrugresistance-associated protein. Proc Natl Acad Sci USA 92:7690–7694

sdfsdf

Chapter 12 Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes Douglas D. Ross and Takeo Nakanishi Abstract Breast cancer resistance protein (BCRP/ABCG2) was discovered in multidrug resistant breast cancer cells having an ATP-dependent transport-based resistance phenotype. This ABC transporter functions (at least in part) as a xenobiotic protective mechanism for the organism: in the gut and biliary tract, it prevents absorption and enhances elimination of potentially toxic substances. As a placental barrier, it protects the fetus; similarly, it serves as a component of blood-brain and blood-testis barrier; BCRP is expressed in stem cells and may protect them from potentially harmful agents. Therefore, BCRP could influence cancer outcomes by (a) endogenous BCRP affecting the absorption, distribution, metabolism, and elimination of anticancer drugs; (b) BCRP expression in cancer cells may directly cause resistance by active efflux of anticancer drugs; (c) BCRP expression in cancer cells could be a manifestation of the activity of metabolic and signaling pathways that impart multiple mechanisms of drug resistance, self-renewal (stemness), and invasiveness (aggressiveness) – i.e. impart a poor prognosis – to cancers. This chapter presents a synopsis of translational clinical studies relating BCRP expression in leukemias, lymphomas, and a variety of solid tumors with clinical outcome. Data are emerging that expression of BCRP, like P-glycoprotein/ABCB1, is associated with adverse outcomes in a variety of human cancers. Whether this adverse prognostic effect results from resistance imparted to the cancer cells as the direct result of BCRP efflux of anticancer drugs, or whether BCRP expression (and also Pgp expression – coexpression of these transporters is common among poor risk cancers) serves as indicators of the activity of signaling pathways that enhance cancer cellular proliferation, metastases, genomic instability, enhance drug resistance, and oppose programmed cell death mechanisms is yet unknown. Key words: BCRP, Expression, Polymorphism, Cancer, Treatment, Outcome

1. Introduction The development of global resistance to therapy in clinical cancers following specific cancer treatment is termed multidrug resistance. For the past 30 years, multidrug resistance (MDR) engendered extensive basic research to uncover underlying J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_12, © Humana Press, a part of Springer Science + Business Media, LLC 2010

251

252

Ross and Nakanishi

biological causes of this phenomenon. Among the enduring mechanisms identified is cellular efflux of anticancer drugs as a result of membrane-based transport proteins. The drug efflux transporter initially identified as a cause of multidrug resistance was P-glycoprotein (Pgp), the product of the MDR1 gene. Pgp is a member of the B subfamily of the seven-family (A–G) ATP-binding cassette (ABC) superfamily of transporter proteins, and is formally designated as ABCB1. Subsequently, a second ABC transporter, the multidrug resistance-associated protein 1 (MRP1/ABCC1), was found to confer resistance to natural product anticancer drugs. MRP1 belongs to a family of nine transporters (all C subfamily members) of which the first eight have demonstrated ability to cause resistance to antineoplastic drugs. By the early 1990s, it became apparent that ATP-dependent transport systems other than Pgp or MRP1 could be associated with multidrug resistance. Our own studies of blast cells from patients with acute leukemia revealed that expression of Pgp or MRP1 could not account for all the transport-based drug resistance observed when measured by functional efflux assays (1). This led our laboratory to investigate the ATP-dependent drug efflux phenotype manifested by MCF-7/AdrVp cells, a multidrug resistant cell line that does not overexpress Pgp or MRP1 that was produced by Chen et al., by selection of MCF-7 human breast cancer cells with doxorubicin and verapamil (2). Using differential display hybridization, we isolated a novel ABC transporter from MCF-7/AdrVp cells, and proved by transfection of MCF-7 cells that the enforced overexpression of the transporter recapitulated the drug efflux and multidrug resistance phenotype of MCF-7/ AdrVp cells in the transfected cells (3). Because of its isolation from multidrug resistant human breast cancer cells, we termed the new transporter the “Breast Cancer Resistance Protein,” or BCRP (3). At about the same time, two other laboratories reported isolating a similar cDNA, which they termed ABCP (4) or MXR (5). Review articles regarding BCRP in normal physiology and neoplastic disease are available (6–55). BCRP consists of 655 amino acid residues, and is a member of the G subfamily of the ABC transporter superfamily, with the formal designation ABCG2. Like all G subfamily members, BCRP is a half-transporter; it possesses only six-transmembrane domains and one ATP-binding domain. For functionality, ABC transporters require at least two ATP-binding domains; hence to be active, BCRP must dimerize or multimerize with another half-transporter. Current evidence, including our own, suggests that BCRP is a homodimer or homomultimer (20, 56). BCRP localizes to the plasma membrane of cells and results in low intracellular accumulation of substrate drugs by effluxing these drugs out of the cell.

Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes

2. Role of BCRP in Normal Physiology

253

Understanding of how BCRP functions physiologically may enable us to appreciate how expression of this transporter is associated with drug resistant, poor prognosis cancers. In polarized cells such as placental syncytiotrophoblast, hepatocytes (bile canaliculi), and intestinal mucosal cells, BCRP is expressed on the apical membranes where it serves to efflux substances out of the fetal circulation, or into the lumen of the bile duct or gut, respectively (20). Hence BCRP has a demonstrated effect on drug clearance, which is of considerable interest regarding potential interactions of BCRP substrate drugs or BCRP inhibitors. The rat homologue of BCRP is expressed on the luminal surface of rat brain capillary endothelial cells, where it may function as a component of the blood brain barrier (20). From the above, it follows that a major function of BCRP is to protect cells from potentially toxic xenobiotics, and to assist the clearance of xenobiotics from the organism. Hence, BCRP is considered to play a prominent role in drug absorption, distribution, metabolism, and elimination (ADME) in normal individuals. Naturally occurring toxic xenobiotics for which BCRP may play a protective role include dietary carcinogens such as PhIP (2-amino-1-methyl-6-phenylimidazo(4,5-b)pyridine), where a higher plasma AUC was observed in Bcrp1-/- mice following oral or IV administration of PhIP (57). Xenobiotics currently known to be substrates for BCRP include many approved cancer chemotherapeutic agents such as mitoxantrone, camptothecin-derived topoisomerase inhibitors (e.g., topotecan and SN-38), methotrexate, methotrexate polyglutamates, and anthracyclines (daunorubicin and doxorubicin), as well as the HER-tyrosine kinase inhibitors CI-1033 and gefitinib (58) (which is also a BCRP inhibitor (59, 60)), and the cyclin-dependent kinase inhibitor flavopiridol (20). The ability of BCRP to efflux anthracyclines is greatly enhanced by the presence of a mutation at codon 482 (R482T or R482G), which has been observed in cells that overexpress BCRP following drug selection (61, 62). Such mutations lose the ability to efflux methotrexate (62). Recently, the BCR-ABL tyrosine kinase inhibitor imatinib mesylate (Gleevec, STI-571) was found to be both a substrate and inhibitor of BCRP (63, 64). Imatinib is currently the most effective drug known for the treatment of the chronic phase of CML. A number of natural or physiological substrates for BCRP have been identified. BCRP transports folic acid, and may play a role in cellular folate homeostasis. In mice, BCRP transports pheophorbide a, a breakdown product of chlorophyll found in

254

Ross and Nakanishi

mouse chow; BCRP knockout mice suffered photosensitivity as the result of tissue accumulation of pheophorbide a (65). This work led to the discovery that BCRP transports protoporphyrin IX, which is structurally related to pheophorbide a. It is known that BCRP binds to hemin, and that BCRP may transport products of the heme biosynthetic or catabolic pathway (66). Conjugates of estrogens (estrone 3 sulfate and estradiol glucuronide) are also substrates (67, 68), but free estrogens are not. Potent and specific inhibitors of BCRP are currently available, with some of the inhibitors in clinical trials or available for clinical use (20). Fumitremorgin C (FTC), a mycotoxin isolated from Aspergillis fumagatis, is highly specific for BCRP and is active at micromolar concentrations. Ko143, a derivative of FTC, is considerably more potent than the parent compound (69). Currently, these specific BCRP inhibitors are useful for functional assays of BCRP transporter activity. GF120918 inhibits both Pgp and BCRP, and is currently undergoing clinical trials as an oral agent. One of these trials demonstrated that coadministration of GF120918 increased the oral bioavailability of topotecan, a BCRP-substrate drug, from 30% to 90% (70). Other BCRP inhibitors include gefitinib (Iressa), flavopiridol, reserpine, certain taxane derivatives (orataxel and tRA96023), estrogens and antiestrogens, imatinib, and certain HIV protease inhibitors (20). Benzimadazoles, commonly used as proton pump inhibitors, are also potent inhibitors of BCRP transport (71). BCRP is also expressed in pluripotent human and murine tissue stem cells and in hematopoietic stem cells. A recently developed and widely used flow cytometric method to identify self-renewing, primitive stem cells is based on the ability of the stem cells to exclude the dye Hoechst 33342 (72, 73). Cell sorting experiments reveal that the cell populations with selfrenewing capacity have low accumulation of Hoechst dye; when Hoechst dye accumulation is plotted as cells with red versus blue fluorescence, the stem cell population appears as a population with low, but slightly greater blue fluorescence than red fluorescence. Because of this, the stem cells with low Hoechst dye content have been called the “side population” (SP), and compose approximately 0.03% of the cells in normal bone marrow (74). Earlier, it was believed that Pgp was responsible for the low accumulation of the dye; however, more recent work revealed that BCRP is strongly expressed in SP cells and may be a major contributor to the SP phenotype in bone marrow cells, in addition to Pgp (73, 75). Indeed, Hoechst 33342 dye is a BCRP substrate. In terms of the function played by BCRP in SP cells, it is clear that BCRP expression is not necessary for stem cell development since Bcrp1-/- mice lack the SP but have normal hematopoiesis. It is presently believed that BCRP functions as

Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes

255

one of the means that protect stem cells from xenobiotics, since Bcrp1-/- mice are more susceptible to the hematopoietic toxicity of mitoxantrone than wild type mice (76).

3. BCRP Impact on Cancer Treatment Outcomes

As a naturally-occurring xenobiotic transporter, BCRP could influence cancer outcomes in a number of ways. One way could be by endogenous BCRP affecting ADME of anticancer drugs, particularly in individuals with nonsynonymous polymorphisms in the coding region of the BCRP gene that result in alterations of BCRP function, or by ingestion of drugs that are BCRP inhibitors. A second way that BCRP could affect cancer treatment outcomes could be by functional expression of BCRP in cancer cells themselves, with the resultant active drug efflux preventing anticancer drugs from reaching their cellular target(s). Thirdly, BCRP expression could be a manifestation of the activity of metabolic and signaling pathways that impart multiple mechanisms of drug resistance, self-renewal (stemness), and invasiveness (aggressiveness) – i.e. impart a poor prognosis – to cancers. We will now consider each of these ways in turn, starting with the effects of BCRP polymorphisms and BCRP expression in human cancers with regard to drug delivery and treatment outcomes.

3.1. Germline BCRP Polymorphisms and Cancer Treatment Outcomes

The pharmacogenetics of the ABC transporters and clinical implications thereof are discussed in Chapter 6. Because a major physiological function of BCRP is xenobiotic defense, functionaltering polymorphisms of BCRP could have an impact on the ADME of anticancer drugs and possibly of dietary carcinogens, and hence these are discussed in this chapter. A number of single nucleotide polymorphisms (SNPs) have been observed in the BCRP gene (77), and of those nonsynonymous SNPs observed in the coding region, the most common and most extensively studied are the G34A (exon 2) and C421A (exon 5) alleles, which cause alterations of amino acids 12 (V12M) and 141 (Q141K), respectively (78–81).

3.1.1. Consequences of BCRP Polymorphisms on Gene Function In Vitro

Imai et al. found that cells transfected with the C376T or C421A allele had markedly decreased expression of BCRP protein and low levels of drug resistance when compared with wild-type BCRP transfected cells (80). Subsequent work by Zamber et al. found four nonsynonymous coding region SNPs of BCRP in DNA obtained from human livers and intestines: G34A, C421A, A616C, and A1758T, with G34A and C421A being the most

256

Ross and Nakanishi

frequent (81). Great heterogeneity in the intestinal expression of BCRP was found, which could not be explained by the presence of the C421A allele. In work done by Mizurai et al., the enforced expression of the G34A or C421A allele in porcine renal LLC-PK1 cells resulted in reduced transporter function, with a reduction in BCRP apical membrane localization for the G34A variant protein and reduced transporter function for the Q141K allele (82). A functional assessment of BCRP polymorphisms and BCRP expression was done by Kobayashi et al. in 100 human placentas (83). The BCRP protein level in A/A homozygotes of the C421A allele was reduced compared with placentas with the C/C wild type configuration. Kondo et al. transfected seven BCRP coding region SNPs into LLC-PK1 cells, and found lower BCRP ­protein expression for the Q141K and S441N alleles (84). Not all BCRP SNPs are associated with diminished transporter activity. Vethanayagam et al. found the I206L allele to have high transporter activity but low protein expression when transfected into human embryonic kidney (HEK) cells, whereas the N590L and D620N had higher expression but lower activity (85). The frequencies of these alleles are generally low. Lee et  al. found four nonsynonymous coding region SNPs among 92 Korean subjects (V12M, Q141K, P269S, Q126Stop), and four SNPs in the BCRP promoter region, one of which was in the HIF-1 response element (C-19031T) (86). Transporter activity was evaluated only for the P296S variant protein, and was found to be diminished. None of the promoter SNPs appeared to affect BCRP promoter activity, including assays done under hypoxic mimetic conditions. Tamura et al. used multicolor fluorescence in situ hybridization to assure uniform mRNA expression of cDNAs of seven BCRP SNPs (V12M, Q141K, F208S, S248P, F431L, S441N and F489L) transduced into Flp-In-293 cells (87, 88). Protein expression from the F208S and S441N variants was found to be low; the V12M and Q141K alleles had IC50s for SN-38 that were approximately half that of the wild-type; all the other alleles examined had significantly lower IC50 values for SN-38, mitoxantrone, doxorubicin, daunorubicin and etoposide when compared with wild type alleles (88). The effects of promoter and noncoding region SNPs on BCRP expression in liver, intestines and lymphoblasts was recently investigated by Poonkuzhali, et  al. (89). Forty one SNPs were found in the promoter region and 49 in the introns; 43 SNPs were novel. Promoter and intron 1 alleles were found that altered BCRP mRNA expression; some (e.g., C–15622T in the promoter and G1143A in intron 1) caused low mRNA expression, and others (e.g., C–15994T in the promoter and C16702T in intron 1) were associated with higher mRNA expression. Tissues expressing the C–15994T genotype had higher clearance of imatinib.

Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes

257

In all, eight functionally significant SNPs were found – 4 in the promoter, and 4 in intron 1. As BCRP is known to utilize multiple promoters (90), these investigators found that transcription from the promoter that generated the exon 1b transcript was lower. 3.1.2. Consequences of BCRP/ABCG2 Polymorphisms on Drug ADME

That BCRP affected the bioavailability of topotecan was demonstrated both in mdr1a-/- mice and in humans (70, 91). Given these data, it is reasonable to ask whether people with polymorphisms of the BCRP/ABCG2 gene can experience significant alterations in the ADME of BCRP substrate drugs. The first report associating a BCRP polymorphism with altered drug exposure found that cancer patients (N = 5) heterozygous for the C421A (Q141K) allele had plasma levels of IV diflomotecan that were almost threefold higher than those of 15 patients harboring the wild-type allele (92). In a subsequent work, the oral bioavailability of topotecan was studied in two patients heterozygous for the C421A allele when compared with ten patients with the wild type allele, and found a 1.34-fold increase in oral bioavailability in the heterozygotes (93). The heterozygous C421A allele did not appear to affect the pharmacokinetics of irinotecan or its metabolite SN-38 (94), despite the fact that BCRP transports SN-38 and its glucuronide (95, 96). Similarly, the G34A or C421A polymorphisms of BCRP were not found to be associated with alterations in the pharmacokinetics of irinotecan, SN-38 or SN-38 glucuronide, or the tumor response rate and toxicity in Korean patients with advanced nonsmall cell lung cancer treated with irinotecan (97). A comprehensive study of SNPs in the ABCB1, ABCC1, ABCC2 and ABCG2 genes by Colombo et al. found no significant alterations in the AUC of a known BCRP substrate – nelfinavir – associated with any polymorphism (98). The G34A or C421A alleles of ABCG2 were not found to alter the pharmacokinetics of doxorubicin in Asian breast cancer patients (99). Higher drug exposure could translate into greater toxicity of BCRP substrate drugs in patients harboring allelic variants. Increased diarrhea but not skin toxicity due to gefitinib was observed in patients heterozygous for the C421A allele of BCRP (Cusatis et al.) (100). Further work by these investigators found higher gefitinib accumulation at steady-state in patients heterozygous for the BCRP C421A allele when compared with patients who were wild-type at this locus (101). Erdelyi et  al. found that the C421A polymorphism of BCRP was associated with toxic encephalopathy due to treatment of childhood ALL, and that the presence of this variant allele synergized with allele variants of ABCB1 in producing this effect (102). The C421A allele of ABCG2 did not appear to be associated with infectious complications of children treated for ALL (103). Kim et  al. found the Q141K polymorphism is

258

Ross and Nakanishi

associated with therapy-related diarrhea in patients with DLBCL who are treated with R-CHOP (104). Given the potential of BCRP polymorphisms to alter drug ADME, a number of compounds have been investigated to serve as in vivo probes of BCRP activity. One candidate probe – nitrofurantoin, which is transported by BCRP – was investigated in Chinese male subjects, but no effects of BCRP SNPs on nitrofurantoin clearance were observed (105). In contrast, two recent papers (Urquhart (106), and Yamasaki (107)) suggest that sulfasalazine could be used as a probe for the effects of BCRP polymorphisms on intestinal absorption in a given subject. 3.1.3 Consequences of BCRP Polymorphisms on Cancer Susceptibility and Prognosis

Korenaga et al. investigated the C421A (Q141K) allele and found that individuals with the wild-type C/C genotype were more at risk for developing nonpapillary renal cell carcinoma (RCC) in a study of 200 Japanese patients with nonpapillary RCC and ageand sex-matched control subjects (108). These authors concluded that BCRP may be a candidate RCC susceptibility gene. Increased susceptibility and poor survival of patients with diffuse large B-cell lymphoma (DLBCL) were observed in persons with the G34A or C421A alleles (109). Gardner et al. found that the C421A allele causes higher intracellular concentrations of the carcinogen PhIP in vitro, but the allele was not associated with increased risk of prostate cancer in human populations (110). A recent study by Hahn et al. found that patients with hormone refractory prostate cancer and the C512A (C/A) allele of BCRP had significantly greater survival beyond 15 months (66%) compared with patients with the wild-type (C/C) genotype (27%) when treated with docetaxel-based combination chemotherapy (111). Similarly, Mueller et al. found worse overall survival (OS) after treatment with platinum-based regimens in a large series of patients with small cell and nonsmall cell lung cancer (SCLC and NSCLC) who carried the 421A allele of BCRP (112).

3.1.4.  BCRP R482T or G mutation: Altered Substrate Specificity

Honjo and colleagues observed a discrepancy among cell lines that overexpressed BCRP in their ability to efflux rhodamine 123. Genomic sequencing of the ABCG2 gene revealed a mutation at codon 482, with replacement of the wild-type arginine by threonine or glycine in the cell lines that were able to efflux rhodamine 123 (113). Furthermore, the R482T mutation conferred greater resistance to anthracyclines, and the R482G mutation appeared to cause less resistance to SN-38 and topotecan, compared with the wild-type arginine (113). Similar findings were made for codon 482 of murine Bcrp1, except that the mutations were to methionine or serine; these authors designated codon 482 a “mutation hot spot” (114). In contrast to the mutant forms, wild-type BCRP is able to transport folic acid (but not folinic acid), as well as methotrexate and some methotrexate polyglutamates

Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes

259

(62). Codon 482 mutations to methionine in human BCRP have also been reported in the CD4+ T-cell line MT-4 selected with doxorubicin (MT-4/DOX500) (115). To date, these mutations have only been found in cultured drug-selected cell lines; no BCRP mutations have been observed in clinical samples of blast cells obtained leukemia patients (116–118). 3.2. BCRP Expression in Human Cancers and Relationship to Cancer Treatment Outcomes 3.2.1. Leukemias AML

Literature reporting investigations of the role of BCRP in human leukemia outcomes is summarized in Table 12.1. Reviews by this author updating the status of investigations of the significance of BCRP expression in acute leukemias were published in 2003, 2004, and 2007 (20, 54, 55). We will briefly recap these reviews and then examine more recent studies. Exploratory studies of BCRP expression in AML began to appear at the turn of the twenty-first Century; initially, many studies measured BCRP mRNA expression. Using semi-quantitative RT-PCR methods, in 2000, our laboratory reported that approximately 30% of AML cases in a series of 20 patients express relatively high levels of BCRP mRNA, and raised the hypothesis that in some patients, BCRP expression may result in chemotherapy resistance (119). Similar variation was found for the expression of MDR1 mRNA among these patients, with a fair correlation (r = 0.66) for coexpression of BCRP and MDR1 mRNA. In a study of 59 newly diagnosed childhood AML patients, using realtime RT-PCR methodology, Steinbach et  al. found that BCRP mRNA expression correlated with poor response to remission induction therapy (120). Van den Heuvel Eibrink et al. studied20 paired AML blast cell samples obtained pre treatment and at time of relapse or refractory disease by. BCRP mRNA was found to be higher in the relapsed or refractory samples (121). Our laboratory found that BCRP mRNA expression measured by real-time RT-PCR correlated with in  vitro resistance to the experimental drug flavopiridol in blast cells from AML patients (116). All patients had the wild-type sequence at codon 482. A study of 40 newly diagnosed AML patients by Abbott et al. (122) attempted to relate the amount of BCRP mRNA expressed to the level of BCRP transporter activity present, using functional assays. Overall, BCRP mRNA expression was higher in the AML samples than in normal bone marrow; however, only 7% of patients expressed what they considered to be “functional” levels of BCRP mRNA. Furthermore, BCRP expression as evidenced by the functional assays and the 5D3 monoclonal antibody was confined to be in small subpopulations of cells, and it was postulated that these subpopulations may represent “primitive leukemic stem cells with intrinsic drug efflux capacity.” This study found no correlations between BCRP expression and outcomes or clinical characteristics of the patients. More recently, work by Benderra

Disease

AML, ALL

AML

AML

AML

Childhood AML

AML

Lead Author, year, citation

Ross 2000 (119)

Sargent 2001 (126)

van den Heuvel-Eibrink 2002 (121)

van der Kolk 2002 (127)

Steinbach 2002 (120)

Abbott 2002 (122)

40 at diagnosis

59 at diagnosis -9 restudied at relapse

20 paired samples, preRx and at rel or RD

20 paired samples, preRx and at rel or RD

20

1 ALL

20 AML

Number studied

mRNA (qRT-PCR)

mRNA (qRT-PCR)

Protein (FC, BXP-34 and BXP-21 MoAbs); FA

mRNA (qRT-PCR)

Protein -IHC (BXP-34 MoAb); in vitro drug sensitivity assays

mRNA, RT-PCR

Parameter, Methodology

Table 12.1 Expression and consequences of BCRP/ABCG2 in human acute leukemias

BCRP mRNA of functional significance – Was > that of normal bone marrow – Observed in only 7% of patients – Did not correlate with clinical outcome BCRP protein expressed in small subpopulations of AML cells-?primitive cells

BCRP mRNA exp – higher in those with no CR – higher at time of relapse

BCRP exp – Correlated with functional assay – Did not increase in relapse samples – Was high in cells with immature phenotype (CD34+ cells)

BCRP mRNA – Increased in rel/RD samples – Correlated with Pgp expression (r = 0.44)

BCRP exp correlated with in vitro sensitivity to daunorubicin but not to doxorubicin or mitoxantrone. No diff between chemo naïve and treated samples

Relatively high BCRP exp in 33%; fair correlation of BCRP with Pgp expression (r = 0.66)

Findings/Conclusions

260 Ross and Nakanishi

AML

AML

AML

AML

AML

AML

AML

AML

van der Pol 2003 (128)

Nakanishi 2003 (116)

Galimberti 2005 (134)

Suvannasankha 2004 (118)

Benderra 2004 (123)

Uggla 2005 (125)

Raaijmakers 2005 (138)

Benderra 2005 (124)

85 de novo patients

22 patients 7 normal donors

40, mostly de novo at diagnosis

149 de novo patients at diagnosis

31 at diagnosis

51 at diagnosis

21 at diagnosis, RD or relapse

45 at diagnosis and at time of relapse or MRD

Protein (FC, BXP-34 MoAb); FA

Protein (FC, BXP-21 MoAb); FA

mRNA (qRT-PCR)

mRNA (qRT-PCR); FA

mRNA (qRT-PCR) Protein (FC, BXP-34, BXP-21, anti-ABCG2 MoAbs); FA

mRNA (qRT-PCR)

mRNA (qRT-PCR)

Protein (FC, BXP-34 and BXP-21 MoAbs); FA; Assays for MRD

(continued)

– BCRP, Pgp, and MRP3 protein/ function correlated with CR, OS – Lower CR, DFS, OS in patients with high expression of 2 or more of these transporters – CD34+ correlated with Pgp but not with BCRP – Higher MRP3 in M5 subtype

BCRP expression and function – Highest in CD34+/CD38− cells in both AML and normal blasts – Ko143 did not increase apoptosis in CD34+/CD38− AML cells

BCRP mRNA at diagnosis – Was not predictive of CR – High expression assoc with shorter OS

BCRP, Pgp prognostic factors for achieving CR. High BCRP expression assoc with lower DFS, OS. Coexpression of Pgp and BCRP had lowest OS, DFS

Poor concordance of BCRP mRNA, protein, and function in AML samples; BCRP found in small subpopulations Wild type sequence found at codon 482

Pgp and BCRP are frequently coexpressed (r = 0.9)

High BCRP mRNA expression correlated with resistance to flavopiridol; no mutations of codon 482 observed

No increase in ABC transporter function at relapse or emergence of MRD. No subpopulations of resistant cells Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes 261

AML in elderly

AML stem cells (LSC)

Childhood AML

Van den Heuvel-Eibrink 2007 (136)

De Figueiredo-Pontes 2008 (133)

Shman 2008 (139)

19 at diagnosis and at Rel Additional 10: 8 de novo, 2 at Rel

26 de novo AML 8 non-AML marrows (4 ITP, 4 orthopedic surg)

154 age > 60, at diagnosis

73 consecutive de novo patients ad diagnosis

mRNA (qRT-PCR)

Protein (FC, BXP-21 MoAb)

mRNA (qRT-PCR)

Protein (FC, BXP-34 MoAb)

52% of relapse cases had increase in % CD34+ cells. Trend, but no statistically significant greater expression of BCRP or Pgp expression in CD34+ versus CD34− cells. Did not test for CD38 expression

– LSCs (CD34+/CD38−/CD123+ cells) had higher Pgp and BCRP protein expression – No difference in Pgp, BCRP, MRP1, or LRP expression among WHO classification subtypes

BCRP mRNA correlated with – Secondary AML, lower WBC – MDR1/ABCB1 mRNA expression CD34 expression and MDR1/BCRP coexpression associated with lower CR, DFS, OS

– High BCRP expression in 33% – BCRP and Pgp frequently coexpressed – High BCRP assoc. with risk of relapse and lower DFS

Genes segregated into 6 prognostic clusters; The cluster containing ABCG2 and ABCB1 had highest RD frequency

Normal karyotypes AML

Gene expression profiling

Damiani 2006 (132)

170 elderly patients at diagnosis

AML

Wilson 2006 (190)

Findings/Conclusions

Disease

Lead Author, year, citation

Parameter, Methodology

Ross and Nakanishi

Table 12.1 (continued) Number studied

262

AML

AML

Childhood ALL

ALL

ALL, adults and children

Adult ALL

Childhood ALL

Ho 2008 (135)

Saraiya 2008 (191)

Sauerbrey 2002 (141)

Stam 2004 (143)

Plasschaert 2003 (117)

Suvannasankha 2004 (142)

Kourti 2007 (144)

49 newly diagnosed

30: 17 B-lineage, 9 B-myeloid, 4 T-lineage

46: 23 B-lineage, 23 T-lineage

13 Infants, 13 Children – all at diagnosis

67: -47 at diagnosis, -20 at relapse

31 with elevated blast counts at diagnosis

34 adult de novo cases, all ages

mRNA (RT-PCR, 35 cycles)

mRNA (qRT-PCR) Protein (FC, BXP-34, BXP-21, 5D3 MoAbs); FA

Protein (FC, BXP-34 MoAb); FA

mRNA (qRT-PCR) in vitro drug sensitivity assays

mRNA (qRT-PCR)

Protein (Western blots, anti-ABCG2 MoAb)

mRNA (qRT-PCR) for all human ABC transporters

(continued)

– MDR1 but not BCRP mRNA expression correlated with the probability of EFS – Frequency of high BCRP expression low compared with expression of MDR1, MRP1, and LRP

– 37% to 47% stained + by one antibody – BCRP function and mRNA levels correlated poorly with antibody staining – BXP-21 staining may predict shorter DFS – No codon 482 mutations found

– Higher BCRP expression and function in B-lineage ALL – No codon 482 mutations found

BCRP mRNA exp – Lower in infant ALL – Correlated with in-vitro ara-C resistance – ara-C not a BCRP substrate Concluded that MDR proteins are noncontributory to resistance of infant ALL

BCRP mRNA exp – Not of prognostic significance – Was not increased at relapse – Was lowest in T-ALL

– In 2 of 4 evaluable patients, BCRP expression increased during a 5-day IV infusion of topotecan and cytarabine

– Transporter expression in total blast population did not predict response – Nonresponders had higher BCRP and/or MDR1 expression in the CD34+/CD38− population

Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes 263

Disease

CML

CML

CML

Lead Author, year, citation

Jordanides 2006 (151)

Jiang 2007 (152)

Wang 2008 (154)

Table 12.1 (continued)

70: 50 CP, 19 AP, 1 BC. All imatinib-naïve Peripheral blood

28: 18 CP, 8 AP, 2 BC, all imatinib-naive Peripheral blood, bone marrow

CCR, PFS, OS – Correlated with pretreatment exp of OCT1 – Did not correlate with Pgp or BCRP exp

CML stem cells (lin− CD34+CD38−) compared with more mature CML cells – Have >tenfold higher proliferation under low growth factor conditions – Produce IL-3 and G-CSF – Express Pgp and BCRP – Have low OCT1 expression – Have higher expression and activity of p210BCR/ABL – Resist killing by imatinib Conclusion: CML stem cells have multiple innate resistance mechanisms to imatinib

mRNA (qRT-PCR)

mRNA (qRT-PCR)

– Functional BCRP is expressed in CD34+ CML cells – BCRP expression in CD34+ normal hematopoietic cells is less than in CML (2 patients) – BCRP does cause imatinib resistance in CD34+ CML cells – Imatinib is an inhibitor of but not a substrate for BCRP in CD34+ CML cells

Findings/Conclusions

mRNA (qRT-PCR); FA

Parameter, Methodology

7, newly diagnosed LPB from chronic phase patients at diagnosis, preRx

Number studied

264 Ross and Nakanishi

CML

11 CP

FA-Dasatinib and imatinib IUR

– Dasatinib is a substrate for BCRP and Pgp – Dasatinib cellular uptake not affected by OCT1

AML acute myelogenous leukemia, ALL acute lymphoblastic leukemia, CML chronic myelogenous leukemia, RT-PCR reverse transcription polymerase chain reaction, qRTPCR quantitative real time RT-PCR, IHC immunohistochemistry, MoAb monoclonal antibody, exp expression, diff difference, FC flow cytometry, FA functional assay, preRx pretreatment, CR complete remission, Rel relapse, RD refractory disease, CCR complete cytogenetic response, PFS progression-free survival, OS overall survival, MRD minimal residual disease, LSC leukemia stem cell, HSC hematopoietic stem cell, CP chronic phase of CML, AP accelerated phase of CML, BC CML in blast crisis, LPB leukapheresed peripheral blood, IUR intracellular uptake and retention

Hiwase 2008 (155)

Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes 265

266

Ross and Nakanishi

et al. (123) evaluated 149 AML patients prior to treatment, using quantitative RT-PCR to detect BCRP mRNA and functional assays to detect Pgp. In contrast to the Abbott study (122), these investigators found that BCRP mRNA expression was a prognostic factor for achieving a complete remission. Furthermore, they found that high BCRP mRNA expression correlated with lower disease-free survival and OS. Patients with both high BCRP mRNA and high Pgp function had the worst prognosis (123). In a follow-up to their initial study of 149 AML patients, Benderra et al. studied an additional 85 de novo AML patients at time of diagnosis for the protein expression and function of Pgp, MRP1, MRP2, MRP3, MRP5, and BCRP (124). The expression or function of BCRP, Pgp, and MRP3 were found to correlate with complete response (CR) rate, and OS. In patients with high expression of two or more of these transporters, lower rates of CR, disease-free survival (DFS), and OS were seen. A study conducted by Uggla et al. on 40 mostly de novo AML patients (median age 57 years) found no effect of BCRP mRNA expression on CR rate, but BCRP expression was associated with shorter OS (125). With the advent of antibodies to BCRP, studies began to appear relating BCRP protein expression to clinical outcome. A study conducted by Sargent et al. on BCRP protein expression in 20 adult AML patients by Sargent et al., using the newly produced BXP34 antibody and immunohistochemical methods, found that BCRP expression correlated with in  vitro blast cell sensitivity to daunorubicin, but not to mitoxantrone or doxorubicin (126). However, no difference in BCRP expression was found between samples from chemotherapy naïve and previously treated patients (126). Since studies of AML patients to date find the wild-type sequence at codon 482 (described above (116–118)), this finding is a bit puzzling, since the native R482 BCRP protein is generally thought to efflux mitoxantrone more efficiently than daunorubicin (62). Using the BXP21 and BXP34 antibodies and functional assays, van der Kolk et  al. studied 20 paired AML blast cell specimens obtained at diagnosis and at time of relapse or resistant disease (127). In this study, BCRP protein expression did correlate with the functional assays, and was found in subpopulations of cells with an immature phenotype (CD34+); however, there was no increase in these BCRP-positive subpopulations at time of relapse, leading the authors to conclude that “BCRP was not consistently upregulated in relapsed/refractory AML.” If one considers the possibility that these subpopulations may represent leukemia stem cells, it would not be necessary for the population to expand at the time of relapse, but merely persist. A somewhat larger study of paired samples was published the next year by van der Pol et al. wherein 45 AML patients were studied at diagnosis

Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes

267

and at time of relapse or development of minimal residual disease (MRD) using functional assays and the BXP34 and BXP21 antibodies to detect BCRP protein expression (128). No increase in BCRP, Pgp, or MRP1 expression or function was found in the relapse samples or at time of MRD. Unlike the study by van der Kolk (127), these investigators did not find that BCRP expression was confined to subpopulations of resistant cells (128). A study by Suvannasankha et  al. investigated pretreatment blast cells from 31 AML patients by measuring BCRP mRNA using real-time RT-PCR, BCRP function, and BCRP protein using flow cytometry and three antibodies to BCRP available at the time (BXP21, BXP34, and 5D3) (118). In cell line controls, expression of BCRP was concordant by all assays employed; however, in the patient samples, BCRP mRNA expression correlated poorly with BCRP protein expression and function. BCRP expression in patient samples was found only in small subpopulations of cells. All mRNA in patient samples had the wild-type sequence at codon 482. It was concluded that the discordance of assays in the patient samples reflected complex biology of BCRP in AML that was not reflected by cell lines. Indeed, a number of biological processes have recently come to light, which may influence the activity– protein ratio of BCRP in living cells. In multidrug resistant human prostate cancer cell lines, the Pim-1  L kinase is upregulated along with BCRP and phosphorylates the threonine of BCRP at codon 362, which results BCRP transporter activation by multimerization of BCRP and translocation of BCRP to the plasma membrane (129). Monomeric or cytoplasmic BCRP may possibly be detected by flow cytometry and internal epitope-recognizing antibodies such as BXP21 or BXP34. Similarly, in murine systems, AKT/PI3K pathway signaling is necessary for BCRP translocation to the plasma membrane, and hence for transporter activity (130). In human CML cells, however, inhibition of AKT/PI3K signaling results in posttranscriptional down regulation of BCRP protein expression (131). Hence, the impact of BCRP phosphorylation by Pim-1 L kinase or oncogenic signaling such as the AKT/PI3K pathway may need to be evaluated in future studies of BCRP expression and activity in human cancers. More recent studies of BCRP expression in AML seem to confirm the notion that BCRP is often coexpressed with Pgp, and connotes a worse prognosis (132–136); furthermore, BCRP and Pgp expressions appear to be associated with subpopulations of cells with primitive characteristics such as expression of CD34 but not CD38 (133, 136–138), although two studies that measured CD34 expression (but not CD38) did not find a statistically significant correlation of BCRP expression with that of CD34 (124, 139). Coexpression of Pgp with BCRP was found in a study 73 consecutive patients with de novo AML and normal karyotype

268

Ross and Nakanishi

by Damiani et al. (132). High expression of BCRP protein was found in approximately one third of these patients. BCRP expression was associated with an increased risk of relapse and lower DFS. A study conducted by van den Heuvel-Eibrink on the expression of BCRP and MDR1 mRNA in 154 previously untreated de novo or secondary AML patients over the age of 60 found a high degree of BCRP and MDR1 mRNA coexpression (136). BCRP and particularly MDR1 mRNA expression correlated with CD34 expression. The most significant poor prognostic indicator was MDR1/BCRP coexpression, which was associated with a lower CR rate. A number of recent studies evaluated BCRP expression in AML blast cell subpopulations with primitive characteristics. Raaijmakers et al. studied BCRP protein expression and function in blast cells from 21 AML patients at the time of diagnosis, and compared these with hematopoietic cells in normal marrow (138). They found the highest BCRP expression and function in the CD34+/CD38− cells in both normal and AML marrows. Interestingly, the BCRP inhibitor Ko143 increased mitoxantrone accumulation but not cytotoxicity in the leukemic CD34+/CD38− cells, leading these investigators to conclude that “selective modulation of BCRP is not sufficient to circumvent resistance of leukemic CD34+/CD38− cells,” suggesting that factors in addition to BCRP may contribute to the drug resistance displayed by these cells. A recent study by Ho et al. evaluated the mRNA expression of the entire family of human ABC transporter genes in malignant blast cells from 18 AML patients who achieved CR and from 13 AML patients who were refractory to induction chemotherapy (135). No difference in ABC transporter expression was observed between the CR and refractory groups; however, when transporter expression was evaluated based on expression of CD34 and CD38, the nonresponders had significantly higher expression of BCRP and/or MDR1 mRNA in their CD34+/ CD38− cells. Finally, de Figueriedo-Pontes et al. isolated subsets of primitive cells from 26 de novo CD34+ AML cases (133). They found that “leukemia stem cells,” defined as CD34+/CD38−/ CD123+ cells, had higher expression of BCRP (BXP21 antibody) and Pgp than the other subsets, leading these authors to conclude that the presence of BCRP and Pgp in leukemia stem cells “reinforces that (multidrug resistance) is one of the mechanisms of treatment failure.” The expression of BCRP, Pgp, MRP1, or LRP did not differ among the various World Health Organization subgroups of AML (133, 140). ALL

Fewer studies have been done in ALL on the impact of BCRP on clinical outcomes than what have been done in AML. No definite trends correlating BCRP expression with prognosis have been found, although it appears that BCRP expression is highest in

Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes

269

B-lineage ALL (117, 141). Two studies evaluated ALL blast cells for mutations at codon 482: no mutations were found (117, 142). A study by Sauerbrey et al. evaluated BCRP mRNA expression in 67 children with ALL, 47 at diagnosis, and 20 at relapse (141). BCRP mRNA was not increased in the specimens from the children in relapse. There was no relationship of BCRP mRNA expression with relapse-free survival in this study. Stam et al. studied BCRP mRNA and in vitro drug sensitivity in 13 infants and 13 children with ALL at the time of diagnosis (143). Surprisingly (since ara-C is not known to be a BCRP substrate), blast cell samples with high BCRP mRNA expression had the highest in vitro resistance to ara-C. Hence, BCRP may be a marker for ara-C resistance in this case, but not a direct cause of the resistance per se. A study by Kourti et al. found that MDR1 mRNA but not BCRP mRNA expression correlated with shorter eventfree survival in 49 newly diagnosed cases of childhood ALL; high BCRP mRNA expression was infrequent, compared to the mRNA expression of MDR1, MRP1, and LRP (144). Using three BCRP antibodies, Suvannasankha et  al. studied thirty adult ALL cases for BCRP function, and mRNA and protein expression. A relatively high frequency of positivity of staining (37–47%) for each antibody was found; although there was poor concordance of antibody staining, mRNA expression, and functional assays in this study, positive staining with the BXP21 antibody was predictive of a shorter DFS (142). CML

Highly effective tyrosine kinase inhibitors (TKI) such as imatinib and dasatinib that are relatively selective for the chimeric BCRABL tyrosine kinase crucial for CML pathogenesis have revolutionized the current treatment for CML. Unfortunately, molecular CRs with imatinib treatment occur in only 35% of chronic phase patients, casting doubt on whether such patients can be cured by TKI therapy alone. It is known that chronic phase CML samples contain quiescent, Philadelphia chromosome positive stem cells that are resistant to imatinib (145). Since normal hematopoietic and certain other stem cells can be identified by their ability to exclude Hoechst 33342 dye by means of expression of BCRP and perhaps other ABC transporters (73), it is reasonable to hypothesize that BCRP expression in CML stem cells may cause efflux of imatinib and hence may be a reason for failure to obtain molecular remissions of CML with TKI therapy (Fig.  12.1). A number of studies suggest that imatinib may be a substrate for and/or inhibitor of BCRP (63, 64, 146–149), although one study did not find resistance to imatinib in osteosarcoma cells transfected to express BCRP (149). Studies in our own laboratory using K562 CML cells transduced to overexpress BCRP found that the BCRPexpressing cells were indeed resistant to imatinib, but to a lesser

270

Ross and Nakanishi

Fig. 12.1. CML stem cell hypothesis. Primitive, self-renewing Philadelphia chromosome positive CML stem cells express BCRP and possibly other mechanisms of resistance to TKI treatment (indicated by granular surface of stem cells), but the more mature CML progenitor cells remain sensitive to TKIs. TKI treatment would result in death of the CML progenitor and mature cell populations, but persistence of the TKI-resistant stem cell population resulting in hematologic (and possibly cytogenetic) CR but not a molecular CR. Acquisition of TKI resistance mutations of BCR-ABL by the stem cell or early progenitor compartment would then result in relapse of the chronic phase, and possibly the appearance of the accelerated or blastic phase.

extent than to mitoxantrone, a well-known substrate for BCRP (131). Imatinib resistance in BCRP-transduced K562 cells was confirmed by Brendel et al. who also found that BCRP-transduced K562 cells were resistant to nilotinib, a novel inhibitor of BCRABL, and were protected from imatinib- and nilotinib-induced downregulation of CRKL phosphorylation (150). Given the CML stem cell hypothesis (Fig.  12.1), a crucial question is whether functional BCRP is expressed in Philadelphia chromosome positive imatinib resistant CML stem cells, and whether BCRP is responsible for the imatinib resistance in these cells. This was investigated by Jordanides et  al. who studied CD34+ CML mononuclear cells obtained by leukapheresis from seven chronic-phase CML patients and found that functional BCRP was expressed in these cells; however, BCRP did not cause imatinib resistance in these stem cells. Instead, imatinib was found to be an inhibitor of and not a substrate for BCRP in these cells, since the BCRP inhibitor FTC enhanced mitoxantrone accumulation but not imatinib accumulation in these CML stem cells (151). The multidrug resistance phenotype of CML stem cells was investigated in lin−/CD34+/CD38− cells collected from 28 CML patients prior to imatinib treatment by Jiang et  al. (152).

Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes

271

Compared with more mature CML cells, the CML stem cells had a higher proliferative capacity, produced IL-3 and G-CSF, expressed Pgp and BCRP, had low expression of human organic cation transporter 1 (OCT1, the transporter for cellular uptake of imatinib), higher expression and activity of the p210BCR/ABL protein, and resisted killing by imatinib. These authors concluded that CML stem cells have multiple mechanisms of resistance to imatinib. Low OCT1 activity in CML cells has been reported to result in a suboptimal response to imatinib (153). High expression of OCT1 mRNA in cells from pretreatment peripheral blood of 70 CML patients was found by Wang et al. to predict a higher complete cytogenetic response rate, longer DFS and OS in response to imatinib treatment compared with patients with low OCT1 mRNA expression (154). These authors found that pretreatment expression of MDR1, BCRP, or MRP1 mRNA did not correlate with imatinib therapeutic outcome; however, these measurements were not made in the CML stem cell compartment. Nevertheless, the authors concluded that “the expression of OCT1, but not efflux transporters, is important in determining the clinical response to imatinib.” Using peripheral blood mononuclear cells from 11 CML patients in chronic-phase, Hiwase et  al. found Dasatinib, a second-generation TKI that is effective against BCRABL mutations that cause resistance to imatinib, to be a substrate for Pgp and BCRP (155). In contrast to imatinib, the cellular uptake of Dasatinib was not found to be affected by OCT1 in these patient samples. 3.2.2. Solid Tumors, Lymphomas, and Myeloma

A synopsis of literature relating the expression and function of BCRP with outcomes in solid tumors is given in Table  12.2. Initial immunohistochemical studies with an anti-BCRP monoclonal antibody (BXP-34), using a panel of human tumors, showed BCRP to be low or undetectable except for one case of small intestine adenocarcinomas (156); however, subsequent investigation by the same investigators (Diestra et  al.), using a newer monoclonal antibody (BXP-21) in formalin-fixed paraffinembedded specimens, demonstrated a high frequency of BCRP immunoreactivity among a panel of 150 untreated human solid tumors comprising 21 tumor types (157). These authors reported that BCRP expression “was seen in all tumor types, but it seemed more frequent in adenocarcinomas of the digestive tract, endometrium, lung, and melanoma” (157). In selected cases, the immunohistochemical data were validated by Western blots. The authors were careful to exclude BCRP in noncancerous tissues, such as the expression of BCRP in venules. As will be seen from the ensuing discussion, many human solid tumors such as lung and esophageal cancers and some lymphomas appear to express BCRP, and frequently, this expression is correlated with adverse prognostic significance. In contrast

Disease

21 tumor types

Aggressive mantle cell lymphoma

Mature T/NK lymphomas

Multiple myeloma

Multiple myeloma

Lead Author, year, citation

Diestra 2002 (157)

Galimberti 2007 (158)

Saglam 2008 (159)

Raaijmakers 2005 (160)

Turner 2006 (161)

mRNA (qRT-PCR) Protein (FC, antiABCG2 MoAb)

31 patients studied before, during, and at relapse after treatment including topotecan FA

Protein (BXP21 MoAb) FA

Protein -IHC (MoAb type not stated)

mRNA (qRT-PCR)

Protein -IHC (BXP-21 MoAb)

Parameter, Methodology

10 newly diagnosed patients Normal bone marrow

119, archival review

20, prior to treatment with R-hyperCVAD

150 untreated human solid tumors

Number studied

– BCRP mRNA and protein expression was higher after treatment or at relapse compared to pretreatment – BCRP promoter was methylated in pretreatment samples

BCRP protein and functional expression higher in normal plasma cells than in myeloma plasma cells

– Relatively high BCRP expression (3-4+) in 78% of samples – MRP1, Pgp, LRP also frequently expressed

BCRP expression – Associated with worse PFS – Correlated with MRD status

Widespread expression of BCRP/ABCG2 in a variety of untreated human solid tumors

Findings/Conclusions

Table 12.2 Expression and consequences of BCRP/ABCG2 in human solid tumors, lymphomas, and myeloma

272 Ross and Nakanishi

Hepatoblas toma

Testicular tumors

Ovarian Cancer

Esophageal squamous cell carcinoma

Locally Advanced Bladder Cancer

NSCLC

Vander Borght 2008 (162)

Bart 2004 (163)

Nakayama 2002 (174)

Tsunoda 2006 (164)

Diestra 2003 (175)

Kawabata 2003 (165)

mRNA (qRT-PCR) Protein -IHC (BXP-21 MoAb) FA

(continued)

22% of lung cancer samples expressed relatively high expression of BCRP

– High levels of BCRP expression found in 28% of the cases – BCRP expression had no prognostic impact – Pgp expression correlated with shorter PFS, but not with OS

Protein (IHC)

83 patients treated with neoadjuvant chemotherapy

23

– BCRP detectable in 61% of cases – Detectable BCRP expression correlated with poorer OS following surgery – BCRP expression an independent prognostic factor

mRNA (qRT-PCR) (N = 33) Protein – IHC, BXP-21 MoAb (N = 100)

100 paraffin sections; 33 frozen tissues. Treatment varied; about 30% rec’d platinum based regimens

Pgp, BCRP expression in germ-cell and testicular lymphomas may contribute to chemoresistance

Protein -IHC (BXP-21 MoAb)

– BCRP expression was not a prognostic indicator – Copper transporter ATP7B expression correlated with poor prognosis

BCRP (but not Pgp) expression increased in all posttreatment samples; BCRP expression highest in areas of hypoxia

Protein -IHC (BXP-21 MoAb)

mRNA (RT-PCR)

82 sampled before cisplatinbased therapy

10 Nonseminoma 10 seminoma 9 testicular lymphoma 10 normal

7 patients sampled pretreatment and at tumor resection, posttreatment with cisplatin, carboplatin and/or doxorubicin

Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes 273

Disease

NSCLC

NSCLC

NSCLC

NSCLC

Lead Author, year, citation

Yoh 2004 (166)

Chen 2008 (170)

Surowiak 2008 (169)

Ota 2008 (167)

Table 12.2 (continued)

156 stage IV patients, prior to platinum-based therapy

32 stage IIIB and IV

10, stages IA to IIIB

72 with stage IIIb or IV disease, prior to platinum-based treatment regimens

Number studied

Protein -IHC

Protein -IHC

Protein -IHC (MoAb type not stated)

Protein -IHC (BXP-21 MoAb)

Parameter, Methodology

High BCRP expression correlated with lower OS, but showed no response to chemotherapy

No prognostic or predictive significance of BCRP expression. BCRP expression correlated with expression of COX2 and Pgp

Isolated CD133+ and CD133− LC cells LC-CD133+ cells had – Stem cell properties, with greater proliferative and tumorigenic capacity – High expression of BCRP – Resistance to chemo and radiotherapy – Dependence on the Oct-4 transcription factor for stem cell phenotype

BCRP expression associated with – Lower response rate – Shorter OS, PFS – Independent prognostic variable for PFS MRP2 expression associated with lower OS No association of Pgp, MRP1, or MRP3 expression with response to treatment or OS

Findings/Conclusions

274 Ross and Nakanishi

NSCLC SCLC

SCLC

Colon Cancer

Colon Cancer Dukes C

Colon Cancer

Melanoma; Neuroendocrine (Merkel) carcinoma

Melanoma

Muller 2009 (112)

Kim 2008 (168)

Candeil 2004 (171)

Glasgow 2005 (173)

Gupta 2006 (172)

Deichmann 2005 (178)

Monzani 2007 (180)

mRNA (qRT-PCR)

(continued)

Human melanomas were found to contain cancer stem cells with enhanced tumorigenic potiential (when transplanted into NOD-SCID mice). These stem cells express CD133, ABCG@/BCRP, notch 4 and other stem cell markers

– BCRP mRNA not higher in melanomas compared to acquired melanocytic nevi – BCRP protein not detected by IHC in melanomas – 3 of the neuroendocrine tumors showed some positive immunostaining for BCRP

mRNA (qRT-PCR) Protein, IHC (BXP21 MoAb)

66 melanomas (RNA extracted from 18) 29 neuroendocrine carcinomas of the skin

7 skin biopsies

– Lower BCRP expression in cancers compared to normal tissues

mRNA (qRT-PCR, northern blots) Protein, IHC (antiABCG2 MoAb)

– 13 colon cancer – 1 hepatic met. from colon cancer Prior to any treatment

BCRP mRNA expression is higher in metastases after irinotecan treatment than in metastases from irinotecan-naïve patients

BCRP expression correlated with poorer response and shorter PFS

Worse OS seen BCRP 421A hetero- or homozygotes in patients treated with platinum-based chemotherapy

– No difference in BCRP mRNA expression between mucinous and nonmucinous tumors – Mucinous tumors overexpressed markers of resistance to 5-FU and oxaliplatin

mRNA (qRT-PCR)

Protein -IHC

mRNA (qRT-PCR), melting curve analysis

mRNA (qRT-PCR)

21 mucinous 30 non-mucinous

42 patients; Liver metastases sampled pre- and post- irinotecan treatment

130 prior to platinum-based therapy

187 NSCLC 161 SCLC 1 mixed

Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes 275

Childhood and adult CNS tumors

Retinoblastoma

Breast Cancer

Breast Cancer

Wilson 2006 (177)

Kanzaki 2001 (181)

Faneyte 2002 (182)

Disease

Valera 2007 (176)

Lead Author, year, citation

Table 12.2 (continued)

25 chemotherapy naïve 27 after anthracycline treatment

43 untreated patients 38 subsequently received anthracycline-based adjuvant therapy after surgery

16 patients after primary enucleation

21 astrocytoma I and II 7 astrocytoma III 21 glioblastomas 17 medulloblastomas; 8 ependymomas; 6 oligodendroglioma

Number studied

BCRP mRNA expression – varied greatly among tumor specimens – not associated with decreased response or survival – no different between chemotherapy naïve and treated groups BCRP protein (IHC) – detected in blood vessels and normal breast epithelium – not detected in breast cancer tumor cells

– BCRP mRNA expression low, with little variability compared to the other genes studied – BCRP mRNA expression did not correlate with the expression of the other genes

mRNA (semi-quantitative RT-PCR) for BCRP, MDR1, MRP1 and LRP

mRNA (qRT-PCR, northern blots) Protein, IHC (BXP34 and BXP21 MoAbs)

– BCRP was detected in vascular endothelium but not in tumor cells in all of the specimens studied

– Higher MDR1 and BCRP mRNA expression in glial tumors than in embryonic tumors – No impact of transporter expression on OS of medulloblastomas or high grade gliomas following multimodality treatment

Findings/Conclusions

Protein: Tissue microarray IHC (BXP-21 MoAb)

mRNA (qRT-PCR) of microdissected samples

Parameter, Methodology

276 Ross and Nakanishi

Breast Cancer

Breast Cancer

Fazeny-Dorner 2003 (192)

Park 2006 (184)

21 pretreatment biopsies prior to neoadjuvant chemotherapy

4 cases following neoadjuvant chemotherapy

59 primary tumor samples obtained prior to chemotherapy

mRNA expression profiling of ABC transporters (GeneChip®)

CGH, cytogenetics; BCRP mRNA, protein or function was not studied

mRNA (qRT-PCR) for BCRP, LRP, MRP1, MRP2 and MDR1

– BCRP/ABCG2 not mentioned as a gene being differentially expressed in responders versus resistant disease group

– 2 patients had amplification in the long arm of chromosome 4 (4q22), the region harboring BCRP

Overall response to anthracycline-based therapy – Significant inverse correlation with MDR1 mRNA expression – Trend to inverse correlation with BCRP mRNA expression – Stronger negative correlation between BCRP mRNA expression and response rate or PFS in patients treated with anthracyclines-based regimens than those treated with CMF

IHC immunohistochemistry, PFS progression-free survival, OS overall survival, MRD minimal residual disease, SCLC small cell lung cancer, NSCLC nonsmall cell lung cancer, preRx pretreatment, LC lung cancer, COX-2 cyclooxygenase-2, Pgp p-glycoprotein, the product of the MDR1 gene, MDR1 multidrug resistance gene 1, MRP1 multidrug-resistance associated protein-1, MRP2 multidrug-resistance associated protein-2, LRP lung resistance protein, 5-FU 5-fluorouracil

Breast Cancer

Burger 2003 (183)

Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes 277

278

Ross and Nakanishi

to the study by Diestra et al., low expression of BCRP was found in malignant melanoma. Furthermore, little or no prognostic value of BCRP expression was found in ovarian cancer, locally advanced bladder cancer, colon cancer (some reports), and ironically, breast cancer. Because of the nature of solid tumor specimens, many of these studies did not evaluate BCRP expression in subpopulations such as the side population; however, when such correlations were made (e.g. in the case of melanoma), BCRP expression was found in subpopulations with enhanced selfrenewal capacity, often with coexpression of CD133, a putative stem cell marker. Such delineations may be important to perform in future studies relating BCRP expression to clinical outcome. Galemberti et  al. studied BCRP mRNA expression in 20 patients with aggressive mantle cell lymphoma prior to treatment, using qRT-PCR (158). BCRP mRNA expression correlated with worse DFS, finding minimal residual disease. In mature T/NK lymphomas, Sagalam et  al. evaluated 119 archival cases using immunohistochemistry (IHC) (159). Relatively high BCRP staining was seen in most of the samples; expression of Pgp, MRP1, and LRP was also seen frequently. Expression of these MDR proteins was not related to survival in this study. In multiple myeloma, a study of 10 newly diagnosed patients prior to treatment by Raaijmakers et al. found that BCRP function and protein expression were lower in multiple myeloma plasma cells compared with bone marrow plasma cells obtained from normal donors (160). A study of 31 multiple myeloma patients by Turner et  al. confirmed low expression of BCRP expression in pretreatment samples, in which the BCRP promoter was found to be methylated (161). In contrast, BCRP mRNA and protein expression were higher after treatment or at time of relapse following a topotecan-based treatment regimen. A study of BCRP protein expression in seven infants or children with hepatoblastoma by Vander Borght et al. found that BCRP expression (but not expression of Pgp, MRP1, MRP2, or MRP3) increased in all samples following treatment with a regimen containing cisplatin or carboplatin and/or doxorubicin (162). Interestingly, expression of BCRP was highest in areas of the tumor that appeared to be hypoxic, consistent with the known hypoxia response element in the BCRP promoter region (34, 66). BCRP expression was evaluated in a variety of testicular tumors by Bart et al., using IHC (163). BCRP and Pgp expression were found in germ cell tumors, testicular lymphomas, and newly formed tumor blood vessels, and “may contribute to chemoresistance.” BCRP mRNA and protein expression were found to be independent adverse prognostic indicators in esophageal squamous cell carcinoma in a study of 133 paraffin or frozen sections by

Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes

279

Tsunoda et al. (164). Treatment in this study varied; approximately 30% received a platinum-based regimen. One of the solid tumors most extensively studied for BCRP expression is lung cancer, particularly NSCLC. In 2003, BCRP expression in lung tumors was confirmed by Kawabata, who used qRT-PCR to detect high levels of BCRP in 6 of 8 nonsmall cell lung cancer cell lines and 5 of 23 (22%) of nonsmall cell lung tumor tissues tested (165). Topotecan efflux in the lung cancer cell lines correlated with the levels of BCRP mRNA expressed. The following year, Yoh et al. published a study of BCRP protein expression in 72 NSCLC patients studied prior to treatment with platinum-based regimens (166). BCRP expression was found to correlate with lower response rate and shorter OS, and was an independent prognostic variable for DFS. In contrast, the study by Ota et al., using IHC in 156 stage IV NSCLC patients prior to treatment with platinum containing regimens, on BCRP protein expression did not find a correlation of BCRP expression with response rate; however, there was a correlation of high BCRP expression and lower OS (167). In SCLC, Kim et al. found a correlation between BCRP protein expression by IHC and poor response and shorter DFS in 130 patients studied prior to treatment with platinum-containing regimens (168). In contrast to these studies, a recent publication by Surowiak et  al. found no prognostic or predictive significance of BCRP protein expression (IHC) among 32 patients with stage IIIB or IV NSCLC (169). In this study, BCRP expression correlated with that of cyclooxygenase-2 and Pgp. A recent study by Chen et al. found evidence for the presence of subsets of cells with stem-cell properties in NSCLC, based on CD133 expression in tumor tissue samples obtained from ten patients (170). The lung cancer CD133+ cells had stem cell properties, with greater proliferative and tumorigenic capacity, high expression of BCRP, resistance to chemo and radiotherapy (including cisplatin, etoposide, doxorubicin and paclitaxel), and a dependence on the Oct-4 transcription factor for stem cell phenotype. The drug resistance profile of these cells includes drugs that are not classical substrates for BCRP (cisplatin, etoposide, paclitaxel). Irinotecan and its active metabolite SN-38 are known substrates for BCRP. A study of clinical biopsy samples of hepatic metastases from 42 patients with colon carcinoma by Candeil et al. demonstrated almost a tenfold increase in BCRP mRNA in metastases obtained postirinotecan treatment compared to metastases obtained pretreatment, or posttreatment with non-BCRP substrate drugs (171). Despite high expression of BCRP in the apical surface of normal small bowel and colon epithelial cells, pretreatment specimens of colon cancer and a hepatic metastasis had lower BCRP mRNA and protein expression compared with normal colon in a study of 13 patients by Gupta et  al. (172).

280

Ross and Nakanishi

Similarly, in 145 paired samples, down-regulation of BCRP expression was seen in cervical carcinoma by these investigators, and also in cancers from breast, lung, ovary, kidney, liver, uterus, rectum, thyroid, testis, and small intestine, compared with normal tissue counterparts. Glasgow et  al. studied BCRP mRNA expression in 21 mucinous compared with 30 nonmucinous Dukes C colon cancer specimens (173) Although mucinous tumors have a worse prognosis than nonmucinous ones, there was no difference in BCRP mRNA expression between mucinous and nonmucinous tumors; however, mucinous tumors overexpressed markers of resistance to 5-FU and oxaliplatin in this study. Together, the above studies imply that BCRP expression in colon carcinoma is low prior to treatment, but may increase following treatment with BCRP substrate drugs. A number of studies do not find a relationship with BCRP expression and adverse outcome in human solid tumors. Nakayama et al. sampled tumor from 82 patients with ovarian cancer prior to cisplatin-based therapy (174). BCRP mRNA expression was not found to be a prognostic indicator. Diestra et  al. studied 83 patients with locally advanced bladder cancer treated with neoadjuvant therapy; although high levels of BCRP protein expression were found in 28% of the cases, BCRP expression had no prognostic impact (175). BCRP mRNA expression in microdissected samples of childhood and adult CNS tumors were investigated by Valera et al. (176). Higher BCRP and MDR1 mRNA expression was found in glial tumors compared with embryonic tumors, but no impact of transporter expression was found on OS of medulloblastomas versus high grade gliomas following multimodality treatment. Wilson et al. used tissue microarrays and IHC to study tumor from 16 patients with retinoblastoma after primary enucleation (177). BCRP was detected in vascular endothelium but not in tumor cells in all the specimens studied. Although the early study by Diestra et al. (157) found BCRP expression in melanoma cells, a subsequent paper by Deichmann et al. found no expression of BCRP protein by IHC in biopsy specimens from 66 melanoma patients (178). Though Deichmann et al. concluded that chemoresistance of melanomas cannot be explained by BCRP expression, subsequent work (e.g. Monzani et al. and others (179, 180)) has found that a subset of CD133+/ABCG2+ cells is present in melanoma specimens, and that this subset expresses angiogenic and lymphangiogenic markers and has increased tumorigenicity when transplanted into NOD-SCID mice. Despite its original isolation from multidrug resistant human breast cancer cells, the level of BCRP expression in clinical breast cancer cases appears low, at least in the “whole cell” population: Kanzaki et  al. studied BCRP, MDR1, MRP1, and LRP mRNA expression in breast cancers form 43 patients prior to treatment (181). Thirty eight of these went on to receive anthracycline-

Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes

281

based adjuvant therapy following surgery. BCRP expression was low, had little variability compared to the other drug resistance markers studied, and did not correlate with the epression of MDR1, MRP1, or LRP. BCRP mRNA and protein expression in breast cancers from 25 chemotherapy naïve patients and 27 patients biopsied following anthracycline treatment were reported by Faynette et al. (182). These investigators found that although BCRP mRNA expression varied greatly among tumor specimens, it was not associated with decreased response or survival. There was no difference in BCRP expression between chemotherapy naïve and treated groups. BCRP protein (assessed by IHC) was detected in blood vessels and normal breast epithelium, but was not detected in breast cancer tumor cells. Burger et  al. studied BCRP, MDR1, MRP1, MRP2, and LRP mRNA expression in 59 primary breast tumor samples obtained prior to chemotherapy and found a significant inverse correlation with MDR1 mRNA expression but only a trend to inverse correlation with BCRP mRNA expression with the overall response to anthracyclinebased therapy (183). There was a stronger negative correlation between BCRP mRNA expression and response rate or DFS in patients treated with anthracycline-based regimens compared with those patients treated with CMF (cytoxan, methotrexate, fluorouracil). A study by Park et al. of mRNA expression profiling in 21 pretreatment breast biopsies prior to neoadjuvant chemotherapy did not mention BCRP among the genes differentially expressed in responders versus nonresponders (184). A number of caveats should be noted in interpreting the above studies of BCRP expression in breast cancer. Some of the patients enrolled in these studies may have experienced antiestrogen treatment, which may affect BCRP expression: recent studies indicate the presence of an estrogen response element in the BCRP promoter, and that antiestrogens such as tamoxifen can oppose transcriptional activation of BCRP expression by estrogen (185). Hence, future studies of BCRP in breast cancers should take into account treatment with antiestrogens at the time of tissue biopsy. Furthermore, SP cells with high drug efflux capacity have been observed in breast cancer and other solid tumors (186), implying a possible role of BCRP and possibly other ABC transporters in breast cancer resistance. 3.3. BCRP Expression as a Manifestation of the Activity of Metabolic and Signaling Pathways That Impart a Poor Prognosis to Cancers

Thirty years ago, the discovery that a membrane ABC transporter (e.g., Pgp) could cause resistance to multiple drugs raised the vain hope that such a transporter could be the sole cause of clinical multidrug resistance. It is now becoming apparent that BCRP and Pgp expression in human cancers could occur as a component of a much larger cancer cellular orchestration of evolution, through acquired genomic instability, toward a genotype of perpetual immortality, proliferation, invasion, and resistance

282

Ross and Nakanishi

to programmed cell death. For example, Chiou et al. found that BCRP was upregulated in oral cancer stem-like cells, along with other prosurvival genes – Oct-4, CD133, CD117, and Nanog (187). BCRP in cancer SP cells (and possibly other cancer stem cells) might aid these cells from evading damage from BCRP substrate drugs; furthermore, via its transport of protoporphyrin IX, BCRP may help cells evade damage because of the reactive oxygen species generated by hypoxia and perhaps other cellular damaging agents (34, 36, 66, 188). It is likely that certain cells with self renewal capacity do not overexpress BCRP, since one study found that BCRP+ and BCRP− cells are similarly tumorigenic (189). BCRP may be just one factor in a host of other determinants of resistance, and inhibition of BCRP alone will probably not be sufficient to sensitize cells with multiple additional resistance mechanisms in place, consistent with the argument made by Raaijmakers et al. that despite showing BCRP overexpression in leukemia stem cells, “selective modulation of BCRP is not sufficient to circumvent resistance of leukemic CD34+/38− cells” (138). There is considerable evidence that BCRP is upregulated along with other drug resistance genes: Wilson et  al. evaluated 170 elderly AML patients before treatment, using gene expression profiling (190). They identified six clusters of genes that varied in disease outcome parameters. The highest rate of resistant disease was found in patients expressing the cluster, which contained BCRP and MDR1. Hence, BCRP expression – along with the expression of other genes marking resistant, self-renewing cancer stem- or initiating-cells – may help lead us to the isolation and characterization of resistant, cancer perpetuating subpopulations, and to discover their unique vulnerabilities, since there will be a need to focus on defeating cancer stem cells without severe damage to normal stem cells. References 1. Ross DD, Doyle LA, Schiffer CA et al (1996) Expression of multidrug resistance-associated protein (MRP) mRNA in blast cells from acute myeloid leukemia (AML) patients. Leukemia 10:48–55 2. Chen YN, Mickley LA, Schwartz AM et al (1990) Characterization of adriamycin-resistant human breast cancer cells which display overexpression of a novel resistance-related membrane protein. J Biol Chem 265:10073–10080 3. Doyle LA, Yang W, Abruzzo LV et al (1998) A multidrug resistance transporter from human MCF-7 breast cancer cells. Proc Natl Acad Sci U S A 95:15665–15670 4. Allikmets R, Schriml LM, Hutchinson A, Romano-Spica V, Dean M (1998) A human pla-

centa-specific ATP-binding cassette gene (ABCP) on chromosome 4q22 that is involved in multidrug resistance. Cancer Res 58:5337–5339 5. Miyake K, Mickley L, Litman T et  al (1999) Molecular cloning of cDNAs which are highly overexpressed in mitoxantrone-resistant cells: demonstration of homology to ABC transport genes. Cancer Res 59:8–13 6. Abbott BL (2003) ABCG2 (BCRP) expression in normal and malignant hematopoietic cells. Hematol Oncol 21:115–130 7. Abbott BL (2006) ABCG2 (BCRP): a cytoprotectant in normal and malignant stem cells. Clin Adv Hematol Oncol 4:63–72 8. Ahmed-Belkacem A, Pozza A, Macalou S et al (2006) Inhibitors of cancer cell multidrug

Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes  resistance mediated by breast cancer resistance protein (BCRP/ABCG2). Anticancer Drugs 17:239–243 9. Allen JD, Schinkel AH (2002) Multidrug resistance and pharmacological protection mediated by the breast cancer resistance protein (BCRP/ABCG2). Mol Cancer Ther 1:427–434 10. Assaraf YG (2007) Molecular basis of antifolate resistance. Cancer Metastasis Rev 26:153–181 11. Bates SE, Robey R, Miyake K et  al (2001) The role of half-transporters in multidrug resistance. J Bioenerg Biomembr 33:503–511 12. Begley DJ (2004) ABC transporters and the blood-brain barrier. Curr Pharm Des 10:1295–1312 13. Breedveld P, Beijnen JH, Schellens JH (2006) Use of P-glycoprotein and BCRP inhibitors to improve oral bioavailability and CNS penetration of anticancer drugs. Trends Pharmacol Sci 27:17–24 14. Bunting KD (2002) ABC transporters as phenotypic markers and functional regulators of stem cells. Stem Cells 20:11–20 15. Calcagno AM, Kim IW, Wu CP, Shukla S, Ambudkar SV (2007) ABC drug transporters as molecular targets for the prevention of multidrug resistance and drug-drug interactions. Curr Drug Deliv 4:324–333 16. Cervenak J, Andrikovics H, Ozvegy-Laczka C et al (2006) The role of the human ABCG2 multidrug transporter and its variants in cancer therapy and toxicology. Cancer Lett 234:62–72 17. Choudhuri S, Klaassen CD (2006) Structure, function, expression, genomic organization, and single nucleotide polymorphisms of human ABCB1 (MDR1), ABCC (MRP), and ABCG2 (BCRP) efflux transporters. Int J Toxicol 25:231–259 18. Coley HM (2008) Mechanisms and strategies to overcome chemotherapy resistance in metastatic breast cancer. Cancer Treat Rev 34:378–390 19. Cusatis G, Sparreboom A (2008) Pharmacogenomic importance of ABCG2. Pharmacogenomics 9:1005–1009 20. Doyle LA, Ross DD (2003) Multidrug resistance mediated by the breast cancer resistance protein BCRP (ABCG2). Oncogene 22:7340–7358 21. Egorin MJ (2000) Adenosine triphosphatebinding cassette proteins and bioavailability: “we can pump you up (or out)”. J Natl Cancer Inst 92:1628–1629

283

22. Ejendal KF, Hrycyna CA (2002) Multidrug resistance and cancer: the role of the human ABC transporter ABCG2. Curr Protein Pept Sci 3:503–511 23. Glavinas H, Krajcsi P, Cserepes J, Sarkadi B (2004) The role of ABC transporters in drug resistance, metabolism and toxicity. Curr Drug Deliv 1:27–42 24. Guo Y, Lubbert M, Engelhardt M (2003) CD34- hematopoietic stem cells: current concepts and controversies. Stem Cells 21:15–20 25. Hadnagy A, Gaboury L, Beaulieu R, Balicki D (2006) SP analysis may be used to identify cancer stem cell populations. Exp Cell Res 312:3701–3710 26. Haimeur A, Conseil G, Deeley RG, Cole SP (2004) The MRP-related and BCRP/ ABCG2 multidrug resistance proteins: biology. substrate specificity and regulation. Curr Drug Metab 5:21–53 27. Han B, Zhang JT (2004) Multidrug resistance in cancer chemotherapy and xenobiotic protection mediated by the half ATP-binding cassette transporter ABCG2. Curr Med Chem Anticancer Agents 4:31–42 28. Hardwick LJ, Velamakanni S, van Veen HW (2007) The emerging pharmacotherapeutic significance of the breast cancer resistance protein (ABCG2). Br J Pharmacol 151:163–174 29. Hirose M, Hosoi E, Hamano S, Jalili A (2003) Multidrug resistance in hematological malignancy. J Med Invest 50:126–135 30. Huang Y (2007) Pharmacogenetics/genomics of membrane transporters in cancer chemotherapy. Cancer Metastasis Rev 26:183–201 31. Ishikawa T, Tamura A, Saito H, Wakabayashi K, Nakagawa H (2005) Pharmacogenomics of the human ABC transporter ABCG2: from functional evaluation to drug molecular design. Naturwissenschaften 92:451–463 32. Kim WS, Weickert CS, Garner B (2008) Role of ATP-binding cassette transporters in brain lipid transport and neurological disease. J Neurochem 104:1145–1166 33. Koshiba S, An R, Saito H et al (2008) Human ABC transporters ABCG2 (BCRP) and ABCG4. Xenobiotica 38:863–888 34. Krishnamurthy P, Schuetz JD (2005) The ABC transporter Abcg2/Bcrp: role in hypoxia mediated survival. Biometals 18:349–358 35. Krishnamurthy P, Schuetz JD (2006) Role of ABCG2/BCRP in biology and medicine. Annu Rev Pharmacol Toxicol 46:381–410 36. Krishnamurthy P, Xie T, Schuetz JD (2007) The role of transporters in cellular heme and porphyrin homeostasis. Pharmacol Ther 114:345–358

284

Ross and Nakanishi

37. Kuo MT (2007) Roles of multidrug resistance genes in breast cancer chemoresistance. Adv Exp Med Biol 608:23–30 38. Lage H, Dietel M (2000) Effect of the breastcancer resistance protein on atypical multidrug resistance. Lancet Oncol 1:169–175 39. Lazarowski A, Czornyj L, Lubienieki F et al (2007) ABC transporters during epilepsy and mechanisms underlying multidrug resistance in refractory epilepsy. Epilepsia 48(Suppl 5): 140–149 40. Lemos C, Jansen G, Peters GJ (2008) Drug transporters: recent advances concerning BCRP and tyrosine kinase inhibitors. Br J Cancer 98:857–862 41. Leonard GD, Fojo T, Bates SE (2003) The role of ABC transporters in clinical practice. Oncologist 8:411–424 42. Lepper ER, Nooter K, Verweij J et al (2005) Mechanisms of resistance to anticancer drugs: the role of the polymorphic ABC transporters ABCB1 and ABCG2. Pharmacogenomics 6:115–138 43. Lin T, Islam O, Heese K (2006) ABC transporters, neural stem cells and neurogenesis–a different perspective. Cell Res 16:857–871 44. Litman T, Druley TE, Stein WD, Bates SE (2001) From MDR to MXR: new understanding of multidrug resistance systems, their properties and clinical significance. Cell Mol Life Sci 58:931–959 45. Lou H, Dean M (2007) Targeted therapy for cancer stem cells: the patched pathway and ABC transporters. Oncogene 26:1357–1360 46. Mao Q (2008) BCRP/ABCG2 in the placenta: expression, function and regulation. Pharm Res 25:1244–1255 47. Marie JP (2001) Drug resistance in hematologic malignancies. Curr Opin Oncol 13: 463–469 48. Perez-Tomas R (2006) Multidrug resistance: retrospect and prospects in anti-cancer drug treatment. Curr Med Chem 13:1859–1876 49. Plasschaert SL, Van Der Kolk DM, De Bont ES et al (2004) Breast cancer resistance protein (BCRP) in acute leukemia. Leuk Lymphoma 45:649–654 50. Polgar O, Bates SE (2005) ABC transporters in the balance: is there a role in multidrug resistance? Biochem Soc Trans 33:241–245 51. Polgar O, Robey RW, Bates SE (2008) ABCG2: structure, function and role in drug response. Expert Opin Drug Metab Toxicol 4:1–15 52. Robey RW, Polgar O, Deeken J, To KW, Bates SE (2007) ABCG2: determining its

relevance in clinical drug resistance. Cancer Metastasis Rev 26:39–57 53. Ross DD (2000) Novel mechanisms of drug resistance in leukemia. Leukemia 14:467–473 54. Ross DD (2004) Modulation of drug resistance transporters as a strategy for treating myelodysplastic syndrome. Best Pract Res Clin Haematol 17:641–651 55. Ross DD (2007) Drug resistance transporters. In: Karp JE (ed) Acute myelogenous leukemia, Humana Press, Totowa, NJ, pp 163–173 56. Nakanishi T, Doyle LA, Hassel B et al (2003) Functional characterization of human breast cancer resistance protein (BCRP, ABCG2) expressed in the oocytes of Xenopus laevis. Mol Pharmacol 64:1452–1462 57. van Herwaarden AE, Jonker JW, Wagenaar E et al (2003) The breast cancer resistance protein (Bcrp1/Abcg2) restricts exposure to the dietary carcinogen 2-amino-1-methyl-6-phenylimidazo(4, 5-b)pyridine. Cancer Res 63:6447–6452 58. Elkind NB, Szentpetery Z, Apati A et al (2005) Multidrug transporter ABCG2 prevents tumor cell death induced by the epidermal growth factor receptor inhibitor Iressa (ZD1839, Gefitinib). Cancer Res 65:1770–1777 59. Nakamura Y, Oka M, Soda H et  al (2005) Gefitinib (“Iressa”, ZD1839), an epidermal growth factor receptor tyrosine kinase inhibitor, reverses breast cancer resistance protein/ ABCG2-mediated drug resistance. Cancer Res 65:1541–1546 60. Yanase K, Tsukahara S, Asada S et al (2004) Gefitinib reverses breast cancer resistance protein-mediated drug resistance. Mol Cancer Ther 3:1119–1125 61. Robey RW, Honjo Y, Morisaki K et al (2003) Mutations at amino-acid 482 in the ABCG2 gene affect substrate and antagonist specificity. Br J Cancer 89:1971–1978 62. Chen ZS, Robey RW, Belinsky MG et  al (2003) Transport of methotrexate, methotrexate polyglutamates, and 17beta-estradiol 17-(beta-D-glucuronide) by ABCG2: effects of acquired mutations at R482 on methotrexate transport. Cancer Res 63:4048–4054 63. Ozvegy-Laczka C, Hegedus T, Varady G et  al (2004) High-affinity interaction of tyrosine kinase inhibitors with the ABCG2 multidrug transporter. Mol Pharmacol 65:1485–1495 64. Burger H, Van Tol H, Boersma AW et  al (2004) Imatinib mesylate (STI571) is a substrate for the BCRP/ABCG2 drug pump. Blood 104:2940–2942

Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes 65. Jonker JW, Buitelaar M, Wagenaar E et  al (2002) The breast cancer resistance protein protects against a major chlorophyll-derived dietary phototoxin and protoporphyria. Proc Natl Acad Sci U S A 99:15649–15654 66. Krishnamurthy P, Ross DD, Nakanishi T et  al (2004) The stem cell marker Bcrp/ ABCG2 enhances hypoxic cell survival through interactions with heme. J Biol Chem 279:24218–24225 67. Suzuki M, Suzuki H, Sugimoto Y, Sugiyama Y (2003) ABCG2 transports sulfated conjugates of steroids and xenobiotics. J Biol Chem 278:22644–22649 68. Imai Y, Asada S, Tsukahara S et  al (2003) Breast cancer resistance protein exports sulfated estrogens but not free estrogens. Mol Pharmacol 64:610–618 69. Allen JD, van Loevezijn A, Lakhai JM et al (2002) Potent and specific inhibition of the breast cancer resistance protein multidrug transporter in vitro and in mouse intestine by a novel analogue of fumitremorgin C. Mol Cancer Ther 1:417–425 70. Kruijtzer CM, Beijnen JH, Rosing H et  al (2002) Increased oral bioavailability of topotecan in combination with the breast cancer resistance protein and P-glycoprotein inhibitor GF120918. J Clin Oncol 20:2943–2950 71. Breedveld P, Zelcer N, Pluim D et al (2004) Mechanism of the pharmacokinetic interaction between methotrexate and benzimidazoles: potential role for breast cancer resistance protein in clinical drug-drug interactions. Cancer Res 64:5804–5811 72. Goodell MA, Brose K, Paradis G, Conner AS, Mulligan RC (1996) Isolation and functional properties of murine hematopoietic stem cells that are replicating in vivo. J Exp Med 183:1797–1806 73. Zhou S, Schuetz JD, Bunting KD et  al (2001) The ABC transporter Bcrp1/ABCG2 is expressed in a wide variety of stem cells and is a molecular determinant of the side-population phenotype. Nat Med 7:1028–1034 74. Goodell MA, Rosenzweig M, Kim H et  al (1997) Dye efflux studies suggest that hematopoietic stem cells expressing low or undetectable levels of CD34 antigen exist in multiple species. Nat Med 3:1337–1345 75. Jonker JW, Freeman J, Bolscher E et  al (2005) Contribution of the ABC transporters Bcrp1 and Mdr1a/1b to the side population phenotype in mammary gland and bone marrow of mice. Stem Cells 23:1059–1065 76. Zhou S, Morris JJ, Barnes Y, Lan L, Schuetz JD, Sorrentino BP (2002) Bcrp1 gene

285

expression is required for normal numbers of side population stem cells in mice, and confers relative protection to mitoxantrone in hematopoietic cells in  vivo. Proc Natl Acad Sci U S A 99:12339–12344 77. Iida A, Saito S, Sekine A et al (2002) Catalog of 605 single-nucleotide polymorphisms (SNPs) among 13 genes encoding human ATP-binding cassette transporters: ABCA4, ABCA7, ABCA8, ABCD1, ABCD3, ABCD4, ABCE1, ABCF1, ABCG1, ABCG2, ABCG4, ABCG5, and ABCG8. J Hum Genet 47:285–310 78. Backstrom G, Taipalensuu J, Melhus H et al (2003) Genetic variation in the ATP-binding cassette transporter gene ABCG2 (BCRP) in a Swedish population. Eur J Pharm Sci 18:359–364 79. Honjo Y, Morisaki K, Huff LM et al (2002) Single-nucleotide polymorphism (SNP) analysis in the ABC half-transporter ABCG2 (MXR/BCRP/ABCP1). Cancer Biol Ther 1:696–702 80. Imai Y, Nakane M, Kage K et  al (2002) C421A polymorphism in the human breast cancer resistance protein gene is associated with low expression of Q141K protein and low-level drug resistance. Mol Cancer Ther 1:611–616 81. Zamber CP, Lamba JK, Yasuda K et al (2003) Natural allelic variants of breast cancer resistance protein (BCRP) and their relationship to BCRP expression in human intestine. Pharmacogenetics 13:19–28 82. Mizuarai S, Aozasa N, Kotani H (2004) Single nucleotide polymorphisms result in impaired membrane localization and reduced atpase activity in multidrug transporter ABCG2. Int J Cancer 109:238–246 83. Kobayashi D, Ieiri I, Hirota T et  al (2005) Functional assessment of ABCG2 (BCRP) gene polymorphisms to protein expression in human placenta. Drug Metab Dispos 33:94–101 84. Kondo C, Suzuki H, Itoda M et  al (2004) Functional analysis of SNPs variants of BCRP/ABCG2. Pharm Res 21:1895–1903 85. Vethanayagam RR, Wang H, Gupta A et  al (2005) Functional analysis of the human variants of breast cancer resistance protein: I206L, N590Y, and D620N. Drug Metab Dispos 33:697–705 86. Lee SS, Jeong HE, Yi JM et  al (2007) Identification and functional assessment of BCRP polymorphisms in a Korean population. Drug Metab Dispos 35:623–632 87. Tamura A, Wakabayashi K, Onishi Y et  al (2006) Genetic polymorphisms of human

286

Ross and Nakanishi

ABC transporter ABCG2: development of the standard method for functional validation of SNPs by using the Flp recombinase system. J Exp Ther Oncol 6:1–11 88. Tamura A, Wakabayashi K, Onishi Y et  al (2007) Re-evaluation and functional classification of non-synonymous single nucleotide polymorphisms of the human ATP-binding cassette transporter ABCG2. Cancer Sci 98:231–239 89. Poonkuzhali B, Lamba J, Strom S et al (2008) Association of breast cancer resistance protein/ABCG2 phenotypes and novel promoter and intron 1 single nucleotide polymorphisms. Drug Metab Dispos 36:780–795 90. Nakanishi T, Bailey-Dell KJ, Hassel BA et  al (2006) Novel 5¢ untranslated region variants of BCRP mRNA are differentially expressed in drug-selected cancer cells and in normal human tissues: implications for drug resistance, tissue-specific expression, and alternative promoter usage. Cancer Res 66:5007–5011 91. Jonker JW, Smit JW, Brinkhuis RF et  al (2000) Role of breast cancer resistance protein in the bioavailability and fetal penetration of topotecan. J Natl Cancer Inst 92:1651–1656 92. Sparreboom A, Gelderblom H, Marsh S et al (2004) Diflomotecan pharmacokinetics in relation to ABCG2 421C>A genotype. Clin Pharmacol Ther 76:38–44 93. Sparreboom A, Loos WJ, Burger H et  al (2005) Effect of ABCG2 genotype on the oral bioavailability of topotecan. Cancer Biol Ther 4:650–658 94. de Jong FA, Marsh S, Mathijssen RH et  al (2004) ABCG2 pharmacogenetics: ethnic differences in allele frequency and assessment of influence on irinotecan disposition. Clin Cancer Res 10:5889–5894 95. Nakatomi K, Yoshikawa M, Oka M et  al (2001) Transport of 7-ethyl-10-hydroxycamptothecin (SN-38) by breast cancer resistance protein ABCG2 in human lung cancer cells. Biochem Biophys Res Commun 288:827–832 96. Schellens JH, Maliepaard M, Scheper RJ et al (2000) Transport of topoisomerase I inhibitors by the breast cancer resistance protein. Potential clinical implications. Ann N Y Acad Sci 922:188–194 97. Han JY, Lim HS, Yoo YK et  al (2007) Associations of ABCB1, ABCC2, and ABCG2 polymorphisms with irinotecanpharmacokinetics and clinical outcome in patients with advanced non-small cell lung cancer. Cancer 110:138–147

98. Colombo S, Soranzo N, Rotger M et  al (2005) Influence of ABCB1, ABCC1, ABCC2, and ABCG2 haplotypes on the cellular exposure of nelfinavir in  vivo. Pharmacogenet Genomics 15:599–608 99. Lal S, Wong ZW, Sandanaraj E et al (2008) Influence of ABCB1 and ABCG2 polymorphisms on doxorubicin disposition in Asian breast cancer patients. Cancer Sci 99:816–823 100. Cusatis G, Gregorc V, Li J et  al (2006) Pharmacogenetics of ABCG2 and adverse reactions to gefitinib. J Natl Cancer Inst 98:1739–1742 101. Li J, Cusatis G, Brahmer J et  al (2007) Association of variant ABCG2 and the pharmacokinetics of epidermal growth factor receptor tyrosine kinase inhibitors in cancer patients. Cancer Biol Ther 6:432–438 102. Erdilyi DJ, Kamory E, Csokay B et al (2008) Synergistic interaction of ABCB1 and ABCG2 polymorphisms predicts the prevalence of toxic encephalopathy during anticancer chemotherapy. Pharmacogenomics J 8:321–327 103. Erdelyi DJ, Kamory E, Zalka A et al (2006) The role of ABC-transporter gene polymorphisms in chemotherapy induced immunosuppression, a retrospective study in childhood acute lymphoblastic leukaemia. Cell Immunol 244:121–124 104. Kim IS, Kim HG, Kim DC et  al (2008) ABCG2 Q141K polymorphism is associated with chemotherapy-induced diarrhea in patients with diffuse large B-cell lymphoma who received frontline rituximab plus cyclophosphamide/doxorubicin/vincristine/ prednisone chemotherapy. Cancer Sci 99:2496–2501 105. Adkison KK, Vaidya SS, Lee DY et al (2008) The ABCG2 C421A polymorphism does not affect oral nitrofurantoin pharmacokinetics in healthy Chinese male subjects. Br J Clin Pharmacol 66:233–239 106. Urquhart BL, Ware JA, Tirona RG et  al (2008) Breast cancer resistance protein (ABCG2) and drug disposition: intestinal expression, polymorphisms and sulfasalazine as an in vivo probe. Pharmacogenet Genomics 18:439–448 107. Yamasaki Y, Ieiri I, Kusuhara H et al (2008) Pharmacogenetic characterization of sulfasalazine disposition based on NAT2 and ABCG2 (BCRP) gene polymorphisms in humans. Clin Pharmacol Ther 84:95–103 108. Korenaga Y, Naito K, Okayama N et  al (2005) Association of the BCRP C421A

Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes polymorphism with nonpapillary renal cell carcinoma. Int J Cancer 117:431–434 109. Hu LL, Wang XX, Chen X et al (2007) BCRP gene polymorphisms are associated with susceptibility and survival of diffuse large B-cell lymphoma. Carcinogenesis 28:1740–1744 110. Gardner ER, Ahlers CM, Shukla S et  al (2008) Association of the ABCG2 C421A polymorphism with prostate cancer risk and survival. BJU Int 102:1694–1699 111. Hahn NM, Marsh S, Fisher W et  al (2006) Hoosier Oncology Group randomized phase II study of docetaxel, vinorelbine, and estramustine in combination in hormone-refractory prostate cancer with pharmacogenetic survival analysis. Clin Cancer Res 12:6094–6099 112. Muller PJ, Dally H, Klappenecker CN et  al (2009) Polymorphisms in ABCG2, ABCC3 and CNT1 genes and their possible impact on chemotherapy outcome of lung cancer patients. Int J Cancer 124:1669–1674 113. Honjo Y, Hrycyna CA, Yan QW et al (2001) Acquired mutations in the MXR/BCRP/ ABCP gene alter substrate specificity in MXR/BCRP/ABCP-overexpressing cells. Cancer Res 61:6635–6639 114. Allen JD, Van Dort SC, Buitelaar M, van Tellingen O, Schinkel AH (2003) Mouse breast cancer resistance protein (Bcrp1/Abcg2) mediates etoposide resistance and transport, but etoposide oral availability is limited primarily by P-glycoprotein. Cancer Res 63:1339–1344 115. Wang X, Nitanda T, Shi M et  al (2004) Induction of cellular resistance to nucleoside reverse transcriptase inhibitors by the wildtype breast cancer resistance protein. Biochem Pharmacol 68:1363–1370 116. Nakanishi T, Karp JE, Tan M et  al (2003) Quantitative analysis of breast cancer resistance protein and cellular resistance to flavopiridol in acute leukemia patients. Clin Cancer Res 9:3320–3328 117. Plasschaert SL, van der Kolk DM, de Bont ES et  al (2003) The role of breast cancer resistance protein in acute lymphoblastic leukemia. Clin Cancer Res 9:5171–5177 118. Suvannasankha A, Minderman H, O’Loughlin KL et al (2004) Breast cancer resistance protein (BCRP/MXR/ABCG2) in acute myeloid leukemia: discordance between expression and function. Leukemia 18:1252–1257 119. Ross DD, Karp JE, Chen TT, Doyle LA (2000) Expression of breast cancer resistance protein in blast cells from patients with acute leukemia. Blood 96:365–368 120. Steinbach D, Sell W, Voigt A et  al (2002) BCRP gene expression is associated with a

287

poor response to remission induction therapy in childhood acute myeloid leukemia. Leukemia 16:1443–1447 121. van den Heuvel-Eibrink MM, Wiemer EA et  al (2002) Increased expression of the breast cancer resistance protein (BCRP) in relapsed or refractory acute myeloid leukemia (AML). Leukemia 16:833–839 122. Abbott BL, Colapietro AM, Barnes Y et  al (2002) Low levels of ABCG2 expression in adult AML blast samples. Blood 100: 4594–4601 123. Benderra Z, Faussat AM, Sayada L et  al (2004) Breast cancer resistance protein and P-glycoprotein in 149 adult acute myeloid leukemias. Clin Cancer Res 10:7896–7902 124. Benderra Z, Faussat AM, Sayada L et al (2005) MRP3, BCRP, and P-glycoprotein activities are prognostic factors in adult acute myeloid leukemia. Clin Cancer Res 11:7764–7772 125. Uggla B, Stahl E, Wagsater D et  al (2005) BCRP mRNA expression v. clinical outcome in 40 adult AML patients. Leuk Res 29:141–146 126. Sargent JM, Williamson CJ, Maliepaard M et al (2001) Breast cancer resistance protein expression and resistance to daunorubicin in blast cells from patients with acute myeloid leukaemia. Br J Haematol 115:257–262 127. van der Kolk DM, Vellenga E, Scheffer GL et al (2002) Expression and activity of breast cancer resistance protein (BCRP) in de novo and relapsed acute myeloid leukemia. Blood 99:3763–3770 128. van der Pol MA, Broxterman HJ, Pater JM et al (2003) Function of the ABC transporters, P-glycoprotein, multidrug resistance protein and breast cancer resistance protein, in minimal residual disease in acute myeloid leukemia. Haematologica 88:134–147 129. Xie Y, Xu K, Linn DE et  al (2008) The 44-kDa Pim-1 kinase phosphorylates BCRP/ ABCG2 and thereby promotes its multimerization and drug-resistant activity in human prostate cancer cells. J Biol Chem 283:3349–3356 130. Mogi M, Yang J, Lambert JF et  al (2003) Akt signaling regulates side population cell phenotype via Bcrp1 translocation. J Biol Chem 278:39068–39075 131. Nakanishi T, Shiozawa K, Hassel BA, Ross DD (2006) Complex interaction of BCRP/ ABCG2 and imatinib in BCR-ABL-expressing cells: BCRP-mediated resistance to imatinib is attenuated by imatinib-induced reduction of BCRP expression. Blood 108:678–684 132. Damiani D, Tiribelli M, Calistri E et  al (2006) The prognostic value of P-glycoprotein

288

Ross and Nakanishi

(ABCB) and breast cancer resistance protein (ABCG2) in adults with de novo acute myeloid leukemia with normal karyotype. Haematologica 91:825–828 133. de Figueiredo-Pontes LL, Pintao MC, Oliveira LC et  al (2008) Determination of P-glycoprotein, MDR-related protein 1, breast cancer resistance protein, and lungresistance protein expression in leukemic stem cells of acute myeloid leukemia. Cytometry B 74B:163–168 134. Galimberti S, Guerrini F, Palumbo GA et al (2004) Evaluation of BCRP and MDR-1 coexpression by quantitative molecular assessment in AML patients. Leuk Res 28:367–372 135. Ho MM, Hogge DE, Ling V (2008) MDR1 and BCRP1 expression in leukemic progenitors correlates with chemotherapy response in acute myeloid leukemia. Exp Hematol 36:433–442 136. van den Heuvel-Eibrink MM, van der Holt B, Burnett AK et  al (2007) CD34-related coexpression of MDR1 and BCRP indicates a clinically resistant phenotype in patients with acute myeloid leukemia (AML) of older age. Ann Hematol 86:329–337 137. Ho RH, Choi L, Lee W et al (2007) Effect of drug transporter genotypes on pravastatin disposition in European- and AfricanAmerican participants. Pharmacogenet Genomics 17:647–656 138. Raaijmakers MH, de Grouw EP, Heuver LH et al (2005) Breast cancer resistance protein in drug resistance of primitive CD34+38cells in acute myeloid leukemia. Clin Cancer Res 11:2436–2444 139. Shman TV, Fedasenka UU, Savitski VP, Aleinikova OV (2008) CD34+ leukemic subpopulation predominantly displays lower spontaneous apoptosis and has higher expression levels of Bcl-2 and MDR1 genes than CD34- cells in childhood AML. Ann Hematol 87:353–360 140. Vardiman JW, Harris NL, Brunning RD (2002) The World Health Organization (WHO) classification of the myeloid neoplasms. Blood 100:2292–2302 141. Sauerbrey A, Sell W, Steinbach D, Voigt A, Zintl F (2002) Expression of the BCRP gene (ABCG2/MXR/ABCP) in childhood acute lymphoblastic leukaemia. Br J Haematol 118:147–150 142. Suvannasankha A, Minderman H, O’Loughlin KL et al (2004) Breast cancer resistance protein (BCRP/MXR/ABCG2) in adult acute lymphoblastic leukaemia: frequent expression and possible correlation with shorter

disease-free survival. Br J Haematol 127:392–398 143. Stam RW, van den Heuvel-Eibrink MM, den Boer ML et  al (2004) Multidrug resistance genes in infant acute lymphoblastic leukemia: Ara-C is not a substrate for the breast cancer resistance protein. Leukemia 18:78–83 144. Kourti M, Vavatsi N, Gombakis N et  al (2007) Expression of multidrug resistance 1 (MDR1), multidrug resistance-related protein 1 (MRP1), lung resistance protein (LRP), and breast cancer resistance protein (BCRP) genes and clinical outcome in childhood acute lymphoblastic leukemia. Int J Hematol 86:166–173 145. Graham SM, Jorgensen HG, Allan E et  al (2002) Primitive, quiescent, Philadelphiapositive stem cells from patients with chronic myeloid leukemia are insensitive to STI571 in vitro. Blood 99:319–325 146. Breedveld P, Pluim D, Cipriani G et al (2005) The effect of Bcrp1 (Abcg2) on the in vivo pharmacokinetics and brain penetration of imatinib mesylate (Gleevec): implications for the use of breast cancer resistance protein and P-glycoprotein inhibitors to enable the brain penetration of imatinib in patients. Cancer Res 65:2577–2582 147. Burger H, Nooter K (2004) Pharmacokinetic resistance to imatinib mesylate: role of the ABC drug pumps ABCG2 (BCRP) and ABCB1 (MDR1) in the oral bioavailability of imatinib. Cell Cycle 3:1502–1505 148. Burger H, van Tol H, Brok M et al (2005) Chronic imatinib mesylate exposure leads to reduced intracellular drug accumulation by induction of the ABCG2 (BCRP) and ABCB1 (MDR1) drug transport pumps. Cancer Biol Ther 4:747–752 149. Houghton PJ, Germain GS, Harwood FC et  al (2004) Imatinib mesylate is a potent inhibitor of the ABCG2 (BCRP) transporter and reverses resistance to topotecan and SN-38 in vitro. Cancer Res 64:2333–2337 150. Brendel C, Scharenberg C, Dohse M et  al (2007) Imatinib mesylate and nilotinib (AMN107) exhibit high-affinity interaction with ABCG2 on primitive hematopoietic stem cells. Leukemia 21:1267–1275 151. Jordanides NE, Jorgensen HG, Holyoake TL, Mountford JC (2006) Functional ABCG2 is overexpressed on primary CML CD34+ cells and is inhibited by imatinib mesylate. Blood 108:1370–1373 152. Jiang X, Zhao Y, Smith C et  al (2007) Chronic myeloid leukemia stem cells possess multiple unique features of resistance to

Impact of Breast Cancer Resistance Protein on Cancer Treatment Outcomes BCR-ABL targeted therapies. Leukemia 21:926–935 153. White DL, Saunders VA, Dang P et al (2007) Most CML patients who have a suboptimal response to imatinib have low OCT-1 activity: higher doses of imatinib may overcome the negative impact of low OCT-1 activity. Blood 110:4064–4072 154. Wang L, Giannoudis A, Lane S et al (2008) Expression of the uptake drug transporter hOCT1 is an important clinical determinant of the response to imatinib in chronic myeloid leukemia. Clin Pharmacol Ther 83: 258–264 155. Hiwase DK, Saunders V, Hewett D et  al (2008) Dasatinib cellular uptake and efflux in chronic myeloid leukemia cells: therapeutic implications. Clin Cancer Res 14:3881–3888 156. Scheffer GL, Maliepaard M, Pijnenborg AC et al (2000) Breast cancer resistance protein is localized at the plasma membrane in mitoxantrone- and topotecan-resistant cell lines. Cancer Res 60:2589–2593 157. Diestra JE, Scheffer GL, Catala I et al (2002) Frequent expression of the multi-drug resistance-associated protein BCRP/MXR/ ABCP/ABCG2 in human tumours detected by the BXP-21 monoclonal antibody in paraffin-embedded material. J Pathol 198:213–219 158. Galimberti S, Nagy B, Benedetti E et  al (2007) Evaluation of the MDR1, ABCG2, Topoisomerases IIalpha and GSTpi gene expression in patients affected by aggressive mantle cell lymphoma treated by the R-Hyper-CVAD regimen. Leuk Lymphoma 48:1502–1509 159. Saglam A, Hayran M, Uner AH (2008) Immunohistochemical expression of multidrug resistance proteins in mature T/NK-cell lymphomas. Apmis 116:791–800 160. Raaijmakers MH, de Grouw EP, Heuver LH et al (2005) Impaired breast cancer resistance protein mediated drug transport in plasma cells in multiple myeloma. Leuk Res 29:1455–1458 161. Turner JG, Gump JL, Zhang C et al (2006) ABCG2 expression, function, and promoter methylation in human multiple myeloma. Blood 108:3881–3889 162. Vander Borght S, van Pelt J, van Malenstein H et  al (2008) Up-regulation of breast cancer resistance protein expression in hepatoblastoma following chemotherapy: A study in patients and in vitro. Hepatol Res 38:1112–1121 163. Bart J, Hollema H, Groen HJ et  al (2004) The distribution of drug-efflux pumps, P-gp, BCRP, MRP1 and MRP2, in the normal

289

blood-testis barrier and in primary testicular tumours. Eur J Cancer 40:2064–2070 164. Tsunoda S, Okumura T, Ito T et  al (2006) ABCG2 expression is an independent unfavorable prognostic factor in esophageal squamous cell carcinoma. Oncology 71:251–258 165. Kawabata S, Oka M, Soda H et  al (2003) Expression and functional analyses of breast cancer resistance protein in lung cancer. Clin Cancer Res 9:3052–3057 166. Yoh K, Ishii G, Yokose T et al (2004) Breast cancer resistance protein impacts clinical outcome in platinum-based chemotherapy for advanced non-small cell lung cancer. Clin Cancer Res 10:1691–1697 167. Ota S, Ishii G, Goto K et  al (2009) Immunohistochemical expression of BCRP and ERCC1 in biopsy specimen predicts survival in advanced non-small-cell lung cancer treated with cisplatin-based chemotherapy. Lung Cancer 64:98–104 168. Kim YH, Ishii G, Goto K et  al (2009) Expression of breast cancer resistance protein is associated with a poor clinical outcome in patients with small-cell lung cancer. Lung Cancer 65:105–111 169. Surowiak P, Pawelczyk K, Maciejczyk A et al (2008) Positive correlation between cyclooxygenase 2 and the expression of ABC transporters in non-small cell lung cancer. Anticancer Res 28:2967–2974 170. Chen YC, Hsu HS, Chen YW et  al (2008) Oct-4 expression maintained cancer stemlike properties in lung cancer-derived CD133-positive cells. PLoS ONE 3:e2637 171. Candeil L, Gourdier I, Peyron D et al (2004) ABCG2 overexpression in colon cancer cells resistant to SN38 and in irinotecan-treated metastases. Int J Cancer 109:848–854 172. Gupta N, Martin PM, Miyauchi S et  al (2006) Down-regulation of BCRP/ABCG2 in colorectal and cervical cancer. Biochem Biophys Res Commun 343:571–577 173. Glasgow SC, Yu J, Carvalho LP et al (2005) Unfavourable expression of pharmacologic markers in mucinous colorectal cancer. Br J Cancer 92:259–264 174. Nakayama K, Kanzaki A, Ogawa K et  al (2002) Copper-transporting P-type adenosine triphosphatase (ATP7B) as a cisplatin based chemoresistance marker in ovarian carcinoma: comparative analysis with expression of MDR1, MRP1, MRP2, LRP and BCRP. Int J Cancer 101:488–495 175. Diestra JE, Condom E, Del Muro XG et al (2003) Expression of multidrug resistance proteins P-glycoprotein, multidrug resistance

290

Ross and Nakanishi

protein 1, breast cancer resistance protein and lung resistance related protein in locally advanced bladder cancer treated with neoadjuvant chemotherapy: biological and clinical implications. J Urol 170:1383–1387 176. Valera ET, Lucio-Eterovic AK, Neder L et al (2007) Quantitative PCR analysis of the expression profile of genes related to multiple drug resistance in tumors of the central nervous system. J Neurooncol 85:1–10 177. Wilson MW, Fraga CH, Fuller CE et al (2006) Immunohistochemical detection of multidrug-resistant protein expression in retinoblastoma treated by primary enucleation. Invest Ophthalmol Vis Sci 47:1269–1273 178. Deichmann M, Thome M, Egner U, Hartschuh W, Kurzen H (2005) The chemoresistance gene ABCG2 (MXR/ BCRP1/ABCP1) is not expressed in melanomas but in single neuroendocrine carcinomas of the skin. J Cutan Pathol 32:467–473 179. La Porta C (2009) Cancer stem cells: lessons from melanoma. Stem Cell Rev 5:61–65 180. Monzani E, Facchetti F, Galmozzi E et  al (2007) Melanoma contains CD133 and ABCG2 positive cells with enhanced tumourigenic potential. Eur J Cancer 43:935–946 181. Kanzaki A, Toi M, Nakayama K et al (2001) Expression of multidrug resistance-related transporters in human breast carcinoma. Jpn J Cancer Res 92:452–458 182. Faneyte IF, Kristel PM, Maliepaard M et  al (2002) Expression of the breast cancer resistance protein in breast cancer. Clin Cancer Res 8:1068–1074 183. Burger H, Foekens JA, Look MP et al (2003) RNA expression of breast cancer resistance protein, lung resistance-related protein, multidrug resistance-associated proteins 1 and 2, and multidrug resistance gene 1 in breast cancer: correlation with chemotherapeutic response. Clin Cancer Res 9:827–836 184. Park S, Shimizu C, Shimoyama T et al (2006) Gene expression profiling of ATP-binding

cassette (ABC) transporters as a predictor of the pathologic response to neoadjuvant chemotherapy in breast cancer patients. Breast Cancer Res Treat 99:9–17 185. Ee PL, Kamalakaran S, Tonetti D et al (2004) Identification of a novel estrogen response element in the breast cancer resistance protein (ABCG2) gene. Cancer Res 64:1247–1251 186. Hirschmann-Jax C, Foster AE, Wulf GG et al (2004) A distinct “side population” of cells with high drug efflux capacity in human tumor cells. Proc Natl Acad Sci U S A 101:14228–14233 187. Chiou SH, Yu CC, Huang CY et al (2008) Positive correlations of Oct-4 and Nanog in oral cancer stem-like cells and high-grade oral squamous cell carcinoma. Clin Cancer Res 14:4085–4095 188. Susanto J, Lin YH, Chen YN et  al (2008) Porphyrin homeostasis maintained by ABCG2 regulates self-renewal of embryonic stem cells. PLoS ONE 3:e4023 189. Patrawala L, Calhoun T, SchneiderBroussard R et al (2005) Side population is enriched in tumorigenic, stem-like cancer cells, whereas ABCG2+ and ABCG2- cancer cells are similarly tumorigenic. Cancer Res 65:6207–6219 190. Wilson CS, Davidson GS, Martin SB et  al (2006) Gene expression profiling of adult acute myeloid leukemia identifies novel biologic clusters for risk classification and outcome prediction. Blood 108:685–696 191. Saraiya B, Gounder M, Dutta J et al (2008) Sequential topoisomerase targeting and analysis of mechanisms of resistance to topotecan in patients with acute myelogenous leukemia. Anticancer Drugs 19:411–420 192. Fazeny-Dorner B, Piribauer M, Wenzel C et  al (2003) Cytogenetic and comparative genomic hybridization findings in four cases of breast cancer after neoadjuvant chemotherapy. Cancer Genet Cytogenet 146: 161–166

Chapter 13 Drug Ratio-Dependent Antagonism: A New Category of Multidrug Resistance and Strategies for Its Circumvention Troy O. Harasym, Barry D. Liboiron, and Lawrence D. Mayer Abstract A newly identified form of multidrug resistance (MDR) in tumor cells is presented, pertaining to the commonly encountered resistance of cancer cells to anticancer drug combinations at discrete drug:drug ratios. In vitro studies have revealed that whether anticancer drug combinations interact synergistically or antagonistically can depend on the ratio of the combined agents. Failure to control drug ratios in vivo due to uncoordinated pharmacokinetics could therefore lead to drug resistance if tumor cells are exposed to antagonistic drug ratios. Consequently, the most efficacious drug combination may not occur at the typically employed maximum tolerated doses of the combined drugs if this leads to antagonistic ratios in vivo after administration and resistance to therapeutic effects of the drug combination. Our approach to systematically screen a wide range of drug ratios and concentrations and encapsulate the drug combination in a liposomal delivery vehicle at identified synergistic ratios represents a means to mitigate this drug ratio-dependent MDR mechanism. The in vivo efficacy of the improved agents (CombiPlex formulations) is demonstrated and contrasted with the decreased efficacy when drug combinations are exposed to tumor cells in vivo at antagonistic ratios. Key words: Multidrug resistance, Synergy, Antagonism, Ratiometric, Drug delivery, Liposomes, Drug screening, Median effect analysis

1. Introduction Combination chemotherapy has been the cornerstone of cancer therapy for over 40 years. Improvements in outcomes for childhood leukemia highlight how development of combination treatments has led to dramatic increases in efficacy over single agents. From response rates of 40% and no cures with methotrexate alone, greater than 95% complete response and 75–80% cure rates could be achieved when methotrexate was administered in combination with J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_13, © Humana Press, a part of Springer Science + Business Media, LLC 2010

291

292

Harasym, Liboiron, and Mayer

asparaginase, daunorubicin, and cytarabine (1, 2). Yet in contrast to these advancements in leukemia, achieving a cure for the majority of solid tumors remains elusive (3). Response rates of pancreatic, esophageal, and recurrent ovarian cancer are still below 20% despite the massive efforts and resources that have been expended to develop superior combination therapies for these indications (4). The failure of many combinations to achieve complete remissions in the majority of cases is frequently attributed to multidrug resistance (MDR), a phenomenon that allows tumor cells to survive and/or flourish under considerable challenge from exogenous cytotoxic agents. Much attention has been paid to identified resistance factors such as P-glycoprotein (Pgp) and multidrug resistance-associated protein (MRP), altered apoptosis mechanisms as well as modifications in enzyme activity (e.g., topoisomerase II, glutathione-Stransferase) in an attempt to elucidate and mitigate these mechanisms in order to improve efficacy (5–8). Efforts to specifically target and neutralize these mechanisms, however, have been largely ineffective (9, 10). Thus, development of new treatments, identification of synergistic drug interactions, and refinement of treatment protocols remain the main strategies for mitigating the effects of MDR and improving the treatment outcomes of cancer patients.

2. Drug–Drug Antagonism: A New Form of MDR

The discovery of favorable drug–drug interactions in combination chemotherapy remains an active field of basic and clinical research. Some researchers have successfully combined agents with different mechanisms of action in which multiple sites in biochemical pathways are attacked, resulting in synergy (11), while others have demonstrated synergy by combining agents that target the same pathway(s) (12–14). Identification and characterization of synergistic drug interactions therefore remains a very active area of research and has resulted in the introduction of several new combination chemotherapies in the past decade (15–18). Intrinsic in the discovery of drug:drug synergy, yet often ignored in these studies, is the identification of drug:drug antagonism, in which the efficacy of the two agents is impaired such that the cytotoxicity of the combined agents is less than what would be expected for the additive activities of the individual agents (19, 20). Extolling the discovery of drug–drug synergy while ignoring the existence of drug–drug antagonism under certain conditions for a given combination can, in fact, lead to compromised efficacy. Exposure of tumor cells to two drugs in combination at a certain ratio and concentration can lead to one of three outcomes: synergistic,

Drug Ratio-Dependent Antagonism

293

additive, or antagonistic activity. Thus, while the literature is quick to identify drug combinations that have augmented activity in vitro, little attention is paid to whether the conditions used in  vitro (e.g., drug ratio and concentration) are pharmacologically relevant in  vivo. Consequently, when synergy is observed, this represents ratios at which tumor cells are susceptible to the combination, but historically identification and avoidance of antagonistic ratios at which tumor cells are resistant to the combined agents is largely ignored. From an empirical perspective, drug:drug antagonism is a form of MDR in that the cellular response is inexplicably less than what would be minimally expected (i.e., additivity); it affects multiple structurally disparate chemotherapeutic agents and it is known to occur across a wide variety of tumor types. For reasons likely related to the interconnected nature of pathway biochemistry and cellular biology, a particular cell line may be less susceptible to a specific drug combination presented at a certain ratio, yet this resistance mechanism can frequently be bypassed through exposure of the same drugs at a different ratio. Further complicating this phenomenon is that some studies reveal that activity of a particular drug ratio may also be concentration-dependent; exhibiting synergy at one total drug concentration and producing additive or antagonistic results at other concentrations. Since drug concentrations decrease over time after in vivo administration, it is important to identify and utilize ratios of drugs that behave synergistically over a broad range of drug concentrations, while avoiding those combinations that exhibit either broad ratio or concentration-specific antagonism. While these ratios can be readily identified through a variety of drug screening techniques, the translation of such information from in  vitro cytotoxicity studies to in  vivo efficacy studies is difficult to achieve using the current treatment paradigm for antitumor combination chemotherapy. 2.1. Current Combination Therapy Development

The empiric process used to advance new combinations in the clinic has evolved little since the concept of combination chemotherapy was pioneered by Frei and coworkers in the 1960s (21). In this process, individual drugs in a combination are escalated to the maximum dose where aggregate toxicity is tolerable with the expectation that maximum therapeutic activity will be achieved at the maximum dose of each agent. Conventional combination therapy, however, frequently fails to account for disparate physiochemical properties of each drug component that will typically result in the rapid and independent distribution and elimination of each agent. Without a means to control the pharmacokinetics (PKs) of each drug, their unique clearance properties will lead to rapidly changing and uncontrolled drug:drug ratios after administration. Therefore, over a

294

Harasym, Liboiron, and Mayer

short period of time, combinations administered as an unbridled cocktail can be present at synergistic, additive, and/or antagonistic ratios if the combination exhibits drug ratio-dependent interactions. Consequently, failure to account for and control drug ratios in the application of drug combinations in vivo could lead to the exposure of tumor cells to antagonistic drug ratios, leading to empirical MDR and corresponding loss of therapeutic activity. Encouragingly, the role of synergistic and antagonistic drug interactions in the mitigation and emergence of MDR is gaining interest in the current literature for a variety of disease states (22–26). 2.1.1. Case Study: Gemcitabine and Cisplatin Pharmacokinetics

The combination of gemcitabine and cisplatin is instructive in highlighting the changes in plasma concentrations and drug ratios that occur following coadministration of the two drugs, using pharmacokinetic parameters presented in Table 13.1. Gemcitabine is infused first over a period of 30 min at a total dose of 1,250 mg/ m2, followed by a 1-h infusion of saline to minimize renal toxicity from cisplatin. Cisplatin is then administered at 100 mg/m2 in 1 L of saline over a 1-h infusion time. The average terminal half-lives of gemcitabine and cisplatin are 69 and 33  min, respectively (product monographs; Eli Lilly 1999, Mayne Pharma, 2003), which represents a 2.1-fold difference in plasma elimination rates. Using the aforementioned PK parameters, volumes of distribution (Table 13.1) and accounting for the differences in both administration start times and durations between gemcitabine and cisplatin, the drug concentrations and associated drug:drug ratios can be estimated. Upon completion of the cisplatin infusion,

Table 13.1 A typical treatment regimen using the combination of cisplatin and gemcitabine for advanced non-small cell lung cancer (NSCLC) Volume of distribution (L/m2)

Terminal half-life (~min)b

50

42–96

41 Prehydrate with 1,000 mL NS over 60 min then cisplatin IV in 1,000 mL NS over 60 min.

20–45

Drug dosea (mg/m2/day)

Administration

Gemcitabine

1,250 on days 1 and 8

IV in 250 mL of NS over 30 min.

Cisplatin

100 on day 1

Drug

Note: Gemcitabine is administered first a Drug dose and treatment schedule from BC Cancer Agency chemotherapy protocol (LUAVPG) available online at http://www.bccancer.bc.ca/HPI/ChemotherapyProtocols/Lung/default.htm b From BC Cancer Agency drug manual and Gemzar and Platinol-AQ product monographs

Drug Ratio-Dependent Antagonism

295

the respective gemcitabine and cisplatin plasma concentrations are 20 and 2 mM, reflecting a 10:1 molar ratio. At 30 min after completion of the cisplatin infusion, only 50% of the postdistribution cisplatin concentration (~1 mM) and 72% of the postdistribution gemcitabine concentration (~14 mM) is expected to remain in the plasma. Hence, the rapidly decreasing plasma drug concentrations will quickly lead to levels that fall below effective cellular cytotoxic values for cisplatin (e.g., the concentration of drug required to kill 50% of target cells in vitro, IC50 value, for gemcitabine and cisplatin in a A549 human non-small cell lung cancer cell line is 0.007 and 4.2  mM, respectively), highlighting that increased drug elimination results in decreased cell kill fractions. Furthermore, given that approximately 50% of the cisplatin and 28% of the gemcitabine is eliminated from the plasma every 30  min postadministration, the gemcitabine:cisplatin ratio increases by 1.44-fold over that time period. Specifically, at t = 0 the molar ratio of gemcitabine:cisplatin is approximately 10:1, whereas after 30 and 60  min, the gemcitabine:cisplatin molar ratio would increase to approximately 14:1 and 21:1, respectively, reflecting a doubling of the original gemcitabine:cisplatin ratio. By 4  h, the gemcitabine:cisplatin ratio will be approximately 185:1, an increase nearly 20-fold from the starting drug ratio. In view of the evidence of drug ratio-dependent synergy in vitro and in  vivo for several drug combinations (13, 14, 27–29), careful consideration should be given to ensure that the in vitro methodologies utilized to evaluate drug combinations for synergy take these pharmacological properties into account. Failure to do so could result in concluding that a given drug combination is synergistic when, in fact, drug ratios and concentrations reflecting those exposed to tumor cells in vivo could be highly antagonistic and susceptible to drug ratio-dependent MDR.

3. In Vivo Avoidance of Drug RatioDependent MDR: The CombiPlex Approach

We have developed an approach in which we identify antagonistic drug:drug ratios for a particular drug combination in vitro and subsequently package the two drugs in a drug delivery vehicle such that a concentration-independent synergistic ratio of the two drugs is delivered directly to the tumor site and antagonistic ratios are avoided. Subsequent in vivo efficacy studies are used to confirm the translation of in vitro drug synergy relationships to actual efficacy improvements over the individual agents, the free drug cocktail and, most importantly, significant increases in efficacy over the two drugs delivered in a vehicle at an antagonistic ratio.

296

Harasym, Liboiron, and Mayer

This review will detail the most common algorithms used to assess ratio-dependent drug:drug synergy/antagonism as well as methods to fix desirable drug ratios in vivo and evaluate the efficacy of fixed ratio combination agents in a variety of tumor xenografts. This ratiometric dosing approach to combination chemotherapy represents a means to augment the in vivo efficacy of combination therapies by maximizing tumor exposure to drugs at their synergistic ratios and avoidance of antagonistic ratios at which the ratio-dependent MDR mechanisms are active. 3.1. Drug Synergy/ Antagonism Screening: Experimental Design Considerations

The most efficient and accurate means to evaluate drug interactions is through in vitro cytotoxicity assays, and several approaches can be taken when structuring how agents are combined experimentally in order to evaluate drug combinations for synergy in vitro. In general, these approaches evaluate drug combinations based upon (1) a nonconstant ratio design where the drug concentrations are chosen arbitrarily based on the features such as the relative in vitro antitumor potencies of the individual agents or plasma concentrations achieved clinically (30, 31), or (2) constant ratio designs where drug concentrations are chosen based on an equipotent activity (i.e., the ratio of the IC50s) and serial dilutions are prepared to obtain a dose–effect (or concentration) range for a given ratio (32, 33). A list of methods used to evaluate drug interactions is found in Table  13.2. Clearly, considerable effort has been given to accurately quantify drug:drug synergy; however, this review will describe the methods most commonly employed in the literature. The most common nonconstant ratio approach to evaluate drug combinations for synergy employs a checkerboard drug combination design. In this method, the drug concentrations and drug ratios are varied. For example, in a 7 × 7 checkerboard layout for a two-drug combination each drug is diluted to generate seven different concentrations. Ideally each individual drug dilution will provide concentrations that provide the full range of tumor cell growth inhibition (e.g., £20% to ³90%). In this design, drug A is diluted vertically and drug B is diluted horizontally in a standard multi-well plate format and combining the two dilution groups results in 49 distinct combinations of the two drugs in addition to the seven concentrations of each individual drug. For most two-drug combinations, basing the dilutions on drug concentrations that span the entire cytotoxicity curve for the individual agents will lead to a rather ad hoc assortment of drug ratios and effective concentrations. A major disadvantage to the checkerboard approach is the large number of fixed ratios that are evaluated at a limited number of effect levels (concentrations). For example, applying the 7 × 7 matrix design described previously to gemcitabine and cisplatin tested in the A549 non-small cell lung cancer cell line results in

Drug Ratio-Dependent Antagonism

297

Table 13.2 Various drug combination interaction methods used for evaluating synergy (adapted from (37)) Evaluation models

References

Isobologram (1870)

(84, 85)

Loewe additivity (1926)

(86–89)

Bliss independence response surface approach (1939)

(90)

Fractional product method of Webb (1963)

(91)

Multivariate linear logistic model (1970)

(92)

Approach of Gessner (1974)

(93)

Method of Valeriote and Lin (1975)

(94)

Method of Drewinko et al. (1976)

(95)

Interaction index calculation of Berenbaum (1977) (35) Method of Steel and Peckman (1979)

(36)

Median-effect method of Chou and Talalay (1984) (41) Method of Berenbaum (1985)

(96)

Method of Greco and Lawrence (1988)

(97)

Method of Pritchard and Shipman (1990)

(98–100)

Bivariate spline fitting (Sühnel 1990)

(101)

Models of Greco et al. (1990)

(38, 102, 103)

Models of Weinstein et al. (1990)

(39)

33 different gemcitabine:cisplatin molar ratios spanning a range from 1:5 to 1:100,000 (Table 13.3). Note that drug concentrations were chosen such that each drug spans its own dose–response curve, from low cell growth inhibition to high cell growth inhibition. The IC50 values for gemcitabine and cisplatin in A549 cells are 0.007 and 4.2 mM, respectively, a difference of approximately 600-fold. At an IC50 matched ratio (1:600) the nearest ratios evaluated by this checkerboard design are 1:500 and 1:750, indicating that only a few ratios are tested near the steepest and most sensitive region of the dose–response curves (i.e., at the IC50 value). Further, of the 33 ratios generated, only 8 are evaluated at more than 1 effect level or concentration (gemcitabine:cisplatin ratios of 1:50, 1:100, 1:200, 1:500, 1:1,000, 1:2,000, 1:5,000, and 1:10,000), with the 1:500 ratio being evaluated at a maximum of only 4 effect levels. Therefore, using a conventional checkerboard

1:100

0.001 mM

0.002 mM

0.005 mM

0.008 mM

0.01 mM

0.02 mM

88.7%

66.9%

36.2%

29.8%

20.9%a

1:25

1:50

1:62.5

1:100

1:250

1:500

1:1,000

0.5 mM

96.5%

1:50

1:100

1:125

1:200

1:500

1:1,000

1:2,000

1.0 mM

77.5%

1:200

1:400

1:500

1:800

1:2,000

1:4,000

1:8,000

4 mM

51.3%

1:300

1:600

1:750

1:1,200

1:3,000

1:6,000

1:12,000

6 mM

37.5%

a

Maximum effect was observed at the 0.02 mM dose; higher concentrations resulted in no further reduction in cell viability

1:5

1:10

1:12.5

1:20

1:50

1:200

0.1 mM

0.0005 mM

Drug Dose

99%

Gemcitabine 100% (IC50 ~ 0.007 mM) 97.4%

Viability

Cisplatin (IC50 ~ 4.2 mM)

1:500

1:1,000

1:1,250

1:2,000

1:5,000

1:10,000

1:20,000

10 mM

8.3%

1:2,500

1:5,000

1:6,250

1:10,000

1:25,000

1:50,000

1:100,000

50 mM

3.4%

Table 13.3 The gemcitabine and cisplatin ratios obtained when using a checkerboard design in the A549 cell line. The individual drug concentrations were chosen to obtain effect levels that spanned the dose–response curves

298 Harasym, Liboiron, and Mayer

Drug Ratio-Dependent Antagonism

299

approach limits the evaluation of the concentration dependency for any particular gemcitabine:cisplatin ratio. Drug combination schemes where drug:drug ratios are maintained by equivalent dilutions of drug A and drug B in a checkerboard design can be utilized; however, this often results in one of the drugs failing to span the full dose–response curve and the corresponding loss of synergy information in this portion of the dose–response curve. This can be particularly important if missing data occurs at high tumor growth inhibition, considering that for chemotherapeutics to be clinically effective multilog tumor cell kill is required. Thus, the synergy results obtained at high cell kill values are likely to be much more relevant than observations at low effect levels such as an effective dose (ED) of 20% or 30% (ED20 and ED30 reflect concentrations that result in 20% and 30% tumor cell growth inhibition, respectively). 3.1.1. Fixed Ratio In Vitro Synergy Analysis: Isobologram, Surface Response, and MedianEffect Methods

In vitro synergy analysis based on fixed drug ratios offers an alternative approach to identifying synergistic drug combinations. The most notable of these are the isobologram, surface response, and median-effect analysis methods, which are described in greater detail below. A constant-ratio approach has the benefit of evaluating the drug combination at a fixed drug:drug ratio over the full range of fraction of affected cells (fa; equivalent to effective dose, ED, the percent tumor growth inhibition relative to control cells) and therefore one can assess whether changes in synergy/antagonism occur for a particular ratio as drug concentrations vary. Isobolograms (or Isobol) are prepared for each fixed ratio from equally effective dose pairs for a single effect level. It should be noted that if one of the individual dose–response curves do not attain the chosen effect level, an isobol cannot be evaluated at that effect level and a lower effect level must be chosen or higher drug concentrations must be evaluated. Figure  13.1 shows an isobologram for the irinotecan:floxuridine combination adapted from Harasym et al. (34) at an ED75 (effective dose required to achieve 75% tumor cell growth inhibition). The ED75 isobol is generated from the dose of irinotecan required to elicit an ED75 plotted on the y-axis and the dose of floxuridine required to generate an ED75 plotted on the x-axis. The straight line joining the two data points on each of the axes is the line of additivity. For experimental combinations the drug concentrations required to generate an ED75 response are then plotted. Data points that lie below the line of additivity are considered synergistic, on the line as additive, and above the line as antagonistic. In this example, five fixed ratios were evaluated at an ED75 and, with the exception of the 10:1 ratio, the four remaining ratios were synergistic. Isobols are simple to generate and are visually easy to interpret; however, the isobols are not readily evaluated statistically and it is

300

Harasym, Liboiron, and Mayer

Fig. 13.1. Isobologram of irinotecan and floxuridine in the HCT-116 colorectal cell line (adapted from data presented in [34]. (a) An ED75 isobol including five fixed ratio data points, which were the drug doses of irinotecan:floxuridine required to achieve an ED75. (b) Three isobols are displayed on a single figure, and fixed ratio data are not shown, for clarity.

difficult to quantify the magnitude of the observed synergistic/ antagonistic response (i.e., the perpendicular distance from the line of additivity). Although onerous to generate, several modified approaches from the original isobologram methodology do allow for more statistical rigor and provide quantitative measures of synergy (35, 36). Further, when required to evaluate large data sets multiple effect level isobolograms where more than three isobols are plotted (i.e., the simultaneous plots of ED50, ED60,

Drug Ratio-Dependent Antagonism

301

ED70 isobols on the same figure) become cumbersome and difficult to visualize due to data congestion (Fig. 13.1b). Thus, for large quantities of data, alternative methods that generate similar results are preferred. Several zero interaction response surfaces (surface response) analysis methods have been described that present data as threedimensional (3-D) concentration effect curves (37–39). The concentrations of drugs A and B are generally plotted on the x-axis and y-axis, and the fa (or other observed effect) is plotted on the z-axis. The 3-D surface that is generated (usually derived from Loewe additivity and/or Bliss independence principles) defines the predicted results of no interaction (zero effect or additivity) for the drug combination. Zero interaction response surfaces have several advantages including: (1) the evaluation of combinations throughout the complete dose range, (2) analysis when one drug dose is fixed and the second drug dose is varied, and (3) analysis when both drugs are simultaneous varied to keep the dose at a fixed ratio. The latter two points represent cross sections through the zero response surfaces. However, disadvantages of surface response analysis include the need for a large number of regularly dispersed data points, the complexity in implementation, the deficiencies in quantifying the measure of the interaction, and the inability to measure statistical uncertainty (33). As an example, Fig. 13.2 shows cytarabine and daunorubicin viability data obtained using the CCRF-CEM leukemia cell line plotted in relationship to a zero response surface (CombiTool, IMB-Jena, adapted from data presented in (40)). Data points shown above the surface are synergistic, near the surface indicate zero interaction (additivity), and below the surface are antagonistic. The 5:1 molar ratio of cytarabine:daunorubicin was readily identified as synergistic due to its noticeable separation from the zero interaction surface compared to the other ratios that were either additive or antagonistic. What is most evident from attempting to analyze this type of data format is the difficultly in identifying the synergism, additivity, or antagonism by visual inspection. To analyze the five fixed ratios in the graph visually it is necessary to rotate the graph to inspect under the surface. Further, it is difficult to quantify the extent of synergy (the distance from the zero response surfaces) and the relative response level (fa) at which synergy/antagonism occurred. Therefore, in the analysis of this data set, surface response analysis failed to efficiently identify important quantitative differences in synergy in a manner that is amenable to high data throughput. Constructing drug combinations for in vitro synergy analysis based on fixed drug ratios has been driven largely by the development of a third method, the median-effect analysis by Chou et al. (33, 41, 42), and this principle has been widely used for combination

302

Harasym, Liboiron, and Mayer

Fig. 13.2. Zero interaction response surfaces (surface response) obtained for cytarabine and daunorubicin in the CCRFCEM cell line. Data were generated from CombiTool. Data points above the zero interaction surfaces indicate synergy and data points below the surface indicate antagonism (adapted from data presented in (40).)

assessment for numerous years (19, 43–45). In this approach, the ratio of two drugs combined has historically been selected based on their relative IC50 values, and serial dilutions of this drug combination are prepared to span the entire in vitro dose–response curve. The degree of tumor cell growth inhibition is compared to that for the individual agents across the same concentration range used in the combination. The approach has the benefit of evaluating the combination at a given fixed drug:drug ratio over the full range of fraction of affected cells and therefore one can assess whether changes in drug concentrations affect whether or not a particular drug ratio is synergistic or antagonistic. Although this information is extremely valuable, the drug ratios reflected by IC50 values of the individual drugs may not reflect those ratios exposed in vivo after systemic administration of a drug combination. Consequently, in order to assess the pharmacological implications of decreasing drug concentrations and changing drug:drug ratios that occur after in vivo administration of conventional drug combinations, multiple ratios must be systematically evaluated in vitro. Historically, this parameter has rarely been investigated in vitro and when it has, the implications of the results obtained with different drug:drug ratios (typically only 2–3) on the in vivo activity of the combination have been unrecognized.

Drug Ratio-Dependent Antagonism 3.1.2. In Vitro Synergy Analysis Guided by In Vivo Parameters: Combination Index-Fraction Affected Analyses

303

The median-effect equation of Chou and Talalay yields identical conclusions for both isobologram and Combination Index (CI)fa data (discussed further later) with the advantage of ease of operation, the ability to analyze large data sets, and graphical representations that allow dose–response interpretations. Specifically, where isobolograms are dose oriented, the CI-fa analyses are effect oriented. Briefly, from dose–response data the median effect (13.1) is log transformed to a linearized equation and thus takes the form of a straight line equation y = mx + b (13.2). m



fa  D  = . f u  Dm 

(13.1)



 f  log  a  = m log( D) − m log( Dm ),  fu 

(13.2)

where fa = fraction of cells affected, fu = fraction of cells unaffected, fa + fu = 1, D = dose of drug, Dm = the median-effect dose (signifies potency, and is usually near the IC50), and m = signifies the shape (sigmodicity) of the dose–effect curve. The latter equation is used to convert monotonic dose–response curves (usually sigmoidal) into straight lines whereby the slope (m) and the x-intercept (log Dm) and hence the Dm value can be obtained and extrapolated for any dose and effect. Using the Combination Index (CI) (13.3) and using the m and Dm values from above the CI value at any effect level (fa) can be determined where:

CI =

( D)1 ( D2 ) + , ( Dx )1 ( Dx ) 2

(13.3)

where CI = combination index; CI < 1, = 1, and >1 indicates ­synergy, additivity, or antagonism, respectively, (D)1 and (D)2 = the drug concentrations in the combination required to inhibit X%, and (Dx)1 and (Dx)2 = the doses of the individual drugs alone required to inhibit X%.It should be noted that the median-effect CI equation is equivalent to the Loewe additivity principle I = da/ DA + db/DB) for mutually exclusive agents (33, 37). The most efficient design for applying the median effect analysis method is to choose a specified fixed ratio as described earlier in the experimental design section, and then perform a series of serial dilutions to obtain a dose–response curve. We use the median-effect equation as our primary method for synergy analysis as it allows for (1) the evaluation of fixed-drug combination ratios at multiple concentrations, (2) the correlation of synergy to the fraction of cells affected, thereby facilitating synergy analysis at high cell kill values, and (3) straightforward generation of synergy

304

Harasym, Liboiron, and Mayer

data through the availability of commercially available software (CalcuSyn, BioSoft). We have utilized the median-effect analysis method to assess the in vitro drug ratio dependency of synergy for several clinically relevant drug combinations (27, 40, 46, 47). Figure 13.3 presents the in vitro synergy results evaluated in this manner for the combination of irinotecan:floxuridine exposed to HCT-116 human colorectal cancer cells (adapted from (34) and (27)). Irinotecan and fluorinated pyrimidine combination therapy are standard of care in metastatic colorectal cancer (48, 49). Taking the traditional approach to analyzing such a combination would result in the generation of a single CI vs. fa curve using an irinotecan:floxuridine ratio that would reflect their relative potencies in vitro (10:1 molar ratio in this case; see Fig. 13.3a). These data indicate that this drug ratio exhibits significant concentration-dependent synergy since at low fa (corresponding to low concentrations of the 10:1 ratio), the combination is nearly additive; however, the combination shifts to strong antagonism when the drug concentration is increased to achieve tumor growth inhibition greater than an fa value of 0.5. Higher fa values (³0.5) are most relevant for cancer applications and under these conditions the 10:1 irinotecan:floxuridine molar ratio was highly antagonistic. Based on these limited data, one could conclude that this combination is, in fact, not desirable from the perspective of drug synergy. However, when four additional drug ratios were evaluated, we observed that other ratios of the same two agents could be identified where strong synergy was obtained over broad drug concentrations reflecting the full range of fa values (Fig. 13.3b). Notably, irinotecan:floxuridine ratios of 1:1, 1:5 and 1:10 were synergistic between fa values of 0.2 and 0.8. We then compared drug ratio-dependent synergy for the different drug ratios by plotting the CI value as a function of drug ratio for an fa value reflecting high cell kill (e.g., fa = 0.8; see Fig. 13.3c). These data highlight the fact that as irinotecan:floxuridine concentrations and drug ratios change (as they would following in vivo administration of conventional aqueous-based drug combinations, see ref. (34)) conditions would likely occur where the two drugs were exposed at antagonistic drug ratios, and this would compromise therapeutic activity. 3.1.3. Automated Screening Methodology

The in vitro evaluations of drug ratio-dependent synergy detailed earlier were performed using five different drug:drug molar ratios ranging from 10:1 to 1:10 in 3–4 cell lines. While this process led to the identification of synergistic drug ratios that could be exploited in vivo via drug delivery systems (see later), the intervals between the different drug ratios are relatively large (up to fivefold) and the screening matrix must be tailored to each particular combination in the context of drug concentrations employed.

Fig. 13.3. Irinotecan and floxuridine analysis using the median effect principle and combination index (CI) values in the HCT-116 colorectal cell line (adapted from data presented in [34]). (a) CI vs. the fraction affected (fa) at the 10:1 ratio. (b) CI vs. fa for five ratios, with the CI values highlighted at the ED80 for each of the five ratios. (c) The indicated CI values from the ED80 for each of the five fixed ratios. CI < 1.0 indicates synergism, CI = 1 indicates additivity, and CI > 1 indicates antagonism.

306

Harasym, Liboiron, and Mayer

Manual cell culture procedures are capable of managing the ­analysis of five fixed ratios plus two individual dose–response profiles using eight serial drug dilutions, triplicate assays, and a minimum of three repeats of each experiment (each experiment represents 630 individual assays). Ideally, however, one would test more drug:drug ratios with smaller intervals in a wider range of tumor cell lines in order to improve the robustness, reliability, and predictability of drug ratio-dependent synergy trends (50– 54). Furthermore, a single drug combination matrix would facilitate the application of robotic liquid handling, thereby increasing throughput and allowing virtually any drug combination to be readily examined for in vitro synergy over a wide range of drug ratios and concentrations. To achieve this goal, we have expanded the number of drug ratios from 5 to 18 where each ratio reflects a twofold increment from its nearest neighbor, have increased the number of concentrations evaluated for each ratio from 8 to 16, have increased the number of experiment repeats from 3 to 4 while maintaining the triplicate replicates within each experiment, and have expanded the number of cell lines screened from 3–4 to 10–20 comprising 5–7 tumor types. In cases where drug combinations are approved for patient use, the in vitro screen is enriched in cell lines representing the clinical indication. The data generated in this process reflect 96,000 individual assays (1,600, 96-well plates). Figure 13.4 presents a schematic illustration of this semiautomated process for systematically evaluating drug combinations for drug ratio-dependent synergy. The process is semiautomated as it relies upon numerous liquid handling robots and workstations (multiplate washers, multiplate readers, etc.) to perform all of the required steps of a cell viability assay. The method evaluates the 18 fixed ratios from a fixed ratio range of 64:1 to 1:2,048. This broad ratio range allows for the collection of fixed-ratio dose–response profiles for drug combinations from agents with similar potencies to those with large differences in potencies (i.e., IC50 differences of approximately 2,000). The viability assay used is the colorimetric tetratzolium assay, MTT (55); however, with minimal programming and procedural adjustments the methodology can be readily adapted to other colorimetric, fluorescent, or luminescent based assays. The liquid handling robots are used to plate cells into 96-well plates, prepare the required drug master plates plus the required serial dilutions, and to add drugs to the cell containing plates. The cell viability results generated are entered into a dose– response matrix for each fixed ratio (see Fig. 13.4; horizontal rows indicate the 18 fixed ratios and the left most vertical column indicates the cell lines). From the dose–response matrix various data analysis options are available to analyze for combination effects, such as

Drug Ratio-Dependent Antagonism

307

Fig. 13.4. Flow diagram of the in vitro drug screening methodology for the identification of a synergistic fixed-ratio range for a two drug combination in a panel of tumor cell lines.

isobologram, zero interaction (surface response), or median effect. As an example, the median effect analysis leads to the median effect matrix, a summation of data that relates the CI vs. fa data sets. Finally, the data are compiled to show the drug combination ratio identification matrix where the combination effect at a defined effect level (i.e., ED80) is displayed. To graphi-

308

Harasym, Liboiron, and Mayer

cally highlight the observed pattern of synergism, antagonism, or additive affects the numeric CI values are color coded. CI values between 0 and 0.9 are synergistic (green), values between 0.9 and 1.1 are additive (yellow), and values greater than 1.1 are antagonistic (red). This “ratio heat map” assists in the identification of regions of consistent synergy and antagonism across multiple tumor cell lines at a defined effect level. It should be noted that the broad drug ratio range typically leads to a significant number of data points off the sensitive range of the dose–response curve (e.g., <10% and >90% cell growth inhibition) for given combinations. Consequently, these data points must be “truncated” in order to avoid erroneous skewing of the synergy analysis using the median-effect algorithm. Table  13.4 presents a “ratio heat map” for the gemcitabine: cisplatin combination. It represents a succinct compilation of the comprehensive data set generated through the automated screening methodology conducted using nine different cell lines and 18 drug ratios. The inclusion of the added cell lines further defines the optimal synergy ratio range near a gemcitabine:cisplatin molar ratio of 1:2, as synergy was observed in all cell lines at this ratio. Higher ratios such as 1:4 for the A2780 ovarian cell line are antagonistic, with widespread drug:drug antagonism observed in several cell lines beyond a drug ratio of 1:128 gemcitabine:cisplatin. Indirectly the tabulated data reveal the different responses of individual cell lines to the gemcitabine:cisplatin combination and such detailed information could be useful for mechanistic correlations. Lastly, the automated screening methodology can be used to generate heat maps for each of the 18 fixed ratios from effect levels of ED5 to high effect levels of ED95 to examine the drug combination effects at any desired effect level. Our primary focus is to identify drug combination synergies at high effect levels (>ED75) and thus at high tumor cell kill. However, this full data set allows synergy to be evaluated at lower effect concentrations, which will be experienced as drugs are eliminated from the body to ensure that significant antagonism does not arise for combinations that may otherwise display synergy at only very high effect levels. 3.2. Translating Drug Ratio-Dependent Synergy In Vivo Using Nanoscale Drug Delivery Systems

The implication of the drug-dependent antagonism (MDR) revealed by the ratiometric screening results described earlier and elsewhere for other drug combinations (27, 47) is that attention must be given to the concentration and ratio of drug combinations that occur over time after systemic administration in vivo. Basing in vitro synergy testing conditions on pharmacologically relevant drug concentrations and drug ratios observed in vivo may increase our understanding of how conventional drug combinations interact therapeutically. More importantly however, this approach has opened the possibility to exploit drug ratio-dependent synergy

1.10

1.10

0.88

1.30

A2780

BxPC-3 Pancreatic 1.30

H460

1.10

9.00

3.00

0.80

Lung

H1299 Lung

IGROV- Ovarian 1

MCF-7 Breast

Ovarian

1.10

A549(3) Lung

0.50

0.79

1.70

0.90

1.10

1.70

H1299 Lung

1.30

32:1

1.80

64:1

Colon

HT-29

Cell line Tumor

1.00

2.90

0.62

0.85

0.83

0.82

0.85

3.60

4.50

16:1

1.10

1.40

0.51

1.40

0.52

0.96

1.10

9.70

2.30

8:1

1.10

1.10

0.49

1.20

0.54

0.73

0.90

0.74

0.62

4:1

1.00

0.70

0.86

0.65

0.94

0.87

0.85

0.95

0.85

2:1

1.00

1.30

0.66

0.96

0.68

0.76

0.88

0.40

0.69

1:1

0.84

0.65

0.62

0.82

0.52

0.76

0.81

0.44

0.50

1:2

0.96

0.87

0.34

0.91

0.38

1.40

0.55

0.58

0.33

1:4

0.81

0.41

0.55

1.10

0.23

1.20

0.71

0.44

0.16

1:8

CI @ ED80

0.42

0.31

0.49

0.67

0.35

3.20

0.41

0.55

0.15

1:16

0.44

0.57

0.27

0.71

0.31

8.30

0.51

0.64

0.14

1:32

0.39

0.64

0.69

1.30

0.48

2.90

0.67

0.18

0.03

1:64

0.38

1.10

2.50

0.76

0.58

4.30

1.20

0.24

0.03

0.81

1.20

1.80

0.92

1.10

3.40

1.90

0.27

0.01

1.30

1.30

4.70

0.93

1.10

2.30

2.60

0.54

0.02

2.20

1.50

0.63

0.86

2.00

6.20

3.40

0.62

0.01

2.90

1.20

3.30

1.00

0.90

5.00

3.80

0.53

0.00

1:128 1:256 1:512 1:1024 1:2048

Table 13.4 A gemcitabine and cisplatin “ratio heat map” displaying the CI values at an ED80 for the standard 18 ratios screened, data from nine cell lines are shown. CI values between 0 and 0.89 are synergistic (green), values between 0.9 and 1.1 are additive (yellow), and values greater than 1.1 are antagonistic (red)

309

310

Harasym, Liboiron, and Mayer

relationships and avoid drug ratio-dependent MDR antagonism through the use of nanoscale drug delivery systems. This is due to the fact that small (20–200  nm diameter) drug delivery vehicles such as liposomes and nanoparticles do not readily distribute into healthy tissues after systemic administration and can be designed to control the rate at which encapsulated drugs are released from the carrier (56, 57). Consequently, such delivery systems can be constructed to deliver drug combinations in vivo such that the formulated drug:drug ratio can be maintained in the body for extended times after i.v. administration (34, 40, 46, 58, 59). An added advantage of nanoscale particulate carriers is that they display enhanced penetration and retention (EPR) effects in solid tumors due to the associated leaky vasculature and poor lymphatic drainage, which results in selective accumulation of the delivery vehicles and their encapsulated contents in sites of tumor growth (56, 60). The ideal drug vehicle for multiple, synergistic agents must coordinate the pharmacokinetics and biodistribution of both drug agents to ensure delivery of both drugs to the tumor site at the desired synergistic ratio. Dissimilar drug leakage rates or tumor distributions between the two drugs could lead to compromised efficacy via drug ratiodependent MDR antagonism. Lastly, the drug carrier should have a relatively prolonged plasma half-life compared to the free agents to allow time for the delivery vehicle to extravasate into tumor tissue and exploit the EPR effect for small particulate drug carriers. Delivery of two agents at a defined ratio through use of a drug delivery vehicle can use one of two strategies. The simplest approach is to formulate each drug component into a separate carrier and combine the two formulated drugs at the desired ratio (Fig. 13.5a). This method has the advantage of facile generation of the desired ratio; however, one must ensure that the drug release rates for both agents are identical and that the biodistribution of both carriers is unaffected by the encapsulation of different agents. These concerns can be eliminated through encapsulation of both active species in a single carrier system (Fig. 13.5b). Such a system presents added complexity to the formulation of a fixed ratio combination agent; iterative variation of internal buffer components and carrier composition is used to coordinate the release of both drugs so that the plasma drug elimination kinetics is matched. For this purpose, liposome technology is better developed than other nanoscale platforms (e.g., nanoparticles, micelles, nanospheres); therefore, the following discussion will focus on development of liposomal carriers. 3.2.1. Designing Liposomes to Deliver Fixed-Ratio Drug Combinations

Liposome drug delivery technology has advanced considerably over the past 25 years to the point where there are now several approved liposomal products of single anticancer agents (e.g., DepoCyt (cytarabine), Doxil (doxorubicin), DaunoXome (daunorubicin)). Liposomes are typically formulated with near equimolar amounts of inert, uncharged lipids such as phosphatidylcholine

Drug Ratio-Dependent Antagonism

311

Fig. 13.5. Formulation strategies for encapsulation of two drug species at a fixed drug ratio in vivo. (a) Mixing of two individual encapsulated agents; (b) Coencapsulation of two agents within a single drug vehicle.

and cholesterol. Addition of polyethyleneglycol (PEG) to create PEG-coated “stealth” liposomes (61) or negatively charged lipids (e.g., distearoylphosphatidylglycerol (DSPG) (62)) greatly enhances the plasma circulation time of the liposomes by increasing stability and preventing opsonization by the immune system. The lipid composition also plays a major role in the retention of drugs encapsulated within the liposome (47). The phase transition temperature of saturated lipids increases with acyl chain length (e.g., from C14 to C18) accompanied by a concomitant increase in drug retention properties (62, 63). Cholesterol composition can also be manipulated to tune drug release rates in vivo (64, 65). We have noted that liposomes comprising high phase transition temperature lipids containing relatively low amounts of cholesterol (£20% mole ratio) are particularly well suited to coordinated retention of multiple drug agents due to their stable gel phase state in vivo. Drug encapsulation of liposomes is achieved through either passive or active loading. Passive loading is conducted by extrusion of the liposomes in the presence of the drug (40, 66, 67). Drug loading efficiencies are usually low (<10%) and drug retention can be poor when injected in vivo (66). Active loading is typically highly efficient (drug loading efficiency >90%) and rapid. An appropriate transmembrane gradient (pH, ion, metal) is established that typically causes a chemical or physical change in the drug once encapsulated. As an example, pH gradients typically cause deprotonation or protonation of a weak acid/base moiety of the drug molecule upon encapsulation such that the drug becomes charged

312

Harasym, Liboiron, and Mayer

within the liposome and thus far less membrane-permeant after encapsulation (66, 68–70). Other mechanisms related to this concept include intraliposomal drug precipitation (71), metal complexation (72–75), and antiport exchange of drugs and excipients (76). Development of a coencapsulated liposomal formulation of floxuridine and irinotecan (CPX-1) illustrates the intricate design– function relationships inherent in delivery of fixed ratio chemotherapeutic agents via a drug delivery vehicle (46). Figure 13.6 depicts the release of irinotecan and floxuridine from several liposomal formulations containing different amounts of cholesterol (Chol) in distearoylphosphatidylcholine (DSPC)/distearoylphosphatidylglycerol (DSPG) liposomes. For encapsulated irinotecan (Fig. 13.6a), as the cholesterol component is increased from 0 to 15% mole ratio (of the total lipid), the plasma half-life increases dramatically from less than 2 h for 80:0:20 DSPC:Chol:DSPG to ~15 h for 65:15:20 (DSPC:Chol:DSPG). This sensitivity of irinotecan release to cholesterol is contrasted by the insensitivity of the second drug floxuridine over this cholesterol range (Fig. 13.6b). Floxuridine leakage rates from the four different liposome formulations are virtually independent of liposome cholesterol content between 0 and 20 mole percent cholesterol. In this case, the 70:10:20 DSPC:Chol:DSPG formulation was selected to coencapsulate irinotecan and floxuridine as it coordinated the plasma PK properties of the two active agents. The internal buffer composition was also instrumental in controlling the drug release properties of CPX-1. The in  vitro release rate of irinotecan was found to be sensitive to the presence of copper in the internal liposomal buffer of CPX-1, while floxuridine release was largely insensitive to copper. Intensive biophysical characterization of both formulations revealed that the aggregation state of irinotecan was modulated by the encapsulated copper gluconate-triethanolamine buffer system; irinotecan formed higher order aggregates in the absence of copper than led to a slower, uncoordinated release relative to floxuridine (77). Careful formulation of multiple drug agents to maintain a particular ratio in  vivo requires iteration of encapsulation techniques to optimize drug encapsulation efficiency, drug retention, and biodistribution properties. The use of in vivo data, particularly pharmacokinetic evaluation of plasma clearance rates of both drugs, is essential to guide the formulation process and lock in the desired drug ratio within the carrier. While coformulation of two active agents into drug vehicles is gaining interest in the literature (3, 40, 46, 78–82), attention must be paid to the release properties of both drugs to ensure that the encapsulated drug ratio is not changing over time. Several recent efforts either failed to measure for drug release rates (79, 81, 82) or to maintain the starting drug ratio (72). Use of drug delivery vehicles that successfully coordinate the delivery of the two active agents at a

Drug Ratio-Dependent Antagonism

313

Fig.  13.6. The in  vivo retention of irinotecan (a) and floxuridine (b) coformulated in various liposomal formulations. Liposomes composed of DSPC:Chol:DSPG (65:15:20 molar ratio); (open rectangle), DSPC:Chol:DSPG (70:10:20 molar ratio); (filled down triangle), DSPC:Chol:DSPG (75:5:20 molar ratio); (open circle), and DSPC:DSPG (80:20 molar ratio); (filled circle) (Reprinted from [46] with permission from Elsevier).

predictable, nonantagonistic ratio is essential to this methodology of mitigating drug ratio-dependent MDR antagonism. 3.3. In Vivo Evaluation of Fixed Ratio Combination Therapies

The importance of avoiding exposure of tumor cells to antagonistic drug ratios is elucidated through in vivo comparisons of antitumor activity of the fixed synergistic ratio combination formulation against the combination in which the ratio is uncontrolled (i.e., a free drug cocktail), and most importantly against a fixed liposome-encapsulated antagonistic ratio of the two active species. The goal of such studies is to establish the efficacy superiority of

314

Harasym, Liboiron, and Mayer

the fixed synergistic ratio formulation of the two agents by ­ensuring that (1) a contribution to the observed efficacy is made by both agents and (2) the exposure of tumor cells to a synergistic drug ratio leads to improved efficacy over formulations in which either the ratio is uncontrolled or set within an antagonistic range. Therefore, evaluation of in vivo efficacy of fixed-ratio combination agents (e.g., CombiPlex formulations) must include assessment of its efficacy against the free drug cocktail, the singly formulated liposomal agents, and finally against a formulation in which the drug ratio is fixed in a region previously identified as antagonistic. Typically, mouse models of cancer are employed for this analysis. Prior to formal efficacy experiments, the maximum tolerable doses of each agent are determined, defined as a single dose or series of doses that results in less than 15% of total body weight loss and no toxicity-related mortality for any one dose or schedule of doses. Human xenograft solid tumors subcutaneously grown in immune-compromised mice may be utilized and selected on the basis of defined genetics and growth attributes. Tumor cells utilized in these experiments can be genetically manipulated or selected to express preferable properties and are injected into mice. Once the tumors have grown to a palpable (measurable) size, delivery vehicle and free drug compositions can be administered, preferably intravenously, and their effects on tumor growth are monitored. It is not readily possible to analyze efficacy data for synergistic or antagonistic interactions by median effect, surface response, or isobologram analysis in an in  vivo model. Generation of an appropriate (i.e., statistically robust) data set using animal models would require a dose titration of each individual liposomal agent and the combined agents at several ratios and using multiple experimental repeats: clearly the number of animals required for such a study would be prohibitive and the reproducibility between experiments would be far less than for in vitro experiments. The efficacy of the synergistic CombiPlex formulation is therefore determined across multiple xenograft tumor models to ensure that the superior in  vitro efficacy of the synergistic drug ratio translates to in vivo efficacy improvements. Secondly, the efficacy of the competing free drug cocktail is optimized by determining the most favorable dose and treatment schedule so that free cocktail in vivo efficacy represents the best possible performance of an unrestricted drug ratio formulation. Doses of all agents are then administered at or near MTD to account for differences in therapeutic index. It is worth noting that this frequently requires somewhat lower doses of the combined CombiPlex agent relative to the individual liposomal drugs due to the aggregate toxicity of the combined agents.

Drug Ratio-Dependent Antagonism

315

Assessment of antitumor activity is most conveniently measured through volume measurement of subcutaneous xenograft tumors or by survival studies for nonsolid tumors. Several methods are available for statistical comparisons between treatment groups. Tumor growth delay (T − C) is measured as the median time in days for a treated group (T) to reach an arbitrarily determined tumor size (for example, 400 mg) minus median time in days for the control group to reach the same tumor size while tumor regression as a result of treatment may also be used as a means of evaluating a tumor model. Results are expressed as reductions in tumor size (mass) over time. The preferred method of calculated cell kill for solid tumor model evaluation involves measuring tumors repeatedly by calipers until all exceed a predetermined size (e.g., 400 mg). The tumor growth and tumor doubling time can then be evaluated. Log10 cell kill parameters can be calculated by (13.4a–13.4c):





log10 cell kill (T − C ) = , Dose (3.32) (Td )(No. of doses)

(

)

log10 cell kill (total) =

log10 cell kill (net) =

(T − C ) , 3.32 (Td )

(

)

((T − C ) − (duration of Rx )), (3.32 (Td ))

(13.4a)

(13.4b)

(13.4c)

where (T − C) = tumor growth delay, Td = Tumor doubling time. Evaluation of nonsolid tumors include measurement of increase in life-span (ILS%), tumor growth delay (T − C, as above) or longterm survivors (cures). Increase in lifespan is calculated by dividing the median survival time of the treatment group by the median survival time of the control group. Long-term survivors are identified as subjects that survive up to and beyond three times the survival of the untreated group. The efficacy advantages of the CombiPlex platform to avoid exposure of tumors to antagonistic drug ratios that may lead to MDR are exhibited in Figs. 13.7 and 13.8. The CombiPlex formulation CPX-1, a liposomal formulation of irinotecan and floxuridine coencapsulated at the synergistic ratio of 1:1, was evaluated for antitumor activity (cell line: pancreatic Capan-1) against the free drug cocktail, the individual liposomal components, and a second coencapsulated formulation of irinotecan and floxuridine at an antagonistic ratio of 1:10 (Fig. 13.7, adapted from data presented in (27)). CPX-1, at a dose of 37 mmol/kg of each agent, showed superior efficacy over all other groups with a calculated log cell kill (LCK) value of 1.8, besting the LCK efficacy of

316

Harasym, Liboiron, and Mayer

­ igh-dose liposomal irinotecan (37  mmol/kg) of 1.4 by nearly h 2.5 times, and the LCK of liposomal floxuridine (37 mmol/kg) of 0.3 by ~62 times. CPX-1 is also clearly superior to both free drugs dosed at their MTD; irinotecan (148 mmol/kg) and floxuridine (1,000  mmol/kg) achieved LCK values of only 0.6 and 0.4, respectively. A 1:1 free drug cocktail formulation of irinotecan and floxuridine (148:148 mmol/kg) was also inferior (LCK value 0.65, a factor of 14 times less than CPX-1), demonstrating the value of encapsulating drug-ratio dependent agents in a drug delivery vehicle that controls drug ratios. Finally, encapsulation of antagonistic drug ratios led to decreased efficacy despite the addition of more drug. A dose of liposomal irinotecan at 7.4 mmol/kg led to an observed LCK of 1.1, while liposomal irinotecan:floxuridine at a dose of 7.4:74  mmol/kg had an LCK of 0.55, nearly 3.6 times less activity despite an identical dose of irinotecan and addition of 74  mmol/kg of floxuridine. This result clearly demonstrates the deleterious effects on in vivo efficacy that occur when tumors are exposed to cytotoxic drugs at an antagonistic ratio (27). Later tumor distribution studies confirmed that the augmented activity of CPX-1 and decreased activity of the 1:10 irinotecan:floxuridine formulation was due to synergistic and antagonistic drug interactions at the tumor site, as both formulations delivered the two drugs at their respective ratios to the solid tumors (34). The drug ratio-dependent antitumor activity of a liposomal formulations of cytarabine:daunorubicin (CPX-351) was demonstrated in P388 ascites tumor-bearing BDF-1 mice (40). The 55-day survival percentages of several liposomal formulations of the two drugs are shown in Fig. 13.8. The highest survival percentage of 100% was observed for the 5:1 ratio of cytarabine and daunorubicin, previously demonstrated to be synergistic in vitro. Survivors steadily decreased for all other formulations, dropping to 83% for a 12:1 ratio formulation (despite 50% more cytarabine and only 25% less daunorubicin), and, strikingly, only 50% survival for a 3:1 cytarabine:daunorubicin formulation that actually administered the same amount of cytarabine (10  mmol/kg) as CPX-351, but nearly twofold more daunorubicin (7.4 mmol/kg for the 3:1 formulation vs. 4  mmol/kg for CPX-351). In this instance, more drug actually led to less activity, despite tumor biodistribution studies demonstrating drug delivery at the established ratios in both cases. Clearly, deviation from synergistic ratio toward ratios known to be antagonistic can lead to compromised efficacy. Further translation of in vitro drug screening informatics to in  vivo efficacy is provided by a study by Abraham et  al. (72). Previous in  vitro cytotoxicity studies had demonstrated that coadministration of doxorubicin and vincristine was antagonistic, although detailed studies to identify drug ratio regions of synergy/

Drug Ratio-Dependent Antagonism

317

Fig.  13.7. Quantitative analysis of CPX-1 in  vivo activity against Capan-1 pancreatic xenograft tumor model (adapted from data presented in [27]). Numbers in brackets represent dose of irinotecan:floxuridine or single agents in mmol/kg. CPX-1 liposomal formulation of irinotecan:floxuridine at synergistic 1:1 molar ratio, ITN irinotecan, Flox floxuridine, Lipo liposomal, antagonistic liposomal formulation of irinotecan:floxuridine at antagonistic 1:10 molar ratio.

Fig.  13.8. Survival of BDF-1 mice bearing P388 ascites tumors at day 55 following Q3Dx3 treatment with saline or coencapsulated liposomal cytarabine:daunorubicin at different drug:drug ratios (adapted from data presented in (40)).

318

Harasym, Liboiron, and Mayer

antagonism were absent. Drug:drug antagonism was confirmed in subsequent in vivo efficacy studies, which demonstrated equivalent efficacy for liposomal doxorubicin:vincristine vs. the single agent of liposomal doxorubicin. This result agreed with an earlier report of antitumor efficacy of liposomal vincristine coadministered with the liposomal doxorubicin agent Doxil. In this study the efficacy of the combined agents was actually less than that observed for Doxil alone, despite the efficacy of both individual liposomal agents being superior to the free drug (83). In both cases, it should be noted that exquisite in  vivo control of drug ratios was lacking: in the former case; the plasma drug:drug ratio changed from the initial 4:1 vincristine:doxorubicin ratio to over 20:1 in 24 h (72) while the pharmacokinetic differences between the individual liposomal formulations of vincristine and doxorubicin were not examined. These studies, while examples of incomplete consideration of drug ratio analysis, demonstrate the translation of in vitro antagonism leading to compromised in vivo efficacy likely due to a drug ratio-dependent MDR mechanism.

4. Conclusion MDR continues to be a significant impediment to improving the outcomes of cancer patients to chemotherapy treatment. Current combination therapies, while typically more effective than single agent treatments, can be subject to a newly identified form of MDR that manifests itself as a resistance to drugs presented at discrete ratios and concentrations when administered concurrently. In the current treatment paradigm, multiple agents are administered in saline-based cocktail that cannot account for the disparate pharmacokinetics of each individual agent, which generally leads to rapidly and widely changing drug ratios postadministration. Therefore, plasma elimination of free drug cocktails can result in tumor exposure to antagonistic drug ratios and corresponding compromised efficacy. In this review, we have detailed methods by which this drug-ratio dependent MDR mechanism can be identified and bypassed, resulting in improved antitumor efficacy. Screening for ratio-dependent drug interactions is best achieved through an approach that assesses drug interactions across a wide range of drug ratios and effect levels. Using these techniques has led to identification of drug ratio-dependent synergistic and antagonistic interactions for numerous drug combinations. Given the apparent broad applicability of drug ratiodependent synergy and the ability to exploit such information

Drug Ratio-Dependent Antagonism

319

in vivo by delivering fixed drug ratios in particulate drug carriers, we have demonstrated the application of automated, ratiometric drug combination screening in order to facilitate the systematic evaluation of multiple drug combinations at numerous ratios under pharmacologically relevant conditions with high throughput. Liposomal- and nanoparticle-based delivery technologies can be developed to deliver synergistic drug combinations at fixed drug ratios and avoid tumor exposure to antagonistic ratios. Codelivery and coordinated release of synergistic drug combinations facilitate the achievement of maximum efficacy by maintaining synergy throughout the pharmacodynamic (dose–response cytotoxicity) profile and avoidance of antagonistic ratios in vivo. References 1. Frei EI, Freireich EJ (1964) Leukemia. Sci Am 210:88–96 2. Freireich EJ, Frei EI (1964) Recent advances in acute leukemia. Prog Hematol 27:187–202 3. Ramsay EC, Dos Santos N, Dragowska WH, Laskin JJ, Bally MB (2005) The formulation of lipid-based nanotechnologies for the delivery of fixed dose anticancer drug combination. Curr Drug Del 2:341–351 4. Ewesuedo RB, Ratain MJ (2003) Principles of cancer therapeutics. In: Vokes EE, Golomb HM (eds) Oncologic therapies. Springer, Secaucus, NJ, pp 19–66 5. Shabbits JA, Krishna R, Mayer LD (2001) Molecular and pharmacological strategies to overcome multidrug resistance. Expert Rev Anticancer Ther 1:89–98 6. Coley HM (2008) Mechanisms and strategies to overcome chemotherapy resistance in metastatic breast cancer. Cancer Treat Rev 34:378–390 7. Shabbits JA, Hu Y, Mayer LD (2003) Tumor chemosensitization strategies based on apoptosis manipulations. Mol Cancer Ther 2: 805–813 8. Zhou SF, Wang LL, Di YM et  al (2008) Substrates and inhibitors of human multidrug resistance associated proteins and the implications in drug development. Curr Med Chem 15:1981–2039 9. Perez-Tomas R (2006) Multidrug resistance: retrospect and prospects in anti-cancer drug treatment. Curr Med Chem 13:1859–1876 10. Nobili S, Landini I, Giglioni B, Mini E (2006) Pharmacological strategies for overcoming multidrug resistance. Curr Drug Targets 7:861–879 11. Shah MA, Schwartz GK (2001) Cell cycle-mediated drug resistance: an emerging concept in cancer therapy. Clin Cancer Res 7:2168–2181

12. Dancey JE, Chen HX (2006) Strategies for optimizing combinations of molecularly targeted anticancer agents. Nat Rev Drug Disc 5:649–659 13. Keith CT, Borisy AA, Stockwell BR (2005) Multicomponent therapeutics for networked systems. Nat Rev Drug Disc 4:1–8 14. Lehar J, Zimmermann GR, Krueger AS et al (2007) Chemical combination effects predict connectivity in biological systems. Mol Sys Biol 3:1–14 15. Cohen MH, Gootenberg J, Keegan P, Pazdur R (2007) FDA drug approval summary: Bevacizumab (Avastin®) plus carboplatin and paclitaxel as first-line treatment of advanced/ metastatic recurrent nonsquamous non-small cell lung cancer. Oncologist 12:713–718 16. Schwartz RN, Vozniak M (2008) Current and emerging treatments in multiple myeloma. J Manag Care Pharm 14:S12–S18 17. Tannock IF, de Wit R, Berry WR et al (2004) Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med 351:1502–1512 18. von der Maase H, Hansen SW, Roberts JT et al (2000) Gemcitabine and cisplatin versus methotrexate, vinblastine, doxorubicin, and cisplatin in advanced or metastatic bladder cancer: results of a large, randomized, multinational, multicenter, phase III study. J Clin Oncol 18:3068–3077 19. Hongrapipat J, Kopeckova P, Liu J, Prakongpan S, Kopecek J (2008) Combination chemotherapy and photodynamic therapy with Fab’ fragment trageted HPMA copolymer conjugates in human ovarian carcinoma cells. Mol Pharm 5:696–709 20. Photiou A, Shah P, Leong LK, Moss J, Retsas S (1997) In vitro synergy of paclitaxel (Taxol) and

320

Harasym, Liboiron, and Mayer

vinorelbine (Navelbine) against human melanoma cell lines. Eur J Cancer 33:463–470 21. Frei EI (1972) Combination cancer therapy: Presidential address. Cancer Res 32: 2593–2607 22. Adams DJ, Sandvold ML, Myhren F et  al (2008) Anti proliferative activity of ELACY (CP-4055) in combination with cloretazine (VNP40101M), idarubicin, gemcitabine, irinotecan and topetecan in human leukemia and lymphoma cells. Leuk Lymph 49:786–797 23. Saiman L (2007) Clinical utility of synergy testing for multidrug-resistant Pseudomonas aeruginosa isolated from patients with cystic fibrosis: “the motion for”. Paediatr Respir Rev 8:249–255 24. Seydel JK, Schaper KJ, Rüsch-Gerdes S (1994) Experimental drugs and combination therapy. Immunobiology 191:569–577 25. Timukaynak F, Can F, Azap OK et al (2006) In vitro activities of non-traditional antimicrobials alone or in combination against multidrug-resistant strains of Pseudomonas aeruginosa and Acinetobacter baumannii isolated from intensive care units. Int J Antimicrob Agents 27:224–228 26. Urban C, Segal-Maurer S, Rahal JJ (2003) Considerations in control and treatment of nosocomial infections due to multidrugresistant Acinetobacter baumannii. Clin Infect Des 36:1268–1274 27. Mayer LD, Harasym TO, Tardi PG et  al (2006) Ratiometric dosing of anticancer drug combinations: controlling drug ratios after systemic administration regulates therapeutic activity in tumor-bearing mice. Mol Cancer Ther 5:1854–1863 28. Mayer LD, Janoff AS (2007) Optimizing combination chemotherapy by controlling drug ratios. Mol Interv 7:216–223 29. Borisy AA, Elliot PJ, Hurst NW et al (2003) Systematic discovery of multicomponent therapeutics. Proc Natl Acad Sci USA 100:7977–7982 30. Kanzawa F, Koizumi F, Koh Y et al (2001) In vitro synergistic interactions between the cisplatin analogue nedaplatin and the DNA topoisomerase I inhibitor irinotecan and the mechanism of this interaction. Clin Cancer Res 7:202–209 31. Raitanen M, Rantanen V, Kulmala J et  al (2002) Supra-additive effect with concurrent paclitaxel and cisplatin in vulvar squamous cell carcinoma in vitro. Int J Cancer 100:238–243 32. Mercalli A, Sordi V, Formicola R et al (2007) A preclinical evaluation of pemetrexed and irinotecan combination as second-line che-

motherapy in pancreatic cancer. Br J Cancer 96:1358–1367 33. Chou TC (2006) Theoretical basis, experimental design, and computerized simulation of synergsim and antagonism in drug combination studies. Pharmacol Rev 58:621–681 34. Harasym TO, Tardi PG, Harasym NL et  al (2007) Increased preclinical efficacy of irinotecan and floxuridine coencapsulated inside liposomes is associated with tumor delivery of synergistic drug ratios. Oncol Res 16:361–374 35. Berenbaum MC (1977) Synergy, additivism and antagonism in immunosuppression. Clin Exp Immunol 28:1–18 36. Steel GG, Peckham MJ (1979) Exploitable mechanisms in combined radiotherapy-chemotherapy: the concept of additivity. Int J Radiat Oncol Biol Phys 5:85–91 37. Greco WR, Bravo G, Parsons JC (1995) The search for synergy: a critical review from a response surface perspective. Pharmacol Rev 47:331–385 38. Greco WR, Park HS, Rustum YM (1990) An application of a new approach for the quantitation of drug synergism to the combination of cis-diamminedichloroplatinum and 1-b-D-arabinofuranosylcytosine. Cancer Res 50:5318–5327 39. Weinstein JL, Bunow B, Weislow OS et  al (1990) Synergistic drug combinations in AIDS therapy. Ann N Y Acad Sci 616:367–384 40. Tardi PG, Johnstone SA, Harasym NL et al (2009) In vivo maintenance of synergistic cytarabine:daunorubicin ratios greatly enhances therapeutic efficacy. Leuk Res 33:129–139 41. Chou TC, Talalay P (1984) Quantitative analysis of dose–effect relationships: the combined effects of multiple drugs or enzyme inhibitors. Adv Enzyme Regul 22:27–55 42. Chou TC (1991) The median-effect principle and the combination index for quantitation of synergism and antagonism. In: Chou TC, Rideout DC (eds) Synergism and antagonism in chemotherapy. Academic, New York, pp 61–102 43. Konecny G, Untch M, Slamon DJ et  al (2001) Drug interactions and cytotoxic effects of paclitaxel in combination with carboplatin, epirubicin or vinorelbine in breast cancer cell lines and tumor samples. Breast Cancer Res Tr 67:223–233 44. Lisztwan J, Pornon A, Chen B, Chen S, Evans DB (2008) The aromatase inhibitor letrozole and inhibitors of insulin-like growth factor I receptor synergistically induce apoptosis in

Drug Ratio-Dependent Antagonism in vitro models of estrogen-dependent breast cancer. Breast Cancer Res 10:R56 45. Saigal B, Glisson BS, Johnson FM (2008) Dose-dependent and sequence-dependent cytotoxicity of erlotinib and docetaxel in head and neck squamous cell carcinoma. Anticancer Drugs 19:465–477 46. Tardi PG, Gallagher RG, Johnstone SA et al (2007) Coencapsulation of irinotecan and floxuridine into low cholesterol-containing liposomes that coordinate drug release in  vivo. Biochim Biophys Acta 1768: 678–687 47. Harasym TO, Tardi PG, Johnstone SA et al (2007) Fixed drug ratio liposome formulations of combination cancer therapeutics. In: Gregoriadis G (ed) Liposome technology volume III: interactions of liposomes with biological milieu, 3rd edn. Informa Healthcare USA Inc., New York, pp 25–46 48. Douillard JY, Cunningham D, Roth AD et al (2000) Irinotecan combined with fluorouracil compared with fluorouracil alone as firstline treatment for metastatic colorectal cancer: a multicenter randomised trial. Lancet 355:1041–1047 49. Saltz LB, Cox JV, Blanke C et  al (2000) Irinotecan plus fluororuracil and leucovorin for metastatic colorectal cancer. N Engl J Med 343:905–914 50. Corbett T, Valeriote F, LoRusso P, et  al (1997) In vivo methods for screening and preclinical testing – use of rodent solid tumors for drug delivery. In: Anticancer drug development guide: preclinical screening, clinical trials, and approval, Humana, Totawa, NJ, pp. 75–99 51. Grindley GB (1982) Multiple models of utility for the rational development of new concepts involved in metabolic modulation. In: Fidler IJ, White RJ (eds) Design of models for testing cancer therapeutic agents. Van Nostrand Reinhold, New York, pp 206–214 52. Harrison S (2002) Perspective on the history of tumor models. In: Teicher BA (ed) Tumor models in cancer research. Humana, Totawa, NJ, pp 3–22 53. Johnson RK (1990) Screening methods in antineoplastic drug discovery. J Natl Cancer Inst 82:1082–1083 54. Kaufmann SH, Peereboom CA, Buckwalter CA et al (1996) Cytotoxic effects of Topetecan combined with various anticancer agents in human cancer lines. J Natl Cancer Inst 88: 734–741 55. Mosmann T (1983) Rapid colorimetric assay for cellular growth and survival: application

321

to proliferation and cytotoxcity assays. J Immunol Methods 65:55–63 56. Maeda H (2001) The enhanced permeability and retention (EPR) effect in tumor vasculature: the key role of tumor-selective macromolecular drug targeting. Adv Enzyme Regul 41:189–207 57. Torchilin VP (2007) Targeted pharmaceutical nanocarriers for cancer therapy. AAPS J 9:E128–E147 58. Batist G, Chi K, Miller W et al (2006) Phase 1 study of CPX-1, a fixed ratio formulation of irinotecan (IRI) and floxuridine (FLOX), in patients with advanced solid tumors. J Clin Oncol 24:2014 59. Batist G, Miller W, Mayer L et  al (2007) Ratiometric dosing of irinotecan (IRI) and floxuridine (FLOX) in a phase I trial: a new approach for enhancing the activity of combination chemotherapy. J Clin Oncol 25:2549 60. Maeda H, Wu J, Sawa T, Matsumura Y, Hori K (2000) Tumor vascular permeability and the EPR effect in macromolecular therapeutics: a review. J Control Release 65:271–284 61. Lasic DD, Martin FJ (1995) Stealth liposomes. CRC, Boca Raton, FL 62. Tari A, Huang L (1989) Structure and function of phosphatidylglycerol in the stabilization of the phosphatidylethanolamine bilayer. Biochemistry 28:7708–7712 63. Boman NL, Mayer LD, Cullis PR (1993) Optimization of the retention properties of vincristine in liposomal systems. Biochim Biophys Acta 1152:253–258 64. Dos Santos N, Mayer LD, Abraham SA et al (2002) Improved retention of idarubicin after intravenous injection obtained for cholesterol-free liposomes. Biochim Biophys Acta 1561:188–201 65. Dos Santos, N., Waterhouse, D., Masin, D. et al (2005) Substantial increases in idarubicin plasma concentration by liposome encapsulation mediates improved antitumor activity. J Control Release 105:89–109 66. Mayer LD, Cullis PR, Bally MB (1994) The use of transmembrane pH gradient-driven drug encapsulation in the pharmacodynamic evaluation of liposomal doxorubicin. J Liposome Res 4:529–553 67. Peleg-Shulman T, Gibson D, Cohen R, Abra R, Barenholz Y (2001) Characterization of sterically stabilized cisplatin liposomes by nuclear magnetic resonance. Biochim Biophys Acta 1510:894–902 68. Cullis PR, Hope MJ, Bally MB et al (1997) Influence of pH gradients on the transbilayer transport of drugs, lipids, peptides, and metal

322

Harasym, Liboiron, and Mayer

ions into large unilamellar vesicles. Biochim Biophys Acta 1331:187–211 69. Drummond DC, Noble CO, Hayes ME, Park JW, Kirpotin DB (2008) Pharmacokinetics and in  vivo drug release rates in liposomal nanocarrier development. J Pharm Sci 97:4696–4740 70. Haran G, Cohen R, Bar LK, Barenholz Y (1993) Transmembrane ammonium sulfate gradients in liposome produce efficient and stable entrapment of amphipathic weak bases. Biochim Biophys Acta 1151:201–215 71. Kirpotin DB (2000) Compound loaded liposomes and methods for their preparation. US Patent 6, 110, 491 72. Abraham SA, McKenzie C, Masin D et  al (2004) In vitro and in vivo characterization of doxorubicin and vincristine coencapsulated within liposomes through use of transition metal ion complexation and pH gradient loading. Clin Cancer Res 10:728–738 73. Li C, Cui J, Li Y et  al (2008) Copper ionmediated liposomal encapsulation of mitoxantrone: the role of anions in drug loading, retention and release. Eur J Pharm Sci 34:333–344 74. Ramsay EC, Alnajim J, Taggar A et al (2006) Transition metal-mediated liposomal encapsulation of irinotecan (CPT-11) stabilizes the drug in the therapeutically active lactone conformation. Pharm Res 23:2799–2808 75. Taggar A, Alnajim J, Anantha M et al (2006) Copper-topetecan complexation mediates drug accumulation into liposomes. J Control Release 114:78–88 76. Dicko A, Tardi PG, Xie X, Mayer LD (2007) Role of copper gluconate/triethanolamine in irinotecan encapsulation inside the liposomes. Int J Pharm 337:219–228 77. Dicko A, Frazier AA, Liboiron BD et  al (2008) Intra and inter-molecular interactions dictate the aggregation state of irinotecan co-encapsulated with floxuridine inside liposomes. Pharm Res 25:1702–1713 78. Liu Y, Lu W-L, Guo J et al (2008) A potential target associated with both cancer and cancer stem cells: a combination therapy for eradication of breast cancer using vinorelbine stealthy liposomes plus parthenolide stealthy liposomes. J Control Release 129:18–25 79. Wang J, Goh B, Lu W et al (2005) In vitro cytotoxicity of stealth liposomes co-encapsulating doxorubicin and verapamil on doxorubicin-resistant tumour cells. Biol Pharm Bull 28:822–828 80. Webb MS, Johnstone S, Morris TJ et  al (2007) In vitro and in vivo characterization of a combination chemotherapy formulation

consisting of vinorelbine and phosphatidylserine. Eur J Pharm Biopharm 65:289–299 81. Wu J, Lu Y, Lee A et al (2007) Reversal of multidrug resistance by transferrin-conjugated liposomes co-encapsulating doxorubicin and verapamil. J Pharm Pharmaceut Sci 10:350–357 82. Zhao X, Wu J, Muthusamy N, Byrd JC, Lee RJ (2008) Liposomal coencapsulated fludarabine and mitoxantrone for lymphoproliferative disorder treatment. J Pharm Sci 97:1508–1518 83. Vaage J, Donovan D, Mayhew E, Uster P, Woodle M (1993) Therapy of mouse mammary carcinomas with vincristine and doxorubicin encapsulated in sterically stabilized liposomes. Int J Cancer 54:959–964 84. Fraser TR (1870–1871) An experimental research on the antagonism between the actions of physostigma and atropia. Proc R Soc Edinb 7:506–511 85. Fraser TR (1871) The antagonism between the actions of active substances. Br Med J 2:485–487 86. Loewe S (1928) Die Quantitation Probleme der Pharmakologie. Ergeb Physiol Biol Chem Exp Pharmakol 27:47–187 87. Loewe S (1953) The problem of synergism and antagonism of combined drugs. Arzneim Forsch 3:285–290 88. Loewe S (1957) Antagonism and antagonists. Pharmacol Rev 9:237–242 89. Loewe S, Muischnek H (1926) Effect of combinations: mathematical basis of problem. Arch Exp Pathol Pharmakol 114:313–326 90. Bliss CI (1939) The toxicity of poisons applied jointly. Ann Appl Biol 26:585–615 91. Webb JL (1963) Effect of more than one inhibitor. In: Webb JL (ed) Enzymes and metabolic inhibitors, Vol. 1. Academic, New York, pp. 66–79, 487–512 92. Cox DR (1970) The analysis of binary data. Methuen, London 93. Gessner PK (1974) The isobolographic method applied to drug interactions. In: Morselli PL, Garattini S, Cohen SN (eds) Drug interactions. Raven, New York, pp 349–362 94. Valeriote F, Lin H (1975) Synergistic interaction of anticancer agents: a cellular perspective. Cancer Chemother Rep 59:895–900 95. Drewinko B, Loo TL, Brown B, Gottlieb JA, Freireich EJ (1976) Combination chemotherapy in vitro with adriamycin. Observations of additive, antagonistic, and synergistic effects when used in two-drug combinations on cultured human lymphoma cells. Cancer Biochem Biophys 1:187–195

Drug Ratio-Dependent Antagonism 96. Berenbaum MC (1985) The expected effect of a combination of agents: the general solution. J Theor Biol 114:413–431 97. Greco WR, Lawrence DL (1988) Assessment of the degree of drug interaction where the response variable is discrete. Am Stat Assoc, Proc Biopharm Sect 183–188 98. Prichard MN, Shipman C Jr (1990) A three dimensional model to analyze drug–drug interactions (review). Antiviral Res 14:181–206 99. Prichard MN, Shipman C Jr (1992) Response to J. Sühnel’s comment on the paper: A three-dimensional model to analyze drug– drug interactions, by Prichard MN, Shipman C Jr, in Antiviral Res 14:181–206, 1990. Antiviral Res 17:95–98 100. Sühnel J (1992) Comment on the paper: A three-dimensional model to analyze drug–

323

drug interactions, by Prichard MN, Shipman C Jr, in Antiviral Res 14:181–186, 1990. Antiviral Res 17:91–93 101. Sühnel J (1990) Evaluation of synergism and antagonism for the combined action of antiviral agents. Antiviral Res 13:23–40 102. Greco WR, Rustum YM (1992) Reply to letters by Berenbaum and Sühnel concerning Greco et  al. 1990, in Cancer Res 50:5318–5327, 1990. Cancer Res 52: 4561–4565 103. Sühnel J (1992) Correspondence regarding W R. Greco et al.: an application of a new approach for the quantitation of drug synergism to the combination of cis-diamminedichloroplatinum and 1-b-D-arabinofuranosylcytosine. Cancer Res 50:5318–5327, 1990. Cancer Res 52: 4560–4561

sdfsdf

Chapter 14 Reversing Agents for ATP-Binding Cassette Drug Transporters Chow H. Lee Abstract The multidrug resistance (MDR) phenotype exhibited by cancer cells is believed to be the major barriers to successful chemotherapy in cancer patients. The major form of MDR phenotype is contributed by a group of ATP-binding cassette (ABC) drug transporters which include P-glycoprotein, multidrug resistance-associated protein 1, and breast cancer resistance protein. There has been intense search for compounds which can act to reverse MDR phenotype in cultured cells, in animal models, and ultimately in patients. The ongoing search for MDR modulators, compounds that act directly on the ABC transporter proteins to block their activity, has led to three generations of drugs. Some of the third-generation MDR modulators have demonstrated encouraging results compared to earlier generation MDR modulators in clinical trials. These modulators are less toxic and they do not affect the pharmacokinetics of anticancer drugs. Significant numbers of natural products have also been identified for their effectiveness in reversing MDR in a manner similar to the MDR modulators. Other MDR reversing strategies that have been studied quite extensively are also reviewed and discussed in this chapter. These include strategies aimed at destroying mRNAs for ABC drug transporters, approaches in inhibiting transcription of ABC transporter genes, and blocking of ABC transporter activity using antibodies. This review summarizes the development of reversing agents for ABC drug transporters up to the end of 2008, and provides an optimistic view of what we have achieved and where we could go from here. Key words: Multidrug resistance, ABC drug transporters, P-glycoprotein, MDR modulators, Reversing agent, mRNA degradation, Transcriptional inhibition

1. Introduction The ATP-binding cassette (ABC) transporters are a large group of membrane proteins found virtually in all species. They are capable of transporting a variety of compounds which include peptides, lipids, and anti-cancer drugs. The common feature amongst the ABC transporters is their ability to transport substrates against a concentration gradient utilizing energy from ATP hydrolysis. The J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_14, © Humana Press, a part of Springer Science + Business Media, LLC 2010

325

326

Lee

human genome has 48 ABC genes which are further categorized into seven distinct subfamilies, ABCA to ABCG, based on sequence homology and domain organization. Three members of the ABC transporter family, P-glycoprotein (Pgp, also known as ABCB1), multidrug resistance-associated protein 1 (MRP1, also known as ABCC1), and breast cancer resistance protein (BCRP, also known as ABCG2), have so far been associated with the multidrug resistance (MDR) phenotype in cancer cells. These transporters, especially Pgp, are believed to be one of the major causes for failure in chemotherapy in cancer patients. Hence, there has been tremendous amount of studies directed at understanding the structure and function of these drug transporters, as well as in finding the best strategy and drugs that can truly reverse MDR in patients. This review chapter summarizes our current knowledge on the development of inhibitors of ABC drug transporters and the strategies in reversing MDR. This review focuses more on Pgp since there are more studies done on this ABC drug transporter. The obstacles to success in reversing clinical MDR will be discussed and opinions concerning the future of this issue presented.

2. Reversing MDR A number of cellular mechanisms are known to lead to the development of drug resistance (1). However, it is now clear that increased drug efflux by overexpression of ABC drugs transporters from cancer cells is the most common mechanism that reduces the effectiveness of anti-cancer drugs due to the reduced accumulation of drug levels in these cells (1). Furthermore, there is evidence to link the ABC drug transporters, especially Pgp, to clinical drug resistance. For instance, overexpression of Pgp correlates with drug resistance in several forms of cancers (2), and expression of Pgp in some tumors predicts poor chemotherapeutic responses in patients (3). Naturally, these observations had sparked intense search for both compounds and new ways to overcome the MDR phenotype in cancer cells. This section summarizes and provides an update up to the end of 2008 on the development of reversing agents for the ABC drugs transporters. 2.1. Inhibition of ABC Transporters Function as a Mean to Sensitize Multidrug-Resistant Cancer Cells

The most logical step in developing or discovering compounds that can reverse MDR phenotype is finding molecules that can directly block ABC drug transporter activity. In fact, most ABC reversing agents discovered or developed to date fall within this category. The following two subsections summarize the historical development of the so-called MDR modulators and the current status of clinical studies of some of these compounds. The intent is not to exhaustively discuss all of the studied compounds but to

Reversing Agents for ATP-Binding Cassette Drug Transporters

327

focus on selected ones which demonstrated somewhat promising results. For a complete list of compounds that have earlier shown to reverse MDR (first- and second-generation MDR modulators), please refer to other earlier reviews (4, 5). 2.1.1. Development of First and Second-Generation MDR Modulators

Verapamil, a calcium channel blocker, was one of the first-generation MDR modulators to be discovered to have the ability to reverse MDR (6). It was found to enhance intracellular accumulation of many anti-cancer drugs in various cell lines. Other MDR modulators were soon discovered and amongst these, cyclosporine A was the most effective and best studied (7). Cyclosporine A was able to completely resensitize a drug-resistant human T-cell acute lymphatic leukemia cell line to anti-cancer drugs and was also effective against doxorubicin resistance in solid tumors (8). Both verapamil and cyclosporine A had entered clinical trials but side effects were seen in patients due to the fact that high doses of the drugs were required (7, 9). The failure of verapamil and cyclosporine A led to the development of second-generation MDR modulators which are mostly derivatives of first-generation MDR modulators with improved efficacy and reduced side effects. Amongst this group of compounds, the cyclosporine A analog SDZ PSC833 (valspodar) was by far the most exciting. SDZ PSC833 was 10- to 20-fold more potent than its predecessor in reversing MDR in cell lines (10, 11), and highly effective against in vivo ascites models and solid tumor MDR models in animals (12). Unfortunately, clinical trials showed that SDZ PSC833 can seriously impair drug metabolism and elimination, resulting in patients overexposing to increased serum concentrations of cytotoxic drugs (13, 14). SDZ PSC833 has recently been tested on patients with multiple myeloma but again it failed to demonstrate any usefulness (15).

2.1.2. Development of Third-Generation MDR Modulators

The third-generation MDR modulators were developed to overcome the limitations that the second-generation MDR modulators exhibited. Most of these drugs which are effective at nanomolar concentration range were developed using structure– activity relationships and combinatorial chemistry. They are not metabolized by cytochrome P450 and they do not affect the pharmacokinetics of anti-cancer drugs. The third-generation MDR modulators which are in clinical trials include LY335979 (Zosuquidar), GF120918 (Elacridar), CBT-1, and XR9576 (Tariquidar). LY335979 (Zosuquidar) is one of the most potent MDR modulator known to date with Ki of 59 nM (16). Studies using cell lines and membrane vesicles have confirmed that LY335979 is not a modulator of either MRP or BCRP, but a highly specific MDR modulator of Pgp (4). In preclinical studies, the compound significantly enhance survival rate and reduce tumor mass of mice

328

Lee

engrafted with MDR-bearing human tumors (16). LY335979 had no effect on the pharmacokinetics of anti-cancer drugs which may in part explained by its lack of inhibitory effect on cytochrome P450 isozymes (16). Clinical studies on LY335979 have been quite promising. A 75% response rate was observed among 16 AML patients who were given LY335979 in combination with daunorubicin and cytarabine, suggesting the possibility of giving LY335979 to AML patients in combination with induction doses of conventional cytotoxic drugs (17). In a recent Phase I/II clinical trial, LY335979 was shown to have little effect on the pharmacokinetics of anti-cancer drugs in patients with non-Hodgkin’s lymphoma (18). LY335979 is therefore suitable for further Phase III clinical trials. GF120918 is another third-generation MDR modulator which has exhibited some promising properties. Unlike LY335979, GF120918 is not specific for Pgp as the drug can completely reverse mitoxantrone resistance in cells overexpressing BCRP (19). However, GF120918 does share one important property with other third-generation MDR modulators, in that it has minimal interactions with anti-cancer drugs. In a recent Phase I clinical trial, coadministration of GF120918 with oral topotecan resulted in complete apparent oral bioavailability of topotecan (20), suggesting that GF120918 should be suitable for further Phase II or III studies. CBT-1 is a bisbenzylisoquinoline plant alkaloid and a relatively new drug developed against Pgp (21). In Phase I trials, CBT-1 showed no effect on the pharmacokinetics of doxorubicin or paclitaxel (22, 23). The drug is currently in Phase II and III trials (21). XR9576, an anthranilamide derivative, demonstrated high potency both in  vitro and in  vivo studies. In mice carrying the intrinsically resistant colon tumors, XR9576 potentiated the antitumor activity of doxorubicin without significant toxicity (24). In addition, XR9576 fully restored the anti-tumor activity of several anti-cancer drugs against two highly resistant MDR human tumor xenografts in nude mice (24). Phase I trials showed that XR9576 had no effect on the pharmacokinetics of paclitaxel, vinorelbine, or doxorubicin when it was administered to patients with solid tumors (25, 26), and it was tolerable to patients at concentrations effective in inhibiting Pgp (27). Unfortunately, Phase II and III trials with XR9576 have not been very encouraging. Studies in patients with non-small-cell lung cancer were terminated due to chemotherapy-related toxicity in patients administered with the drug (28). For a complete list and updates on the third-generation MDR modulators, please refer to Table 14.1.

Reversing Agents for ATP-Binding Cassette Drug Transporters

329

Table 14.1 Third-generation MDR modulators

2.1.3. Natural Products as MDR Modulators

MDR modulator

Targeted ABC transporter(s)

Current stage of studies

References

CBT-1

Pgp

III

(29)

Tesmilifene

?

III

(30)

MS209 (Dofequidar)

Pgp, MRP1

III

(31)

PSC833 (Valspodar)

Pgp

III

(15)

ONT-093

Pgp

II

(32)

Annamycin

?

II

(33)

Mitotane

Pgp

II

(34)

R101933 (Laniquidar)

Pgp

II

(35)

VX710 (Biricodar)

Pgp, MRP1

II

(36, 37)

LY335979 (Zosuquidar)

Pgp

I, II

(18)

XR9576 (Tariquidar)

Pgp, MRP1

I, II

(36, 38)

GF120918 (Elacridar) Pgp, BCRP

I

(20)

Sulindac

MRP1

I

(39, 40)

S9788

Pgp

I

(41)

The search for MDR modulators has also extended to the natural products. The rationale is that natural products and their derivatives will be less toxic and more potent than the disappointed first- and second-generation MDR modulators. A very large number of varying sources of natural products capable of reversing MDR phenotype have been found and continued to be found. A list of this group of MDR modulators discovered in the last 3 years is shown in Table 14.2. Perhaps, the most widely studied amongst this group of compounds is curcumin (21, 42). Curcumin and its derivatives can inhibit the function of all three major ABC transporters, Pgp, MRP, and BCRP. Its low bioavailability when given orally and its rapid metabolism have prompted investigators to assess the effect of encapsulating curcumin with liposome. Apparently, liposomal curcumin can overcome its bioavailability problems when given intravenously (21). Its toxicity is relatively low and comparable to third-generation MDR modulators, and it has been shown to be effective anti-tumor activity in animal studies

330

Lee

Table 14.2 Natural product MDR modulators Year

MDR modulators

Targeted ABC transporter(s)

References

2005

5-Bromo tetrandine

Pgp

(43)

Kavalactones

Pgp

(44)

Curcumin

MRP1

(45)

Flavonoids

MRP1

(46)

Myricetin

MRP1

(47)

6-Prenylchrysin

BCRP

(48)

Piperazinobenzopyranones and BCRP phenylalkylaminobenzopyrazones

(49)

Stilbenoids

BCRP

(50)

Tectochrysin

BCRP

(48)

Coumarins

Pgp

(51)

Diterpenes, cycloartane triterpenes, carotenoids

Pgp

(52)

Curcumin

BCRP

(53)

Eupatin

BCRP

(54)

Ginsenosides

BCRP

(55)

Curtisii root extract

Pgp, MRP1

(56)

Deoxyschizandrin

Pgp

(57)

Kaempferia parviflora extracts

Pgp, MRP1

(58)

Schisandrol A

Pgp

(59)

Tryptanthrin

Pgp

(60)

Vitamin E TPGS

Pgp

(61)

Dihydro-b-agarofuran sesquiterpenoids

Pgp

(62)

3¢,4¢,7-Trimethoxyflavone

BCRP

(63)

Rotenoids

BCRP

(64)

Tetrahydrocurcumin

BCRP

(65)

N-hexane root extracts

Pgp

(66)

2006

2007

2008

Chokeberry and mulberry leaves Pgp, MRP1

(67)

Reversing Agents for ATP-Binding Cassette Drug Transporters

331

(21). With such encouraging results, it would not be a surprise that curcumin may be assess for ability to reverse clinical MDR in the very near future. 2.2. Degradation of mRNA as a Mean to Suppress Expression of ABC Transporters

Targeting specific mRNA for degradation as a mean to inactivate gene expression has been long realized as a potential powerful therapeutic approach. mRNAs for MDR-associated genes have been targeted as a mean to reverse the MDR phenotype. However, unlike the more rigorous search for agents that could directly modulate ABC-transporter’s activity, much less studies focused on discovering agents capable of reversing MDR via destruction of mRNAs for ABC-drug transporters. This is in large due to our lack of basic understanding on cellular components which control the degradation of mammalian mRNAs. I anticipate that as we increase our basic knowledge in this field, especially in identifying key enzymes that control mammalian mRNA decay, new technologies aimed at manipulating mRNA levels will emerge. The approaches that have been used to target mRNAs for ABC-drug transporters are antisense oligonucleotides, ribozymes, and more recently the small interfering RNA (siRNA).

2.2.1. Antisense Oligonucleotides and Ribozymes

Antisense oligonucleotide (AON) was the first mRNA degradation technology to be developed and therefore it is not surprising that it was first to be assessed for efficacy in reversing the MDR phenotype. The earlier version of AON such as phosphorothioate oligonucleotide is believed to function through RNase H-mediated degradation of complementary mRNA. The AON has worked remarkably well in reducing Pgp expression and chemosensitizing drug-resistant cells in culture (4, 68). AONs have also been successful against MRP- and BCRP-mediated drug resistance in cell lines (69–71). Several other newer generation of AONs have been developed and these include modifications at 2' OH, locked nucleic acids, peptide nucleic acids, morpholino compounds, and hexitol nucleic acids, all of which do not utilize the activity of RNase H. Amongst these, peptide nucleic acid (72) and AON with methoxyethoxy group substitution at 2' position (73) have demonstrated modest improvements over phosphorothioate AON against Pgp in cell lines. A recent report showed that when conjugated to doxorubicin, AON was able to further increase drug accumulation in cells as compared to doxorubicin alone or the AON alone (74). This suggests that conjugated AON in combination with anti-cancer drugs may offer a more powerful treatment for MDR. Catalytic RNA hammerhead ribozyme was the second mRNA degradation technology that was used against mRNA for ABC transporters. It is designed to endonucleolytically cleave a specific mRNA at a defined position in trans-containing a NUX motif, in which N is any nucleotide and X is A, C, or U (75).

332

Lee

Catalytic RNAs have been successfully used to reverse MDR phenotypes in Pgp-, MRP-, and BCRP-overexpressing cancer cell lines (76–79). In addition, multitarget multiribozyme containing all trans-acting ribozymes against all three ABC transporters were highly successful in decreasing the expression of ABC transporters and in reversing the drug-resistance phenotype that the transporters conferred (80). To date, there are no reports on the use of AON- and ribozyme-based gene therapy against ABC transporters in preclinical and clinical trials. Such studies would most likely proceed if promising results arise from clinical studies on bcl-2 and VEGF mRNAs using AON and ribozyme, respectively (75). Thus, the clinical benefit of reversing MDR using AON and ribozyme remains unknown. 2.2.2. siRNA and Short-Hairpin RNA

The latest mRNA degradation technology uses the RNA interference (RNAi) posttranscriptional mechanism. RNAi is mediated by double-stranded RNA, which is cleaved by the endoribonuclease DICER into 21- to 23-nucleotide duplexes known as siRNAs. The antisense strand of siRNA duplexes, which is part of the RNA-induced silencing complex (RISC), then guides RISC to destroy the target RNA by the second endoribonuclease, Argonaute2. Since 2003, there have been numerous reports on the use of siRNAs which have transient effect, as well as the plasmid expression vector-based short-hairpin RNAs (shRNAs) which have long-term effect, against ABC transporters. For instance, siRNAs and shRNAs have been successfully used to target Pgp, MRP, and BCRP in various cell types (81–85). The use of siRNA and shRNA in reversing MDR has been extended to animal models (86–89) using innovative nucleic acid delivery system. For instance, Stein et al. (86) recently showed significant suppression of Pgp mRNA and reversal of MDR phenotype in vivo using an intratumoral jet-injection delivery of shRNA-expressing vector against Pgp. Using a mouse model-bearing human tongue squamous cell cancer, Jiang et al. (88) reported successful inhibition of Pgp expression and reversing drug resistance by delivering siRNA against Pgp using attenuated Salmonella typhimurium. Significant reversal of MDR phenotype has also been reported in animal models when retroviral-mediated shRNA and electric pulse delivery of siRNA against Pgp were used (87, 89). It does appear that it is feasible to reverse MDR in  vivo using mRNA degradation technology. However, such approach has several challenges (90) and foremost would rely heavily on an efficient and harmless mode of delivery system.

2.3. Other Approaches in Reversing MDR

Many other approaches have been assessed for feasibility in reversing MDR. These have been discussed in some detail in recent reviews (5, 21). In this section, I discuss two approaches which have been

Reversing Agents for ATP-Binding Cassette Drug Transporters

333

studied more extensively and that have shown some degree of promises. 2.3.1. Inhibition at the DNA Level

Several methods to inhibit the transcription of Pgp as a mean to reverse MDR have been reported. The transcriptional decoy strategy has been used against Pgp. The 12 nucleotides, double stranded, transcriptional decoy which corresponds to MED-1 sequence element located upstream of MDR1 gene has been shown to sensitize drug-resistant leukemic cell line against vinblastine (91). Others have shown that LANCL2 can be used to transcriptionally suppress Pgp promoter activity leading to reduction in Pgp mRNA and protein levels (92). Another approach utilized transcriptional regulator containing Cys2-His2 type zinc finger (Zif) which recognizes Pgp promoter in addition to Zifs from the SP1 or Zif268 transcription factors (93). The five Zif chimeric proteins were able to suppress Pgp promoter activity, reduce Pgp expression, and sensitize cells to doxorubicin (93, 94). Significant inhibition on Pgp transcription followed by restoration of drug sensitivity has also been demonstrated with b–naphthoflavone (95) and with a natural marine product Et743 (96). To date, there are no reports on the use of transcriptional inhibition strategy as a mean to reverse MDR in animal studies.

2.3.2. Antibodies

To date, only antibodies against Pgp have been assessed for ability to reverse MDR phenotype. Tsuruo et al. demonstrated the ability of Pgp-specific monoclonal antibody MRK16 to inhibit tumor growth in athymic mice (97). This antibody which recognizes an external epitope of human Pgp was later shown to inhibit Pgpmediated MDR in cell line and transgenic mice expressing human Pgp (98, 99). Monoclonal antibodies against Pgp in combination with other MDR modulators (100, 101) or conjugated to cytotoxic agent (102) have also shown successes in inhibiting Pgpmediated MDR. Interestingly, palmitoylated synthetic peptides against extracellular loops of Pgp reconstituted in liposomes were able to induce production of specific autoantibodies and improved chemotherapy in mice (103). The sera from immunized mice were also effective in reducing cellular resistance to vinblastine and doxorubicin (103). Antibodies against ABC transporters have not been assess for ability to reverse MDR in clinical trials.

3 . Major Obstacles to Success in Reversing Clinical MDR

The obstacles confronting development of drugs to reverse clinical MDR are no different from any drug development program against any other diseases. In many respects, the problems are not

334

Lee

unique. Nonspecific toxicity of MDR modulators to cancer patients is a common problem. Another common obstacle is the unexpected and undesired pharmacokinetic interactions between the modulators and the anti-cancer drugs used for the treatment of patients. Often enough, the modulators blocked the clearance and/or metabolism of anti-cancer drugs resulting in accumulation of the drugs in plasma and hence enhanced toxicity to cancer patients. It is unfortunate that animals or preclinical studies are not good predictor or first round screen for these potential obstacles. Consequently, when tested in clinical trials, a high number of MDR modulators would proved to be disappointment. Another common barrier is the lack of drug delivery system that would target drugs to specific cells, tissues, or organs. The expectation of modern medicine is high. Given that we are able to find the molecular targets and design drugs that would counter attack targeted protein or mRNA, the drug would be useless if it is not able to be delivered to the target cells. Another hurdle which has only recently emerged is the discovery of single nucleotide polymorphisms (SNPs) amongst the ABC drug transporters (104, 105). Genetic polymorphisms in Pgp have been shown to change the mRNA expression, protein expression, and function of the protein, and this is believed to contribute to variations in drug responses observed in different individuals and ethnic groups. The presence and abundance of multiple ABC drug transporters pose another hurdle. Similarly, the presence of MDR mechanisms other than that conferred by the ABC drug transporters would pose further complications and challenges.

4. Conclusion It appears to be daunting when one took a first glance at the problems discussed above and at the rather slow progress that we have made over the last 20  years in finding effective drugs for reversing clinical MDR. However, over the years of experience we have understand the problems and we have taken initiatives to overcome these problems. For instance, we have become smarter in designing drugs that are less toxic and those that do not affect the pharmacokinetics of anti-cancer drugs. This is evidence from the higher rate of success amongst the third-generation MDR modulators. We have become aware of the existence of multiple ABC drug transporters and SNPs amongst the ABC drug transporters, and the potential problems that we faced. It would not be unusual to administer more than one drug to reverse clinical MDR should there be more than one ABC transporters present

Reversing Agents for ATP-Binding Cassette Drug Transporters

335

or other MDR mechanisms exist in a particular patient. In fact, it has becoming clearer that successful practice in clinical oncology would usually rely on the use of a combination of drugs as exemplified by the use of antiangiogenic drugs (106). Variant oncoproteins for Bcr-Abl which are resistant to Gleevec and variants for EGF-R which are resistant to Iressa are known to exist (107). New drugs have also been developed against these variants with considerable success. Thus, any variants for ABC transporters, as a result of SNPs, which are resistant to conventional MDR modulators can be potentially counter attack by development of new drugs. As with any other drug development program, one of the most important feature to successful clinical MDR is the ability to deliver MDR modulators, whether it be nucleic acid- or nonnucleic acid-based, to target cells. Research into drug delivery is actively on-going (108, 109) and improvements in this area can potentially overcome some of the obstacles discussed above. In summary, there is no doubt that our knowledge on how to develop better MDR modulators has increased considerably. We have become aware of the existing and future potential problems and how to counter attack these problems. This has and will continue to lead us to better wisdom in developing a wide range of new and better MDR modulators. References 1. Cole SPC, Tannock IF (2005) Drug resistance. In: Tannock IF, Hill RP, Bristow RG, Harrington L (eds) The basic science of oncology. McGrawHill, New York, pp 376–399 2. Goldstein LJ, Galski H, Fojo A et  al (1989) Expression of multidrug resistance gene in human cancers. J Natl Cancer Inst 81: 116–124 3. Chan HS, Haddad G, Thorner PS et al (1991) P-glycoprotein expression as a predictor of the outcome of therapy for neuroblastoma. N Engl J Med 325:1608–1614 4. Lee CH (2004) Reversing agents for ATP-binding cassette (ABC) transporters: application in modulating multidrug resistance (MDR). Curr Med Chem Anticancer Agents 4:43–52 5. Wu C-P, Calcagno AM, Ambudkar SV (2008) Reversal of ABC drug transporter-mediated multidrug resistance in cancer cells: evaluation of current strategies. Curr Mol Pharmacol 1:93–105 6. Tsuruo T, Iida H, Tsukagoshi S, Sakurai Y (1981) Overcoming of vincristine resistance in P388 leukemia in vivo and in vitro through enhanced cytotoxicity of vincristine and vinblastine by verapamil. Cancer Res 41:1967–1972

7. Tan B, Piwnica-Worms D, Ratner L (2000) Multidrug resistance transporters and modulation. Curr Opin Oncol 12:450–458 8. Twentyman PR, Fox NE, White DJ (1987) Cyclosporine A and its analogues as modifiers as adriamycin and vincristine resistance in a multidrug resistant human lung cancer cell line. Br J Cancer 56:55–57 9. Thomas H, Coley HM (2003) Overcoming multidrug resistance in cancer: an update on the clinical strategy of inhibiting p-glycoprotein. Cancer Control 10:159–165 10. Twentyman PR, Bieehen NM (1991) Resistance modification by PSC-833, a novel non-immunosuppressive cyclosporine. Eur J Cancer 27:1639–1642 11. te Borkhorst PA, van Kapel J, Schoester M, Sonneveld P (1992) Reversal of typical multidrug resistance by cyclosporin and its nonimmunosuppressive analogue SDZ PSC 833 in Chinese hamster ovary cells expressing the mdr1 phenotype. Cancer Chemother Pharmacol 30:238–242 12. Krishna R, Mayer LD (2000) Multidrug resistance (MDR) in cancer. Mechanisms, reversal using modulators of MDR and the

336

Lee

role of MDR modulators in influencing the pharmacokinetics of anticancer drugs. Eur J Pharm Sci 11:265–283 13. Bates S, Kang M, Meadows B et al (2001) A Phase I study of infusional vinblastine in combination with the P-glycoprotein antagonist PSC 833 (valspodar). Cancer 92:1577–1590 14. Wandel C, Kim RB, Kajiji S et  al (1999) P-glycoprotein and cytochrome P-450 3A inhibition: dissociation of inhibitory potencies. Cancer Res 59:3944–3948 15. Friedenberg WR, Rue M, Blood EA et  al (2006) Phase III study of PSC-833 (valspodar) in combination with vincristine, doxorubicin, and dexamethasone (valspodar/ VAD) versus VAD alone in patients with recurring or refractory multiple myeloma (E1A95): a trial of the Eastern Cooperative Oncology Group. Cancer 106:830–838 16. Dantzig AH, Law KL, Starling JJ (2001) Reversal of multidrug resistance by the P-glycoprotein modulator, LY335979, from the bench to the clinic. Curr Med Chem 8:39–50 17. Gerrard G, Payne E, Baker RJ et  al (2004) Clinical effects and P-glycoprotein inhibition in patients with acute myeloid leukemia treated with zosuquidar trihydrochloride, daunorubucin and cytarabine. Haematologica 89:782–790 18. Morschhauser F, Zinzani PL, Burgess M et al (2007) Phase I/II trial of a P-glycoprotein inhibitor, Zosuquidar. 3HCl trihydrochloride (LY335979), given orally in combination with the CHOP regimen in patients with non-Hodgkin’s lymphoma. Leuk Lymphoma 48:708–715 19. Allen JD, Brinkhuis R, Wijnholds J, Schinkel A (1999) The mouse Bcrp1/Mxr/Abcp gene: amplification and overexpression in cell lines selected for resistance to topotecan, mitoxantrone, or doxorubicin. Cancer Res 59: 4237–4241 20. Kuppens IELM, Witteveen EO, Jewell RC et al (2007) A Phase I, randomaized, open-label, parallel-cohort, dose-finding study of elacridar (GF120918) and oral topotecan in cancer patients. Clin Cancer Res 13:3276–3285 21. Shukla S, Wu C-P, Ambudkar SV (2008) Development of inhibitors of ATP-binding cassette drug transporters – present status and challenges. Expert Opin Drug Metab Toxicol 4:205–223 22. Oldham RK, Reid WK, Preisler HD, Barnett D (1998) A Phase I and pharmacokinetic study of CBT-1 as a multidrug resistance modulator in the treatment of patients with

advanced cancer. Cancer Biother Radiopharm 13:71–80 23. Oldham RK, Reid WK, Barnett D (2000) Phase I study of CBT-1 and Taxol in patients with Taxol resistant cancers. Cancer Biother Radiopharm 15:153–159 24. Mistry P, Stewart AJ, Dangerfield W et  al (2001) In vitro and in  vivo reversal of P-glycoprotein-mediated multidrug resistance by a novel potent modulator, XR9576. Cancer Res 61:749–758 25. Abraham J, Edgerly M, Wilson R et al (2001) A Phase I study of the novel P-glycoprotein (Pgp) antagonist, XR9576 in combination with vinorelbine. Proc Am Soc Clin Oncol 20:287 Abstract 26. Ferry D, Price L, Atsmon J (2001) A phase IIa pharmacokinetic and pharmacodynamic study of the P-glycoprotein inhibitor, XR9576 in patients treated with doxorubicin chemotherapy. Proc Am Assoc Cancer Res 42:950 Abstract 5106 27. Stewart A, Steiner J, Mellows G et al (2000) Phase I trial of XR9576 in healthy volunteers demonstrates modulation of P-glycoprotein in CD56+ lymphocytes after oral and intravenous administration. Clin Cancer Res 6:4186–4191 28. Nobili S, Landini I, Giglioni B, Mini E (2006) Pharmacological strategies for overcoming multidrug resistance. Curr Drug Targets 7:861–879 29. Oldham R (2007) Safety and efficacy study of CBT-1 and paclitaxel with carboplatin in patients with advanced inoperable non-small cell lung cancer. Available at http://www. clinicaltrial.gov 2007 30. Reyno L, Seymour L, Tu D et  al (2004) Phase III study of N, N-diethyl-2-[4(phenylmethyl)phenoxylethanamine (BMS217380–01) combined with doxorubicin versus doxorubicin alone in metastatic/ recurrent breast cancer: National Cancer Institute of Canada clinical trials group study MA.19. J Clin Oncol 22:269–276 31. Saeki T, Nomizu T, Toi M et  al (2007) Dofequidar fumrate (MS-209) in combination with cyclophosphamide, doxorubicin, and fluorouracil for patients with advanced or recurrent breast cancer. J Clin Oncol 25:411–417 32. Mistry PP, Folkes AA (2002) ONT-093 (Ontogen). Curr Opin Investig Drugs 3:1666–1671 34. Berruti A, Terzolo M, Sperone P et al (2005) Etoposide, doxorubicin and cisplatin plus mitatane in the treatment of advanced adreno-

Reversing Agents for ATP-Binding Cassette Drug Transporters cortical carcinomas: a large prospective Phase II trial. Endocr Relat Cancer 12:657–666 35. van Zuylen L, Sparreboom A, van Der Gaast A et  al (2000) The orally administered P-glycoprotein inhibitor R101933 does not alter the plasma pharmacokinetics of docetaxel. Clin Cancer Res 6:1365–1371 36. Fox E, Bates SE (2007) Tariquidar (XR9576): a P-glycoprotein drug efflux pump inhibitor. Expert Rev Anticancer Ther 7:447–459 37. Rowinsky EK, Smith L, Wang YM et  al (1998) Phase I and pharmacokinetic study of paclitaxel in combination with biricodar, a novel agent that reverses multidrug resistance conferred by overexpression of both MDR1 and MRP. J Clin Oncol 16:2964–2976 38. Agrawal M, Abraham J, Balis FM et al (2003) Increased 99mTc-sestambi accumulation in normal liver and drug-resistant tumors after the administration of the glycoprotein inhibitor, XR9576. Clin Cancer Res 9:650–656 40. O’Connor R, O’Leary M, Ballot J et al (2007) A Phase I clinical and pharmacokinetic study of the multi-drug resistance protein-1 (MRP1) inhibitor sulindac, in combination with epirubicin in patients with cancer. Cancer Chemother Pharmacol 59:79–87 41. Tranchand B, Catimel G, Lucas C et  al (1998) Phase I clinical and pharmacokinetic study of S9788, a new multidrug resistance reversal agent given alone and in combination with doxorubicin to patients with advanced solid tumors. Cancer Chemother Pharmacol 41:281–291 42. Limtrakul P (2007) Curcumin as chemosensitizer. Adv Exp Med Biol 595:269–300 43. Jin J, Wang FP, Wei H, Liu G (2005) Reversal of multidrug resistance of cancer through inhibition of P-glycoprotein by 5-bromotetrandrine. Cancer Chemother Pharmacol 55:179–188 44. Weiss J, Sauer A, Frank A, Unger M (2005) Extracts and kavalactones of Piper methysticum G. Forst (kava-kava) inhibit P-glycoprotein in vitro. Drug Metab Dispos 33:1580–1583 45. Chearwae W, Wu C-P, Chu H et  al (2006) Curcuminoids purified from tumeric powder modulate the function of human multidrug resistance protein 1 (ABCC1). Cancer Chemother Pharmacol 57:376–388 46. Wu C-P, Calcagno AM, Hladky SB, Ambudkar SV, Barrand MA (2005) Modulatory effects of plant phenols on human multidrug-resistance proteins 1, 4 and 5 (ABCC1, 4 and 5). FEBS J 272:4725–4740 47. van Zanden JJ, De Mul A, Wortelboer HM et  al (2005) Reversal of in  vitro cellular

337

MRP1 and MRP2 mediated vincristine resistance by the flavonoid myricetin. Biochem Pharmacol 69:1657–1665 48. Ahmed-Belkacem A, Pozza A, MunozMartinez F et al (2005) Flavonoid structureactivity studies identify 6-prenylchrysin and tectochrysin as potent and specific inhibitors of breast cancer resistance protein ABCG2. Cancer Res 65:4852–4860 49. Boumendjel A, Nicolle E, Moraux T et  al (2005) Piperazinobenzopyranones and phenalklyaminobenzopyranones: potent inhibitors of breast cancer resistance protein (ABCG2). J Med Chem 48:7275–7281 50. Morita H, Koyama K, Sugimoto Y, Kobayashi J (2005) Antimitotic activity and reversal of breast cancer resistance protein-mediated drug resistance by stilbenoids from Bletilla striata. Bioorg Med Chem Lett 15:1051–1054 51. Raad I, Terreux R, Richomme P et al (2006) Structure-activity relationship of natural and synthetic coumarins inhibiting the mutidrug transporter P-glycoprotein. Bioorg Med Chem 14:6979–6987 52. Molnar J, Gyemant N, Tanaka M et al (2006) Inhibition of multidrug resistance of cancer cells by natural diterpenes, triterpenes and carotenoids. Curr Pharm Des 12:287–311 53. Chearwae W, Shukla S, Limtrakul P, Ambudkar SV (2006) Modulation of the function of the multidrug resistance-linked ATP-binding cassette transporter ABCG2 by the cancer chemopreventive agent curcumin. Mol Cancer Ther 5:1995–2006 54. Henrich CJ, Bokesch HR, Dean MB et  al (2006) A high-throughput cell-based assay for inhibitors of ABCG2 activity. J Biomol Screen 11:176–183 55. Jin J, Shahi S, Kang HK, Van Veen HW, Fan TP (2006) Metabolism of ginsenosides as novel BCRP inhibitors. Biochem Biophys Res Commun 345:1308–1314 56. Limtrakul P, Siwanon S, Yodkeeree S, Duangrat C (2007) Effect of Stemona curtisii root extract on P-glycoprotein and MRP-1 function in multidrug-resistant cancer cells. Phytomedicine 14:381–389 57. Yoo HH, Lee M, Lee MW et al (2007) Effects of Schisandra lignase on P-glycoproteinmediated drug efflux in human intestinal Caco-2. Planta Med 73:444–450 58. Patanasethanont D, Nagai J, Yumoto R et al (2007) Effects of Kaempferia parviflora extracts and their flavone constituents on P-glycoprotein function. J Pharm Sci 96:223–233 59. Fong W-F, Wan C-K, Zhu G-Y et al (2007) Schisandrol A from Schisandra chinensis

338

Lee

reverses P-glycoprotein-mediated multidrug resistance by affecting Pgp-substrate complexes. Planta Med 73:212–220 60. Yu S-T, Chen T-M, Tseng S-Y, Chen Y-H (2007) Tryotanthrin inhibits MDR1 and reverses doxorubicin resistance in breast cancer cells. Biochem Biophys Res Commun 358:79–84 61. Collnot EM, Baldes C, Wempe MF et  al (2007) Mechanism of inhibition of P-glycoprotein mediated efflux by vitamin E TPGS: influence on ATPase activity and membrane fluidity. Mol Pharmacol 4:465–474 62. Gao JM, Wu WJ, Zhang JW, Konishi Y (2007) The dihydro-b-agarofuran sesquiterpenoids. Nat Prod Rep 24:1153–1189 63. Katayama K, Masuyama K, Yoshioka S et al (2007) Flavonoids inhibit breast cancer resistance protein-mediated drug resistance: transporter specificity and structure-activity relationship. Cancer Chemother Pharmacol 60:789–797 64. Ahmed-Belkacem A, Macalou S, Borrelli F et  al (2007) Nonprenylated totenoids, a new class of potent breast cancer resistance protein inhibitors. J Med Chem 50: 1933–1938 65. Limtrakul P, Chearwae W, Shukla S, Phisalphong C, Ambudkar SV (2007) Modulation of function of three ABC transporters, P-glycoprotein (ABCB1), mitoxanthrone resistance protein (ABCG2) and multidrug resistance protein 1 (ABCC1) by tetrahydrocurcumin, a major metabolite of curcumin. Mol Cell Biochem 296:85–95 66. Romiti N, Pellati F, Nieri P et  al (2008) P-glycoprotein inhibitory activity of lipophilic constituents of Echinacea pallida roots in a human proximal tubular cell lines. Planta Med 74:264–266 67. Skupien K, Kostrzewa-Nowak D, Oszmianski J, Tarasiuk J (2008) In vitro antileukemia activity of extracts from chokeberry (Aronia melanocarpa [Michx] Elliot) and mulberry (Morus alba L.) leaves against sensitive and multidrug resistant HL60 cells. Phytother Res 22:689–694 68. Nadali F, Pourfathollah AA, Alimoghaddam K et  al (2007) Multidrug resistance inhibition by antisense oligonucleotide against MDR1/mRNA in P-glycoproein expressing leukemic cells. Hematology 12:393–401 69. Stewart AJ, Canitrot Y, Baracchini E et  al (1996) Reduction of expression of the multidrug resistance protein (MRP) in human tumor cells by antisense phosphorothioate oligonucleotides. Biochem Pharmacol 51:461–469

70. Lautier D, Canitrot Y, Deeley RG, Cole SP (1996) Multidrug resistance mediated by the multidrug resistance protein (MRP) gene. Biochem Pharmacol 52:967–977 71. Kawabata S, Oka M, Shiozawa K et al (2001) Breast cancer resistance protein directly confers SN-38 resistance of lung cancer cells. Biochem Biophys Res Commun 280: 1216–1223 72. Wei HL, Wu YJ, Jing T, Bai DC, Ma LF (2003) Sensitization and apoptosis augmentation of K562/ADM cells by anti-multidrug resistance gene peptide nucleic acid and antisense oligodeoxyribonucleotide. Acta Pharmacol Sin 24:805–811 73. Alahari SK, DeLong R, Fisher MH et  al (1998) Novel chemically modified oligonucleotides provide potent inhibition of P-Glycoprotein expression. J Pharmacol Exp Ther 286:419–428 74. Ren Y, Wang Y, Zhang Y, Wei D (2008) Overcoming multidrug resistance in human carcinoma cells by an antisense oligodeoxynucleotide-doxorubicin conjugate in  vitro and in vivo. Mol Pharmacol 5:579–587 75. Tafech A, Bassett T, Sparanese D, Lee CH (2006) Destroying RNA as a therapeutic approach. Curr Med Chem 13:863–881 76. Kobayashi H, Dorai T, Holland JF, Ohnuma T (1994) Reversal of drug sensitivity in multidrug-resistant tumor cells by an MDR1 (PGY1) ribozyme. Cancer Res 54:1271–1275 77. Kowalski P, Stein U, Scheffer GL, Lage H (2002) Modulation of the atypical multidrug-resistant phenotype by a hammerhead ribozyme directed against the ABC transporter BCRP/MXR/ABCG2. Cancer Gene Ther 9:579–586 78. Materna V, Liedert B, Thomale J, Lage H (2005) Protection of platinum-DNA adduct formation and reversal of cisplatin resistance by anti-MRP2 hammerhead ribozymes in human cancer cells. Int J Cancer 115:393–402 79. Gao P, Zhou GY, Guo LL et  al (2007) Reversal of drug resistance in breast carcinoma cells by anti-mdr1 ribozyme regulated by the tumor-specific MUC-1 promoter. Cancer Lett 256:81–89 80. Kowalski P, Surowiak P, Lage H (2005) Reversal of different drug-resistant phenotypes by an autocatalytic multitarget multiribozyme directed against the transcripts of the ABC transporters MDR1/P-gp, MRP2, and BCRP. Mol Ther 11:508–522 81. Wu H, Hait WN, Yang J-M (2003) Small interfering RNA-induced suppression of MDR1 (P-Glycoprotein) restores sensitivity

Reversing Agents for ATP-Binding Cassette Drug Transporters to multidrug-resistant cancer cells. Cancer Res 63:1515–1519 82. Duan Z, Brakora KA, Seiden MV (2004) Inhibition of ABCB1 (MDR1) and ABCB4 (MDR3) expression by small interfering RNA and reversal of paclitaxel resistance in human ovarian cancer cells. Mol Cancer Ther 3:833–838 83. Tian X, Zamek-Gliszcznski MJ, Zhang P, Brouwer KL (2004) Modulation of multidrug resistance-associated protein 2 (Mrp2) and Mrp3 expression and function with small interfering RNA in sandwich-cultured rat hepatocytes. Mol Pharmacol 66:1004–1010 84. Ee PL, He X, Ross DD, Beck WT (2004) Modulation of breast cancer resistance protein (BCRP/ABCG2) gene expression using RNA interference. Mol Cancer Ther 3:1577–1583 85. Stierle V, Laigle A, Jolles B (2005) Modulation of MDR1 gene expression in multidrug resistant MCF7 cells by low concentration of small interfering RNAs. Biochem Pharmacol 70:1424–1430 86. Stein U, Walther W, Stege A et  al (2008) Complete in  vivo reversal of the multidrug resistance phenotype by jet-injection of antiMDR1 short hairpin RNA-encoding plasmid DNA. Mol Ther 16:178–186 87. Xiao H, Wu Z, Shen H et al (2008) In vivo reversal of P-glycoprotein-mediated multidrug resistance by efficient delivery of stealth RNAi. Basic Clin Pharmacol Toxicol 103: 342–348 88. Jiang Z, Zhao P, Zhou Z et al (2007) Using attenuated salmonella typhi as tumor targeting vector for MDR1 siRNA delivery: an experimental study. Cancer Biol Ther 6: 555–560 89. Pichler A, Zelcer N, Prior JL, Kuil AJ, PiwnicaWorms D (2005) In vivo RNA interferencemediated ablation of MDR1 P-glycoprotein. Clin Cancer Res 11:4487–4494 90. Sepp-Lorenzino L, Ruddy MK (2008) Challenges and opportunities for local and systemic delivery of siRNA and antisense oligonucleotides. Nature 84:628–632 91. Marthinet E, Divita G, Bernaud J, Rigal D, Baggetto LG (2000) Modulation of the typical multidrug resistance phenotype by targeting the MED-1 region of human MDR1 promoter. Gene Ther 7:1224–1233 92. Park S, James CD (2003) Lanthionine synthetase components C-like 2 increases cellular sensitivity to adriamycin by decreasing the expression of P-glycoprotein through a transcription-mediated mechanism. Cancer Res 63:723–727

339

93. Xu D, Ye D, Fisher M, Juliano RL (2002) Selective inhibition of P-glycoprotein expression in multidrug-resistant tumor cells by a designed transcriptional regulator. J Pharmacol Exp Ther 302:963–971 94. Bartsevich VV, Juliano RL (2000) Regulation of the MDR1 gene by transcriptional repression selected using peptide combinatorial libraries. Mol Pharmacol 58:1–10 95. Nwankwo JO (2006) Significant transcriptional down-regulation of the human MDR1 gene by 0naphthoflavone: a proposed hypothesis linking potent CYP gene induction to MDR1 inhibition. Med Hypotheses 68:661–669 96. Jin S, Gorfajn B, Faircloth G, Scotto KW (2000) Ecteinascidin 743, a transcriptiontargeted chemotherapeutic that inhibits MDR1 activation. Proc Natl Acad Sci USA 97:6775–6779 97. Tsuruo T, Hamada H, Sato S, Heike Y (1989) Inhibition of multidrug-resistant human tumor growth in athymic mice by anti-P-glycoprotein monoclonal antibodies. Jpn J Cancer Res 80:627–631 98. Mickisch GH, Pai LH, Gottesman MM, Pastan I (1992) Monoclonal antibody MRK16 reverses the multidrug resistance of multidrug-resistant transgenic mice. Cancer Res 52:4427–4432 99. Iwahashi T, Okochi E, Ariyoshi K (1993) Specific targeting and killing activities of antiP-glycoprotein monoclonal antibody MRK16 directed against intrinsically multidrug-resistant human colorectal carcinoma cell lines in the nude mouse model. Cancer Res 53: 5475–5482 100. Watanabe T, Naito M, Kokubu N, Tsuruo T (1997) Regression of established tumors expressing P-glycoprotein by combination of adriamycin, cyclosporine derivatives, and MRK16 antibodies. J Natl Cancer Inst 89:512–518 101. Goda K, Fenyvesi F, Basco Z (2007) Complete inhibition of P-glycoprotein by simultaneous treatment with a distinct class of modulators and the UIC2 monoclonal antibody. J Pharmacol Exp Ther 320: 81–88 102. Efferth T, Volm M (1993) P-glycoproteinmediated multidrug resistance by monoclonal antibodies, immunotoxins or antisense oligonucleotides in kidney carcinoma and normal kidney cells. Oncology 50:303–308 103. Madoulet C, Perrin L, Tosi PF, Albert P (2006) Anti-tumour immunotherapy against mulitdrug resistance. Ann Pharm Fr 64: 87–96

340

Lee

104. Sharom FJ (2008) ABC multidrug transporters: structure, function and role in chemoresistance. Pharmocogenomics 9:105–127 1 05. Sauna ZE, Kim I-W, Ambudkar SV (2007) Genomics and the mechanism of P-glycoprotein (ABCB1). J Bioenerg Bio­ membr 39:481–487 106. Kerbel RS (2006) Antiangiogenic therapy: a universal chemosensitization strategy for cancer? Science 312:1171–1174 107. Weinberg RA (2007) The rationale treatment of cancer. In: Weinberg RA (ed) The

biology of cancer. Garland Science, New York, pp 725–796 108. Xiong X-B, Uludag H, Lavasanifar A (2008) Biodegradable amphiphilic poly (ethylene oxide)-block-polyesters with grafted polyamines as supramolecular nanocarriers for efficient siRNA delivery. Biomaterials 30(2):242–253 109. Juliano R, Alam MR, Dixit V, Kang H (2008) Mechanisms and strategies for effective delivery of antisense and siRNA oligonucleotides. Nucleic Acid Res 36:4158–4171

Chapter 15 Overcoming Multidrug Resistance in Cancer: Clinical Studies of P-Glycoprotein Inhibitors Helen M. Coley Abstract Chemotherapy remains the mainstay in the treatment and management of many cancers. However, this treatment modality is fraught with difficulties associated with toxicity and also the emergence of chemotherapy resistance is a considerable problem. Cancer scientists and oncologists have worked together for some time to find ways of understanding anticancer drug resistance and also to develop pharmacological strategies to overcome that resistance. The greatest focus has been on the reversal of the multidrug resistance (MDR) phenotype by inhibition of the ATP-binding cassette (ABC) drug transporters. Inhibitors of ABC transporters – termed MDR modulators – have in the past been numerous and have occupied industry and academia in drug discovery programs. The field has been fraught with difficulties and disappointments but, nonetheless, we are currently considering the fourth generation of MDR modulator development with much data pending from the clinical trials with the third-generation modulators. Firstgeneration MDR modulator compounds were very diverse and broad spectrum pharmacological agents which fuelled the excitement surrounding the research into the MDR phenotype in cancer at the time. Second-generation agents were very heavily evaluated in mechanistic studies and formed the basis for a number of oncology portfolios of big pharmaceutical companies. Given this input, a number of clinical trials were carried out, the results of which were somewhat disappointing. Even with the modest evidence of active combinations, trial data were considered promising enough to warrant development of the third-generation of modulators. A number of key molecules have been identified with potent, long lasting MDR reversal properties, and minimal pharmacokinetic interaction with the co-administered cytotoxic agent. The results from a number of these trials are eagerly awaited and there are many in the cancer research community who remain committed to this area of anticancer drug discovery. Key words: MDR modulator, ABC transporters, P-glycoprotein, Pharmacokinetic interaction

1. Introduction The notion of anticancer drug resistance in the clinical and laboratory setting goes back as far as the 1950s, a few years following the first use of chemotherapeutic drugs. Elucidation into the underlying mechanisms began to emerge in the 1970s (1). What J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_15, © Humana Press, a part of Springer Science + Business Media, LLC 2010

341

342

Coley

remains clear is that the adaptive response of cancer cells to ­produce a drug-resistant phenotype is a complex process that involves numerous pathways. Indeed, it should be noted that drug resistance in a cancer cell line model, or ultimately in the cancer patient, is likely to be due to multifactorial and complex mechanisms, rather than due to a single mechanism. To consider this in more detail, the molecular biology of the cancer cell has recently revealed to us the complexity of their associated survival pathways and their interactions. Together with its intrinsic genetic instability and the genotoxic nature of the cancer drugs themselves, we can only expect the underlying basis of anticancer drug resistance to be a result of a number of mechanisms. In order to ensure the effectiveness of an anticancer drug, a number of processes should be considered: (a) cellular pharmacokinetics e.g. uptake and retention, (b) evasion of sequestration or metabolic inactivation, (c) interaction with the drug target, (d) evasion of DNA repair pathways (particularly relevant for alkylating agents), and (e) effective induction of cell death (e.g. via apoptosis). Anticancer drug resistance may impede one or more of these processes – but this list is not exhaustive and we must consider anticancer drug resistance as something multifactorial and still awaiting full characterisation. There may be altered drug pharmacokinetic properties (e.g. distribution, uptake, metabolism, and elimination) either at the intra-tumour or at a cellular level. Further, the pharmacodynamic properties of anticancer drugs may be attenuated to prevent the desired drug action (e.g. alteration in the drug target). Finally, downstream signalling of drug actions to elicit an apoptotic response may be dampened in resistant tissue e.g. a lack of the mismatch repair enzyme gene hMLH1 has been shown to reduce the apoptotic response in cells treated with cisplatin (2). One of the biggest problems within the field of anticancer drug resistance has been the poor translation of the key mechanisms identified by in vitro studies to the clinical scenario. Drug resistance may be regarded as a multifaceted and dynamic phenotype which ultimately results in enhanced tumour cell survival and reduced chemo-responsiveness, regardless of the specific mechanism(s) involved. In specific instances, there will be tissue- or drug-dependent predisposing influences towards particular resistance mechanisms. For example, the activity of enzymes such as methyl-guanine-methyl transferase (MGMT) is crucial in the responsiveness of glioma and other brain tumours to the chloroethyl-nitrosourea group of compounds (e.g. CCNU, BCNU). Indeed, competitive inhibition of the enzyme can be shown to bring about enhanced chemo-responsiveness in the clinical setting (3). A specific form of anticancer drug resistance is termed multidrug resistance (MDR) and it is considered by many to represent a significant obstacle to the success of chemotherapy in many ­cancers

Overcoming Multidrug Resistance in Cancer: Clinical Studies of P-Glycoprotein Inhibitors

343

in the clinical setting. MDR is a phenomenon best described as tumour cells in vitro that have been exposed to one cytotoxic agent (e.g. doxorubicin) developing cross-resistance to a range of structurally and functionally unrelated compounds, but usually of natural product origin (e.g. epipodophyllotoxins, taxanes, anthracyclines, etc). The common correlate in MDR cell lines was shown some time ago to be increase (or over-expression) of membrane glycoproteins, the ATP-binding cassette (ABC) transporters. The most widely described and first to be associated with anticancer drug resistance was the 170 kDa P-glycoprotein (Pgp) (1). Drug resistance in the form of MDR may occur intrinsically in some cancers without previous exposure to chemotherapy agents (4). As previously stated, the cytotoxic drugs that are most frequently associated with MDR are hydrophobic, amphipathic natural products, such as the taxanes (paclitaxel, docetaxel), vinca alkaloids (vinorelbine, vincristine, vinblastine), anthracyclines (doxorubicin, daunorubicin, epirubicin), mitoxantrone, epipodophyllotoxins (etoposide, teniposide, topotecan), dactinomycin, and mitomycin C. Pgp belongs to the ABC family of transporters, which are associated with several (in excess of 40) family members that share sequence and structural homology (see Fig. 15.1 for a simplified schematic diagram of Pgp). It is believed that, while this class of transporters have a large number of members, only ten or so are reported to confer the drug-resistant phenotype. These transporters use the energy that is released when they hydrolyse ATP to drive the movement of various (exogenous and endogenous) molecules across the cell membrane. In addition to their physiological expression in normal tissues, many are shown to be expressed, and, importantly, over-expressed, in human tumours. A number of ABC transporters and the chemotherapy drugs to which they have been shown to confer resistance are listed in Table 15.1. In addition to cytotoxic drugs, Pgp also transports several other classes of pharmacological agents including digoxin, ­opiates,

Fig. 15.1. Simplified schematic diagram of P-glycoprotein.

344

Coley

Table 15.1 Anticancer drugs known to be substrates of P-glycoprotein and other efflux transporters Anticancer agents

Pgp [ABCB1]

MRP1 [ABCC1]

BCRP (MXR) [ABCG2]

Anthracyclines (doxorubicin, daunorubicin, epirubicin, mitoxantrone)

+

+

+

Topoisomerase inhibitors (etoposide, teniposideazatoxin)

+

+

Topoisomerase inhibitors (topotecan, irinotecan, SN-38, indolocarbazole NB-506, indolocarbazole J-107088)

Vinca alkaloids (vincristine, vinblastine)

+

+

−

Alkaloids (cepharanthine, homoharringvtonine)

+

−

−

Taxanes (paclitaxel, docetaxel)

+

−

−

Antitumor antibiotics (actinomycin D, mitomycin C)

+

+

−

Antimetabolites (cytarabine)

+

(Methotrexate)

(Methotrexate)

Acridines (amsacrine)

+

Anthracenes (bisantrene)

−

−

+

Flavopiridol

−

−

+

Pgp P-glycoprotein, MRP multidrug resistance protein, BCRP breast cancer resistance protein

polycyclic aromatic hydrocarbons, and technetium (Tc-99m) ­sestamibi that has been used in imaging techniques involving MDR modulators (5). In cancerous tissue, the expression of Pgp is usually highest in tumours that are derived from tissues that normally express Pgp, such as epithelial cells of the colon, kidney, adrenal, pancreas, and liver, resulting in the potential for resistance to some cytotoxic agents before chemotherapy is initiated. In other tumours, the expression of Pgp may be low at the time of diagnosis, but may be induced after exposure to chemotherapy agents, thereby resulting in the development of MDR in those cells (4). There is a body of evidence, albeit at times inconsistent, that links the failure of certain chemotherapeutic agents to the expression of Pgp (6–8). Moreover, the induction of MDR1 RNA can be very rapid following exposure of tumour cells to chemotherapy (9).

Overcoming Multidrug Resistance in Cancer: Clinical Studies of P-Glycoprotein Inhibitors

345

2. Generations of Pgp Inhibitors The 30-year period since the discovery of Pgp has seen an enormous effort to generate clinically applicable inhibitors to restore sensitivity of cancer cells to chemotherapy. Inhibitor development has been through three distinct generations, classified according to the strategy employed in their discovery. 2.1. First-Generation Pgp Inhibitors

Many agents that modulate the Pgp transporter were identified in the 1980s, including verapamil, cyclosporin A (CsA), tamoxifen, and several calmodulin antagonists (10). These agents often produced disappointing results in  vivo because their low binding affinities necessitated the use of high doses, resulting in unacceptable toxicity (10, 11). Many of the first chemosensitizers identified were themselves substrates for Pgp and thus worked by competing with the cytotoxic compounds for efflux by the Pgp pump; therefore, high serum concentrations of the chemosensitizers were necessary to produce adequate intracellular concentrations of the cytotoxic drug (12). In addition, many of these chemosensitizers are substrates for other transporters and enzyme systems, and their use results in unpredictable pharmacokinetic interactions in the presence of chemotherapy agents. To overcome these limitations, several novel analogues of these early chemosensitizers were tested and developed, with the aim of ­finding Pgp modulators with less toxicity and greater potency. Verapamil was rapidly entered clinical trials in 1985 in spite of a lack of knowledge of Pgp and verapamil interactions (13). These clinical trials were complicated by the cardiotoxicity produced by verapamil as a function of its ability to produce hypotension through calcium channel blockade. The immunosuppressant CsA, known to possess effects on the cell membrane, was also considered as an MDR modulator in the 1980s. CsA was shown to potentiate the cytotoxicity of vincristine and doxorubicin via a mechanism that increased the intracellular concentration of anthracyclines (14, 15) and the vinca alkaloids (16). The mechanism underlying the inhibition appeared to be by reducing the membrane interaction of anticancer drugs with Pgp (16, 17) and also via competitive transport since the Pgp expressing cells also showed reduced CsA accumulation (18). CsA was shown to bind to an identical site to vinblastine (16) using a method of azidopine labelling (17). The results from those studies provided the first evidence for multiple sites of drug interaction on Pgp. Perhaps, the biggest impetus for pursuing the early use of MDR modulators in the clinical setting was provided by the work of the Toronto group headed by Chan et al. who first showed that the expression of Pgp was a significant prognostic marker in certain childhood malignancies (19, 20). This group then went on

346

Coley

to use CsA in combination with chemotherapy in retinoblastoma patients and achieved a high cure rate (91% of previously untreated patients remained relapse-free, with salvage therapy combining CsA and chemotherapy prolonging survival in those previously untreated with CsA) (21). Although these trials were limited in size, they raised substantial interest in the cancer research community. The mounting enthusiasm for pharmacological modulator use enabled rapid progression of CsA to phase I clinical trials (22, 23). The most salient points from those trials were that considerable pharmacokinetic changes were seen when CsA was given in combination with agents such as etoposide and highlighted the difficulties in managing the consequences of this. Several drug classes were examined for the ability to behave as chemosensitizing agents in combination with chemotherapy drugs in cultured cancer cell lines. The anti-oestrogen tamoxifen was considered in this regard and this was of considerable interest as it clearly has implications for the treatment of breast cancer (24–27). The overall conclusion following the clinical trials of the firstgeneration MDR modulators was that their clinical potency was too low to act as effective modulating agents, and this was further complicated by the considerable pharmacokinetic interaction of modulating agents with chemotherapeutic agents. 2.2. Second-Generation Pgp Inhibitors

The experiences gained from studies of the first-generation inhibitors provided the grounds for more rational drug discovery programs to search for new MDR modulators. However, the first-generation agents such as verapamil were used as lead compounds in drug design. The calcium channel blockers, which include verapamil, were amongst the classes of drugs that were focused on. A standard clinical preparation of verapamil is composed of a racemic mix of L- and D-forms and thus it was decided to investigate the MDR reversing capacity of the individual isomers. The MDR-reversing capacity of the racemic and individual isomers was shown to be equal (28, 29). Perhaps, the most salient point to be gained form these studies was that the D-isomer of verapamil – dexverapamil – has a tenfold lower calcium channel blocking activity than the L-isomer and for this reason shows significantly less cardiotoxicity. The toxicity profile of dexverapamil in a phase I/II trial combining vinblastine in renal cell carcinoma patients still indicated cardiotoxicity but to a more manageable level (30, 31). However, clinical studies failed to produce any responses in cancer patients treated with drug combinations involving dexverapamil and thus this approach was not pursued any longer (30, 31). Following the success of the first-generation Pgp inhibitor CsA, particularly in the trials from the Toronto group (see above), the search continued for related compounds or derivatives as

Overcoming Multidrug Resistance in Cancer: Clinical Studies of P-Glycoprotein Inhibitors

347

potential new MDR modulators. As the immunosuppressive activity was not a prerequisite for inhibition of Pgp, a number of derivatives devoid of immunosuppressive activity were examined for the ability to overcome resistance to anticancer drugs in vitro (32). The cyclosporin derivative PSC-833 (subsequently named Valspodar™) and the cyclopeptolide SDZ280-446 emerged as lead compounds from these drug discovery programs (33–36). The reversal of drug resistance was observed using a wide range of anticancer drugs including vinca alkaloids, anthracyclines, colchicine, and paclitaxel in combination with the cyclosporins. Pgp inhibition via PSC-833 pulse dosage was pronounced and long-lived, still evident at 48 h (37). The circumvention of drug resistance was shown to be related to an enhancement in drug accumulation and cellular retention of anticancer drugs in cell lines (38), but the effects were also observed in monolayer polarised intestinal cells (39). Such effects have clear implications for the pharmacokinetics of orally administered drugs. Further studies showed that PSC-833 was relatively unselective in its inhibition of ABC transporters (40) and this was considered to be a potential disadvantage as normal tissues may be affected due to their relative and specific expression levels of various ABC family members. However, for some this was considered to be a possible advantage as multiple ABC transporters may coexist in high amounts in some tumours. Valspodar was by far the most rigorously researched of the second-generation MDR modulators, with trials being conducted through phase I–III. Certain patients with acute myeloid leukaemia (AML) emerge as the group that may gain most benefit from MDR modulation as a treatment modality, as witnessed in a number of phase III trials. A trial incorporating daunorubicin, cytarabine and etoposide with valspodar in elderly chemotherapy-naïve AML patients failed to show any benefit (41). The results from that study were mirrored in a very similar trial with the same drug combinations and the same patient group (>65 years of age) (42). An interesting study by Kolitz et al. (43) indicated that a combination of daunorubicin, etoposide and cytarabine with valspodar in previously untreated patients less than 45 years of age provided a survival benefit. Another trial in previously treated AML patients (with no specific age cohort considered) suggested that valspodar in combination with mitoxantrone, etoposide, and cytarabine was ineffective (44). Ovarian cancer patients showing chemorefractory disease were treated with a combination of paclitaxel with valspodar but no clinical benefit was demonstrated (45). Some drug discovery programs focused around computational investigations in an attempt to identify particular affinities of Pgp for chemical moieties, and information was gained regarding the physical and chemical character necessary for drug recognition (46, 47). Several second-generation inhibitors including

348

Coley

the triazine-aminopiperidine S9788, resulted from such studies (48, 49). However, again the clinical efficacy of such agents was never proven, with cardiac toxicity being noted when used in combination with doxorubicin (50). VX-710 (Biricodar) was developed through synthetic chemistry to interact with the FK-506 (Tacrolimus) binding protein and also demonstrated the ability to reverse MDR (51). VX-710 was shown to reverse resistance to doxorubicin, vincristine, etoposide, and paclitaxel in  vitro. The mechanisms underlying the chemosensitisation were shown to be via competitive binding with Pgp-mediated drug efflux, as demonstrated by photoaffinity labelling (51). Subsequent studies with a range of resistant cancer cell lines demonstrated that VX-710 was not selective for Pgp (ABCB1). Both MRP (ABCC1) (52) and BCRP (ABCG2) (53) are also inhibited by VX-710 in cell lines thereby promoting this compound as a “broad-spectrum” modulator of efflux-mediated resistance pathways. A phase II trial in small cell lung cancer patients using doxorubicin and vincristine showed promising results as the combinations indicated only mild or low inherent toxicity of VX-710 (54). The second-generation Pgp inhibitors yielded more information into the mechanisms underlying MDR modulation and of course built upon the knowledge gained from the first-generation modulator studies. The significant advances were the identification and development of compounds with high potency and the ability to reverse the MDR phenotype, with associated lower (but not negligible) toxicity. The identification of the affinity of MDR modulators for multiple ABC transporters began to be considered as problematic, however, as this highlighted the potential for more pharmacokinetic interactions. The potential problems associated with MDR modulator strategies were at this stage becoming more apparent from the reports of the clinical trials emerging in the mounting literature on MDR research. 2.3. Third-Generation Pgp Inhibitors

The development of third-generation of Pgp inhibitors used combinatorial chemistry approaches with lead compounds derived from pharmacological information already available on drug–Pgp interaction. More rational design was possible at this stage of MDR modulator development based on data from the previous generations of drug development. A number of criteria were identified as being essential features for new reversal agents; these included reversal of the resistant phenotype at low dose levels with a high level of affinity interaction with Pgp. Other features required for candidate molecules were an indication of selectivity for a particular ABC transporter (e.g. ABCB1 vs. ABCG2) and a reduced susceptibility for metabolic biotransformation by the cytochrome P450 (CYP) isoform CYP3A4.

Overcoming Multidrug Resistance in Cancer: Clinical Studies of P-Glycoprotein Inhibitors

349

Elacridar – GF120918/GG918 – was the first of this new generation of compounds, developed by Glaxo laboratories from a combinatorial chemistry program (55). The structure of the compound is an acridonecarboxamide which in  vitro demonstrated reversal of MDR by chemosensitisation in combination with both vincristine and doxorubicin (55). Reversal of drug resistance was accompanied by enhancement of drug accumulation of anticancer with the potency of elacridar to reverse drug resistance in the nanomolar range. Moreover, the potency of elacridar was also borne out in studies using cell lines with very high levels of drug resistance (56). The inhibition of anticancer drug efflux was proposed to be via allosteric binding on Pgp (57). Elacridar was shown to enhance the cytotoxicity of anticancer drugs in cells expressing BCRP and with similar potency to that observed with Pgp (58). The ability of Elacridar to interact with BCRP was demonstrated in experiments where topotecan accumulation was enhanced (59). Ex vivo restoration of chemosensitivity to anthracyclines was shown in clinical samples of acute myelogenous leukaemia (AML) (60) and subsequently a number of in vivo studies were undertaken with the drug. The development of a number of modulators was accompanied by the development of so-called “functional assays” to look for demonstration of MDR modulation in an ex vivo setting. One approach was to analyse of the CD56+ subset of peripheral blood lymphocytes by making use of their intrinsic functional Pgp expression and their reduced ability to accumulate the fluorescent dye rhodamine123, using rapid flow cytometric analysis. Ex vivo data revealed that incubation of cancer cells with elacridar increased rhodamine123 accumulation and these results can be regarded as a surrogate marker for Pgp activity in vivo (61). In vivo studies showed that IV administration of elacridar demonstrated its good bioavailability and reversed doxorubicin resistance (55). Importantly, ex vivo analysis in myeloma and AML samples indicated the concentrations of the modulator necessary to inhibit Pgp were clinically attainable and thus provided the rationale for further studies (62). A drug discovery program carried out by Xenova PLC (Slough, UK) based on a diketopiperazine compound produced a potent MDR modulator XR9051 (63). However, there were shortcomings in the physico-chemical properties of this molecule (e.g. hydrophobicity) and subsequently a series of modifications to the anthranilimide nucleus were undertaken (64). The most potent of the resultant derivatives was XR9576 (tariquidar), which was able to completely overcome resistance to doxorubicin, paclitaxel, etoposide, and vincristine at a concentration range of 25–80 nM in a number of MDR cell lines (65). Resistance was overcome through inhibition of efflux as demonstrated by the ability of tariquidar to increase the cellular accumulation of the Pgp substrates rhodamine123 and [3H]-daunomycin. A very

350

Coley

important feature of tariquidar was that its pharmacological effects persisted for longer than 22 h following its removal from the assay system, which compares favourably with the persistence of cyclosporine and verapamil, being around 1 h. The antitumour activities of paclitaxel, vincristine, and etoposide were restored in whole mice, in highly resistant tumour xenografts (65) and in a tumour spheroid model (66). The high potency of tariquidar interaction with Pgp was demonstrated by a KD ~ 5 nM and moreover, the drug did not appear to act as a substrate for Pgp itself (67). The ability of tariquidar to inhibit vinblastine and paclitaxel transport was shown to occur via allosteric reductions in the binding affinity of anticancer drugs on Pgp (68). The interaction of tariquidar with Pgp was shown in some elegant studies to be characterised by a slow rate of dissociation and perturbation of cellular drug compartmentalization (68). The awareness of the role of the hepatic CYP3A4 isoform in the pharmacokinetic profiles seen in trials incorporating MDR modulators and cytotoxic drugs became a specific and important issue; it was shown that it shared a considerable substrate specificity with Pgp. Significantly, tariquidar was shown to not be subjected to metabolism via the CYP3A4 isoform but involvement of the 1A2, 2C8, and 2C9 isoforms was reported (69). This was heralded as a considerable advantage and positioned tariquidar very favourably compared with the majority of other MDR modulators. BCRP interaction was noted but the affinity of tariquidar for this transporter was shown to be considerably lower than for Pgp (70). The first phase I study using tariquidar indicated no effect on the pharmacokinetics of vinorelbine and no novel toxicities were reported (5). Twenty-five patients enrolled in this trial, of which 13 experienced disease stabilization and 2 had a partial response. Interestingly, the maximum tolerated dosage for vinorelbine administered alone is 30 mg/m2, but in combination with tariquidar it was shown to be 22.5 mg/m2. This reduction may occur as a result of a pharmacodynamic interaction (possibly at the level of Pgp in the bone marrow) that led to increased toxicity of the drug without increase in systemic exposure. In other phase I/II studies combining tariquidar with paclitaxel or doxorubicin for the treatment of solid tumours, no adverse pharmacokinetic interactions were reported (71, 72). However, in spite of the promising pre-clinical and phase I clinical trial data, subsequent trials with tariquidar were terminated owing to toxicities (serious in some cases) associated with the chemotherapy in the tariquidar arm (73). This may have been a reflection on the previous pharmacodynamic effects suggested in the phase I trial with vinorelbine (5). There has been some discussion as to why vinorelbine was administered at a slightly elevated starting dose of 25  mg/m2, and paclitaxel dose at a higher than usual dose was 200  mg/m2 (74). In spite of these

Overcoming Multidrug Resistance in Cancer: Clinical Studies of P-Glycoprotein Inhibitors

351

initial setbacks, the National Cancer Institute (NCI) began further phase I/II trials and these remain described as ongoing status (at the time of going to press). A phase I study of tariquidar plus chemotherapy in treating children with various solid tumours has finished patient enrolment, and results are pending. There are two phase II trials, one combining tariquidar with docetaxel in patients with lung, ovarian, cervical, or kidney cancer and another with combination chemotherapy and surgical resection in the treatment of adrenocortical cancer are both still recruiting. Another third-generation MDR modulator is LY335979, otherwise known as zosuquidar, which was based on the quinoline MS-073, previously shown to be an effective modulator of Pgp (75). In vitro studies showed that zosuquidar restored sensitivity to vinblastine, doxorubicin, etoposide, and paclitaxel in drug-resistant CEM/VLB100 cells (76) and these effects were apparent following removal of the modulator from the culture environment. Zosuquidar was shown to directly inhibit Pgp and it also showed excellent potency some 500- to 1500-fold greater than that seen for CsA and verapamil (77), with a binding affinity KD of 73 nM (78). It appears that zosuquidar is not transported by Pgp but it can show effects on accumulation of anticancer drugs possibly via allosteric pathways (79). Importantly, zosuquidar did not demonstrate binding to ABCC1 or ABCG2 (78–81) with a lower affinity for CYP3A compared to that for Pgp (78). These data suggest therefore that zosuquidar is selective for Pgp, with low potential for interfering with the metabolism of anticancer drugs. In vivo studies have demonstrated that zosuquidar does not alter the pharmacokinetics of etoposide or doxorubicin in mice (76, 79), with small changes to the pharmacokinetics of paclitaxel (82, 83). In a phase I clinical trial orally administered zosuquidar in combination with doxorubicin was given to patients with advanced non-haematological malignancies (84). The maximum tolerated dose of the inhibitor was 300 mg/m2 (administered every 12 h for 4  days) with doxorubicin doses at either 45 or 75  mg/m2. There were no reported toxicities attributable to zosuquidar, nor was there any alteration in the pharmacokinetics of doxorubicin. Another phase I study administered the inhibitor intravenously, again, with doxorubicin. In this case, the maximal doses administered were 640 and 75 mg/m2 f or zosuquidar and doxorubicin, respectively. At the higher doses of the inhibitor (exceeding 500 mg), modest pharmacokinetic interactions were seen with a decreased doxorubicin clearance (c. 20%) (85). Although the reduced clearance was associated with an enhanced leucopaenia and thrombocytopenia, this effect was not considered clinically significant. Docetaxel chemotherapy combined with zosuquidar was also well-tolerated, with minimal increases in docetaxel plasma peak concentrations and AUC levels. Based on those results, a

352

Coley

phase II trial using the same combination treatment has been started in breast carcinoma (86), the results of which are eagerly awaited. A phase I trial in AML patients concluded that zosuquidar could be given safely in combination with induction doses of daunorubicin and cytosine arabinoside (87). Importantly, 11 out of the 16 patients in this study achieved a complete remission. In contrast, a phase I trial of zosuquidar in combination with vinorelbine in patients with solid tumours showed no objective responses, and the inhibitor reduced the clearance of vinorelbine (88). Results from the majority of phase I/II trials, have shown an absence, or only modest effects, of zosuquidar on the pharmacokinetics of anti-cancer drugs, supporting its progression into phase III trials. However, to date, only one such trial has been initiated looking at AML and high-risk myelodysplastic syndrome. A total of 442 patients were randomized to standard chemotherapy (cytosine arabinoside and daunorubicin) with or without zosuquidar at 550 mg/m2. Disappointingly, treatment with the inhibitor did not provide any benefit, even in the subset of patients with measurable Pgp expression, with overall survival in the experimental and treatment arm being 8 and 9 months, respectively (89). Recently, the MDR modulators tariquidar, along with elacridar were shown to be very effective in modulating the bloodbrain-barrier in a glioblastoma tumour model treated with a combination involving paclitaxel (90), which suggest there may be a chance for the use of such modulating agents in more specific clinical scenarios. Another significant member of the first-MDR modulators is OC144-093 (ONT-093) (91). This compound is based on a diarylimidazole structure and it was identified as a potent Pgp modulator using high throughput cell screening. ONT-093 showed activity in MDR Pgp expressing cells of breast, ovarian, uterine, lymphoma, and colorectal cancer lines. Effective chemosensitisation via incubation of cells with ONT-093 in combination with either doxorubicin, paclitaxel or vinblastine was achieved at a dose of only 30 nM. The inhibition of Pgp is reversible, but long-lived (>12 h) which is a common feature in the first-generation inhibitors. In pharmacokinetic studies, ONT-093 was shown to be primarily converted to an O-de-ethylated metabolite by liver microsomes, but importantly, this was shown not to occur via the CYP3A4 isoform (92). Administration of paclitaxel with ONT-093 co-administered via the oral route enhanced paclitaxel bioavailability, which is consistent with inhibition of Pgp in the GI tract. In summary, the third-generation inhibitors represent significant improvements on the previous generation compounds showing enhanced potency with minimal pharmacokinetic interactions. Definitive clinical trial data are awaited in order for possible clinical development of this treatment modality.

Overcoming Multidrug Resistance in Cancer: Clinical Studies of P-Glycoprotein Inhibitors

2.4. Fourth Generation: Natural Sources

353

An increased pharmacological characterisation of multidrug efflux transporters has revealed their important role in the pharmacokinetics of a large number of prescribed medications, quite aside from their role in the pharmacokinetics of cancer chemotherapy. This knowledge has stemmed from the expression of ABC transporters in normal tissues, particularly those involved in drug absorption and excretory pathways. In addition, it is well known that components of food can also interfere with the oral bioavailability of many drugs and that the drug–food interactions may involve Pgp. A number of established foods, in particular fruits such as orange, grapefruit, and strawberry, can inhibit Pgp function (93, 94) and have been shown to effect the transport of drugs such as vinblastine in Caco-2 cells. Generally, the potencies of natural extracts for MDR reversibility are low (i.e. high micromolar) and therefore these compounds are unlikely candidates for clinical modulation of Pgp activity. However, even given the low potency there is the opportunity that active components will influence drug bioavailabilities and other pharmacokinetic parameters. Natural products such as curcumin and the flavonoids kaempferol and quercetin have been shown to influence Pgp function (95) in human cancer cell lines with the MDR phenotype. Flavonoid dimers have been the subject of some more recent drug discovery programs (96). Mechanistic studies have shown that developmental flavonoids can stimulate the ATPase activity of Pgp. Competition for transport or inhibition of the process would influence drug absorption or elimination at various tissues in the body. The characterisation of natural product MDR modulators is currently not a high priority for the cancer research community. However, the active components of food/plant extracts already identified could be exploited as lead compounds for chemical modification to generate novel, selective, and high affinity Pgp inhibitors.

3. Final Comments and Conclusions With a complex and sometimes fraught past history, the state of clinical development of Pgp inhibitors is currently relatively inactive. Considerable resources and time have been spent on the development of the third-generation inhibitors. However, these efforts have failed to produce clinical trial data with the desired outcomes, due to issues with pharmacokinetic or pharmacodynamic interactions and toxicities. Unfortunately, there has been poor dissemination of trial results by the pharmaceutical industry which conflicts with fruitful interaction between industry and academic groups. In future trials there should be more rigour associated with clinical trial design with more use of surrogate

354

Coley

markers and imaging techniques. In spite of the issues MDR modulator development poses, the problem of clinical anticancer drug resistance remains a significant issue and thus we should continue our efforts to overcome this. References 1. Juliano RL, Ling V (1976) A surface glycoprotein modulating drug permeability in Chinese hamster ovary cell mutants. Biochim Biophys Acta 455:152–162 2. Plumb JA, Strathdee G, Sludden J, Kaye SB (2000) Reversal of drug resistance in human tumour xenografts by 2¢-deoxy-5-azacytidineinduced demethylation of the hMLH gene promoter. Cancer Res 60:6039–6044 3. Marathi UK, Dolan ME (1994) Anti-neoplastic activity of sequenced administration of O6-benzylguanine, streptozotocin and 1, 3-bis(2-chlroethy)-1-nitrosourea in  vitro and in vivo. Biochem Pharmacol 48:2127–2134 4. Scotto KW (2002) Transcriptional regulation of ABC transporters. Oncogene 22:7496–7511 5. Agrawal M, Abraham J, Balis FM et al (2003) Increased 99mTc-sestamibi accumulation in normal liver and drug resistant tumours after the administration of the glycoprotein inhibitor, XR9576. Clin Cancer Res 9:650–656 6. Linn SC, Pinedo HM, van Ark-Otte J (1997) Expression of drug resistance proteins in breast cancer, in relation to chemotherapy. Int J Cancer 71:787–795 7. Rudas M, Filipits M, Taucher S (2003) Expression of MRP1, LRP and Pgp in breast carcinoma patients treated with preoperative chemotherapy. Breast Cancer Treat 81: 149–157 8. Mechetner E, Kyshtoobayeva A, Zonis S (1998) Levels of multidrug resistance (MDR1) P-glycoprotein expression in human breast cancer correlate with in vitro resistance to Taxol and doxorubicin. Clin Cancer Res 4:389–398 9. Abolhoda A, Wilsin AE, Ross H et al (1999) Rapid activation of MDR1 gene expression in human metastatic sarcoma after in vivo exposure to doxorubicin. Clin Cancer Res 5: 3352–3356 10. Krishna R, Mayer LD (2000) Multidrug resistance (MDR) in cancer. Mechanisms, reversal using modulators of MDR and the role of MDR modulators in influencing the pharmacokinetics of anticancer drugs. Eur J Pharm Sci 11:265–283 11. Ferry DR, Trauneker H, Kerr DJ (1996) Clinical trials of P-glycoprotein reversal in solid tumours. Eur J Cancer 32A:1070–1081

12. Ambudkar SV, Dey S, Hrycyna CA et  al (1999) Biochemical, cellular and pharmacological aspects of the multidrug transporter. Ann Rev Pharmacol Toxicol 39:361–398 13. Benson AB, Trump DL, Koeller JM et  al (1985) Phase I Study of vinblastine and verapamil given by concurrent i.v. infusion. Cancer Treat Rep 69:795–799 14. Coley HM, Twentyman PR, Workman P (1989) Improved cellular accumulation is characteristic of anthracyclines which retain high activity in multidrug resistant cell lines, alone or in combination with verapamil or cyclosporine A. Biochem Pharmacol 38: 4467–4475 15. Silbermann MH, Boersma AW, Janssen AL et al (1989) Effects of cyclosporin A and verapamil on the intracellular daunorubicin accumulation in Chinese hamster ovary cells with increasing levels of drug-resistance. Int J Cancer 44:722–726 16. Tamai I, Safa AR (1990) Competitive interaction of cyclosporins with the Vinca alkaloid-binding site of P-glycoprotein in multidrug-resistant cells. J Biol Chem 265: 16509–16513 17. Tamai I, Safa AR (1991) Azidopine noncompetitively interacts with vinblastine and cyclosporin A binding to P-glycoprotein in multidrug resistant cells. J Biol Chem 266: 16796–16800 18. Goldberg H, Ling V, Wong PY, Skorecki K (1998) Reduced cyclosporin accumulation in multidrug-resistant cells. Biochem Biophys Res Commun 152:552–558 19. Chan HS, Thorner PS, Haddad G (1990) Immunohistochemical detection of P-glycoprotein: prognostic correlation in soft tissue sarcoma of childhood. J Clin Oncol 8:689–704 20. Chan HS, Haddad G, Thorner PS (1991) P-glycoprotein expression as a predictor of the outcome of therapy for neuroblastoma. N Engl J Med 325:1608–1614 21. Chan HS, DeBoer G, Thiessen JJ (1996) Combining cyclosporin with chemotherapy controls intraocular retinoblastoma without requiring radiation. Clin Cancer Res 2: 1499–1508

Overcoming Multidrug Resistance in Cancer: Clinical Studies of P-Glycoprotein Inhibitors 22. Lum BL, Kaubisch S, Yahanda AM et  al (1992) Alteration of etoposide pharmacokinetics and pharmacodynamics by cyclosporine in a phase I trial to modulate multidrug resistance. J Clin Oncol 10:1635–1642 23. Yahanda AM, Alder KM, Fisher GA et  al (1992) Phase I trial of etoposide with cyclosporine as a modulator of multidrug resistance. J Clin Oncol 10:1624–1634 24. Callaghan R, Higgins CF (1995) Interaction of tamoxifen with the multidrug resistance P-glycoprotein. Br J Cancer 71:294 25. Fine RL, Sachs CW, Albers M (1991) Tamoxifen potentiates the cytotoxicity of vinblastine by increasing intracellular drug accumulation. Proc Am Assoc Cancer Res 32:375 26. Ramu A, Glaubiger D, Fuks Z (1984) Reversal of acquired resistance to doxorubicin in P388 murine leukemia cells by tamoxifen and other triparanol analogues. Cancer Res 44: 4392–4395 27. Stuart NS, Philip P, Harris AL et  al (1992) High-dose tamoxifen as an enhancer of etoposide cytotoxicity. Clinical effects and in  vitro assessment in p-glycoprotein expressing cell lines. Br J Cancer 66:833–839 28. Damiani D, Michieli M, Michelutti A et  al (1993) P170-associated resistance to doxorubicin, daunorubicin and idarubicin. Anticancer Drugs 4:173–180 29. Haussermann K, Benz B, Gekeler V, Schumacher K, Eichelbaum M (1991) Effects of verapamil enantiomers and major metabolites on the cytotoxicity of vincristine and daunomycin in human lymphoma cell lines. Eur J Clin Pharmacol 40:53–59 30. Motzer RJ, Lyn P, Fischer P et  al (1995) Phase I/II trial of dexverapamil plus vinblastine for patients with advanced renal cell carcinoma. J Clin Oncol 13:1958–1965 31. Warner E, Hedley D, Andrulis I et al (1998) Phase II study of dexverapamil plus anthracycline in patients with metastatic breast cancer who have progressed on the same anthracycline regimen. Clin Cancer Res 4:1451–1457 32. Gaveriaux C, Boesch D, Boelsterli JJ et al (1989) Overcoming multidrug resistance in Chinese hamster ovary cells in  vitro by cyclosporin A (Sandimmune) and non-immunosuppressive derivatives. Br J Cancer 60:867–871 33. Jachez B, Nordmann R, Loor F (1993) Restoration of taxol sensitivity of multidrugresistant cells by the cyclosporine SDZ PSC 833 and the cyclopeptolide SDZ 280-446. J Natl Cancer Inst 85:478–483 34. Keller RP, Altermatt HJ, Nooter K et al (1992) SDZ PSC 833, a non-immunosuppressive

35.

36.

37.

38.

39.

40.

41.

42.

43.

355

cyclosporine: its potency in overcoming P-glycoprotein-mediated multidrug resistance of murine leukemia. Int J Cancer 50:593–597 Pourtier-Manzanedo A, Didier AD, Muller CD, Loor F (1992) SDZ PSC 833 and SDZ 280-446 are the most active of various resistance-modifying agents in restoring rhodamine-123 retention within multidrug resistant P388 cells. Anticancer Drugs 3:419–425 Te Boekhorst PA, van Kapel J, Schoester M, Sonneveld P (1992) Reversal of typical multidrug resistance by cyclosporin and its nonimmunosuppressive analogue SDZ PSC 833 in Chinese hamster ovary cells expressing the mdr1 phenotype. Cancer Chemother Pharmacol 30:238–242 Boesch D, Loor F (1994) Extent and persistence of P-glycoprotein inhibition in multidrug-resistant P388 cells after exposure to resistance-modifying agents. Anticancer Drugs 5:229–238 Michieli M, Damiani D, Michelutti A et  al (1994) Restoring uptake and retention of daunorubicin and idarubicin in P170-related multidrug resistance cells by low concentration D-verapamil, cyclosporin-A and SDZ PSC 833. Haematologica 79:500–507 Zacherl J, Hamilton G, Thalhammer T et  al (1994) Inhibition of P-glycoprotein-mediated vinblastine transport across HCT-8 intestinal carcinoma monolayers by verapamil, cyclosporine A and SDZ PSC 833 in dependence on extracellular pH. Cancer Chemother Pharmacol 34:125–132 Barrand MA, Rhodes T, Center MS, Twentyman P (1993) Chemosensitisation and drug accumulation effects of cyclosporin A, PSC-833 and verapamil in human MDR large cell lung cancer cells expressing a 190 k membrane protein distinct from P-glycoprotein. Eur J Cancer 29A:408–415 Van der Holt B, Lowenberg B, Burnett AK et al (2005) The value of the MDR1 reversal agent PSC-833 in addition to daunorubicin and cytarabine in the treatment of elderly patients with previously untreated acute myeloid leukemia (AML), in relation to MDR1 status at diagnosis. Blood 106:2646–2654 Baer MR, George SL, Dodge RK et al (2002) Phase 3 study of the multidrug resistance modulator PSC-833 in previously untreated patients 60 years of age and older with acute myeloid leukemia: Cancer and Leukemia Group B Study 9720. Blood 100:1224–1232 Kolitz JE, George SL, Dodge RK et al (2004) Dose escalation studies of cytarabine, daunorubicin, and etoposide with and without multidrug resistance modulation with PSC-833 in

356

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

Coley untreated adults with acute myeloid leukemia younger than 60 years: final induction results of Cancer and Leukemia Group B Study 9621. J Clin Oncol 22:4290–4301 Greenberg PL, Lee SJ, Advani R et al (2004) Mitoxantrone, etoposide, and cytarabine with or without valspodar in patients with relapsed or refractory acute myeloid leukemia and highrisk myelodysplastic syndrome: a phase III trial (E2995). J Clin Oncol 22:1078–1086 Fracasso PM, Brady MF, Moore DH et  al (2001) Phase II study of paclitaxel and valspodar (PSC 833) in refractory ovarian carcinoma: a gynecologic oncology group study. J Clin Oncol 19:2975–2982 Zamora JM, Beck WT (1986) Chloroquine enhancement of anticancer drug cytotoxicity in multiple drug resistant human leukemic cells. Biochem Pharmacol 35:4303–4310 Wang RB, Kuo CL, Lien LL, Lien EJ (2003) Structure-activity relationship: analyses of p-glycoprotein substrates and inhibitors. J Clin Pharm Ther 28:203–228 Dhainaut A, Regnier G, Atassi G et al (1992) New triazine derivatives as potent modulators of multidrug resistance. J Med Chem 35:2481–2496 Hill BT, van der Graaf WT, Hosking LK et al (1993) Evaluation of S9788 as a potential modulator of drug resistance against human tumour sublines expressing differing resistance mechanisms in  vitro. Int J Cancer 55:330–337 Punt CJ, Voest EE, van Oosterom AT et  al (1997) Phase 1B study of doxorubicin in combination with the multidrug resistance reversing agent S9788 in advanced colorectal cancer and renal cell cancer. Br. J. Cancer 76:1376–1381 Germann UA, Shlyakhter D, Mason VS (1997) Cellular and biochemical characterization of VX-710 as a chemosensitizer: reversal of P-glycoprotein-mediated multidrug resistance in vitro. Anticancer Drugs 8:125–140 Germann UA, Ford PJ, Shlyakhter D, Mason VS, Harding MW (1997) Chemosensitization and drug accumulation effects of VX-710, verapamil, cyclosporin A, MS-209 and GF120918 in multidrug resistant HL60/ADR cells expressing the multidrug resistance-associated protein MRP. Anticancer Drugs 8:141–155 Minderman H, O’Loughlin KL, Pendyala L, Baer MR (2004) VX-710 (biricodar) increases drug retention and enhances chemosensitivity in resistant cells overexpressing P-glycoprotein, multidrug resistance protein, and breast cancer resistance protein. Clin Cancer Res 10:1826–1834

54. Gandhi L, Harding MW, Neubauer M et  al (2007) A phase II study of the safety and efficacy of the multidrug resistance inhibitor VX-710 combined with doxorubicin and vincristine in patients with recurrent small cell lung cancer. Cancer 109:924–932 55. Hyafil F, Vergely C, Du Vignaud P, GrandPerret T (1993) In vitro and in vivo reversal of multidrug resistance by GF120918, an acridonecarboxamide derivative. Cancer Res 53:4595–4602 56. Myer MS, Joone G, Chasen MR, van Rensburg CE (1999) The chemosensitizing potential of GF120918 is independent of the magnitude of P-glycoprotein-mediated resistance to conventional chemotherapeutic agents in a small cell lung cancer line. Oncol Rep 6:217–218 57. Martin C, Berridge G, Mistry P et al (2000) Drug binding sites on P-glycoprotein are altered by ATP binding prior to nucleotide hydrolysis. Biochemistry 39:11901–11906 58. De Bruin M, Miyake K, Litman T, Robey R, Bate SE (1999) Reversal of resistance by GF120918 in cell lines expressing the ABC half-transporter, MXR. Cancer Lett 146: 117–126 59. Maliepaard M, van Gastelen MA, Tohgo A et  al (2001) Circumvention of breast cancer resistance protein (BCRP)-mediated resistance to camptothecins in  vitro using nonsubstrate drugs or the BCRP inhibitor GF120918. Clin Cancer Res 7:935–941 60. Elgie AW, Sargent JM, Williamson CJ, Lewandowicz GM, Taylor CG (1999) Comparison of P-glycoprotein expression and function with in vitro sensitivity to anthracyclines in AML. Adv Exp Med Biol 457: 29–33 61. Witherspoon SM, Emerson DL, Kerr BM et al (1996) Flow cytometric assay of modulation of P-glycoprotein function in whole blood by the multidrug resistance inhibitor GG918. Clin Cancer Res 2:7–12 62. den Ouden D, van den Heuvel M, Schoester M, van Rens G, Sonneveld P (1996) In vitro effect of GF120918, a novel reversal agent of multidrug resistance, on acute leukemia and multiple myeloma cells. Leukemia 10: 1930–1936 63. Dal IL, Tuffley W, Callaghan R et al (1998) Reversal of P-glycoprotein-mediated multidrug resistance by XR9051, a novel diketopiperazine derivative. Br J Cancer 78:885–892 64. Roe M, Folkes A, Ashworth P et  al (1999) Reversal of P-glycoprotein mediated multidrug resistance by novel anthranilamide derivatives. Bioorg Med Chem Lett 9:595–600

Overcoming Multidrug Resistance in Cancer: Clinical Studies of P-Glycoprotein Inhibitors 65. Mistry P, Stewart AJ, Dangerfield W et  al (2001) In vitro and in  vivo reversal of P-glycoprotein-mediated multidrug resistance by a novel potent modulator, XR9576. Cancer Res 61:749–758 66. Walker J, Martin C, Callaghan R (2004) Inhibition of P-glycoprotein function by XR9576 in a solid tumour model can restore anticancer drug efficacy. Eur J Cancer 40: 594–605 67. Martin C, Berridge G, Mistry P et al (1999) The molecular interaction of the high affinity reversal agent XR9576 with P-glycoprotein. Br J Pharmacol 128:403–411 68. Martin C, Berridge G, Mistry P et al (2000) Drug binding sites on P-glycoprotein are altered by ATP binding prior to nucleotide hydrolysis. Biochemistry 39:11901–11906 69. Labrie P, Maddaford SP, Lacroix J et al (2006) In vitro activity of novel dual action MDR anthranilamide modulators with inhibitory activity at CYP-450. Bioorg Med Chem 14:7972–7987 70. Robey RW, Steadman K, Polgar O et  al (2004) Pheophorbide A is a specific probe for ABCG2 function and inhibition. Cancer Res 64:1242–1246 71. Pusztai L, Wagner P, Ibrahim N et al (2005) Phase II study of tariquidar, a selective P-glycoprotein inhibitor, in patients with chemotherapy-resistant advanced breast cancer. Cancer 104:682–691 72. Thomas H, Coley HM (2003) Overcoming multidrug resistance in cancer: an update on the clinical strategy of inhibiting P-glycoprotein. Cancer Control 10:159–165 73. Nobili S, Landini I, Giglioni B, Mini E (2006) Pharmacological strategies for overcoming multidrug resistance. Curr Drug Targets 7:861–879 74. Fox E, Bates SE (2007) Tariquidar (XR9576): a P-glycoprotein drug efflux pump inhibitor. Expert Rev Anticancer Ther 7:447–459 75. Sato W, Fukazawa N, Suzuki T, Yusa K, Tsuruo T (1991) Circumvention of multidrug resistance by a newly synthesized quinoline derivative, MS-073. Cancer Res 51:2420–2424 76. Dantzig AH, Shepard RL, Cao J et al (1996) Reversal of P-glycoprotein-mediated multidrug resistance by a potent cyclopropyldibenzosuberane modulator, LY335979. Cancer Res 56:4171–4179 77. Green LJ, Marder P, Slapak CA (2001) Modulation by LY335979 of P-glycoprotein function in multidrug-resistant cell lines and human natural killer cells. Biochem Pharmacol 61:1393–1399

357

78. Dantzig AH, Shepard RL, Law KL et  al (1999) Selectivity of the multidrug resistance modulator, LY335979, for P-glycoprotein and effect on cytochrome P-450 activities. J Pharmacol Exp Ther 290:854–862 79. Starling JJ, Shepard RL, Cao J et  al (1997) Pharmacological characterization of LY335979: a potent cyclopropyldibenzosuberane modulator of P-glycoprotein. Adv Enzyme Regul 37:335–347 80. Lehne G, Morkrid L, den Boer M, Rugstad HE (2000) Diverse effects of P-glycoprotein inhibitory agents on human leukemia cells expressing the multidrug resistance protein (MRP). Int J Clin Pharmacol Ther 38:187–195 81. Shepard RL, Cao J, Starling JJ, Dantzig AH (2003) Modulation of P-glycoprotein but not MRP1- or BCRP-mediated drug resistance by LY335979. Int J Cancer 103:121–125 82. Bardelmeijer HA, Ouwehand M, Beijnen JH, Schellens JH, van Tellingen O (2004) Efficacy of novel P-glycoprotein inhibitors to increase the oral uptake of paclitaxel in mice. Invest New Drugs 22:219–229 83. Callies S, de Alwis DP, Harris A et al (2003) A population pharmacokinetic model for paclitaxel in the presence of a novel Pgp modulator, Zosuquidar Trihydrochloride (LY335979). Br J Clin Pharmacol 56:46–56 84. Rubin EH, de Alwis DP, Pouliquen I et  al (2002) A phase I trial of a potent P-glycoprotein inhibitor, Zosuquidar. 3HCl trihydrochloride (LY335979), administered orally in combination with doxorubicin in patients with advanced malignancies. Clin Cancer Res 8:3710–3717 85. Sandler A, Gordon M, De Alwis DP et  al (2004) A Phase I trial of a potent P-glycoprotein inhibitor, zosuquidar trihydrochloride (LY335979), administered intravenously in combination with doxorubicin in patients with advanced malignancy. Clin Cancer Res 10:3265–3272 86. Fracasso PM, Goldstein LJ, de Alwis DP et al (2004) Phase I study of docetaxel in combination with the P-glycoprotein inhibitor, zosuquidar, in resistant malignancies. Clin Cancer Res 10:7220–7228 87. Gerrard G, Payne E, Baker RJ et  al (2004) Clinical effect and P-glycoprotein inhibition in patients with acute myeloid leukaemia treated with zosuquidar trihydrochloride, daunorubicin and cytarabine. Haematologica 89:782–790 88. Le LH, Moore MJ, Siu LL et al (2005) Phase I study of the multidrug resistance inhibitor zosuquidar administered in combination with vinorelbine in patients with advanced

358

89.

90.

91.

92.

Coley solid tumours. Cancer Chemother Pharmacol 56:154–160 Cripe LD, Li X, Litzow M et al (2006) A randomized, placebo-controlled, double blind trial of the MDR modulator, zosuquidar, during conventional induction and post-remission therapy for pts > 60 years of age with newly diagnosed acute myeloid leukemia (AML) or high-risk myelodysplastic syndrome (HR-MDS): ECOG 3999. ASH Annu Meet Abstr 108:423 Hubesack M, Muller C, Hocherl P et  al (2008) Effect of ABCB1 modulators elacridar and tariquidar on the distribution of pacltaxel in nude mice. J Cancer Res Clin Oncol 134: 597–607 Newman MJ, Roadarte JC, Benbatoul KD et  al (2000) Discovery and characteri­ sation of OC144-093, a novel inhibitor of P-glycoprotein-mediated multidrug resistance. Cancer Res 60:2964–2972 Guns ES, Bullock PL, Reimer ML et al (2001) Assessment of the involvement of CYP3A in the vitro metabolism of a new modulator of

93.

94.

95.

96.

MDR in cancer chemotherapy, OC144-193, by human liver microsomes. Eur J Drug Metab Pharmacokinet 26:273–282 Deferme S, Van Gelder J, Augustijns P (2002) Inhibitory effect of fruit extracts on P-glycoprotein-related efflux carriers: an in-vitro screening. J Pharm Pharmacol 54:1213–1219 Honda Y, Ushigome F, Koyabu N et al (2004) Effects of grapefruit juice and orange juice components on P-glycoprotein- and MRP2mediated drug efflux. Br J Pharmacol 143:856–864 Limtrakul P, Khantamat O, Pintha K (2005) Inhibition of P-glycoprotein function and expression by kaempferol and quercetin. J Chemother 17:86–95 Chan KF, Zhao Y, Burkett BA et  al (2006) Flavonoid dimmers as bivalent modulators for P-glycoprotein-based multidrug resistance: synthetic apigenin homodimers linked with defined-length poly(ethylene glycol) spacers increase drug retention and enhance chemosensitivity in resistant cancer cells. J Med Chem 49:6742–6759

Chapter 16 Pharmacokinetic and Pharmacodynamic Implications of P-Glycoprotein Modulation Jeannie M. Padowski and Gary M. Pollack Abstract Modulation of P-glycoprotein (Pgp)-mediated transport has significant pharmacokinetic implications for Pgp substrates. Pharmacokinetic alterations may be at the systemic (blood concentrations), regional (organ or tissue concentrations), or local (intracellular concentrations) level. Regardless of the particular location of Pgp modulation, changes in substrate pharmacokinetics will have the potential to alter the magnitude and duration of pharmacologic effect (pharmacodynamics). It is important to understand each of the aspects of Pgp modulation for a given Pgp substrate in order to predict the degree to which Pgp modulation may affect that substrate, to minimize untoward effects associated with that modulation, or to exploit that modulation for specific therapeutic advantage. Key words: P-glycoprotein, Substrate, Transport, Pharmacokinetics, Pharmacodynamics

1. Introduction Transport proteins that operate in the efflux direction (against substrate uptake into a particular organ or tissue) have garnered significant attention in recent years within the pharmaceutical sciences community, and have been the subject of numerous comprehensive reviews (1–3). These proteins can act at one or more steps in the cascade of pharmacokinetic and pharmacodynamic events that ultimately lead to the biologic response, either beneficial or detrimental, to a pharmacologic agent. Transport proteins operating in the efflux direction in the gastrointestinal tract can impede absorption (4), decreasing the rate and, potentially, the extent of presentation of the substrate to the systemic circulation. Outwardly directed transport in organs of elimination, primarily the liver (5) and kidney (6), can facilitate the irreversible removal of substrate molecules from the systemic circulation. J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_16, © Humana Press, a part of Springer Science + Business Media, LLC 2010

359

360

Padowski and Pollack

Finally, and perhaps most importantly from a biologic response standpoint, efflux proteins can serve as barriers to substrate distribution into organs (7), tissues (8), or cellular spaces (9) that represent important pharmacologic targets. Such barrier functions may appear at the interface between the organ and the vasculature, or may be present on the cell surface. In either case, the net result of efflux transport is to diminish the amount of substrate available at the site of action to interact with the biologic receptor. Although a variety of efflux transport proteins are expressed in mammalian tissues, the most well-characterized, and perhaps the most important, of these systems is P-glycoprotein (Pgp). Pgp is a 170-kDa protein that is a member of the ABC superfamily of energy-dependent transport systems (10). It is localized to cellular membranes, and originally was identified by its ability to confer multidrug resistance to mammalian cells (11). Although Pgp often is viewed in terms of its ability to transport bulky hydrophobic cations, it is in actuality a relatively indiscriminate system that can interact with compounds from a wide range of chemical classes and with a large degree of structural and physicochemical dissi­ milarity (12). Indeed, attempts to develop structure–transport relationships for Pgp have yielded mixed, generally negative, results due the lack of well-defined structural specificity. This lack of specificity may be due, in part, to the presence of multiple substrate recognition sites on the protein (13). In addition to complexities associated with substrate recognition, the location of the binding sites on Pgp relative to the cellular membrane has been debated. Although this is a biologically subtle point, it is of critical importance to the kinetics of substrate transport, as it will determine the apparent driving-force concentration for the efflux process (Fig. 16.1). Although both extracellular and intracellular locations have been suggested, the most widely accepted model for Pgp-mediated transport is with the

Fig. 16.1. Depending on the degree of intra- versus extracellular drug sequestration and the rate and extent of drug equilibration across each leaflet of the cell membrane, the driving force for Pgp-mediated flux may be the extracellular (C1), intramembrane (C2), or intracellular (C3) drug concentration.

Pharmacokinetic and Pharmacodynamic Implications of P-Glycoprotein Modulation

361

substrate-binding site localized within the cellular membrane (14). From a pharmacokinetic standpoint, the proximate driving-force concentration for substrate flux may be the extracellular concentration (e.g., if binding to extracellular proteins is low and if the extracellular concentration is in rapid equilibrium with substrate concentration in the cell membrane) or the intracellular concentration (if sequestration within cellular organelles is minimal and if the cytosolic substrate concentration is in rapid equilibrium with substrate concentration in the cell membrane). It also is conceivable that neither the extracellular nor the intracellular concentration will appear to be the driving force for Pgp-mediated flux if neither locale serves as an adequate surrogate for concentrations in the cellular membrane. Given these brief introductory comments, the purpose of this review is to focus on the pharmacokinetic and pharmacodynamic implications of modulation of Pgp-mediated transport. Approaches or mechanisms through which Pgp activity may be modified will be considered first; the impact of those changes on drug disposition then will be reviewed. Finally, the outcome of altered substrate flux in terms of biologic response will be discussed with respect to several different types of drug response.

2. Modulation of Pgp Activity Changes in substrate flux by Pgp may be the result of chemical modulation (inhibition of transport activity or induction of expression) or due to genetic polymorphisms. When considering the potential outcome of Pgp modulation, it is useful to keep in mind that the effects of chemical inhibitors/inducers or genetic changes may not be specific. For example, inhibitors of Pgp may also modulate other transport proteins; genetic differences in Pgp may result in changes in other pharmacokinetic mechanisms or pathways. The most common and well-studied chemical modulation of Pgp is inhibition of transport. Pgp inhibition has received substantial attention within the pharmaceutical industry as a means to enhance pharmacologic response to existing drugs or new chemical entities that may be significant substrates for this transport protein (15). The initial impetus for identifying or developing compounds that would effectively inhibit Pgp was to reverse multidrug resistance in cancer cells, thereby enhancing antitumor effects (16). Several existing therapeutic agents (e.g., quinidine, verapamil, and cyclosporine) were identified as Pgp inhibitors. Subsequently, several new compounds (e.g., PSC833 and GF120918) were developed for the express purpose of limiting Pgp-mediated flux. Despite major efforts to develop a Pgp inhibitor with clinical efficacy, to

362

Padowski and Pollack

date no such compound has become part of the therapeutic armamentarium for cancer or any other disorder. The inability to translate this fundamental idea into a clinically useful approach is largely due to difficulties associated with targeting delivery of potential inhibitors to the appropriate site of Pgp-mediated transport (i.e., the desired target tissue, if Pgp serves as a barrier transporter for the therapeutic agent in question; an organ of elimination, if the intent is to prolong residence time in the systemic circulation; or the gastrointestinal tract, if the goal was to enhance absorption). Because Pgp is an inhibitable transporter, because it can affect substrate disposition at a variety of pharmacokinetic steps, and because it interacts with such a wide variety of substrates, drug–drug interactions at the level of Pgp-mediated transport are a likely phenomenon (17, 18). The high likelihood of unintended inhibition of Pgp has led the pharmaceutical industry to screen new compounds for their Pgp transport and inhibition properties, in much the same way that compounds are routinely screened for lability in the presence of various isoforms of cytochrome P450 (19). Specific examples of drug–drug interactions in Pgp transport will be considered in subsequent sections of this review. Although the focus of much less attention than inhibition, the potential induction of Pgp by drugs or other xenobiotics also represents a potential form of drug–drug interaction that could have pharmacokinetic or pharmacodynamic consequences (20, 21). The first suggestion that Pgp might evidence biologically significant induction resulting in overt pharmacokinetic and/or pharmacodynamic changes was a report that expression of Pgp in rat brain tissue was increased following subchronic exposure to morphine (22). Morphine and many other opioids are substrates for Pgp, albeit to varying degrees (23, 24). One logical hypothesis, therefore, was that induction of Pgp by morphine at the blood–brain barrier would decrease concentrations of morphine in brain tissue, leading to the emergence of functional tolerance to the antinociceptive effects of the drug. It is currently unclear as to what extent Pgp is generally inducible and, if so, in what organ systems such induction might occur. This topic will be touched upon briefly in the section on pharmacodynamics later in this review. Significant recent attention has been paid to genetic polymorphisms of Pgp, and how those polymorphisms may influence drug disposition or action in vivo. This complicated line of investigation has yielded contradictory results. To date, there is little consensus on whether genetic polymorphisms in humans result in a consistent and clinically relevant change in Pgp-mediated transport (25). This issue also will be discussed in more detail in the following sections of this review.

Pharmacokinetic and Pharmacodynamic Implications of P-Glycoprotein Modulation

3. Pharmacokinetic Consequences of Pgp Modulation 3.1. Gastrointestinal Absorption

363

Pgp is expressed on the apical membrane of intestinal epithelial cells, oriented in a manner to secrete substrates from the epithelial cell into the intestinal lumen. As such, Pgp serves as a functional barrier to absorption of a variety of substrates across the intestinal wall, including the cardiovascular agents digoxin, quinidine, verapamil, and talinolol, and the opioid loperamide (26). The expression of Pgp appears to vary along the length of the human intestine in a manner somewhat inverse to that of the primary intestinal isoform of cytochrome P450, CYP3A4 (27). Pgp and CYP3A4 have common substrates and inhibitors, and there is evidence that the two proteins may act in a concerted fashion to limit systemic absorption of some compounds (28). Orally administered drugs that are substrates of both Pgp and CYP3A4 have three possible fates: they may cross the gut wall and reach the systemic circulation, they may be biotransformed by CYP3A4 in the enterocyte, or they may be effluxed by Pgp back into the intestinal lumen. Of course, molecules that reappear in the lumen of the intestine may be reabsorbed distally, with subsequent opportunities to be absorbed, metabolized, or effluxed. The entry and exit of a compound across the intestinal epithelium multiple times prior to either absorption into the systemic circulation or metabolism in the gut wall constitutes a lumen-to-enterocyte recycling process (27) which results in an increased luminal mean residence time and a decrease in the overall rate of intestinal absorption. In general, intestinal Pgp appears to influence the peak concentration of an orally administered substrate in the systemic circulation (Cmax) more significantly than overall systemic exposure (area under the concentration–time curve or AUC) (29). Pgp-mediated efflux will decrease the extent of systemic absorption if the substrate also is metabolized in the intestinal epithelium (by CYP3A4 or another enzyme) or if absorption of the substrate occurs only at a specific site in the intestine. There is limited but compelling evidence that intestinal Pgp can serve an excretory rather than absorptive barrier function, and can actually facilitate flux of substrates from the systemic circulation into the intestinal lumen (30, 31). Thus, Pgp in the intestine will play a dual role for some substrates, by limiting flux from the intestinal lumen into blood and by stimulating flux from the blood into the intestinal lumen. Several lines of evidence suggest that drug–drug interactions at the level of Pgp in the gut may be biologically significant. For example, the calcium channel blocker verapamil, one of the first compounds shown to inhibit Pgp-mediated flux, increased systemic exposure in rats to orally administered talinolol, a model Pgp substrate (32). Because talinolol is not metabolized extensively,

364

Padowski and Pollack

it was suggested that this interaction occurred at the level of efflux transport. A similar interaction between digoxin and quinidine (increased systemic exposure to digoxin in the presence of quinidine) in humans has been known for many years, but only recently was attributed to Pgp (33). In addition to drug–drug interactions based on inhibition of Pgp, the potential for drug-associated induction of intestinal Pgp also has been examined. For example, subchronic oral administration of rifampin decreased the ability of oral verapamil to permeate the rat small intestine (34). This interaction was interpreted as induction of intestinal Pgp, resulting in increased secretion of verapamil from the enterocyte into the intestinal lumen. In a commonly cited clinical study, the single-dose pharmacokinetics of digoxin was compared before and after administration of rifampin in healthy volunteers (35). As illustrated in Fig. 16.2, the systemic exposure to digoxin, expressed as the AUC in blood, was reduced during rifampin treatment, in part due to an increase in digoxin biliary clearance, and in part due to a trend toward increased digoxin secretion from blood into the gut lumen. Rifampin increased intestinal Pgp (protein content); among individual subjects, the rifampin-associated decrease in AUC correlated with intestinal Pgp content after oral, but not intravenous, digoxin. A similar study design was used to examine the pharmacokinetics of talinolol in the presence or absence of rifampin administration (36). Rifampin decreased systemic exposure to oral talinolol by approximately 33%, and increased the systemic clearance of intravenous talinolol by nearly 30%. These observations suggest that Pgp-mediated drug–drug interactions at the level of intestinal absorption may be relatively common.

Fig. 16.2. Digoxin nonrenal clearance (block bars) and AUC0-96 h (gray bars) in humans following a 1  mg dose of digoxin, either alone (before rifampin) or with concomitant administration of 600  mg/day rifampin for 14  days (during rifampin). Bars represent means ± SD. Data obtained from ref. (35).

Pharmacokinetic and Pharmacodynamic Implications of P-Glycoprotein Modulation

3.2. Biliary Excretion

365

Pgp is one of several transport proteins located on the canalicular membrane of hepatocytes that serve to mediate substrate flux from hepatocytes into bile. Several compounds of clinical importance undergo Pgp-mediated biliary excretion, including anticancer agents such as doxorubicin and paclitaxel (37), cardiac agents such as digoxin and quinidine (38), the immunosuppressive agent cyclos­ porine (39), and antivirals such as ritonavir and saquinavir (40). Drug–drug interactions at the level of Pgp-mediated biliary excretion are also of potential clinical importance. For example, quinidine has been shown to decrease the biliary excretion of digoxin in humans (Fig. 16.3) by nearly 50% (41), and the Pgp inhibitor GF120918 significantly reduced the biliary clearance of doxorubicin in an isolated perfused rat liver model (42). Any change in the excretory transport of a compound undergoing substantial biliary clearance would, by definition, alter the systemic exposure to this compound. In addition, modulation of excretion into bile would change substrate accumulation in the hepatocellular compartment, a site of potential toxicity or, in a limited number of cases, the site of the intended pharmacologic response. Finally, altered biliary clearance would change substrate presentation to the intestinal lumen, another site of potential drug toxicity, via the common bile duct. If the compound undergoes enterohepatic recycling, the change in luminal exposure may be substantial. Most drugs are metabolized in the liver. In some cases, metabolism produces toxic species that may be eliminated into bile. If the metabolite was a substrate for Pgp, inhibition of this protein would decrease metabolite excretion, potentially resulting in hepatotoxicity. Due largely to difficulties in obtaining bile from humans, the clinical studies evaluating the role of Pgp in biliary excretion have been limited. Numerous in  vitro and preclinical in vivo studies have been conducted, and have led to an increased

Fig. 16.3. Influence of quinidine on the biliary clearance of digoxin in humans, based upon pooled data from two clinical studies. Bars represent means ± SD. Data obtained from ref. (41).

366

Padowski and Pollack

Fig. 16.4. Influence of dexamethasone pretreatment on the rate of rhodamine-123 biliary clearance in rats. Data obtained from ref. (44).

understanding of the possible mechanisms underlying some clinically important drug–drug interactions. Induction of Pgp on the canalicular membrane of hepatocytes may increase the rate of biliary clearance for some drugs. For example, rifampin pretreatment enhanced the biliary clearance of digoxin in humans, presumably through increased expression of Pgp (35), and pretreatment of rats with phenothiazine to induce hepatic Pgp expression stimulated the rate of biliary excretion of the peptide octreotide and the antineoplastic vincristine (43). As shown in Fig. 16.4, dexamethasone pretreatment in rats increased the biliary clearance of rhodamine-123, a Pgp substrate, by nearly fourfold (44). Tamoxifen and its biotransformation products are excreted into bile; tamoxifen pretreatment increased biliary excretion of tamoxifen and its metabolites approximately sixfold, and increased the expression of mRNA which encodes Pgp in the rat (mdr1b), suggesting that tamoxifen displays autoinduction of Pgp-mediated biliary excretion (45). 3.3. Urinary Excretion

A significant body of evidence suggests that Pgp can mediate substrate secretion into urine, and therefore can be a determinant of renal clearance for some drugs such as digoxin (46). Pgp is expressed on the luminal surface of cells lining the proximal tubule of the kidney, the region of the nephron in which most active secretion in the blood-to-urine direction occurs. Evidence for Pgp-mediated secretory clearance in the kidney has been generated with the Pgp substrate digoxin in combination with known modulators of Pgp. Quinidine decreased the renal secretion of digoxin in isolated dog (47) and rat (48) models of renal excretion. Consistent with these observations in preclinical species, coadministration of quinidine with digoxin in humans increased digoxin serum concentrations and correspondingly decreased digoxin renal clearance (49, 50) (Fig. 16.5). A similar interaction between quinidine and digoxin was observed in vitro

Pharmacokinetic and Pharmacodynamic Implications of P-Glycoprotein Modulation

367

Fig. 16.5. Plasma concentration–time profiles for [3H]-digoxin in a representative patient before (open circles) and during (closed circles) quinidine administration. Data obtained from ref. (49).

(38). Clinical studies also have demonstrated drug–drug interactions with digoxin at the level of Pgp-mediated renal secretion. For example, coadministration of PSC 833 (valspodar), a Pgp inhibitor, to healthy volunteers decreased the renal clearance of digoxin by approximately 75% (51). Ritonavir, an inhibitor of Pgp in vitro, markedly decreased renal, as well as nonrenal, clearance of digoxin in human subjects (52). The Pgp inhibitor cyclosporine also decreased renal secretion of digoxin in isolated perfused rat kidneys (53). The role of Pgp in this interaction was confirmed using Pgp overexpressing LLC-PK1 renal epithelial cells (54). Clarithromycin (55) and itraconazole (54) also have been shown to inhibit Pgp-mediated renal excretion of digoxin. While it is clear, based upon the aggregate results of studies of digoxin renal clearance, that Pgp is involved in the renal excretion of this cardiac glycoside, these efforts also suggest an important and clinically relevant conclusion. Digoxin has a narrow therapeutic window, and even modest changes in bioavailability or systemic clearance can lead to untoward side effects. A multitude of drug–drug interactions with digoxin have been cataloged over the years, and yet the proximal mechanism(s) underlying these interactions remained elusive until fairly recently. It is now clear that modulation of Pgp-mediated transport in the kidney is one explanation for drug–drug interactions involving alterations in the renal clearance of digoxin. Because Pgp expressed in the intestine (which limits the rate and extent of digoxin absorption) and the liver (which mediates biliary excretion of digoxin) also determines systemic exposure to digoxin, it is obvious that global modulation of Pgp function can have significant effects on the disposition and action of this important cardiac agent. 3.4. Extravascular Distribution

Pgp serves as a barrier to the flux of substrates from the systemic circulation into protected organs or tissues. These so-called “sanctuary sites” include the placenta (56), testes (57), and brain (58). The protective role of Pgp relative to tissue exposure has been most

368

Padowski and Pollack

well-studied in the central nervous system. Arguably, the most significant biologic role of Pgp, at least relative to protecting mammalian systems from chemical insult, is as a functional component of the blood–brain barrier. The structural and biochemical features of the blood–brain interface, particularly with respect to drug delivery to central nervous system (CNS) targets, have been the subject of numerous reviews (19, 59, 60). The blood–brain barrier is composed of a single layer of endothelial cells with intimate cell-to-cell communication afforded by complex tight junctions. These tight junctions limit paracellular permeability of hydrophilic molecules and bulky compounds. Substrate penetration of the blood–brain barrier therefore requires flux through the luminal membrane of capillary endothelia, the endothelial cytoplasm, and the abluminal membrane prior to entry into the extracellular fluid compartment of the brain. The close association of pericytes and astrocytic foot processes with the abluminal endothelial membrane presents an additional barrier to xenobiotic flux. The endothelial cells lining the brain microvasculature also contain systems that serve as functional, rather than structural, impediments to substrate translocation into the central nervous system (2). Cellular components of the blood–brain barrier express metabolic enzymes that biotransform drugs, limiting brain exposure to the intact parent (61). However, enzyme expression (at the BBB or in the brain as a whole) does not provide a significant route of drug elimination or a general barrier to substrate influx into brain for most substrates. More importantly, the blood–brain barrier endothelial cells express numerous membrane transport proteins that mediate substrate influx into and/or efflux from the brain. These various transport systems have been reviewed previously (62). From the standpoint of drug delivery to the brain, transport proteins mediating influx could be used to facilitate uptake of hydrophilic compounds across the blood–brain barrier. These systems mediate nutrient (glucose, for example) uptake into brain; attempts to engineer drug delivery modalities to make use of these systems are ongoing (63). Protein-mediated efflux, on the other hand, serves to limit brain exposure primarily to lipophilic substrates that otherwise would permeate the blood–brain barrier readily. Pgp, the primary barrier transporter at the blood–brain interface, is expressed on the luminal surface of brain capillary endothelial cells (58). However, other structural models of Pgp localization have been suggested. For example, Golden and Pardridge demonstrated that Pgp is expressed on foot processes of astrocytes in close proximity to capillary endothelium (i.e., on the abluminal side of endothelial cells) (64). As discussed earlier in this review, although the difference in anatomical location is minor, the driving forces responsible for Pgp-mediated transport will be markedly different between these two sites (Fig. 16.1), resulting in differ-

Pharmacokinetic and Pharmacodynamic Implications of P-Glycoprotein Modulation

369

ences in substrate transport kinetics between the brain tissue and the systemic circulation. Experiments performed in mice lacking Pgp expression, including genetic knockouts and naturally occurring Pgp-deficient animals have provided voluminous evidence that Pgp can significantly impede drug translocation across the blood–brain interface (2, 65). Pharmacokinetic experiments in transport-deficient mice form the foundation of the current understanding of attenuated blood–brain barrier translocation by Pgp. This experimental model will be highlighted in the subsequent section on pharmacodynamics. The fact that Pgp can mediate efflux of lipophilic molecules of varying pharmacologic classes and chemical structures complicates the design of effective CNS-targeted drugs. The ability of numerous lipophilic molecules to diffuse through the blood–brain barrier is counterbalanced by Pgp, resulting in low brain penetration despite otherwise favorable physicochemical characteristics. Several studies have demonstrated the role of Pgp in the exposure of brain tissue to pharmacologically relevant compounds. One of the earliest examples focused on the opioid peptide [d-penicillamine2,5] enkephalin (DPDPE). Although DPDPE enters the brain rapidly (66), only a limited fraction of the administered dose (<0.1%) reaches brain tissue, and brain-to-serum partitioning is thus low. These characteristics (rapid uptake; low partitioning at apparent distribution equilibrium) suggest that DPDPE may undergo active efflux in the brain-to-blood direction. Subsequent dose-ranging studies revealed that DPDPE is a substrate for a saturable efflux process at the blood–brain interface (66). The identity of the efflux transport system as Pgp was confirmed by experiments in Pgp-deficient mdr1a(−/−) mice (67), which evidenced enhanced brain uptake of DPDPE compared to Pgpcompetent controls. Pretreatment of Pgp-competent mdr1a(+/+) mice with the Pgp inhibitor GF120918 (68) produced similar results (Fig.  16.6). These studies illustrate the standard experimental paradigm for documenting the influence of Pgp on brain uptake, and demonstrate clearly that Pgp plays a pivotal role in determining brain exposure to this opioid pentapeptide. The list of drugs for which Pgp serves as a barrier transporter at the blood–brain interface has grown considerably over the past decade. The most important of these are CNS-targeted agents, which include, but are not limited to, antidepressants (amitriptylline, doxepine, nortriptylline, and venlafaxine), antipsychotics (chlorpromazine, questiapine, and risperidone), antiemetics (ondansetron), analgesics (morphine), and anticonvulsants (phenytoin) (69). Of course, a variety of other agents are used to treat systemic disorders that may express some CNS involvement. Brain exposure is limited by blood–brain barrier Pgp to many of these agents, including anticancer drugs (e.g., doxorubicin and paclitaxel), antivirals (indinavir, nelfinavir, and saquinavir), antibiotics (erythromycin, levofloxacin, and rifampin), and steroids (dexamethasone) (69).

370

Padowski and Pollack

Fig. 16.6. Influence of the Pgp inhibitor GF120918 or genetic deficiency of mdr1a Pgp on the brain-to-blood distribution ratio of DPDE in mice. Data obtained from ref. (67, 68).

Limited brain exposure to therapeutic agents often is viewed as a negative characteristic. However, limited brain exposure can be a positive attribute when peripheral response is desired but the agent can elicit undesirable effects in the CNS. To the extent that Pgp expressed in the blood–brain barrier may abate or limit such responses, the classic example is the opioid loperamide. Loperamide is used therapeutically in the treatment of diarrhea; despite reasonable opioid receptor potency, loperamide is devoid of central opioid effects because it is excluded by Pgp from permeating the blood–brain barrier. When Pgp-mediated transport is abolished, loperamide elicits significant central opioid effects (70). A more recent example involves nonsedating antihistamines. The lack of central antihistaminic effects, which include drowsiness, is conferred by Pgp-mediated efflux at the blood–brain interface (71). Thus, mitigation of brain uptake of Pgp substrates can be a beneficial result in many therapeutic instances.

4. Pharmacodynamic Consequences of Pgp Modulation

Pharmacodynamics, or the magnitude and time course of pharmacologic effect, clearly can be impacted by Pgp modulation. In considering the pharmacodynamic implication of changes in Pgp function, it is useful to keep in mind that the unbound substrate concentration at the site of action will, in most cases, drive pharmacologic response. Modulation of Pgp function, which generally means a decrease in intrinsic transport activity, can increase substrate concentration at the receptor site in one of three ways (Fig. 16.7). First, a decrease in Pgp-mediated transport can increase systemic concentrations of a Pgp substrate, either by increasing systemic bioavailability (for a substrate administered orally) or by decreasing systemic clearance (for a substrate eliminated significantly by biliary and/or renal clearance). Next, substrate concentrations at the site of

Pharmacokinetic and Pharmacodynamic Implications of P-Glycoprotein Modulation

371

Fig. 16.7. Negative modulation of Pgp function may increase substrate concentration at a target site in three ways. Pgp inhibition can increase substrate concentrations at the systemic level, with marginal impact on target exposure (thin line), at the organ level, with moderate impact on target exposure (medium line), or at the cellular level, with maximum impact on target exposure (thick line).

action could be increased secondary to a decrease in Pgp-mediated flux if the receptor is located in a “sanctuary” organ, such as the brain, for which Pgp has an important role as a barrier transporter. In this case, systemic concentrations might not be affected by Pgp modulation, but the end-organ-to-blood concentration ratio (i.e., tissue-to-blood partition coefficient) would be increased. Finally, if Pgp is located on the target cell membrane (cancer cells, for example), substrate concentration in the target intracellular space might be increased following Pgp inhibition even though systemic substrates concentrations and substrate concentrations in the bulk tissue containing the cellular target were not increased. It should be noted that two, or perhaps all three, of these mechanisms may be in play for a given substrate. Thus, concentrations at the target site might be elevated more than would be predicted by a simple change in the systemic concentration to dose relationship, the target organ to blood concentration ratio, or the cellular target to end-organ extracellular fluid concentration relationship. 4.1. Changes in Systemic Concentration Leading to Altered Pharmacologic Response

In general, inhibition of Pgp-mediated transport results in relatively modest changes in systemic concentrations of Pgp substrates compared to changes associated with inhibition of metabolic enzymes (72). Thus, one would expect only a modest influence of altered Pgp function on biologic response when the locus of the

372

Padowski and Pollack

Fig. 16.8. Influence of ritonavir on digoxin total clearance (black bars), renal clearance (gray bars), and nonrenal clearance (open bars) in human subjects. Bars represent means ± SEM. Data obtained from ref. (52).

Pgp-mediated interaction is at the level of systemic bioavailability or clearance. Nevertheless, there are several examples in the clinical literature of drug–drug interactions, likely at the level of Pgpmediated transport, that enhance biologic effect (usually toxicity). The cause of this enhanced activity in the presence of Pgp modulation is an increase in systemic exposure (peak concentration, AUC, or mean residence time) at a fixed dose of the pharmacologic agent. For example, the antiviral drug ritonavir, which can inhibit Pgp-mediated transport, increases the AUC for digoxin, prolongs its residence time (Fig. 16.8), and potentially increases its toxicity (52, 73). Diltiazem, another Pgp inhibitor, increases the systemic concentrations and associated toxicity of the immunosuppressive agent tacrolimus (74). 4.2. Changes in Target Organ Concentration Leading to Altered Pharmacologic Response

The most significant pharmacodynamic impact of Pgp modulation appears to occur when inhibition of Pgp-mediated flux increases target-organ drug concentrations compared to drug concentrations in the systemic circulation. In these cases, systemic drug concentrations no longer are predictive of concentrations in the target organ. Although this situation can occur for any target organ that represents a “sanctuary site” protected by Pgp, the most common examples involve the CNS. With respect to understanding how CNS pharmacodynamics are perturbed secondary to Pgp modulation, opioids have served as a useful pharmacologic class. Many opioids are excluded from the brain to some extent by blood–brain barrier Pgp (Table 16.1). The fact that the Pgp effect, i.e., the ratio of brain uptake in the absence versus presence of functional Pgp, varies significantly among this compound class (from approximately 1.2 for morphine to more than 10 for DPDPE and loperamide) (23), coupled with the ability to measure a clear pharmacologic endpoint (antinociception or respiratory depression), makes this an ideal compound set with which to probe the pharmacodynamic implications of Pgp modulation.

Pharmacokinetic and Pharmacodynamic Implications of P-Glycoprotein Modulation

373

Table 16.1 Influence of mdr1a on the net uptake of opioids into mouse brain Compound

Opioid receptor subtype

Pgp effect

Meperidine

µ

0.98 ± 0.27

Fentanyl

µ

1.24 ± 0.17

Morphine

µ

1.24 ± 0.08

Bremazocine

k

1.50 ± 0.21

Deltorphin II

d

1.58 ± 0.36

Methadone

m

2.61 ± 0.55

Naltrindole

d

4.44 ± 0.93

Loperamide

m

10.4 ± 1.9

DPDPE

d

11.6 ± 6.4

Pgp effects (mean ± SD) were calculated as brain uptake clearance during in situ perfusion in mdr1a (−/−)/brain uptake clearance in mdr1a (+/+) mice. Data obtained from ref. (23)

A relatively early report in humans suggested that inhibition of Pgp-mediated transport by quinidine could unmask central opioid effects for loperamide, an opioid normally devoid of antinociceptive and respiratory depressant effects (75). In this experiment, coadministration of quinidine with loperamide led to modest respiratory depression compared with the administration of loperamide alone. Because loperamide does not appreciably enter the brain under normal circumstances, one interpretation of these results was that quinidine inhibited Pgp-mediated efflux at the blood–brain interface, allowing sufficient accumulation of loperamide in brain to elicit a central pharmacologic response. Although clinical data on opioid responsivity secondary to Pgp modulation are limited, animal experimentation has provided a wealth of information. Early studies in this area showed that the dose–response relationship for DPDPE-associated antinociception was shifted approximately 30-fold to the left (i.e., demonstrating enhanced sensitivity) in Pgp-deficient [mdr1a(−/−) mice] compared with their Pgp-competent counterparts (Fig.  16.9a), and that the leftward shift remained even when effect was related to blood concentration as opposed to dose (67). This latter result (Fig. 16.9b) confirmed that the primary pharmacodynamic influence of Pgp was on distribution of DPDPE between blood and brain as opposed to systemic elimination of the opioid. Similar results were obtained with chemical inhibition, as opposed to genetic disruption, of the transporter (68), indicating that the

374

Padowski and Pollack

Fig. 16.9. Influence of Pgp on antinociceptive response to DPDPE in mdr1a (−/−) (open circles) and mdr1a (+/+) (closed circles) mice, determined based upon latency to response in hotplate assay (a). Associated blood concentrations in mdr1a (−/−) (closed circles) and mdr1a (+/+) (open circles) mice, did not change based upon Pgp status (b). Data obtained from ref. (67).

pharmacodynamic outcome was specific to loss of Pgp function as opposed to a nonspecific change in the gene-deficient animals. Because DPDPE had a very high Pgp effect (>tenfold), loss of Pgp function had a substantial impact on the magnitude and duration of antinociception. In contrast, the Pgp effect for morphine was modest (<1.5-fold) (76, 77), and the corresponding influence of Pgp disruption on morphine-associated antinociception was measurable but limited. This large difference in Pgp-related modulation of opioid pharmacology suggested that this class of compounds presents unique advantages in dissecting the role of Pgp in the blood–brain barrier. Subsequent experimentation in this area was directed at several different goals. The first was to determine the degree of variability in Pgp effect among a wide range of opioid compounds. This goal can be accomplished using a variety of experimental paradigms, including in situ brain perfusion in Pgp-competent versus Pgp-deficient mice, in  vivo brain-to-blood partitioning studies (either pre- or postestablishment of distribution equilibrium)

Pharmacokinetic and Pharmacodynamic Implications of P-Glycoprotein Modulation

375

in Pgp-competent versus Pgp-deficient mice, and through use of chemical modulators of Pgp function in any animal species (typically mice or rats, but also moving up in species complexity all the way to humans, as in the loperamide-quinidine example cited above). Based on the results of in situ brain perfusion experiments, opioids were classified as having no Pgp effect (meperidine), low Pgp effect (morphine, fentanyl), intermediate Pgp effect (methadone, naltrindole), or high Pgp effect (SNC 121, DPDPE, and loperamide) . These designations were borne out by in vivo experiments (24), although the absolute magnitude of the Pgp effect differed to some extent between the two experimental approaches. This observation leads to an important question that is not completely resolved: does determination of the Pgp effect prior to the attainment of distribution equilibrium predict the Pgp at substrate equilibration across the blood–brain barrier? Across a compound set, it is possible to develop such categorical predictions, although there are distinct quantitative differences in estimating the Pgp effect pre- and postdistribution equilibrium. A related but separate issue concerns the precision with which such measurements may be made. Intuitively, one may expect that measurements made at distribution equilibrium would be more reproducible than measurements made prior to distribution equilibrium, simply because in the former case the system is under a stable, time-independent condition, whereas in the latter the system is still in dynamic flux. These experimental issues require further investigation. Having categorized the influence of Pgp on brain uptake of opioids as a class, the next step was to determine the degree to which the pharmacodynamics of opioids could be altered by Pgp modulation. This goal was also addressed predominantly with mouse models (genetic or chemical disruption of Pgp) and measurement of opioid-associated antinociception (response to a thermal stimulus). In general, disruption of Pgp-mediated transport shifted the dose–response and blood concentration–response data to the left (24, 67, 77), consistent with an increase in opioid partitioning into brain. The degree of leftward shift in these relationships was, in general, consistent with the Pgp effect characterization of the opioids. At a fixed administered opioid dose, Pgp disruption increased the magnitude of antinociceptive effect, the duration of antinociception, and the area under the effect versus time curve. Finally, in general, Pgp disruption did not alter the relationship between antinociceptive effect and bulk brain tissue concentration, as illustrated in Fig. 16.10 (24, 77). This latter observation is important in that it suggests that the only influence of Pgp with respect to central opioid pharmacology is its influence on opioid distribution into brain. Once the opioid molecule penetrates the blood–brain barrier, Pgp no longer has an influence on the ability of that molecule to access the opioid receptor.

376

Padowski and Pollack

Fig. 16.10. Influence of Pgp-mediated efflux on the antinociceptive action of morphine (a) and alfentanil (b). Closed circles indicated mice with intact Pgp function; open symbols indicate mice lacking Pgp expression (circles) or receiving verapamil as a Pgp modulator (triangles). Solid lines indicate result of linear (r2 = 0.624, p < 0.01; panel a) or hyperbolic (r2 = 0.868, p < 0.001; panel b) regression; dashed lines (panel a) indicate 95% confidence interval on the linear regression. Data obtained from ref. (24, 77).

DPDPE is an exception to these general observations (67). Ablation of Pgp-mediated flux (in Pgp-deficient animals) caused an approximate 30-fold leftward shift in the DPDPE dose–response relationship. When antinociceptive response was related to brain tissue concentration, as opposed to dose or blood concentration, an approximate tenfold leftward shift remained (Fig. 16.11). This observation suggested that Pgp-deficient mice were more sensitive to DPDPE than their Pgp-competent counterparts, even at a fixed concentration of the opioid peptide in bulk brain tissue. There currently is no adequate explanation for the apparently anomalous behavior of DPDPE. Initial speculation focused on the possible expression of Pgp at parenchymal locales in brain, consistent with observations in human brain tissue (64). As illustrated in Fig. 16.12, modeling the DPDPE pharmacodynamics in this manner indeed rectified the difference between Pgp-deficient and Pgp-competent animals (67). Other explanations, however,

Pharmacokinetic and Pharmacodynamic Implications of P-Glycoprotein Modulation

377

Fig. 16.11. Antinociceptive effect versus brain concentrations of DPDPE in mdr1a (−/−) (open circles) versus mdr1a (+/+) (closed circles) mice. Data obtained from ref. (67).

Fig. 16.12. Distribution model that rectifies DPDPE disposition and antinociceptive action in Pgp-competent (top) and Pgp-deficient (bottom) mice. Solid arrows indicate apparent Pgp-mediated flux; broken arrows indicate non-Pgp flux processes. Mathematical modeling of concentration versus time (blood and brain) and effect versus time data suggested two distinct influences of P-pg: modulation of DPDPE distribution between blood and brain tissue and modulation of DPDPE distribution between bulk brain tissue and the receptor biophase. Modified from ref. (67).

also must be considered. One reasonable aspect to consider is the fact that Pgp function appears heterogeneous throughout the brain (78) (Fig. 16.13). If Pgp function is regional, and if regionality of substrate transport corresponds with regionality of receptor expression, then use of bulk brain tissue concentration as a surrogate for concentration at the receptor site may be misleading. Continued experimentation is required to address this point, and once again use of opioids as a pharmacologic class may provide fruitful results. 4.3. Changes in Intracellular Concentration Leading to Altered Pharmacologic Response

Pgp originally was identified based upon its ability to confer cellular resistance to the effects of several drugs (11), leading to its designation as a multidrug resistance transport protein. In this capacity, Pgp regulates the distribution of pharmacologic agents not between the systemic circulation and the extracellular fluid space of an organ or tissue, as it does at the blood–brain interface,

378

Padowski and Pollack

Fig. 16.13. Influence of Pgp on the distribution of [3H]-verapamil within mouse brain. Apparent Pgp effects were obtained by simultaneous modeling of [3H]-verapamil concentration–time data obtained from Pgp-competent and Pgp-deficient mice. Data obtained from ref. (78).

but rather between the extracellular and the intracellular fluid spaces. If the relevant pharmacologic target is expressed intracellularly, Pgp has the potential to provide very close regulation of the amount of substrate that can reach that target at a fixed extracellular concentration of the substrate The most obvious and well-studied influence of Pgp on intracellular accumulation and/or the pharmacologic response to Pgp substrates at the cellular level is with respect to anticancer drugs. This topic has been the subject of numerous and ongoing comprehensive reviews (79–82). The ability to inhibit Pgp (or other efflux transport proteins that confer cellular resistance to anti-cancer agents) substantially would represent a significant therapeutic gain. From the standpoint of cellular pharmacodynamics, the system is relatively simple: inhibition of a barrier transporter leads to enhanced cellular accumulation of the pharmacologic agent, which in turn causes a more significant or prolonged pharmacologic effect. In the case of tumor cells in vivo, this effect would translate, in theory, to an enhanced anti-tumor response. Despite the simplicity of the cellular pharmacodynamic system, translation of Pgp inhibition into a meaningful therapeutic approach for the treatment of cancer is fraught with difficulties. One way in which tumor cell accumulation of an anticancer agent can be increased is simply by increasing the administered dose of the agent. Dose escalation would increase circulating concentrations of the drug, ultimately resulting in increased drug accumulating in the tumor cells. This simplistic approach is unworkable due to the systemic toxicity associated with anticancer agents. The advantage associated with a Pgp inhibition strategy is thus the potential to achieve enhanced tumor cell accumulation of an anticancer agent in the absence of increased systemic toxicity.

Pharmacokinetic and Pharmacodynamic Implications of P-Glycoprotein Modulation

379

Unfortunately, the relatively ubiquitous expression of functional Pgp throughout the body has effectively limited this strategy. Because Pgp is an important determinant of systemic exposure at multiple pharmacokinetic steps (absorption, biliary clearance, and renal clearance), inhibition of Pgp may lead to increased systemic concentrations of Pgp substrates, the same outcome associated with a simple increase in dose of a Pgp substrate. Moreover, because Pgp plays a protective role for several important organs and tissues, indiscriminate inhibition of the transporter may lead to enhanced accumulation, and toxicity, in those organs. Although results to date have been generally disappointing, it is premature to abandon the Pgp (or other barrier transporter) inhibition strategy. Undoubtedly, this will represent a fruitful area of research in the future. Experimental goals, however, must shift from identifying increasingly potent or specific Pgp inhibitors to developing strategies for targeting those inhibitors to specific sites of action. In the case of cancer chemotherapy, strategies for targeting Pgp inhibitors specifically, or at least preferentially, to the tumor would limit the degree of systemic drug–drug interaction, increasing the potential utility of the transport–inhibition strategy. A less specific, but possibly feasible, drug delivery strategy for Pgp inhibitors would be to preferentially (or selectively) target the inhibitor to the end-organ of interest. In the case of cancer chemotherapy, this might increase exposure of the anticancer agent to an entire organ or tissue space (the brain, for example, in treating CNS tumors) while sparing exposure throughout the rest of the body. One recently explored approach for targeting Pgp inhibitors to the blood–brain barrier is delivery of the inhibitory compounds via the nasal route (78). While this approach appears to result in selective delivery to the brain (Table 16.2), at least in rodents, it also is associated with numerous challenges, such as appropriate formulation of these generally lipophilic agents to allow mass delivery sufficient to exert a meaningful degree of inhibition (83).

Table 16.2 Systemic exposure, brain exposure, and brain partitioning for rifampin administered by the nasal or intraperitoneal route in mice Rifampin route (dose)

Intranasal (4 mg/kg)

Intraperitoneal (50 mg/kg)

IN/IP ratio

AUC0-60 min, brain (nM min)

1,560

173

9.02

AUC0-60 min, plasma (nM min)

18,200

1,590,000

Kp, brain

0.0860

0.000110

794

Exposure was expressed as area under the concentration–time curve (AUC) through 60 min postdose. Brain partitioning (Kp, brain) was calculated as AUC in brain/AUC in plasma. The influence of administration route was expressed as an intranasal to intraperitoneal ratio. Data obtained from ref. (83)

380

Padowski and Pollack

4.4. Pharmacodynamic Implications of P-Glycoprotein Induction

It is logical to presume that upregulation of Pgp function would result in pharmacodynamic changes opposite to those of Pgp inhibition, that is, lower concentrations of Pgp substrates at the target site, with a resulting decrease in the magnitude or duration of pharmacologic response. This clearly is the case for multidrug resistance at the cellular level (i.e., a decrease in the intracellular to extracellular concentration ratio for a pharmacologic agent that also is a Pgp substrate). In addition, when systemic clearance is increased secondary to Pgp upregulation in clearing organs, or when systemic availability is decreased in response to increased Pgp function in the intestine, concentrations at the target site may decrease. Such decreases are not indicative of enhanced exclusion from the target site, but merely of decreased concentrations in the systemic circulation. A more interesting situation occurs when pharmacologic response is mediated in a “sanctuary organ” protected by Pgp, such as the brain. Although examples of pharmacodynamic consequences of upregulating Pgp are rare when the transporter is serving an organ-level distributional barrier function (Fig. 16.7), a recent report illustrates the anticipated result (21). In general, murine blood–brain barrier Pgp is unresponsive to standard Pgp inducers such as rifampin. However, transgenic mice expressing human pregnane X receptor (hPXR) demonstrate measurable induction of Pgp expression and function. Exposure of isolated brain capillaries to rifampin in vitro increased Pgp expression (as determined by immunostaining of the protein) and transport function (as measured by the luminal concentration of a fluorescent Pgp substrate (Fig.  16.14a). Both Pgp expression and function were increased by rifampin exposure relative to control in a rifampin concentration-dependent manner. Rifampin-associated increases in Pgp-mediated transport at the blood–brain interface appear to have potential pharmacodynamics implications. Trans­ genic mice pretreated with rifampin evidenced a substantially lower antinociceptive effect after administration of methadone (Fig. 16.14b), a moderate Pgp substrate. The decrease in methadoneassociated pharmacologic response after rifampin pretreatment is consistent with enhanced barrier function, presumably by induced Pgp. Indeed, rifampin did not change the systemic concentrations of methadone in these animals (indicating that the reduction in response was not due to enhanced elimination of methadone from the systemic circulation) or the relationship between antinociception and methadone brain concentrations (indicating that rifampin pretreatment did not have a nonspecific effect on the pharmacologic endpoint). Although these data are somewhat narrow in scope, they clearly indicate that Pgp induction at the interface between the systemic circulation and a “protected” organ can affect pharmacologic response mediated by target receptors within that organ.

Pharmacokinetic and Pharmacodynamic Implications of P-Glycoprotein Modulation

381

Fig. 16.14. Response of isolated brain capillaries to rifampin in vitro (a) and influence of rifampin on the time course of methadone-associated antinociception (b) in transgenic (hPXR) mice. In panel (a), black bars depict luminal membrane Pgp immunofluorescence; gray bars represent luminal accumulation of [N-(4-nitrobenzofurazan-7-yl)-D-Lys8]cyclosporine A, a specific probe of Pgp function. In panel (b), open circles represent vehicle-pretreated mice, and solid circles represent mice pretreated with rifampin (50 mg/kg/day orally; 3 days). Antinociception was measured by electrical stimulation vocalization after a 3-mg/kg subcutaneous dose of methadone. Modified from ref. (21).

References 1. Zhou SF, Wang LL, Di YM et  al (2008) Substrates and inhibitors of human multidrug resistance associated proteins and the implications in drug development. Curr Med Chem 15:1981–2039 2. Miller DS, Bauer B, Hartz AM (2008) Modu­ lation of P-glycoprotein at the blood-brain barrier: opportunities to improve central nervous system pharmacotherapy. Pharmacol Rev 60:196–209 3. Gradhand U, Kim RB (2008) Pharmacogenomics of MRP transporters (ABCC1–5) and BCRP (ABCG2). Drug Metab Rev 40:317–354 4. Glaeser H, Fromm MF (2008) Animal models and intestinal drug transport. Expert Opin Drug Metab Toxicol 4:347–361

5. Shitara Y, Horie T, Sugiyama Y (2006) Trans­ porters as a determinant of drug clearance and tissue distribution. Eur J Pharm Sci 27: 425–446 6. Launay-Vacher V, Izzedine H, Karie S et  al (2006) Renal tubular drug transporters. Nephron Physiol 103:p97–p106 7. Liu X, Chen C, Smith BJ (2008) Progress in brain penetration evaluation in drug discovery and development. J Pharmacol Exp Ther 325: 349–356 8. Lum BL, Gosland MP (1995) MDR expression in normal tissues. Pharmacologic implications for the clinical use of P-glycoprotein inhibitors. Hematol Oncol Clin North Am 9:319–336

382

Padowski and Pollack

9. Fojo T, Coley HM (2007) The role of efflux pumps in drug-resistant metastatic breast cancer: new insights and treatment strategies. Clin Breast Cancer 7:749–756 10. Teodori E, Dei S, Martelli C, Scapecchi S, Gualtieri F (2006) The functions and structure of ABC transporters: implications for the design of new inhibitors of Pgp and MRP1 to control multidrug resistance (MDR). Curr Drug Targets 7:893–909 11. Juliano RL, Ling V (1976) A surface glycoprotein modulating drug permeability in Chinese hamster ovary cell mutants. Biochim Biophys Acta 455:152–162 12. Ha SN, Hochman J, Sheridan RP (2007) Mini review on molecular modeling of P-glycoprotein (Pgp). Curr Top Med Chem 7:1525–1529 13. Ma Q, Lu AY (2008) The challenges of dealing with promiscuous drug-metabolizing enzymes, receptors and transporters. Curr Drug Metab 9:374–383 14. Siarheyeva A, Lopez JJ, Glaubitz C (2006) Localization of multidrug transporter substrates within model membranes. Biochemistry 45:6203–6211 15. Shukla S, Wu CP, Ambudkar SV (2008) Development of inhibitors of ATP-binding cassette drug transporters: present status and challenges. Expert Opin Drug Metab Toxicol 4:205–223 16. Wigler PW, Patterson FK (1993) Inhibition of the multidrug resistance efflux pump. Biochim Biophys Acta 1154:173–181 17. Aszalos A (2007) Drug-drug interactions affected by the transporter protein, P-glyco­ protein (ABCB1, MDR1) II. Clinical aspects. Drug Discov Today 12:838–843 18. Aszalos A (2007) Drug-drug interactions affected by the transporter protein, P-glyco­ protein (ABCB1, MDR1) I. Preclinical aspects. Drug Discov Today 12:833–837 19. Sun H, Chow EC, Liu S, Du Y, Pang KS (2008) The Caco-2 cell monolayer: usefulness and limitations. Expert Opin Drug Metab Toxicol 4:395–411 20. Matheny CJ, Ali RY, Yang X, Pollack GM (2004) Effect of prototypical inducing agents on P-glycoprotein and CYP3A expression in mouse tissues. Drug Metab Dispos 32:1008–1014 21. Bauer B, Yang X, Hartz AM et al (2006) In vivo activation of human pregnane X receptor tightens the blood-brain barrier to methadone through P-glycoprotein up-regulation. Mol Pharmacol 70:1212–1219 22. Aquilante CL, Letrent SP, Pollack GM, Brouwer KL (2000) Increased brain P-glycoprotein in morphine tolerant rats. Life Sci 66:PL47–PL51

23. Dagenais C, Graff CL, Pollack GM (2004) Variable modulation of opioid brain uptake by P-glycoprotein in mice. Biochem Pharmacol 67:269–276 24. Kalvass JC, Olson ER, Cassidy MP, Selley DE, Pollack GM (2007) Pharmacokinetics and pharmacodynamics of seven opioids in P-glycoprotein-competent mice: assessment of unbound brain EC50, u and correlation of in  vitro, preclinical, and clinical data. J Pharmacol Exp Ther 323:346–355 25. Macdonald N, Gledhill A (2007) Potential impact of ABCB1 (p-glycoprotein) polymorphisms on avermectin toxicity in humans. Arch Toxicol 81:553–563 26. Adachi Y, Suzuki H, Sugiyama Y (2003) Quantitative evaluation of the function of small intestinal P-glycoprotein: comparative studies between in situ and in  vitro. Pharm Res 20:1163–1169 27. Badhan R, Penny J, Galetin A, Houston JB (2008) Methodology for development of a physiological model incorporating CYP3A and P-glycoprotein for the prediction of intestinal drug absorption. J Pharm Sci 98(6):2180–2197 28. Kivisto KT, Niemi M, Fromm MF (2004) Functional interaction of intestinal CYP3A4 and P-glycoprotein. Fundam Clin Pharmacol 18:621–626 29. Kitamura Y, Koto H, Matsuura S et al (2008) Modest effect of impaired P-glycoprotein on the plasma concentrations of fexofenadine, quinidine, and loperamide following oral administration in collies. Drug Metab Dispos 36:807–810 30. Gramatte T, Oertel R (1999) Intestinal secretion of intravenous talinolol is inhibited by luminal R-verapamil. Clin Pharmacol Ther 66:239–245 31. Ballent M, Lifschitz A, Virkel G, Sallovitz J, Lanusse C (2006) Modulation of the P-glycoprotein-mediated intestinal secretion of ivermectin: in vitro and in vivo assessments. Drug Metab Dispos 34:457–463 32. Spahn-Langguth H, Baktir G, Radschuweit A et  al (1998) P-glycoprotein transporters and the gastrointestinal tract: evaluation of the potential in  vivo relevance of in  vitro data employing talinolol as model compound. Int J Clin Pharmacol Ther 36:16–24 33. Igel S, Drescher S, Murdter T et  al (2007) Increased absorption of digoxin from the human jejunum due to inhibition of intestinal transporter-mediated efflux. Clin Pharmacokinet 46:777–785 34. Sandstrom R, Lennernas H (1999) Repeated oral rifampicin decreases the jejunal permeability

Pharmacokinetic and Pharmacodynamic Implications of P-Glycoprotein Modulation of R/S-verapamil in rats. Drug Metab Dispos 27:951–955 35. Drescher S, Glaeser H, Murdter T et al (2003) P-glycoprotein-mediated intestinal and biliary digoxin transport in humans. Clin Pharmacol Ther 73:223–231 36. Westphal K, Weinbrenner A, Zschiesche M et al (2000) Induction of P-glycoprotein by rifampin increases intestinal secretion of talinolol in human beings: a new type of drug/drug interaction. Clin Pharmacol Ther 68:345–355 37. Advani R, Fisher GA, Lum BL et al (2001) A phase I trial of doxorubicin, paclitaxel, and valspodar (PSC 833), a modulator of multidrug resistance. Clin Cancer Res 7:1221–1229 38. Fromm MF, Kim RB, Stein CM, Wilkinson GR, Roden DM (1999) Inhibition of P-glyco­ protein-mediated drug transport: a unifying mechanism to explain the interaction between digoxin and quinidine. Circulation 99: 552–557 39. Marie JP, Helou C, Thevenin D, Delmer A, Zittoun R (1992) In vitro effect of P-glycoprotein (Pgp) modulators on drug sensitivity of leukemic progenitors (CFU-L) in acute myelogenous leukemia (AML). Exp Hematol 20:565–568 40. Kim RB, Fromm MF, Wandel C et al (1998) The drug transporter P-glycoprotein limits oral absorption and brain entry of HIV-1 protease inhibitors. J Clin Invest 101:289–294 41. Angelin B, Arvidsson A, Dahlqvist R, Hedman A, Schenck-Gustafsson K (1987) Quinidine reduces biliary clearance of digoxin in man. Eur J Clin Invest 17:262–265 42. Booth CL, Brouwer KR, Brouwer KL (1998) Effect of multidrug resistance modulators on the hepatobiliary disposition of doxorubicin in the isolated perfused rat liver. Cancer Res 58:3641–3648 43. Yamada T, Kato Y, Kusuhara H, Lemaire M, Sugiyama Y (1998) Characterization of the transport of a cationic octapeptide, octreotide, in rat bile canalicular membrane: possible involvement of P-glycoprotein. Biol Pharm Bull 21:874–878 44. Micuda S, Fuksa L, Mundlova L et al (2007) Morphological and functional changes in p-glycoprotein during dexamethasone-induced hepatomegaly. Clin Exp Pharmacol Physiol 34:296–303 45. Riley J, Styles J, Verschoyle RD et al (2000) Association of tamoxifen biliary excretion rate with prior tamoxifen exposure and increased mdr1b expression. Biochem Pharmacol 60:233–239 46. Tanigawara Y (2000) Role of P-glycoprotein in drug disposition. Ther Drug Monit 22: 137–140

383

47. De Lannoy IA, Koren G, Klein J, Charuk J, Silverman M (1992) Cyclosporin and quinidine inhibition of renal digoxin excretion: evidence for luminal secretion of digoxin. Am J Physiol 263:F613–F622 48. Hori R, Okamura N, Aiba T, Tanigawara Y (1993) Role of P-glycoprotein in renal tubular secretion of digoxin in the isolated perfused rat kidney. J Pharmacol Exp Ther 266:1620–1625 49. Schenck-Gustafsson K, Dahlqvist R (1981) Pharmacokinetics of digoxin in patients subjected to the quinidine-digoxin interaction. Br J Clin Pharmacol 11:181–186 50. Schenck-Gustafsson K, Jogestrand T, Nord­ lander R, Dahlqvist R (1981) Effect of quinidine on digoxin concentration in skeletal muscle and serum in patients with atrial fibrillation. Evidence for reduced binding of digoxin in muscle. N Engl J Med 305:209–211 51. Kovarik JM, Rigaudy L, Guerret M, Gerbeau C, Rost KL (1999) Longitudinal assessment of a P-glycoprotein-mediated drug interaction of valspodar on digoxin. Clin Pharmacol Ther 66:391–400 52. Ding R, Tayrouz Y, Riedel KD et  al (2004) Substantial pharmacokinetic interaction between digoxin and ritonavir in healthy volunteers. Clin Pharmacol Ther 76:73–84 53. Okamura N, Hirai M, Tanigawara Y et  al (1993) Digoxin-cyclosporin A interaction: modulation of the multidrug transporter P-glycoprotein in the kidney. J Pharmacol Exp Ther 266:1614–1619 54. Ito S, Koren G, Harper PA, Silverman M (1993) Energy-dependent transport of digoxin across renal tubular cell monolayers (LLC-PK1). Can J Physiol Pharmacol 71:40–47 55. Hirata S, Izumi S, Furukubo T et  al (2005) Interactions between clarithromycin and digoxin in patients with end-stage renal disease. Int J Clin Pharmacol Ther 43:30–36 56. Ceckova-Novotna M, Pavek P, Staud F (2006) P-glycoprotein in the placenta: expression, localization, regulation and function. Reprod Toxicol 22:400–410 57. Fromm MF (2004) Importance of P-glyco­ protein at blood-tissue barriers. Trends Pharmacol Sci 25:423–429 58. Golden PL, Pollack GM (2003) Blood-brain barrier efflux transport. J Pharm Sci 92:1739–1753 59. Pardridge WM (1998) CNS drug design based on principles of blood-brain barrier transport. J Neurochem 70:1781–1792 60. Ohtsuki S, Terasaki T (2007) Contribution of carrier-mediated transport systems to the bloodbrain barrier as a supporting and protecting

384

Padowski and Pollack

interface for the brain; importance for CNS drug discovery and development. Pharm Res 24:1745–1758 61. Meyer RP, Gehlhaus M, Knoth R, Volk B (2007) Expression and function of cytochrome p450 in brain drug metabolism. Curr Drug Metab 8:297–306 62. Graff CL, Pollack GM (2004) Drug transport at the blood-brain barrier and the choroid plexus. Curr Drug Metab 5:95–108 63. Pardridge WM (2008) Re-engineering biopharmaceuticals for delivery to brain with molecular Trojan horses. Bioconjug Chem 19:1327–1338 64. Golden PL, Pardridge WM (2000) Brain microvascular P-glycoprotein and a revised model of multidrug resistance in brain. Cell Mol Neurobiol 20:165–181 65. Chen C, Liu X, Smith BJ (2003) Utility of Mdr1-gene deficient mice in assessing the impact of P-glycoprotein on pharmacokinetics and pharmacodynamics in drug discovery and development. Curr Drug Metab 4: 272–291 66. Chen C, Pollack GM (1997) Blood-brain disposition and antinociceptive effects of -D-penicillamine2, 5-enkephalin in the mouse. J Pharmacol Exp Ther 283:1151–1159 67. Chen C, Pollack GM (1998) Altered disposition and antinociception of [D-penicillamine(2, 5)] enkephalin in mdr1a-gene-deficient mice. J Pharmacol Exp Ther 287:545–552 68. Chen C, Pollack GM (1999) Enhanced antinociception of the model opioid peptide [D-penicillamine] enkephalin by P-glycoprotein modulation. Pharm Res 16:296–301 69. Thuerauf N, Fromm MF (2006) The role of the transporter P-glycoprotein for disposition and effects of centrally acting drugs and for the pathogenesis of CNS diseases. Eur Arch Psychiatry Clin Neurosci 256:281–286 70. Schinkel AH, Wagenaar E, Mol CA, van Deemter L (1996) P-glycoprotein in the blood-brain barrier of mice influences the brain penetration and pharmacological activity of many drugs. J Clin Invest 97:2517–2524 71. Chen C, Hanson E, Watson JW, Lee JS (2003) P-glycoprotein limits the brain penetration of nonsedating but not sedating H1-antagonists. Drug Metab Dispos 31:312–318

72. Lin JH (2007) Transporter-mediated drug interactions: clinical implications and in vitro assessment. Expert Opin Drug Metab Toxicol 3:81–92 73. Phillips EJ, Rachlis AR, Ito S (2003) Digoxin toxicity and ritonavir: a drug interaction mediated through p-glycoprotein? Aids 17: 1577–1578 74. Hebert MF, Lam AY (1999) Diltiazem increases tacrolimus concentrations. Ann Pharmacother 33:680–682 75. Sadeque AJ, Wandel C, He H, Shah S, Wood AJ (2000) Increased drug delivery to the brain by P-glycoprotein inhibition. Clin Pharmacol Ther 68:231–237 76. Letrent SP, Pollack GM, Brouwer KR, Brouwer KL (1999) Effects of a potent and specific P-glycoprotein inhibitor on the bloodbrain barrier distribution and antinociceptive effect of morphine in the rat. Drug Metab Dispos 27:827–834 77. Zong J, Pollack GM (2000) Morphine antinociception is enhanced in mdr1a gene-deficient mice. Pharm Res 17:749–753 78. Graff CL, Zhao R, Pollack GM (2005) Pharmacokinetics of substrate uptake and distribution in murine brain after nasal instillation. Pharm Res 22:235–244 79. Pallis M, Turzanski J, Higashi Y, Russell N (2002) P-glycoprotein in acute myeloid leukaemia: therapeutic implications of its association with both a multidrug-resistant and an apoptosis-resistant phenotype. Leuk Lymphoma 43:1221–1228 80. Takara K, Sakaeda T, Okumura K (2004) Carvedilol: a new candidate for reversal of MDR1/P-glycoprotein-mediated multidrug resistance. Anticancer Drugs 15:303–309 81. Lee JJ, Swain SM (2005) Development of novel chemotherapeutic agents to evade the mechanisms of multidrug resistance (MDR). Semin Oncol 32:S22–S26 82. Coley HM (2008) Mechanisms and strategies to overcome chemotherapy resistance in metastatic breast cancer. Cancer Treat Rev 34:378–390 83. Padowski JM (2008) A multi-factorial approach to understanding and predicting brain exposure to pharmacologic agents. Doctoral Dissertation, University of North Carolina at Chapel Hill. UMI No. UNC:3270, pp 95–112

Chapter 17 Examination of CYP3A and P-Glycoprotein-Mediated Drug–Drug Interactions Using Animal Models Punit H. Marathe and A. David Rodrigues Abstract With the advent of polytherapy for cancer treatment it has become prudent to minimize, as much as possible, the potential for drug–drug interactions (DDI). Toward this end, the metabolic and transporter pathways involved in the disposition of a drug candidate (phenotyping) and potential for inhibition and induction of drug-metabolizing enzymes and transporters are evaluated in vitro. Such in vitro human data can be made available prior to human dosing and enable in vitro to in vivo-based predictions of clinical outcomes. Despite some success, however, in vitro systems are not dynamic and sometimes fail to predict drug–drug interactions for a variety of reasons. In comparison, relatively less effort has been made to evaluate predictions based on data derived from in vivo animal models. This chapter will attempt to summarize different examples from the literature where animal models have been used to predict cytochrome P450 3A (CYP3A)- and P-glycoprotein-based DDI. When employing data from animal models one needs to be aware of species differences in enzyme- and transporter-activity leading to differences in pharmacokinetics, clearance pathways as well as species differences in selectivity and affinity of probe substrates and inhibitors. Because of these differences, in vivo animal studies alone, cannot be predictive of human DDI. Despite these caveats, the information obtained from validated in vivo animal models may prove useful when used in conjunction with in vitro–in vivo extrapolation methods. Such an integrated data set can be used to select drug candidates with a reduced DDI potential. Key words: Pharmacokinetics, Drug–drug interactions, In vivo animal studies, Phenotyping

1. Introduction For a life-threatening disease like cancer, concomitant ­administration of two or more drugs is a common practice and the potential for drug–drug interactions (DDI) has to be carefully monitored. Case reports of DDI resulting in adverse effects have been published for many drugs including temsirolimus and ketoconazole (1), capecitabine and warfarin (2), felodipine and erythromycin (3), mibefradil and terfenadine (4) and gemfibrozil and J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_17, © Humana Press, a part of Springer Science + Business Media, LLC 2010

385

386

Marathe and Rodrigues

cerivastatin (5, 6). As a result, most pharmaceutical companies now consider it prudent to minimize, as much as possible, the potential for DDI. Pharmacokinetic DDI occur when more than one drug shares a common clearance pathway involving a drug-metabolizing enzyme or transporter. In general, adverse reactions are generally associated with drugs able to induce or inhibit drug-metabolizing enzymes, or transporters, and drugs with a narrow therapeutic window. To understand the metabolic pathways involved in the disposition of a given drug candidate, in vitro reaction phenotyping methods are used (7, 8). Once the metabolic pathways are understood, there is a need to assess the potential impact of inhibition of the important pathways. Likewise, in vitro screening of new chemical entities for enzyme inhibition and induction is widely conducted at the discovery stage (9–12). At the same time, screening of drug candidates as substrates and inhibitors of various transporters is also becoming routine (13, 14). P-glycoprotein (Pgp) is particularly studied in the field of oncology, because of its involvement in chemoresistance. In addition, Pgp modulates drug pharmacokinetics, its importance in mediating DDI involving absorption, governing tissue-to-plasma ratios, and excretion is now accepted (15–18). Although several methods have been published describing the prediction of DDI on the basis of in vitro data, one has to accept that in vitro systems are not dynamic and sometimes are not predictive. For example, it is possible that other metabolic pathways and pharmacokinetic processes might also be altered by the inhibitor or may provide an escape from the pathway inhibited by the inhibitor. It is also likely that the true inhibitor concentrations cannot be predicted based on plasma concentrations alone, especially if the inhibitor accumulates in the liver because of active transport (19, 20). This poses a major problem when the extrapolations of the in vitro data are carried out ahead of human (clinical) studies. In the field of oncology, the situation is more difficult because of the challenges of conducting clinical DDI studies. For ethical reasons it may be difficult to administer suitable probe substrates that do not provide therapeutic benefit to cancer patients. However, several less invasive approaches have been described to investigate CYP3A-mediated DDI (e.g. urine 6b-OH cortisol/cortisol ratio, plasma 4b-OH-cholesterol, saliva analysis after oral administration of midazolam) (21–24). Owing to the logistical difficulties associated with conducting meaningful well-controlled DDI studies in cancer patients, additional tools such as in  vivo animal models could be employed. Data generated from these models could increase comfort that a drug will not be an inhibitor/inducer and on the basis of the strength of the data one can delay the DDI study until after proof of concept in the clinic.

Examination of CYP3A and P-Glycoprotein-Mediated Drug–Drug Interactions

387

When compared to predictions based on in vitro human data, relatively less effort has been made to evaluate predictions based on data derived from (in vivo) animal models. Therefore, this chapter will attempt to summarize different examples from the literature where animal models have been used to predict enzyme- and transporter-based DDI. The focus will be on CYP (CYP3A subfamily), as the oxidative enzyme(s), and Pgp as the efflux transporter. Compounds of interest are characterized as substrates or inhibitors of these proteins. Potentially, the information obtained from validated in  vivo tests in animal models can be used in conjunction with in vitro–in vivo extrapolation methods. Moreover, the increasing availability of transgenic (humanized) rodent models will further enable the bridging of in vitro and in vivo data preclinically.

2. Focus on CYP3A Enzymes and Pgp In the human liver and small intestine, cytochrome P450 3A (CYP3A) is the most important subfamily of CYPs that catalyzes the biotransformation of a wide variety of exogenous and endogenous substances (25). It has been estimated that CYP3A4, the most abundant human liver and gut CYP, plays a significant role in the metabolism of about half of the marketed drugs (26) and can serve as the locus of numerous metabolic drug interactions (27). Pgp is an efflux protein found in the small intestine, eliminating organs (liver, kidney), and target organs (brain, tumor). It has been noted that most CYP3A4 substrates are also substrates of Pgp, which demonstrates the mutually broad substrate selectivity of these proteins (28). The chapter focuses on studies described in the mouse, rat, and monkey although examples using the rabbit and dog are also found in the literature (29–31). 2.1. Examples of Interaction Studies Involving CYP3A Enzymes

Midazolam is extensively metabolized in humans and its clearance and oral bioavailability are governed by CYP3A4 (32–34). Ketoconazole, a potent CYP3A4 inhibitor, impairs the clearance and first pass extraction of midazolam. This leads to marked elevations in the midazolam area under the plasma concentration vs. time curve (AUC) and increased CNS side effects (35, 36). As a result of these clinical findings, several groups have conducted animal studies using midazolam as a CYP3A substrate and ketoconazole as a CYP3A inhibitor to assess the potential applicability of an experimental pharmacokinetic model for study of interactions involving substrates and inhibitors of CYP3A (37). It is worth noting that midazolam is not a Pgp substrate. Therefore, although ketoconazole can effectively inhibit both CYP3A and Pgp, the interaction between midazolam and ketoconazole can be essentially attributed to effects on CYP3A.

388

Marathe and Rodrigues

Another useful CYP3A probe is triazolam. An interaction between triazolam and ketoconazle was described (38) in Pgpdeficient and wild type mice. Brain concentrations of triazolam and the brain-to-serum concentration ratio were not different in the Pgp-deficient (vs. wild type) mice. In addition, the brain-toserum concentration ratio was not increased by ketoconazole. This indicates that triazolam is not a substrate of mouse Pgp. Co-administration of ketoconazole increased triazolam concentrations in the serum, brain, and liver both in wild type and Pgpdeficient mice. This increase in triazolam exposure is attributed to impairment of clearance in the presence of ketoconazole. 2.1.1. Transgenic Mouse

Because of the cross-species differences in the metabolism and disposition of CYP3A substrates, several CYP gene knockout and humanized (CYP3A4) mouse lines have been established for studying drug metabolism, pharmacology, and toxicity. Granvil et al. were able to show that the rate of 1¢-OH-midazolam and 4-OH-midazolam formation was significantly higher in the intestinal microsomes of the humanized (transgenic) Tg-CYP3A4 mice compared to those from wild type mice and humans. Moreover, oxidation of midazolam was inhibited markedly by ketoconazole, but not by inhibitors selective for other CYP forms (e.g., furafylline, sulfaphenazole and quinidine). Pretreatment with ketoconazole led to significantly increased Cmax and AUC of orally administered midazolam and the effect was more pronounced in the Tg-CYP3A4 mice (39).

2.1.2. Rat as a Model

Kotegawa et al. studied the interactions of midazolam and ketoconazole in the male Sprague Dawley rat after intraperitoneal (IP) administration of ketoconazole and intravenous (IV) and intragastric administration of midazolam (40). Ketoconazole increased the AUC (sixfold) of midazolam dosed intragastrically. Through IV and intragastric administration of midazolam, the authors concluded that the low oral bioavailability of midazolam in the rat is the result of hepatic rather than gastrointestinal metabolism. In comparison, the effect on midazolam AUC (16fold increase) is more pronounced in human subjects because of gut and liver first pass effects (41). These same authors showed that ketoconazole inhibits midazolam 1¢- and 4-hydroxylation in rat liver microsomes, with inhibition constant (Ki) values in the submicromolar range. The authors also showed that whereas 4-OH midazolam formation is mediated almost exclusively by CYP3A1 and CYP3A2, the formation of 1¢-OH-midazolam is catalyzed by CYP3A isoforms along with CYP2C and possibly by CYP2E1. Therefore, formation of 1¢-OH-midazolam in the rat may not be as sensitive to inhibition of CYP3A as it is in the human. However, since 4-OH midazolam formation accounts for a majority of midazolam intrinsic clearance, the change in

Examination of CYP3A and P-Glycoprotein-Mediated Drug–Drug Interactions

389

midazolam pharmacokinetics in the male rat may not be as sensitive to the formation of 1¢-OH-midazolam. Kanazu et al. employed DEX-pretreated female rats to study the interaction between midazolam and ketoconazole (42, 43). This animal model is thought to be more representative of human CYP3A4-mediated metabolism owing to very low involvement of other Cyps (e.g. Cyp 2C11). Pretreatment with dexamethasone increases the level of Cyp3A in female rat liver microsomes similar to the levels in human liver microsomes. Using this model, midazolam was shown to be primarily metabolized by the liver (both in vivo and in vitro) and the extraction ratio decreased from 93 to 77% in the presence of ketoconazole. For quantitative prediction of any interaction, it is essential to measure plasma concentrations of the inhibitor. Towards this end, Yamano et al., predicted inhibition of midazolam in rats based on plasma and liver concentrations of the inhibitors (itraconazole and ketoconazole)(44). The in vivo study was supplemented by measurement of Ki values for inhibitors and unbound concentrations in plasma and liver. The predicted increase in midazolam concentrations, based on the in  vitro Ki data, was considerably underestimated when using plasma unbound concentrations of inhibitors and best predicted after consideration of the concentrative uptake of the inhibitors in the liver and the unbound concentrations in the liver. Similarly, the interaction between nevirapine and two CYP inhibitors (ketoconazole and fluconazole) (45) was best predicted when the maximum unbound concentration of each inhibitor in the portal vein (46) was used along with data on the inhibition of 12-hydroxylation of nevirapine in rat liver microsomes. Recently, the rat was shown to be a useful animal model to rank-order compounds in the lead optimization stage based on the magnitude of increase in AUC in the presence of ketoconazole relative to midazolam (47). In vitro, in liver microsomes, there was no measurable turnover of these compounds and reaction phenotyping efforts were not discriminatory, although high involvement of CYP3A4 was suspected. 2.1.3. Non-human Primate as a Model

The animal model considered to have the greatest potential is the monkey because the CYP enzymes in these animals are thought to be similar to those in humans on the basis of sequence similarity (48, 49). However, the high cost, specialized training required to handle monkeys, and ethical concerns limit the use of this animal model in preclinical studies. The midazolam-ketoconazole pair has been used to characterize the CYP3A-mediated DDI in the monkey by various authors (43, 50). In a more recent study, the cynomolgus monkey was evaluated as an animal model for predicting intestinal DDI related to CYP3A4 (51). The monkey intestine was the predominant site for the first pass metabolism of

390

Marathe and Rodrigues

midazolam because the intestinal bioavailability (2.3%) was much lower than the hepatic bioavailability (70%). The authors showed a 22-fold increase in midazolam (1 mg/kg po) AUC in the presence of ketoconazole (20 mg/kg po) but no change in the midazolam AUC when administered intravenously. Additionally the Ki values of ketoconazole for midazolam 1¢-hydroxylation in monkey intestinal and liver microsomes were comparable to those in the respective human samples. For illustrative purposes, the utility of the cynomolgus monkey model is presented in Fig. 17.1. In this instance, the model was used in a preclinical setting to assess the inhibitory potential of an oncologic prior to first-in-human studies. BMS-1 was shown to be a potent inhibitor of CYP3A4 in human liver microsomes and CYP3A activity in cynomolgus monkey liver microsomes. Consequently, an in vivo monkey DDI study was conducted with midazolam as the CYP3A probe and the impact of BMS-1 on the ratio of 1¢-OH-midazolam to midazolam in plasma was assessed. The ratio was then compared with that obtained after concomitant

Fig. 17.1. Evaluation of an oncologic NCE as a potential CYP3A inhibitor.

Examination of CYP3A and P-Glycoprotein-Mediated Drug–Drug Interactions

391

administration of midazolam and ketoconazole. BMS-1 was deemed to be a weak CYP3A inhibitor based on a relatively small ratio (~1.3 vs. 5.0 with ketoconazole) and a clinical DDI study was postponed until after the proof-of-concept in the clinic. Overall, it was possible to rationalize the weak inhibition in vivo because of high protein binding. 2.2. Examples of Interaction Studies Involving PgpMediated Transport 2.2.1. Mouse as a Model

2.2.2. Rat as a Model

The impact of Pgp on drug disposition has been well characterized due to the availability of Pgp (mdr1a) knock-out mice (52, 53). However, such mice exhibit compensatory increases in CYP3A and mdra1b expression to compensate for the loss of Pgp (54). Therefore, the potential for underestimating the influence of Pgp in knock-out mice cannot be ruled out and some have resorted to using specific Pgp substrates and inhibitors. Towards this end, a number of investigators have employed Pgp inhibitors such as valspodar (PSC833), elacridar (GF120918), and zosuquidar (LY335979)(55–57). LY335979 is thought to be a more selective Pgp inhibitor owing to its low affinity for CYP enzymes and other drug transporting proteins such as MRP1, MRP2 and BCRP (58, 59). For example, an intraperitoneal injection of LY335979 prior to dosing has been shown to increase the brain-to-plasma ratio of Gleevec (threefold) over wild-type control mice (56). A similar increase in the brain-to-plasma ratio was observed in Pgp-knockout mice (vs. wild type). In a similar study, uptake of paclitaxel into the brain was determined in the absence and presence of LY335979 (60). When administered orally prior to the paclitaxel dose, LY335979 increased the paclitaxel brain AUC (3.5- to 5.0-fold) compared to untreated mice. LY335979 also increased the paclitaxel concentrations in plasma and tissues to levels similar to those observed in Pgp-knockout mice suggesting inhibition of Pgp. Male Sprague–Dawley rats have also been used as animal models to study Pgp-mediated DDI. For example, Yumoto et al. evaluated the effect of DEX and cyclosporin pretreatment on the Pgpmediated intestinal exsorption and biliary clearance of Rho123 (61). Rho123 has been widely used as a marker to study Pgp function in multidrug-resistance and normal cells. Because Rho123 is not a substrate of CYP3A, it is an attractive probe to evaluate Pgp function in vivo. The authors were able to show that DEX pretreatment increased Pgp expression (approximately twofold) in the intestine, as measured by western blot analysis, but not in the liver. In DEX-pretreated rats, the intestinal exsorption clearance of Rho123 increased approximately twofold, but biliary clearance was not increased in accordance with the changes in Pgp expression. Cyclosporin A, given intravenously, decreased intestinal exsorption clearance and biliary clearance of Rho123 in both untreated and DEX-pretreated rats.

392

Marathe and Rodrigues

2.2.3. Non-human Primate

In a monkey model to evaluate the effect of Pgp, GF120918 when co-administered with erythromycin (62, 63), showed a significant increase in both absorption and systemic exposure of erythromycin. Although the extent of absorption of erythromycin improved significantly, the hepatic extraction was essentially unchanged. No changes in either the absorption or hepatic extraction of propranolol (negative control) were noted. Hepatic portal and systemic pharmacokinetics of GF120918 were also characterized in this study. At maximum concentrations observed in vivo, GF120918 was shown not to inhibit any CYP enzymes, thus demonstrating its utility as an inhibitor of Pgp/BCRP. The maximum hepatic portal plasma concentrations of GF120918 were in the micromolar range, similar to those used in the in vitro transport studies of Pgp substrates (64). The cynomolgus monkey has also been used as a model to evaluate intestinal first-pass effect of fexofenadine (a Pgp substrate) (51). Concomitant administration of ketoconazole increased the AUC of fexofenadine threefold.

2.3. Examples of Interaction Studies Implicating both CYP3A and Pgp

Oral bioavailability of drugs is often dependent on intestinal efflux and CYP3A-mediated first-pass metabolism. However, the ability to directly assess the potential importance of these two intestinal processes in  vivo is technically challenging. Pgp is thought not only to affect the secretion of a variety of drugs in the small intestine, colon, liver, and kidney, but it can also affect the metabolism and disposition of drugs since it is an important factor in controlling cellular drug concentrations and the residence time of drugs inside cells. In fact, many investigators are now assessing the interplay between drug-metabolizing enzymes such as CYP3A4 and transporters like Pgp (65–67). Lin et al. utilized indinavir as the dual CYP3A and Pgp substrate in order to investigate the inducing effects of DEX (68). Pretreatment of rats with DEX had little effect on the systemic pharmacokinetics of indinavir; whereas it markedly affected its oral bioavailability. The decreased oral bioavailability was attributed to an increase in both intestinal and hepatic first-pass metabolism after DEX pretreatment. In fact, DEX pretreatment caused a significant induction of CYP3A and Pgp expression in both the small intestine and liver. In vitro data revealed that the increased intestinal and hepatic metabolism by DEX was due to an increase in Vmax without a significant change in Km. The predicted hepatic extraction, based on the in  vitro data, was in reasonably good agreement with the observed in vivo hepatic extraction with and without DEX pretreatment. However, a significant discrepancy was noted in the predicted (<
Examination of CYP3A and P-Glycoprotein-Mediated Drug–Drug Interactions

393

prolonging the intracellular residence time through the repetitive processes of exsorption and reabsorption. Cummins et  al. used a single pass rat intestinal perfusion model to provide an isolated in vivo system to study the role of Pgp in controlling the extent of intestinal metabolism by CYP3A (69). They examined the absorption and metabolism of a dual substrate of Pgp and CYP3A (K77) and a substrate of CYP3A but not of Pgp (midazolam) across a segment of rat ileum. In addition, transporter function was modulated using the Pgp inhibitor GF120918. The absorptive permeability of K77 was increased in presence of GF120918 whereas midazolam permeability was unchanged under similar conditions. Measurement of metabolite concentrations allowed assessment of the role of Pgp in modulating the extent of intestinal metabolism. A significant decrease in the fraction metabolized was noted for K77 while no change in this parameter was observed for midazolam. Thus, although GF120918 was used as a selective inhibitor of Pgp, the study demonstrated a role of Pgp in modulating the extent of intestinal metabolism in vivo by controlling drug access to the enzyme. Ward et al. employed the cynomolgus monkey as an in vivo animal model to investigate the combined role of Pgp and CYP3A (62, 63). Ketoconazole concentrations were measured in the portal and systemic circulation after intraduodenal dosing and were in the range of concentrations encountered with usual therapeutic doses of ketoconazole in humans (70). Erythromycin, a dual substrate of Pgp and CYP3A was used as a positive control and propranolol served as a negative control. In the absence of ketoconazole, the absorption and systemic exposure of erythromycin was poor. Co-administration of ketoconazole produced an increase in absorption and systemic exposure. The elimination half-life of erythromycin was also prolonged in the presence of ketoconazole. In contrast, ketoconazole had no significant effect on either the absorption, or systemic exposure of propranolol. The model was used in a drug discovery setting to identify compounds with acceptable absorption and hepatic extraction, with minimized potential for combined DDI involving Pgp and CYP3A inhibitors since contribution from CYP3A vs. Pgp could not be delineated.

3. Industrial Perspective In drug discovery today, numerous in vitro studies with human systems are being performed in order to proactively assess the risk associated with DDI liabilities. The eventual clinical significance of a discovery compound displaying an in vitro IC50 in the low micromolar (£1 µM) range for a human CYP or transporter is

394

Marathe and Rodrigues

difficult to assess. In fact, interpretation of such data is influenced by myriad factors including therapeutic concentration, protein binding, route and duration of administration, likelihood of drug co-administration, and the affinities of the inhibitor and substrate for the CYP or transporter in question, relative to their concentrations at the active site. Additional factors (for a substrate) such as relative importance of a particular CYP or transporter in the disposition of the NCE need to be considered when assessing the clinical impact of a co-administered inhibitor. In vivo animal interaction studies, if properly designed and interpreted with appropriate caveats can provide useful information to build upon the findings of in  vitro studies. Towards this end, an idealized decision tree is provided for the use of in vitro and in vivo data in preclinical species to predict human DDI (Fig. 17.2). 3.1. Establishing In Vitro–In Vivo Connectivity

When in vitro studies demonstrate that a new chemical entity is primarily a CYP3A4 substrate, reaction phenotyping studies should be conducted in animal liver microsomes with ketoconazole to confirm CYP3A involvement. In vivo pharmacokinetic studies should establish the role of metabolism vs. excretion in the disposition of the NCE in the animal model under therapeutically relevant conditions. This should be followed by an in vivo interaction study with ketoconazole and the change in AUC should be compared to that of a CYP3A probe (e.g. midazolam). The change in the in  vivo AUC obtained with ketoconazole should be correlated to the in vitro effect of ketoconazole on the metabolism of NCE and an in vitro–in vivo relationship (IVIVR) should be established. This IVIVR can then be extended to predict the potential for human DDI on the basis of human in vitro findings. Figure 17.2 describes a similar proposal for an NCE that is primarily a Pgp substrate or a CYP3A4/Pgp inhibitor. Potential issues that need particular attention in the design and interpretation of animal DDI studies include different affinity or selectivity of a probe substrate/inhibitor for animal versus human CYP/transporter, different extents of plasma protein binding of the substrate and/or inhibitor in animal versus human, different liver to plasma distribution of the substrate and/or inhibitor between species, differences in the metabolic and excretion patterns between species, differences in the fraction metabolized through the CYP of interest between species, and potential irreversible inhibition of animal and/or human CYP. Potential differences in the distribution of specific drug metabolizing enzymes and transporters across different tissues in different species should also be considered as another caveat. Midazolam has been extensively used as a CYP3A probe in the rat. However, the rat differs from the human with respect to the quantitative formation of the two CYP3A-mediated metabolites, clearance relative to hepatic blood flow, and contribution of

Examination of CYP3A and P-Glycoprotein-Mediated Drug–Drug Interactions

395

Fig. 17.2. An idealized decision tree for the use of in vitro and in vivo preclinical data for prediction of human DDI.

liver vs. intestine to presystemic extraction of midazolam. Particular attention should also be paid to the dose of the probe substrate to ensure it does not cause saturation of the first pass metabolism in the intestine and/or the liver which may lead to under-prediction of the DDI.

396

Marathe and Rodrigues

CYP specificity of a probe substrate for the animal enzyme should be confirmed ideally using a panel of recombinant enzymes due to very limited knowledge of selective CYP inhibitors and limited availability of selective antibodies. Availability of recombinant enzymes from dog and monkey is limited (71, 72). A suitable, specific probe substrate should have a low extraction ratio such that a decrease in intrinsic clearance due to enzyme inhibition will result in a more readily observed decrease in systemic clearance. When conducting in vivo experiments in the rat, the probe substrate can be dosed to bile-duct cannulated rats to verify that its systemic clearance is predominantly (>50%) via hepatic metabolism; if a large component is excretory or non-CYP mediated, the robustness of the interaction is lessened. Conducting such studies in higher animals is not always feasible. If the compound of interest is a CYP inhibitor, characterization of its affinity to the animal enzyme mediating the selected probe oxidation and the human CYP of concern will be important to understand potential species differences. For example, ketoconazole inhibits formation of both triazolam metabolites and 4-OH-midazolam formation in mouse liver microsomes. Unlike triazolam, however, ketoconazole does not inhibit 1’-OH midazolam formation in the mouse (Ki values ~300 times higher compared with human liver microsomes). Formation of 1¢-OH-midazolam has a major CYP2C component in addition to CYP3A in mice, demonstrating that metabolic profiles of drugs in animals cannot be assumed to reflect human metabolic patterns even with closely related substrates (73). In this regard triazolam appears to be a better probe for CYP3A in the mouse compared to midazolam. 3.2. Which Inhibitor Concentration is Most Relevant?

Measurement of the plasma concentrations of the inhibitor in vivo in the animal model is essential to confirm that the concentrations exceed the inhibitory potency in  vitro. There is considerable debate about which concentrations of the inhibitor best predict the magnitude of the in vivo drug interaction. On the basis of the free drug hypothesis, free plasma concentrations of the inhibitor have been used to predict the magnitude of interaction. However, free (unbound) concentrations of the inhibitor often severely under-predict the interaction and therefore, using total plasma concentrations is considered a more conservative approach (74). Use of concentrative uptake in the liver and unbound concentrations in the liver have also been proposed for quantitative predictions (44). In clinical practice, it is not practical to measure unbound concentrations in the liver. However, a good correlation was observed between cell to medium concentration ratios in isolated rat hepatocytes and liver to blood unbound concentration ratios (44). It may be possible to predict liver unbound concentrations in humans on the basis of cell to medium ratios of drugs in isolated human hepatocytes in order to avoid cross-species

Examination of CYP3A and P-Glycoprotein-Mediated Drug–Drug Interactions

397

comparison in quantitative predictions. The predicted unbound concentrations in the liver can aid in quantitative prediction of DDI on the basis of in  vivo animal interaction studies. This assumes that there are no species differences in the distribution of unbound drugs in the liver between human and rat and there is no difference in the activity of influx and efflux transporters. Yet another concentration thought to be best predictive is the maximum unbound concentration of the inhibitor in the portal vein during absorption (75) and has been employed to predict the magnitude of DDI in animals on the basis of in vitro data. 3.3. Taking Pgp into Account

Unlike CYP3A, confirming different affinity or selectivity of a probe substrate/inhibitor for animal vs. human Pgp is a much bigger challenge due to unavailability of cell lines transfected with animal specific Pgp. A few studies have used cell lines transfected with mouse mdr1a but the corresponding cell lines from other animal species are lacking. Availability of such cell lines and knowledge of specificity of the probes used will enable proper design and interpretation of animal DDI studies involving Pgp-mediated disposition. More investigation is needed in terms of which concentrations are relevant for predicting Pgp-mediated interactions in vivo. Pgp serves as the efflux transporter in many organs and the substrate/inhibitor concentrations achieved at the brush border membrane of the small intestine will be very different from those achieved at the apical membrane of brain or the bile canalicular membrane of hepatocytes. Knowledge of the concentrationresponse kinetics for a Pgp substrate and inhibitor will help us assess the extent of DDI with respect to absorption, brain penetration, renal clearance, biliary secretion etc., the net results of which will be the observed change in plasma AUC. Animal interaction studies in the area of Pgp inhibitors are useful to enable oral therapy with anticancer drugs like paclitaxel and their efficient delivery to the brain. Significant overlap exists for the substrate and inhibitor specificity between CYP3A and Pgp making it difficult to identify whether the observed in vivo interaction is a result of CYP3A-mediated or Pgp-mediated interaction or both. It is therefore, important to choose the right substrate-inhibitor pair. For example use of ritonavir to assess CYP3A mediated interactions is appropriate only if the substrate is specific for CYP3A and other CYP enzymes and transporters are not involved. Substantial effort has been expended to identify specific modulators of either Pgp or CYP3A, with varying degrees of success. Given the potentially important role of transporters such as Pgp in mediating multidrug resistance to oncology treatments, a number of transport inhibitors have been discovered and progressed through development as adjuncts in cancer chemotherapy (76). Among these is GF120918, which has been used as a selective inhibitor of Pgp (reported Ki of 35 nM (77)) without a potent

398

Marathe and Rodrigues

inhibitory effect on CYP3A (reported Ki of 10 µM (78)), although, it is also known to affect other related efflux proteins such as BCRP (77, 79, 80). 3.4. Measurement of Metabolite Concentrations In vivo

This technique can further aid in delineating the effect of the inhibitor on metabolism vs. a transporter-mediated pathway such as direct excretion. In validating an animal model, it is useful to show similar circulating metabolites, a decrease in the metabolite concentrations and a decrease in the AUC ratio of the metabolite to the parent when an inhibitor is on board. These changes are more convincing of inhibition of a CYP-mediated pathway rather than changes in the parent exposure alone. Midazolam, with its well characterized oxidative metabolites 1¢-OH and 4-OH-midazolam offers a valuable tool in this regard. Changes in the disposition of a metabolite, however, are not always easy to interpret depending on the mechanism of its subsequent elimination and how it is influenced by the presence of the inhibitor. Comparison of the formation of metabolites when dosed alone and in presence of a Pgp inhibitor may allow assessment of the role of Pgp in modulating the extent of intestinal and liver metabolism (69).

3.5. Transgenic and Other Animal Models

The knockout animals are another valuable tool to sort out the contribution of related enzymes and transporters with relatively nonselective inhibitors. Ketoconazole is an inhibitor of both CYP3A and Pgp. However, it can be used as a CYP3A inhibitor in Pgp knockout mice to study CYP3A-mediated DDI (38). Similarly, GW120918 was used in a Mrp2-deficient rat to assess impact of Pgp on the biliary excretion of [D-Pen2, D-Pen5]Enkephalin (81). Transgenic mice with humanized PXR and CAR are being investigated for study of inductive DDI (82–84). The PXR humanized mice respond to human PXR activators such as rifampicin but not to the rodent activator pregnenalone 16a-carbonitrile. A chimeric mouse model in which the liver could be replaced by more than 80% with human hepatocytes has been established (85). The human CYP3A4 expressed in these mice was induced by rifabutin and shown to be a useful animal model to estimate and predict the in vivo induction of CYPs in human. Metabolism and toxicity of docetaxel was investigated in Cyp3a−/− mice that showed severely impaired detoxification capacity for docetaxel (86). To determine the relative importance of intestinal vs. hepatic Cyp3a in first pass metabolism, the authors generated transgenic Cyp3a−/− mice expressing human CYP3A4 in either the intestine or the liver. Expression of CYP3A4 in the intestine dramatically decreased absorption of docetaxel into the bloodstream, whereas hepatic expression aided systemic docetaxel clearance. More transgenic lines expressing human CYP1A1/ CYP1A2, CYP2E1, CYP2D6, and CYP3A7 have been generated

Examination of CYP3A and P-Glycoprotein-Mediated Drug–Drug Interactions

399

which offer broad utility in the evaluation and prediction of xenobiotic metabolism and DDI (87–89). Because of its relevance to human physiology, the rat may prove to be a highly predictable model for the pharmaceutical industry. Until recently, the lack of efficient tools to manipulate the rat genome had drastically limited the use of this research model. Recent advances in gene expression and transgenic systems may enable creation of “humanized rats” in which endogenous rat genes are replaced by their human counterparts providing novel in vivo animal models for predicting DDI in humans (90). Unlike the CYP enzymes, humanized transporter models are still not available for efflux transporters (e.g. Pgp). Availability of additional transgenic mice and rats with humanized CYPs and transporters will be an excellent tool for study of DDI as the twenty first century unfolds. Caution should be exercised in making quantitative predictions of DDI since quantitative analysis of kinetics may be impeded by the complexity of gene transcription, gene translation, and differences in the expression levels and tissue expression patterns (91, 92).

4. Conclusions Because of species differences in drug disposition, in vivo animal studies alone, cannot be predictive of human DDI. However, results of in vivo animal DDI studies could be used in a coordinated fashion with the in vitro results derived from the same species and the observed correlations could be integrated to enable predictions of human interactions on the basis of human in vitro results (Fig. 17.2). More simplistically, discovery compounds may be rank- ordered based on the magnitude of the in vivo interaction in an appropriately validated animal model and the compounds progressed further depending on the assessment of the risk of the likely DDI. Such strategies will become more meaningful as transgenic (humanized) rodent models become increasingly available. This will be important as the industry strives to integrate data related to drug-metabolizing enzymes and transporters and assess the net effect of their interplay on drug disposition and drug-interaction profile of new chemical entities. References 1. Boni JP, Leister C, Burns J, Hug B (2008) Differential effects of ketoconazole on exposure to temsirolimus following intravenous infusion of temsirolimus. Br J Cancer 98:1797–1802

2. Yildirim Y, Ozyilkan O, Akcali Z, Basturk B (2006) Drug interaction between capecitabine and warfarin: a case report and review of the literature. Int J Clin Pharmacol Ther 44:80–82

400

Marathe and Rodrigues

3. Bailey DG, Bend JR, Arnold JM, Tran LT, Spence JD (1996) Erythromycin-felodipine interaction: magnitude, mechanism, and comparison with grapefruit juice. Clin Pharmacol Ther 60:25–33 4. Alfaro CL (2001) Emerging role of drug interaction studies in drug development: the good, the bad, and the unknown. Psychopharmacol Bull 35:80–93 5. Jones PH, Davidson MH (2005) Reporting rate of rhabdomyolysis with fenofibrate + statin versus gemfibrozil + any statin. Am J Cardiol 95:120–122 6. Kind AH, Zakowski LJ, McBride PE (2002) Rhabdomyolysis from the combination of a statin and gemfibrozil: an uncommon but serious adverse reaction. WMJ 101:53–56 7. Emoto C, Murase S, Iwasaki K (2006) Approach to the prediction of the contribution of major cytochrome P450 enzymes to drug metabolism in the early drug-discovery stage. Xenobiotica 36:671–683 8. Williams JA, Bauman J, Cai H et al (2005) In vitro ADME phenotyping in drug discovery: current challenges and future solutions. Curr Opin Drug Discov Devel 8:78–88 9. Dambach DM, Andrews BA, Moulin F (2005) New technologies and screening strategies for hepatotoxicity: use of in vitro models. Toxicol Pathol 33:17–26 10. Hutzler M, Messing DM, Wienkers LC (2005) Predicting drug–drug interactions in drug discovery: where are we now and where are we going? Curr Opin Drug Discov Devel 8:51–58 11. Kim MJ, Kim H, Cha IJ et al (2005) Highthroughput screening of inhibitory potential of nine cytochrome P450 enzymes in  vitro using liquid chromatography/tandem mass spectrometry. Rapid Commun Mass Spectrom 19:2651–2658 12. Persson KP, Ekehed S, Otter C et  al (2006) Evaluation of human liver slices and reporter gene assays as systems for predicting the cytochrome P450 induction potential of drugs in vivo in humans. Pharm Res 23:56–69 13. Balimane PV, Patel K, Marino A, Chong S (2004) Utility of 96 well Caco-2 cell system for increased throughput of P-gp screening in drug discovery. Eur J Pharm Biopharm 58:99–105 14. Stouch TR, Gudmundsson O (2002) Progress in understanding the structure-activity relationships of P-glycoprotein. Adv Drug Deliv Rev 54:315–328 15. Ho RH, Kim RB (2005) Transporters and drug therapy: implications for drug disposition and disease. Clin Pharmacol Ther 78:260–277

16. Yu DK (1999) The contribution of P-glycoprotein to pharmacokinetic drug–drug interactions. J Clin Pharmacol 39:1203–1211 17. Aszalos A (2007) Drug–drug interactions affected by the transporter protein, P-glycoprotein (ABCB1, MDR1) II. Clinical aspects. Drug Discov Today 12:838–843 18. Aszalos A (2007) Drug–drug interactions affected by the transporter protein, P-glycoprotein (ABCB1, MDR1) I. Preclinical aspects. Drug Discov Today 12:833–837 19. Hara Y, Nakajima M, Miyamoto KI, Yokoi T (2005) Inhibitory effects of psychotropic drugs on mexiletine metabolism in human liver microsomes: prediction of in  vivo drug interactions. Xenobiotica 35:549–560 20. Kanamitsu S, Ito K, Sugiyama Y (2000) Quantitative prediction of in vivo drug–drug interactions from in vitro data based on physiological pharmacokinetics: use of maximum unbound concentration of inhibitor at the inlet to the liver. Pharm Res 17:336–343 21. Bodin K, Bretillon L, Aden Y et  al (2001) Antiepileptic drugs increase plasma levels of 4beta-hydroxycholesterol in humans: evidence for involvement of cytochrome p450 3A4. J Biol Chem 276:38685–38689 22. Galteau MM, Shamsa F (2003) Urinary 6beta-hydroxycortisol: a validated test for evaluating drug induction or drug inhibition mediated through CYP3A in humans and in animals. Eur J Clin Pharmacol 59:713–733 23. Kanebratt K, Diczfalusy U, Backstrom T et al (2008) Cytochrome P450 induction by rifampicin in healthy subjects: determination using the Karolinska cocktail and the endogenous CYP3A4 marker 4beta-hydroxycholesterol. Clin Pharmacol Ther 84:589–594 24. Link B, Haschke M, Grignaschi N et al (2008) Pharmacokinetics of intravenous and oral midazolam in plasma and saliva in humans: usefulness of saliva as matrix for CYP3A phenotyping. Br J Clin Pharmacol 66:473–484 25. Guengerich FP (1999) Cytochrome P-450 3A4: regulation and role in drug metabolism. Annu Rev Pharmacol Toxicol 39:1–17 26. Liu YT, Hao HP, Liu CX, Wang GJ, Xie HG (2007) Drugs as CYP3A probes, inducers, and inhibitors. Drug Metab Rev 39:699–721 27. Zhou SF (2008) Drugs behave as substrates, inhibitors and inducers of human cytochrome P450 3A4. Curr Drug Metab 9:310–322 28. Wacher VJ, Silverman JA, Zhang Y, Benet LZ (1998) Role of P-glycoprotein and cytochrome P450 3A in limiting oral absorption of peptides and peptidomimetics. J Pharm Sci 87:1322–1330

Examination of CYP3A and P-Glycoprotein-Mediated Drug–Drug Interactions 29. Kuroha M, Azumano A, Kuze Y, Shimoda M, Kokue E (2002) Effect of multiple dosing of ketoconazole on pharmacokinetics of midazolam, a cytochrome P450 P450 3A substrate in beagle dogs. Drug Metab Dispos 30:63–68 30. Kuroha M, Kayaba H, Kishimoto S et  al (2002) Effect of oral ketoconazole on firstpass effect of nifedipine after oral administration in dogs. J Pharm Sci 91:868–873 31. Sinko PJ, Kunta JR, Usansky HH, Perry BA (2004) Differentiation of gut and hepatic first pass metabolism and secretion of saquinavir in ported rabbits. J Pharmacol Exp Ther 310:359–366 32. Gorski JC, Jones DR, Haehner-Daniels BD et  al (1998) The contribution of intestinal and hepatic CYP3A to the interaction between midazolam and clarithromycin. Clin Pharmacol Ther 64:133–143 33. Thummel KE, Shen DD, Podoll TD et  al (1994) Use of midazolam as a human cytochrome P450 3A probe: I. In vitro–in  vivo correlations in liver transplant patients. J Pharmacol Exp Ther 271:549–556 34. Wandel C, Bocker RH, Bohrer H et al (1998) Relationship between hepatic cytochrome P450 3A content and activity and the disposition of midazolam administered orally. Drug Metab Dispos 26:110–114 35. Lam YW, Alfaro CL, Ereshefsky L, Miller M (2003) Pharmacokinetic and pharmacodynamic interactions of oral midazolam with ketoconazole, fluoxetine, fluvoxamine, and nefazodone. J Clin Pharmacol 43:1274–1282 36. Pea F, Furlanut M (2001) Pharmacokinetic aspects of treating infections in the intensive care unit: focus on drug interactions. Clin Pharmacokinet 40:833–868 37. Marathe PH, Rodrigues AD (2006) In vivo animal models for investigating potential CYP3A- and Pgp-mediated drug–drug interactions. Curr Drug Metab 7:687–704 38. von Moltke LL, Granda BW, Grassi JM et al (2004) Interaction of triazolam and ketoconazole in P-glycoprotein-deficient mice. Drug Metab Dispos 32:800–804 39. Granvil CP, Yu AM, Elizondo G et al (2003) Expression of the human CYP3A4 gene in the small intestine of transgenic mice: In vitro metabolism and pharmacokinetics of midazolam. Drug Metab Dispos 31:548–558 40. Kotegawa T, Laurijssens BE, von Moltke LL et  al (2002) In vitro, pharmacokinetics, and pharmacodynamic interactions of ketoconazole and midozolam in the rat. J Pharmacol Exp Ther 302:1228–1237 41. Tsunoda SM, Velez RL, von Moltke LL, Greenblatt DJ (1999) Differentiation of

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

401

intestinal and hepatic cytochrome P450 3A activity with use of midazolam as an in  vivo probe: effect of ketoconazole. Clin Pharmacol Ther 66:461–471 Kanazu T, Okamura N, Yamaguchi Y, Baba T, Koike M (2005) Assessment of the hepatic and intestinal first-pass metabolism of midazolam in a CYP3A drug–drug interaction model rats. Xenobiotica 35:305–317 Kanazu T, Yamaguchi Y, Okamura N, Baba T, Koike M (2004) Model for the drug–drug interaction responsible for CYP3A enzyme inhibition. I: evaluation of cynomolgus monkeys as surrogates for humans. Xenobiotica 34:391–402 Yamano K, Yamanoto K, Kotaki K, Sawada Y, Iga T (1999) Quantitative prediction of metabolic inhibition of midazolam by itraconazole and ketoconazole in rats: implication of concentrative uptake of inhibitors of into liver. Drug Metab Dispos 27:395–402 Kishimoto W, Takano J, Senda C et al (2000) Quantitative prediction of in vivo drug interactions between nevirapine and antifungal agents from in vitro data in rats. Biol Pharm Bull 23:1027–1032 Ito K, Iwatsubo T, Kanamitsu S, Nakajima Y, Sugiyama Y (1998) Quantitative prediction of in  vivo drug clearance and drug interactions from in  vitro data on metabolism, together with binding and transport. Annu Rev Pharmacol Toxicol 38:461–499 Mandlekar SV, Rose AV, Cornelius G et  al (2007) Development of an in vivo rat screen model to predict pharmacokinetic interactions of CYP3A4 substrates. Xenobiotica 37: 923–942 Igarashi T, Sakuma T, Isogai M, Nagata R, Kamataki T (1997) Marmoset liver cytochrome P450s: study for expression and molecular cloning of their cDNAs. Arch Biochem Biophys 339:85–91 Mankowski DC, Laddison KJ, Christopherson PA et  al (1999) Molecular cloning, expression, and characterization of CYP2D17 from cynomolgus monkey liver. Arch Biochem Biophys 372:189–196 Komura H, Iwaki M (2008) Species differences in in  vitro and in  vivo small intestinal metabolism of CYP3A substrates. J Pharm Sci 97:1775–1800 Ogasawara A, Kume T, Kazama E (2007) Effect of oral ketoconazole on intestinal firstpass effect of midazolam and fexofenadine in cynomolgus monkeys. Drug Metab Dispos 35:410–418 Schinkel AH, Smit JJ, van Tellingen O et  al (1994) Disruption of the mouse mdr1a

402

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

Marathe and Rodrigues P-glycoprotein gene leads to a deficiency in the blood–brain barrier and to increased sensitivity to drugs. Cell 77:491–502 Schinkel AH, Wagenaar E, van Deemter L, Mol CA, Borst P (1995) Absence of the mdr1a P-Glycoprotein in mice affects tissue distribution and pharmacokinetics of dexamethasone, digoxin, and cyclosporin A. J Clin Invest 96:1698–1705 Schuetz EG, Umbenhauer DR, Yasuda K et al (2000) Altered expression of hepatic cytochromes P-450 in mice deficient in one or more mdr1 genes. Mol Pharmacol 57:188–197 Bardelmeijer HA, Ouwehand M, Beijnen JH, Schellens JHM, van Tellingen O (2004) Efficacy of novel P-glycoprotein inhibitors to increase the oral uptake of paclitaxel in mice. Invest New Drugs 22:219–229 Dai H, Marbach P, Lemaire M, Hayes M, Elmquish WF (2003) Distribution of STI571 to the brain is limited by P-glycoproteinmediated efflux. J Pharmacol Exp Ther 304:1085–1092 Salama NN, Kelly EJ, Bui T, Ho RJ (2005) The impact of pharmacologic and genetic knockout of P-glycoprotein on nelfinavir levels in the brain and other tissues in mice. J Pharm Sci 94:1216–1225 Dantzig AH, Shepard RL, Law KL et  al (1999) Selectivity of the multidrug resistance modulator, LY335979, for P-glycoprotein and effect on cytochrome P-450 activities. J Pharmacol Exp Ther 290:854–862 Shepard RL, Cao J, Starling JJ, Dantzig AH (2003) Modulation of P-glycoprotein but not MRP1- or BCRP-mediated drug resistance by LY335979. Int J Cancer 103:121–125 Kemper EM, Cleypool C, Boogerd W, Beijnen JH, van Tellingen O (2004) The influence of the P-glycoprotein inhibitor zosuquidar trihydrochloride (LY335979) on the brain penetration of paclitaxel in mice. Cancer Chemother Pharmacol 53:173–178 Yumoto R, Murakami T, Sanemasa M et  al (2001) Pharmacokinetic interaction of cytochrome P450 3A-related compounds with rhodamine 123, a P-glycoprotein substrate, in rats pretreated with dexamethasone. Drug Metab Dispos 29:145–151 Ward KW, Proksch JW, Levy MA, Smith BR (2001) Development of an in vivo preclinical screen model to estimate absorption and bioavailability of xenobiotics. Drug Metab Dispos 29:82–88 Ward KW, Stelman GJ, Morgan JA et  al (2004) Development of an in vivo preclinical screen model to estimate absorption and firstpass hepatic extraction of xenobiotics. II. Use

of ketoconazole to identify P-glycoprotein/ CYP3A-limited bioavailability in the moneky. Drug Metab Dispos 32:172–177 64. Kim WY, Benet LZ (2004) P-glycoprotein (P-gp/MDR1)-mediated efflux of sex-steroid hormones and modulation of P-gp expression in vitro. Pharm Res 21:1284–1293 65. Brimer C, Dalton JT, Zhu Z et  al (2000) Creation of polarized cells coexpressing CYP3A4, NADPH cytochrome P450 reductase and MDR1/P-glycoprotein. Pharm Res 17:803–810 66. Cummins CL, Jacobsen W, Benet LZ (2002) Unmasking the dynamic interplay between intestinal P-glycoprotein and CYP3A4. J Pharmacol Exp Ther 300:1036–1045 67. Kivisto KT, Niemi M, Fromm MF (2004) Functional interaction of intestinal CYP3A4 and P-glycoprotein. Fundam Clin Pharmacol 18:621–626 68. Lin JH, Chiba M, Chen IW et al (1999) Effect of dexamethasone on the intestinal first-pass metabolism of indinavir in rats: evidence of cytochrome P4503A and P-glycoprotein induction. Drug Metab Dispos 27:1187–1193 69. Cummins CL, Samphati L, Reid MJ, Benet LZ (2003) In vivo modulation of intestinal CYP3A metabolism by P-glycoprotein: studies using the rat single-pass intestinal perfusion model. J Pharmacol Exp Ther 305:306–314 70. Venkatakrishnan K, von Moltke LL, Greenblatt DJ (2000) Effects of the antifungal agents on oxidative drug metabolism: clinical relevance. Clin Pharmacokinet 38:111–180 71. Hichiya H, Takemi C, Tsuzuki D et al (2002) Complementary DNA cloning and characterization of cytochrome P450 2D29 from Japanese monkey liver. Biochem Pharmacol 64:1101–1110 72. Zhang L, Fitzloff JF, Engel LC, Cook CS (2001) Species difference in stereoselective involvement of CYP3A in the mono-N-dealkylation of disopyramide. Xenobiotica 31:73–83 73. Perloff MD, von Moltke LL, Court MH et al (2000) Midazolam and triazolam biotransformation in mouse and human liver microsomes: relative contribution on CYP3A and CYP2C isoforms. J Pharmacol Exp Ther 292:618–628 74. Bachmann KA, Lewis JD (2005) Predicting inhibitory drug–drug interactions and evaluating drug interaction reports using inhibition constants. Ann Pharmacother 39:1064–1072 75. Ito K, Iwatsubo T, Kanamitsu S et al (1998) Prediction of pharmacokinetic alterations caused by drug–drug interactions: metabolic interaction in the liver. Pharmacol Rev 50:387–412

Examination of CYP3A and P-Glycoprotein-Mediated Drug–Drug Interactions 76. Robert J, Jarry C (2003) Multidrug resistance reversal agents. J Med Chem 46:4805–4817 77. Wallstab A, Koester M, Bohme M, Keppler D (1999) Selective inhibition of MDR1 P-glycoprotein-mediated transport by the acridone carboxamide derivative GG918. Br J Cancer 79:1053–1060 78. Ward KW, Azzarano LM (2004) Preclinical pharmacokinetic properties of the P-glycoprotein inhibitor GF120918A (HCl salt of GF120918, 9, 10-dihydro-5-methoxy-9-oxoN-[4-[2-(1, 2, 3, 4-tetrahydro-6, 7-dimethoxy2-i soquinolinyl)ethyl]phenyl]-4-acridinecarboxamide) in the mouse, rat, dog, and monkey. J Pharmacol Exp Ther 310:703–709 79. Evers R, Kool M, Smith AJ et  al (2000) Inhibitory effect of the reversal agents V-104, GF120918 and Pluronic L61 on MDR1 Pgp-, MRP1- and MRP2-mediated transport. Br J Cancer 83:366–374 80. Jonker JW, Smit JW, Brinkhuis RF et al (2000) Role of breast cancer resistance protein in the bioavailability and fetal penetration of topotecan. J Natl Cancer Inst 92:1651–1656 81. Hoffmaster KA, Zamek-Gliszczynski MJ, Pollack GM, Brouwer KLR (2004) Hepatobiliary disposition of the metabolically stable opioid peptide [D-Pen2, D-Pen5]enkaphalin (DPDPE): pharmacokinetic consequences of the interplay between multiple transport systems. J Pharmacol Exp Ther 311:1203–1210 82. Dai G, Wan YJ (2005) Animal models of xenobiotic receptors. Curr Drug Metab 6:341–355 83. Robertson GR, Field J, Goodwin B et al (2003) Transgenic mouse models of human CYP3A4 gene regulation. Mol Pharmacol 64:42–50

403

84. Xie W, Yeuh MF, Radominska-Pandya A et al (2003) Control of steroid, heme, and carcinogen metabolism by nuclear pregnane X receptor and constitutive androstane receptor. Proc Natl Acad Sci USA 100:4150–4155 85. Katoh M, Matsui T, Nakajima M et al (2004) Expression of human cytochromes P450 in chimeric mice with humanized liver. Drug Metab Dispos 32:1402–1410 86. van Herwaarden AE, Wagenaar E, van der Kruijssen CM et al (2007) Knockout of cytochrome P450 3A yields new mouse models for understanding xenobiotic metabolism. J Clin Invest 117:3583–3592 87. Gonzalez FJ (2004) Cytochrome P450 humanised mice. Hum Genomics 1:300–306 88. Gonzalez FJ (2007) CYP3A4 and pregnane X receptor humanized mice. J Biochem Mol Toxicol 21:158–162 89. Gonzalez FJ, Cheung C (2008) Humanized mouse lines and their application for prediction of human drug metabolism and toxicological risk assessment. J Pharmacol Exp Ther 327:288–299 90. Cozzi J, Fraichard A, Thiam K (2008) Use of genetically modified rat models for translational medicine. Drug Discov Today 13:488–494 91. Lin JH (2008) Applications and limitations of genetically modified mouse models in drug discovery and development. Curr Drug Metab 9:419–438 92. van Waterschoot RA, van Herwaarden AE, Lagas JS et al (2008) Midazolam metabolism in cytochrome P450 3A knockout mice can be attributed to up-regulated CYP2C enzymes. Mol Pharmacol 73:1029–1036

sdfsdf

Chapter 18 Generating Inhibitors of P-Glycoprotein: Where to, Now? Emily Crowley, Christopher A. McDevitt, and Richard Callaghan Abstract The prominent role for the drug efflux pump ABCB1 (P-glycoprotein) in mediating resistance to chemotherapy was first suggested in 1976 and sparked an incredible drive to restore the efficacy of anticancer drugs. Achieving this goal seemed inevitable in 1982 when a series of calcium channel blockers were demonstrated to restore the efficacy of chemotherapy agents. A large number of other compounds have since been demonstrated to restore chemotherapeutic sensitivity in cancer cells or tissues. Where do we stand almost three decades since the first reports of ABCB1 inhibition? Unfortunately, in the aftermath of extensive fundamental and clinical research efforts the situation remains gloomy. Only a small handful of compounds have reached late stage clinical trials and none are in routine clinical usage to circumvent chemoresistance. Why has the translation process been so ineffective? One factor is the multifactorial nature of drug resistance inherent to cancer tissues; ABCB1 is not the sole factor. However, expression of ABCB1 remains a significant negative prognostic indicator and is closely associated with poor response to chemotherapy in many cancer types. The main difficulties with restoration of sensitivity to chemotherapy reside with poor properties of the ABCB1 inhibitors: (1) low selectivity to ABCB1, (2) poor potency to inhibit ABCB1, (3) inherent toxicity and/or (4) adverse pharmacokinetic interactions with anticancer drugs. Despite these difficulties, there is a clear requirement for effective inhibitors and to date the strategies for generating such compounds have involved serendipity or simple chemical syntheses. This chapter outlines more sophisticated approaches making use of bioinformatics, combinatorial chemistry and structure informed drug design. Generating a new arsenal of potent and selective ABCB1 inhibitors offers the promise of restoring the efficacy of a key weapon in cancer treatment – chemotherapy. Key words: Multidrug resistance, ABC drug efflux pump, Combinatorial chemistry, Drug design, Pharmacophore modeling, Homology modeling, High resolution structure, Pharmacokinetic interactions

1. Introduction Despite its widespread use and applicability in treating all stages of cancer, i.e. from front-line therapies to palliation, the efficacy of chemotherapy remains suboptimal. One of the main reasons for the underwhelming success of chemotherapy is the resistant J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_18, © Humana Press, a part of Springer Science + Business Media, LLC 2010

405

406

Crowley, McDevitt, and Callaghan

phenotype, which may be inherent to cancerous tissue or arise following drug administration (1–6). Resistance arises by virtue of the adaptability of cancer cells to a variable local environment (7–11). One of the types of adaptation is a network of often synergistic pathways that negate the efficacy of anticancer drugs (12, 13). These resistance pathways can range from individual factors through to the 3D tissue organization and impact on drug efficacy by alterations to drug distribution within solid tumors, affecting cellular uptake, increasing intracellular metabolism/excretion, specific mutations within target molecules, evasion of repair mechanisms, and dampening of pathways aimed at initiating cell death. Resistance mechanisms can be initiated by host factors such as high cell density, hypoxia, or stress response pathways. One of the most widespread mechanisms of resistance is the expression of efflux pumps such as ABCB1 (P-glycoprotein), ABCC1 (MRP1) and ABCG2 (BCRP) (14–16). Expression of these proteins at the plasma membrane reduces intracellular drug concentration and is therefore a first line of cellular defense. The mechanism of resistance is a generic one owing to the ability of efflux pumps to transport an extraordinary number and range of chemicals (17–19). For example, the multidrug efflux pump ABCB1 is known to interact with over 200 compounds. The broad multispecificity of these transporters is a hallmark of their origin as environmental xenobiotic protection pathways. ABCB1 is normally expressed in a number of healthy tissues, particularly those involved in secretory roles (e.g. liver and GI tract) or in a barrier capacity (e.g. blood–brain and blood–testes) (20–23). Expression of ABCB1 in these tissues is regulated by endogenous transcription factors such as the nuclear orphan receptor family (24). However, cancers arising from these tissues frequently display inherent resistance, which is present prior to chemotherapy exposure. Consequently, overexpression of ABCB1 in cancer cells following exposure to chemotherapy agents is thought to be achieved by virtue of stress response pathways rather than a classical induction process (24). Overexpression of ABCB1 has been demonstrated to generate a resistant phenotype in cultured cancer cell lines and various tumor models (13, 17). In addition, expression of ABCB1 has been cataloged in a large number of human cancer types including several leukemia types and solid tumors from the breast, colon, and adrenal tissues (25–28). A link between expression and a resistant phenotype is well established in acute myelogenous (AML), myelodysplastic syndrome (MDS), and acute lymphoblastic (ALL) leukemias (29, 30). However, the role of ABCB1, and other multidrug efflux pumps, in conferring resistance in many solid tumors continues to be vigorously debated (4, 15, 31, 32). The inability to unequivocally quantify the role of ABCB1 in clinical drug resistance has arisen for a number of reasons including:

Generating Inhibitors of P-Glycoprotein: Where to, Now?

407

(1) variability in the methods used to detect ABCB1 expression, (2) the presence of other, often synergistic resistance pathways, (3) the poor inhibition of its activity in vivo, (4) poor study design, and (5) variability of ABCB1 expression patterns in tumors. A comprehensive understanding of resistance pathways in the clinical setting is therefore required to confidently assign the relative contributions of specific resistance factors. Despite the controversy surrounding the contribution of ABCB1 to resistance in cancer, there is an apparent need to modulate its behavior. Inhibition of the efflux protein leads to increased drug accumulation in cultured cancer cells and improved intratumor distribution in animal models and patients (33–37). Consequently, the development of potent inhibitors of ABCB1 could prove highly beneficial in chemotherapeutic management of cancer. In addition, the ability of ABCB1 to influence drug pharmacokinetics in a nononcology setting renders it a target for specific modulation to regulate drug absorption, distribution, and elimination. Inhibition of ABCB1 was first reported in 1982 using the calcium channel blocker verapamil and this strategy was rapidly progressed to clinical trials (38–40). Unfortunately, the use of verapamil was doomed owing to its poor potency to inhibit ABCB1, whereas its effects on calcium channels (particularly in cardiac tissue) occurred at low plasma concentrations. Similar effects were reported with a number of other ABCB1 “inhibitors” that were already in clinical usage for various unrelated settings (41–44). These compounds inhibited ABCB1 primarily by acting as substrates that could compete for transport by the protein. Unfortunately, drugs belonging to this class of ABCB1 inhibitor were united in displaying poor potency of action, which directly resulted in unacceptable levels of systemic toxicity (45–47). Subsequent generations of ABCB1 inhibitors (Fig.  18.1) have been explored using chemical modification of the first generation agents, combinatorial chemistry to identify new chemical moieties, and, more recently, the use of natural products to uncover novel lead compounds (48–56). Despite these significant efforts, only a small selection of compounds have progressed through to late stage clinical trials; of particular note are Tariquidar (XR9576) (36, 57–59) and the nonimmunosuppresive cyclosporin A derivative, Valspodar (PSC833) (60–62). The various generations of ABCB1 inhibitors have failed to deliver a method of clinical intervention to restore sensitivity of chemotherapy for a number of reasons: 1. Poor selectivity leading to unwanted actions. 2. Low affinity for ABCB1 requiring high plasma concentrations, thereby producing toxic side-effects. 3. Interaction of drugs with other ABC transporters – for example, the perturbation of bile formation or toxicity to stem cells.

408

Crowley, McDevitt, and Callaghan

Fig. 18.1. Inhibitors of ABCB1. A flow-chart outlining the four generations of inhibitors against the multidrug efflux pump ABCB1 that have been developed over the last 30 years.

4. Interaction with cytochrome P450, thereby producing elevated systemic concentrations of anticancer drug, which necessitates dose reduction. 5. Inability to modulate ABCB1 function in vivo. 6. Decreased elimination of anticancer drugs because of ABC transporter inhibitions in “physiological sites”. The failure of clinical trials thus far has engendered a degree of pessimism regarding the ability to inhibit ABCB1 effectively in vivo, and some skepticism regarding its role in drug resistance. This does, however, seem rather premature given the small number of compounds that have been subjected to exhaustive characterization. In addition, the power of combinatorial chemistry has not been fully exploited and the newest generation of compounds

Generating Inhibitors of P-Glycoprotein: Where to, Now?

409

(from natural sources) not yet fully characterized, although there have been suggestions that inhibitor development is redundant because of the recent emergence of novel nongenotoxic anticancer agents that are directed against specific cellular targets rather than the more generic proliferation process. However, the majority of these compounds are cytostatic and many of these novel cytostatic anticancer drugs are themselves subjected to efflux by transporters such as ABCB1 (25, 63, 64). Consequently, the role of genotoxic drugs in cancer treatment in the near future should not be dismissed. Thus the problem of ABCB1 mediated transport appears to be a phenomenon that must be dealt with and the desirability of a clinically useable and efficacious inhibitor remains high in oncological circles. Providing a greater knowledge of the nature of drug–ABCB1 interaction remains an important goal for future anticancer drug development. Understanding pharmacophoric elements of ABCB1 substrates and elucidating the molecular interactions with protein structural elements in the drug binding site will prove useful to ensure new compounds can evade the influence of this protein. The focus of this review is twofold. First we shall compile the data available on defining the pharmacophoric elements of ABCB1 substrates and strategies to improve the number of compounds available for testing. The second half of the review will focus on compiling data about the molecular properties of the drug binding sites of ABCB1 and exploring the possibilities for using structural data to inform inhibitor design.

2. Drug Design for Inhibition of ABCB1

Understanding the factors that determine substrate specificity is crucial for successful drug targeting and in the rationale for the design of novel inhibitors. High resolution structural data coupled with the large volume of functional biochemical data on ABCB1 would serve as the ideal template for understanding drug–protein interaction with a view to create a design of novel inhibitors. However, the refractory nature of membrane proteins to atomic structure resolution studies has meant that high resolution data for ABCB1 has not yet been obtained. In its absence, two distinct lines of investigation have been employed to explore ABCB1–drug interactions. The first has extrapolated a molecular model of ABCB1 enabling docking studies to characterize the drug–protein interactions. This approach has been facilitated by high resolution structural templates, such as the bacterial ABC transporter Sav1866 (65), which have identical topology and a high level of sequence or structural homology. The homology modeling approach of protein-structure based drug design will

410

Crowley, McDevitt, and Callaghan

be addressed in Subheading 18.3 of this chapter. The second line of investigation is a protein-structure independent method that instead exploits knowledge of the different substrates, their physicochemical parameters or affinities for the protein to generate a model of drug interaction. 2.1. What Defines Substrate Recognition?

The first explanation for substrate recognition was proposed by Emil Fischer in 1894 who provided a structural rationale for the interaction between an enzyme and its substrate (66). The elegantly simple Lock-and-Key model postulated that the enzyme contained a rigid binding pocket that interacted with a specific ligand and allowed subsequent release of the enzymatic product. Although this model has been refined over the intervening decades, classic Lock-and-Key type ligand interactions cannot account for the broad multispecificity exhibited by tranporters such as ABCB1. Currently, there are three general models that address multispecific ligand interactions. The first, proposes that the binding pocket, although still an essentially rigid region, contains different interaction sites that allow a range of structurally distinct ligands to bind (67). The second model, based on Koshland’s Induced-Fit model (68), proposes that conformational flexibility within the protein allows the binding pocket to reconfigure and accommodate structurally diverse ligands. The third and most recent model is based on the the observation that a single ligand may bind to a protein in multiple and different orientations (69). The Differential Ligand Positioning model proposes that a single ligand might be able to interact with a number of spatially distinct regions of the binding site thus allowing multiple ligand molecules to bind simulatenously. Although the precise mechanism(s) employed by ABCB1 to interact with ligands has not been elucidated the last three decades of biochemistry have provided significant insight. The multipilicity of ABCB1 drug ligand binding sites was first shown by Tamai and Safa (70) and, to date, there are at least four known distinct drug binding sites (71–77). Biochemical studies have established that the drug binding sites show a range of different behaviors with noncompetitive inhibition for certain substrates, indicative of overlapping substrate specificities; competitive inhibition for other drug ligands, such as vinblastine and doxorubicin; and cooperative allostery between certain substrates, e.g. ATP, vinblastine and verapamil (73, 78–81). In addition there appear to be multiple binding sites on individual transmembrane segments that have the ability to simultaneously bind distinct drugs or multiple molecules of the same drug (71, 82–84). As a consequence, it is likely that the TM segments that contribute to the drugbinding pocket have a high degree of conformational mobility to allow drug molecules to form the required binding sites and to allow for diffe­rent orientations of drug molecules within the binding

Generating Inhibitors of P-Glycoprotein: Where to, Now?

411

pocket. Thus, it is likely that ABCB1 employs a combination of the general multispecific interaction models described above, thereby maximizing the range of ligands with which it can interact. However, this mobility and flexibility in substrate binding creates unique challenges for inhibitor design. 2.2. What Governs ABCB1–Drug Ligand Interactions?

Since ABCB1 was first discovered by Juliano and Ling (85), many studies have sought to clarify the basic functional and structural features of ligands that govern interaction. Identification of the basic chemical feature responsible for mediating ligand–protein interaction is key to developing a framework for interpretative and prognostic evaluations of new compounds. Structure–activity relationships (SAR) have exploited three decades of pharmacological studies on ABCB1 in an attempt to correlate substrate activity with specific molecular descriptors. However, interpreting the large volume of data collected on ABCB1 is highly complex due to a number of intrinsic issues. These stem from largely technical issues including the use of different assays, parameters reported (e.g. IC50, KD, KM & KI), drug solubilities, and variable drug partition coefficients. Despite this, some general features of ABCB1 substrates that were first noted remain relevant in that they tend to be lipophilic and amphiphilic, have a large molecular volume, contain electronegative and hydrogen bonding groups, and occasionally a weakly cationic group. More detailed molecular descriptors have since been revealed by a number of different approaches (for detailed reviews see (86–88)). Prior to the introduction of automated and semiautomated computational pharmacophoric and 3D quantitative structure activity relationships (3D-QSAR), modeling techniques SARs were determined by correlation of substrate activities with molecular descriptors. Zamora and coworkers provided one of the first SAR studies and described the requirement of a basic nitrogen atom and two planar aromatic domains based on investigations using verapamil, indole alkaloids, lysosomotrophic agents and amines (89). This feature set was further probed by Pearce and and coworkers in 1989 using a series of reserpine and yohimbine analogs that demonstrated that these domains also adopted welldefined conformations (90). However, the requirement of the basic nitrogen atom was called into question by a number of studies that used a broader array of ligands and showed that compounds, such as steroid hormones, could also interact with ABCB1 (91–93). In 1997 Bain and coworkers examined 44 compounds, mostly pesticides, and proposed that substrates and inhibitors could be differentiated on the basis of the number of rings, molecular weight, and hydrogen bonding potential (94). They suggested that transported substrates displayed higher molecular weight and hydrogen bonding potential than nontransported

412

Crowley, McDevitt, and Callaghan

substrates. In addition, the transported substrates acted primarily as hydrogen-donors rather than acceptors. A study by Seeling examined the structure of a hundred chemically diverse compounds and sought to more clearly define the number of electron donor groups and their fixed spatial distance (95). Seelig’s analysis proposed a general pattern for ABCB1 substrate recognition comprising two or three electron-donor (or hydrogen-bonding acceptor) groups with a fixed spatial separation of 2.5 ± 0.3 Å (as a type-I pattern) or 4.6 ± 0.6 Å (as a type-II pattern), respectively. Ecker and coworkers (96) subsequently followed Seelig’s work and suggested a correlation between the total electron donating strength of a ligand and its potency as an inhibitor. Ultimately, although SAR data has provided valuable insight into the molecular descriptors of known substrates and inhibitors, it has not provided a platform for the a priori development of novel ligands. SAR studies are constrained by the chemical data upon which they are constructed and, as a consequence, have a limited application for directing ligand screening beyond existing ABCB1 SAR chemical space. This is an issue of critical importance for a multispecific transporter such as ABCB1 and has driven the development of computational tools for applying substrate structure to new inhibitor design. 2.3. From Substrates to Templates – How Can We Design New Inhibitors?

Substrate based inhibitor design exploits the “learnt” rules for ligand–protein interactions and applies them in inhibitor selection and design. But what are the rules for ABCB1, which has defied a simple classification for ligand recognition elements and demonstrated a breadth of acceptable substrate types? It contains several distinct binding sites and may interact with a broad range of compounds without strict structural constraints. Various clinically used compounds were investigated for their ability to inhibit ABCB1 in  vivo and a number of potential modulators were identified. Early attempts with these compounds to block ABCB1 in cultured cell lines and in vitro assays were highly successful and led to the first phase I clinical trials in 1985 (38). However, this and many subsequent trials with first generation ABCB1 inhibitors were plagued by failure in restoring anticancer drug efficacy. The clinical failure of these inhibitors led to the first SAR studies and provided the first insight into the molecular features crucial for interaction with ABCB1. Zamora and coworkers (89) provided the first SAR derived descriptors, however, these were not sufficiently stringent to be applied to drug development. Although they had failed clinically, the first generation ABCB1 inhibitors were effective in  vitro, and thus they were used as the templates for the second generation of inhibitors designed through quantitative structure relationships (QSAR) studies.

Generating Inhibitors of P-Glycoprotein: Where to, Now? 2.3.1. QSAR in Inhibitor Design

413

QSAR is based on the pharmacological principle that drug structure does not necessarily correlate with biological activity (97). QSAR studies examine a range of related compounds for their quantitative effects on a specific target (i.e. degree of agonism or antagonism). 3D-QSAR modeling determines a mathematical model that describes drug potency as a function of the three dimensional interactions with protein based on an aligned training set of compounds. The relationship between the change in the 3D spatial interaction fields and experimentally determined variations in the target feature is calculated by statistical analysis. A number of 3D-QSAR approaches are available and include comparative molecular field analysis (CoMFA) (98), comparative molecular similarity index analysis (CoMSIA) (99), and GOLPE (100) (for detailed discussion and reviews of these techniques see (101, 102)). Quantitative models such as 3D-QSAR can be applied to de novo computational screening to lead the synthesis of higher potency lead compounds. In combination with in vitro testing and analysis, for refinement of the quantitative model, this technique has been used in the design of improved inhibitors and higher affinity ligands (103). Activity predictions by 3D-QSAR models require that the ligands be accurately aligned and, consequently, this limits their application in automated chemical compound database screening. Although 2D-QSAR models can be used for database screening, they lack highly useful 3D information crucial for subsequent drug design. QSAR studies led to modified versions of several first generation lead compounds including indoles such as reserpine (104) and 1,4 dihydropyridines (53), phenothiazine derivatives such as transflupentixal (49), a nonimmunosuppresive cyclosporin A derivative PSC833 (56), and the verapmil derived, triazine-based S9788 (105). Detailed in  vitro assays provided information on the affinity of interaction with the drug and ABCB1 compared to first generation compounds and was used to further refine and optimize the design of second generation ABCB1 inhibitors. Phase I and II clinical trials were undertaken with the most promising second generation ABCB1 inhibitors. However, unfavorable pharmacokinetic interactions led to elevated drug plasma levels and reduced systemic clearance of anticancer drugs, producing significant toxicity in patients (61, 106–108). This necessitated a reduction in the administered dose of chemotherapeutic drugs, which in turn reduced the overall efficacy of anticancer drug treatment. Concomitant inhibition of ABCB1 and cytochrome P450-3A isoform (CYP3A), which is responsible for the metabolism of almost 50% of all clinically employed drugs by the second generation inhibitors, resulted in higher and prolonged plasma levels of anticancer drugs because of impaired metabolism and elimination. It was subsequently determined that ABCB1 and CYP3A have a considerable overlap in substrate specificities

414

Crowley, McDevitt, and Callaghan

and this highlighted one of the limitations of drug design methodologies, namely, the inability to predict undesirable pharmacokinetic interactions (109, 110). Consequently, although 3D-QSAR strategies led to an improvement of first generation compounds, the second generation compounds ultimately failed to provide an ideal route to ABCB1 inhibition without compromising anticancer drug efficacy. Thus, it was necessary to approach inhibitor development from a broader chemical space approach such as pharmacophore modeling. 2.3.2. What Is Pharmacophore Modeling?

The pharmacophore concept was first introduced by Paul Erhlich in the early 1900s (111). The pharmacophore is a description of the molecular framework which contains the essential features responsible for a drug’s biological activity. With the benefit of nearly a century’s additional knowledge, the underlying concept has been expanded to include our understanding of three dimensional substrate structures and the arrangement of their essential molecular features. In current terminology, the pharmacophore is a representation of the spatial arrangement of structural features required for biological activity. The determination of a pharmacophore requires knowledge of (1) the three dimensional structure and bioactive conformation of molecules, (2) key atomic features, and (3) the determination of the relationship between those features and biological activity. Once developed, pharmacophoric models can be highly valuable tools to provide insight into drug molecule interactions and aid in the design of higher potency inhibitors. Seelig’s SAR data (95) was suggestive of a fairly simple pharmacophoric distinction between substrates and inhibitors of ABCB1, whereas the spatial requirements would be indicative of discrete binding sites. SAR data can be analyzed and interpreted for small numbers of compounds (<500), but as datasets grow in size and complexity computational approaches are better suited to generating pharmacophore models. There are a range of programs that are widely used for pharmacophore generation including ALADDIN, COMPASS, SCAMPI, PARM, and DANTE (112); however, the most commonly used are DISCO (113), GASP (114), and Catalyst/HIPHOP (115). These software packages utilize different algorithms to determine a common set of molecular features on the basis of comparisons of interacting compounds (substrates or inhibitors). A consequence of this is that most of the pharmacophore models generated are based on the alignment of a small number (i.e. the training set) of energy minimized conformations of known substrates. However, this means that the dynamic nature of biologically active substrates cannot be fully predicted. One program, Catalyst/HypoGen, employs a combination of QSAR and pharmacophore methods (116). This requires a wide range of interacting and noninteracting

Generating Inhibitors of P-Glycoprotein: Where to, Now?

415

compounds combined with experimentally determined activity data, which provides a more robust pharmacophore. The Catalyst/ HypoGen pharmacophore is capable of predicting the potential capacity for a query compound to interact, as in a traditional pharmacophore model. However, it can also estimate its potential activity based on a regression of the training dataset, as with the 3D-QSAR model. Many pharmacophore models reported in current literature claim to be reasonably accurate at predicting ABCB1 substrates. Ekins and coworkers generated pharmacophore-QSAR models to rank ABCB1 inhibitors on the basis of modulating substrate transport (117). A single substrate pharmacophore was produced by overlaying verapamil and digoxin based structures, followed by fitting vinblastine, and the generated pharmacophore revealed multiple hydrophobic and hydrogenbond acceptor features as important characteristics of ABCB1 substrates. An ensemble model of 100 pharmacophores was generated by Penzotti and coworkers and consisted of a set of 2, 3, and 4-point pharmacophores for discrimination between interacting and noninteracting ABCB1 compounds with potential ABCB1 ligands required to match at least 20% of the pharmacophores in the ensemble (118). Screening of ligands, also referred to as virtual screening, is a data mining approach that applies the pharmacophoric model to screen commercial chemical compound databases to identify molecules that can potentially interact with the protein. Potential compounds can then be purchased and directly tested using in vitro assays. Consequently, this approach has become a frequently used strategy for the identification of novel lead ligands. Several ABCB1 pharmacophores have been used in screening databases. Rebizter and coworkers used a propafenone based pharmacophore model to screen the Derwent World Drug Index (119). This identified 19 new potential ABCB1 substrates but the study did not report subsequent experimental verification (120). More recently, a pharmacophore model generated from 131 propafenone ABCB1 inhibitors was used to screen the SPECS database (134,000 compounds) and successfully identified two lead compounds with submicromolar range affinities (121). Despite these promising leads, none of these compounds have yet made the transition from the laboratory to the clinic. 2.3.3. Limitations of In Silico Approaches to Inhibitor Design

Regardless of the type of model used for drug/inhibitor design, the predictive and interpretative qualities are ultimately constrained by the dataset upon which they were constructed. Acquiring robust datasets is especially important in these studies, where a variety of expression systems and experimental models are available. The majority of QSAR and pharmacophore studies on ABCB1 have focused on datasets gathered from a single species or cell type, and frequently from a single laboratory (122– 125). The promiscuity of transport exhibited by ABCB1 means

416

Crowley, McDevitt, and Callaghan

that there are relatively few noninteracting compounds included in the development of pharmacophoric datasets. As a consequence, this has meant that the models generated have facilitated understanding of ABCB1–ligand interactions; they have not been highly effective in prospective ligand discoveries. Despite this, the information gained from these in silico studies have improved the consensus picture for ABCB1 inhibitor design, although it still remains somewhat broad. Strong inhibitors are characterized by high lipophilicity (and/or molar refractivity) and possess at least two H-bond acceptors. Other features, such as H-bond donors and p–p-stacking, are also proposed to serve as additional interaction features. Pharmacophoric models indicate that there are also steric constraints for interaction (120, 126–130). 2.3.4. Combinatorial Chemistry

Combinatorial chemistry was responsible for the a priori development of the third generation of ABCB1 inhibitors generated with the objective of improving potency without unwanted pharmacokinetic interactions. Four promising lead compounds (Elacridar (50), Zosuquidar (131), Tariquidar (54), and Ontogen (52)) were developed by high throughput screening approaches using SAR analyzes. Their nanomolar potency and efficacy in experimental systems (in vitro and in vivo) led to a rapid progression to clinical trials. Tariquidar garnered the greatest interest because of its high potency (100–1,000-fold greater than the previous generation inhibitors), its discrimination between ABCB1 and ABCC1, and its long effective duration (35, 132, 133). However, despite its success in phase I and II clinical trials, phase III trials were suspended due to unfavorable toxicity reports in the treatments of lung carcinoma and the future of this inhibitor is currently unclear. Clinical trials for other third generation ABCB1 inhibitors are proceeding and although the initial reports are promising, with minimal adverse pharmacokinetic interactions reported, these trials are still at relatively early stages with small patient sample sizes and no unambiguous reports on improvements in anticancer drug efficacy (134–137). Although the pharmaceutical armory is small, there remains a paucity of extensive inhibitor characterization in the clinical setting. More attention should be devoted to trials with large patient populations and a broad range of cancer types, with detailed information on the class of resistance and greater use of surrogate assays.

2.3.5. Nucleotide Binding Domain Targeted Inhibition

Drug binding sites within the transmembrane domain of ABCB1 have been the main target of substrate based inhibitor design. However the inherent plasticity of these sites has rendered it difficult to identify compounds that can modulate drug efflux by this route. However, targeting the drug binding site need not be the only strategy to attain pharmacological inhibition of ABCB1.

Generating Inhibitors of P-Glycoprotein: Where to, Now?

417

The high conservation of the nucleotide binding domains (NBDs) amongst ABC transporters and their fundamental requirement to provide the mechanistic driving force for efflux indicates that they are ideal targets for the inhibition of ABCB1. This is underpinned by a wealth of biochemical and structural data which has meant that our understanding of the catalytic cycle of the NBDs is well understood (for review see (138–140)). For example, the distinct and specific motifs contained within the NBDs are amenable to in silico ligand design techniques. In addition, the presence of two NBDs per transporter also increases the number of potential sites for inhibitor binding, whereas inactivation of only a single NBD would be required to impair ABCB1 mediated drug transport. Several classes of drug, such as the flavonoids, have been observed to interact with the NBDs (141–143). Flavonoids are a large class of naturally occurring compounds widely present in the green plant world with more than 6,500 different compounds described (143, 144). Some naturally occurring flavonoids and their hydrophobic derivatives (e.g. aurones) have been observed to inhibit the transport function of ABCB1 by interaction with NBD 2 and the cytosolic regions of the protein. It has been shown that although some flavonoids can inhibit the labeling of ABCB1 with their photoactive substrates (145–147), indicating that they may bind directly to the substrate binding site, others bind directly to the purified recombinant C-terminal nucleotide-binding domain from mouse ABCB1 (NBD2). Moreover, it appears that the binding domain may overlap the ATP binding site and vicinal steroid binding site (142). However, flavonoids are also potent inhibitors of drug metabolizing enzymes (148, 149) and pharmacokinetic interactions with anticancer drugs are likely to prevent a clinical application in their current form. 2.4. ABCB1 Inhibitors – Where to Now?

Systematic chemical modification and combinatorial chemistry produced the first three generations of potent ABCB1 inhibitors. Unfortunately, the majority of these inhibitors have also been reported to have caused undesirable pharmacokinetic anticancer drug interactions thus limiting their clinical application. Rapid technological advancements have made automated and semiautomated in silico approaches, such as pharmacophore and QSAR modelling, feasible for screening vast compound databases and developing higher potency inhibitors. However, despite the identification of a number of potential lead compounds these approaches have not yet led to the production of compounds for clinical trials. Because of these obstacles to ABCB1 inhibition some recent studies have returned to screening herbal and fruit extracts for lead compounds. These approaches are reminiscent of the first generation inhibitor screening methods of broadly sampling the existing chemical space in an attempt to identify lead compounds.

418

Crowley, McDevitt, and Callaghan

A more rational approach may be the utilization of protein structure modelling exploiting high resolution homologous template structures. By developing models of protein interaction with potential ligands, in combination with complementary physicochemical (QSAR) and pharmacophoric descriptors this may offer a different platform for the design of potential lead inhibitors.

3. Structure Informed Drug Design 3.1. Has Structure Informed Drug Design Been Employed Successfully?

High resolution crystal structures have been used for rational drug design as they can provide detailed molecular information on interactions between the substrate and protein at the active site. This method was used to develop the well known antiinfluenza drugs Relenza and Tamiflu as well as the anticancer drug Imatinib. All three compounds function by targeting the active site of a critical protein involved in illness progression and are the success stories of structure-informed drug design. The antiinfluenza drugs Relenza and Tamiflu target the enzyme neuraminidase, which is responsible for viral release from sites of infection such as the lungs (150). The enzyme cleaves sialic acid residues on a surface receptor involved in anchoring newly formed influenza viral particles, thereby facilitating virus release from infected cells. Several high resolution crystal structures of neuraminidase have been solved, but of particular significance were those structures containing bound sialic acid substrate (151–153). This provided critical insight into specific residues that interact with sialic acid at the enzyme active site. A combination of computational chemistry and examination of the crystal structures of neuraminidase revealed that the C-4 hydroxyl group of sialic acid provided a significant contact point between the substrate and the protein (154). Substitution of the C-4 hydroxyl group with a larger, basic guanidinyl group gave rise to 4-deoxy4-guanidino-Neu5Ac2en (commonly referred to as Zanamivir or Relenza), which displayed antiviral activity (155). Moreover, structural (156) and molecular modelling (157) data of the complex demonstrated that Zanamivir bound directly at the active site of neuraminidase. Subsequent clinical trials demonstrated that Zanamivir is clinically effective for the treatment and prevention of influenza (158, 159). The development and success of Zanamivir provided a platform from which to produce other neuraminidase-targeted antiinfluenza drugs. Further structure–activity relationship studies with Neu5Ac-based derivatives led to a number of improvements to the compound (160, 161). These included mimicking the

Generating Inhibitors of P-Glycoprotein: Where to, Now?

419

sialic acid transition state, optimizing the hydrophobic nature of the drug thereby increasing membrane association, and, finally, the development of a prodrug form, converted to an active form in vivo. The final outcome was the antiinfluenza drug Oseltamivir or Tamiflu. Structure-based drug design has therefore been successfully applied to therapy against influenza by virtue of the high resolution crystal structures of neuraminidase solved in complex with the substrate and the inhibitor. Furthermore, identification of neuraminidase as a target for structure-informed drug design was possible owing to a thorough understanding of both the viral lifecycle and the mechanism of substrate recognition by the protein. 3.2. Can the Approach Be Used for ABCB1?

The success of structure-based drug design relies upon having (1) high resolution protein structures (2) detailed information on the substrate binding site(s), and (3) an understanding of the substrate–protein interactions. These three requirements are essential for the successful development of inhibitors and our progress towards these goals for ABCB1 will be discussed.

3.2.1. Structural Information on ABCB1

A myriad of challenges face structural biologists when attempting to crystallize membrane proteins. These include protein expression, efficient protein extraction from the lipid bilayer, high purity, and sample homogeneity. Such issues are responsible for the lack of crystal structures for any full-length eukaryotic ABC tranporters. In the interim however, electron microscopy (EM) and homology modelling have been used to provide low to medium resolution structural information of ABCB1. Electron microscopy of ABCB1 in the presence and absence of nucleotides revealed that the protein undergoes a significant reorganization in the transmembrane domains (162–167). This was interpreted as corresponding to the opening of a central pore (167), thought to allow hydrophobic drug access to the extracellular environment during the drug translocation process of ABCB1. Currently the highest resolution structure of ABCB1, determined using 2D crystals and cryo-electron microscopy, is approximately 8 Å, which is too low for structure informed drug design (164–167). However, the use of EM to monitor multiple conformational states has been of great benefit in understanding the dynamic aspects of the drug translocation process and this structural information still plays a crucial role in the validation of homology models.

3.2.2. Homology Models of ABCB1

To date EM has provided the only direct structural information on ABCB1. In constrast several high resolution crystal structures of prokaryotic ABC transporters have been solved in recent years (65, 168–172). These structures in conjuction with the parameters obtained from EM studies have been avidly used to produce

420

Crowley, McDevitt, and Callaghan

homology models of ABCB1 (173–177). Homology models provide an invaluable starting point for the interpretation of biochemical data in the absence of high resolution structures (130, 173). The quality and accuracy of a model is entirely dependent on the resolution of the template crystal structure and the template-sample sequence alignment. As a consequence, protein homology models are viewed as approximations of protein structures, comparable to medium resolution images and not suitable for use in structure-informed drug design. Furthermore, the reliability of ABCB1 homology models was recently confounded by the withdrawal of three MsbA structures. In addition, the only available structure for an ABC multidrug efflux pump, namely Sav1866, revealed an unexpected domain organization (65). This called for the refinement and reinterpretation of all previous ABCB1 models (65, 177–179). Despite this, the quality of subsequent models has been improved owing to the higher resolution of the Sav1866 structure (~3 Å) and better sequence alignment with ABCB1 (56% and 52.8% for transmembrane domains 1 and 2 of Sav1866 and ABCB1, respectively) (175). Furthermore, crosslinking data suggest that the unusual “domain swapping” architecture of Sav1866 is also adopted by ABCB1 (180). The crystal structure of Sav1866, therefore, appears to provide a more reliable template for the computational modelling of ABCB1. Two groups have currently produced homology models of ABCB1 using the Sav1866 crystal structure as a template (173, 175). Both homology models were qualitatively compared to the ABCB1 EM model and were shown to be in reasonable correspondence (165). O’Mara and coworkers suggested that in the ATP bound state, the translocation pore is lined with polar residues. In contrast, the nucleotide free configuration has hydrophilic residues shifted to the interhelical regions with the hydrophobic residues being exposed to the translocation pore (175). This considerable molecular rearrangement is in agreement with substantial biochemical data demonstrating dramatic conformational changes in the TMDs caused by ATP binding (for review (18)). A large hydrophobic cavity within the transmembrane region of the protein was observed in the homology model created by Globisch and coworkers (173). Their study incorporated data from numerous investigations on cross-linking within ABCB1 TM segments to validate the model. According to the cross-linking data the transition from the “open to inside” to “closed to inside” conformation is thought to occur during the early stages of the ATP hydrolytic cycle. In the latter state the main protein cavity, as determined by SiteID and SiteFinder programmes, has mainly hydrophobic to neutral surface properties. It is tempting to postulate that based on both models ABCB1 assumes a conformation corresponding to an intermediate stage

Generating Inhibitors of P-Glycoprotein: Where to, Now?

421

in the translocation process (driven by events at the NBDs) wherein the central pore develops a low/intermediate affinity for hydrophobic drugs. Such conformational changes in ABCB1 could generate the changes in substrate affinity during drug translocation (80). However, the specific conformational transitions have yet to be shown experimentally. Therefore, whereas homology models have thus far been unsuitable for structure-based inhibitor design, they have provided a means for interpreting conformational changes within the translocation pore of the protein. Such information can help in the quest to understand how ABCB1 can recognize and transport its vast array of substrates. 3.3. Substrate and Modulator Binding Sites: Biochemical Approaches

Although the homology models reveal topology and conformational shifts during translocation, they do not reveal molecular detail or location of the drug binding sites. Understanding the interaction between the protein and its substrate is crucial for structure-informed inhibitor design. The drug binding sites lie within the TM region of ABCB1 and multiple residues within the TM helices appear to contribute to the drug binding site(s) (181– 183). This section outlines our current understanding of the underlying mechanism for drug recognition by ABCB1.

3.3.1. Location and Number of Drug Binding Sites

Locating the drug binding site(s) in ABCB1 is complicated by its promiscuity and the complexity of the drug–protein interactions. For example, (1) how many sites are there, (2) are they located at distinct regions or within a single large domain, and (3) can drugs bind to more than one site? One technique that has proven invaluable in the quest to uncover the drug binding sites is cysteinescanning mutagenesis. This technique requires a cysteine-less protein template, and fortuitously, ABCB1 is fully functional in the absence of cysteine (184). The substitution of residues at specific positions in ABCB1 with cysteine was used to determine the residues that are critical for protein function and drug binding. Using an array of ABCB1 single-cysteine mutants and thiolreactive substrates, (e.g. MTS-verapamil and dibromobimane), Loo and Clarke outlined a potential drug binding domain within ABCB1 (83, 185–190). The investigations suggested that TM helices 4–6 of TMD1 and 9–12 of TMD2 contributed residues to a drug binding pocket in ABCB1 (Fig. 18.2). This was in agreement with earlier findings, using photoaffinity labeling, which indicated that the N- and C-terminal ends of the TMDs are involved in drug binding (82, 181, 183). In addition, certain residues (e.g. residue Ser222) are within the binding site of more than one unrelated substrate, suggesting a degree of redundancy between regions of interaction. These combined studies provided the field with valuable information about the drug binding regions and the substrate-induced conformational changes. However, the highly reactive nature of the thiol-labeling compounds used, the

422

Crowley, McDevitt, and Callaghan

Fig. 18.2. Potential drug binding sites in ABCB1. A simplified view of the transmembrane helices that are suggested to contribute to the drug binding site(s) of ABCB1. The black circle represents the drug substrate. (a) Represents the helices demonstrated by Loo et al. to form the drug binding site. (b) Illustrates the helices demonstrated by Ecker et al. to be involved in drug binding. The dashed lines demonstrate the interface at which the drug is believed to interact.

complexities of the substrate–ABCB1 interaction, and the conformational flexibility of the protein make it difficult to distinguish between binding to the “true” sites from binding at intermediate stages of the translocation pathway. An alternative method used to identify the drug binding sites of ABCB1 involved propafenone-analogs and matrix-assisted laser desorption-ionization-time-of-flight (MALDI-TOF) mass spectrometry. This technique identified residues in transmembrane helices 3, 5, 6, 8, 10, 11 and 12 as contributing to a putative drug-binding domain for propafenone analogs although the primary sites appear to be formed at the interfaces between TM5-8 and TM3-11 (Fig. 18.2) (84, 191). The binding regions, although distinct for these compounds, also encompass residues proposed to be involved in the binding of other ligands such as vinblastine, cyclosporin, verapamil, and colchicine (192). Predictive methods have also been used to locate the substrate binding regions in ABCB1. Globisch and coworkers used 3D structural information along with SiteID and Site Finder programs in an attempt to pinpoint binding regions and pockets (173). The programs located three binding regions and a central binding cavity. A number of residues in the vicinity of the predicted binding regions have previously been identified to contribute to the binding domain of propafenone and its chemical derivatives. Unfortunately, the binding sites identified are large, dispersed regions rather than distinct “sites” within ABCB1, thereby precluding any realistic attempts at drug-docking. The substrate-induced fit model, in which substrate binding induces unique conformational changes in the flexible TMDs resulting in the formation of a unique drug binding site for that

Generating Inhibitors of P-Glycoprotein: Where to, Now?

423

substrate, has been used to rationalize the large number of residues involved (189). Furthermore, an alternative substrate would induce a different conformational change creating a binding pocket specific for that substrate but within the same binding region. The model was supported by the fact that the TMs are quite flexible at 37 °C (from cross-linking data) and it was suggested that substrate binding reduces the flexibility of the TMs, therefore ensuring localisation of specific residues to the binding pocket(s) (193, 194). A challenge for the research field will be to verify, or refute, this intriguing potential mechanism. 3.3.2. Modulator Sites – Are They Distinct from the Drug Binding Site(s)?

The previous experiments suggested that there is one generic site in ABCB1 that accommodates many different compounds. However, a series of experiments, carried out by Dey and coworkers, demonstrated that the ABCB1 inhibitor cis-(Z)-flupentixol binds to a site distinct from the substrate binding domain, thereby preventing translocation and promoting substrate dissociation (195). Furthermore, conformational changes generated by flupentixol binding to ABCB1, as demonstrated by altered susceptibility to proteolytic digestion and UIC2 antibody binding, are distinct from that induced by ABCB1 substrates or competitive modulators (195–197). cis-(Z)-flupentixol does not interact with ABCB1 substrates in a classical competitive manner. Given that it binds at a distinct site, the interaction with the substrates is defined as an allosteric one. Earlier radioligand binding studies also identified a site that bound nontransported modulators of ABCB1 (Nicardipine and GF120918), which was distinct from the [3H]-vinblastine interaction site (73). Similarly, studies on Hoechst33342 transport also identified a potential modulator specific site that recognizes prazosin and progesterone (198). The discovery of a potentially less promiscuous, allosteric modulator site(s) in ABCB1 may provide an alternative and possibly less complicated avenue along which structure-based inhibitor design may proceed.

4. Conclusion Currently, there remains a certain degree of pessimism as to whether the activity of ABCB1 can be modulated pharmacologically to restore the efficacy of chemotherapy. The pessimism has intensified following the failure of the much vaunted Tariquidar in clinical trials. However, this was one of the very few molecules that have reached advanced clinical trials. The prominent role of this transporter in healthy tissue (pharmacokinetic regulator) and in disease (e.g. resistance in cancer and epilepsy) surely warrants greater efforts to produce novel inhibitors.

424

Crowley, McDevitt, and Callaghan

The power of chemistry and bioinformatics appear to have met their match with the promiscuity exhibited by ABCB1 with respect to substrate binding. Identifying the pharmacophoric elements of substrates may prove untenable if the drug binding domain has a seemingly limitless plasticity or malleability. Generating detailed mechanistic information and protein elements of the binding domain are a priority for future rational inhibitor design. Clearly, we cannot yet pinpoint the site of the drug or modulator binding to ABCB1 with any surety. Similarly, the precise molecular mechanism by which the protein can recognize such a wide array of compounds remains elusive. Biochemical data has brought us to the cusp of this “holy grail” of information. Provision of a high resolution structure containing a bound substrate/modulator would provide an enormous stimulus to reveal the hidden secrets of drug binding to ABCB1 and facilitate structure-informed inhibitor design. References 1. Boyer J, Allen WL, McLean EG et al (2006) Pharmacogenomic identification of novel determinants of response to chemotherapy in colon cancer. Cancer Res 66:2765–2777 2. Cunningham L, Aplenc R (2007) Pharma­ cogenetics of acute lymphoblastic leukemia treatment response. Expert Opin Pharma­ cother 8:2519–2531 3. Ferraldeschi R, Baka S, Jyoti B et al (2007) Modern management of small-cell lung cancer. Drugs 67:2135–2152 4. Gonzalez-Angulo AM, Morales-Vasquez F, Hortobagyi GN (2007) Overview of resistance to systemic therapy in patients with breast cancer. Adv Exp Med Biol 608:1–22 5. Longley DB, Allen WL, Johnston PG (2006) Drug resistance, predictive markers and pharmacogenomics in colorectal cancer. Biochim Biophys Acta 1766:184–196 6. Surowiak P (2006) Prediction of the response to chemotherapy in ovarian cancers. Folia Morphol (Warsz) 65:285–294 7. Calzada MJ, del Peso L (2007) Hypoxiainducible factors and cancer. Clin Transl Oncol 9:278–289 8. Dang CV, Kim JW, Gao P, Yustein J (2008) The interplay between MYC and HIF in cancer. Nat Rev Cancer 8:51–56 9. De Luca A, Carotenuto A, Rachiglio A et al (2008) The role of the EGFR signaling in tumor microenvironment. J Cell Physiol 214:559–567 10. Fukumura D, Jain RK (2007) Tumor microvasculature and microenvironment: targets

for anti-angiogenesis and normalization. Microvasc Res 74:72–84 11. Tlsty TD, Coussens LM (2006) Tumor stroma and regulation of cancer development. Annu Rev Pathol 1:119–150 12. Luqmani YA (2005) Mechanisms of drug resistance in cancer chemotherapy. Med Princ Pract 14(Suppl 1):35–48 13. Mellor HR, Callaghan R (2008) Resistance to chemotherapy in cancer: a complex and integrated cellular response. Pharmacology 81:275–300 14. Doyle LA, Ross DD (2003) Multidrug resistance mediated by the breast cancer resistance protein BCRP (ABCG2). Oncogene 22: 7340–7358 15. Leonard GD, Fojo T, Bates SE (2003) The role of ABC transporters in clinical practice. Oncologist 8:411–424 16. Modok S, Mellor HR, Callaghan R (2006) Modulation of multidrug resistance efflux pump activity to overcome chemoresistance in cancer. Curr Opin Pharmacol 6:350–354 17. Ambudkar SV, Dey S, Hrycyna CA et  al (1999) Biochemical, cellular, and pharmacological aspects of the multidrug transporter. Annu Rev Pharmacol Toxicol 39:361–398 18. Callaghan R, Ford RC, Kerr ID (2006) The translocation mechanism of P-glycoprotein. FEBS Lett 580:1056–1063 19. Pauwels EK, Erba P, Mariani G, Gomes CM (2007) Multidrug resistance in cancer: its mechanism and its modulation. Drug News Perspect. 20:371–377

Generating Inhibitors of P-Glycoprotein: Where to, Now? 20. Cordon-Cardo C, O’Brien JP, Boccia J et  al (1990) Expression of the multidrug resistance gene product (P-glycoprotein) in human normal and tumor tissues. J Histochem Cytochem 38:1277–1287 21. Cordon-Cardo C, O’Brien JP, Casals D et  al (1989) Multidrug-resistance gene (P-glycoprotein) is expressed by endothelial cells at blood-brain barrier sites. Proc Natl Acad Sci USA 86:695–698 22. Fromm MF (2003) Importance of P-glyco­ protein for drug disposition in humans. Eur J Clin Invest 33(Suppl 2):6–9 23. van der Valk P, van Kalken CK, Ketelaars H et al (1990) Distribution of multi-drug resistance-associated P-glycoprotein in normal and neoplastic human tissues. Analysis with 3 monoclonal antibodies recognizing different epitopes of the P-glycoprotein molecule. Ann Oncol 1:56–64 24. Callaghan R, Crowley E, Potter S, Kerr ID (2008) P-glycoprotein: so many ways to turn it on. J Clin Pharmacol 48:365–378 25. Burger H, van Tol H, Boersma AW et  al (2004) Imatinib mesylate (STI571) is a substrate for the breast cancer resistance protein (BCRP)/ABCG2 drug pump. Blood 104: 2940–2942 26. Chan HS, Grogan TM, Haddad G, DeBoer G, Ling V (1997) P-glycoprotein expression: critical determinant in the response to osteosarcoma chemotherapy. J Natl Cancer Inst 89:1706–1715 27. Chan HS, Haddad G, Thorner PS et  al (1991) P-glycoprotein expression as a predictor of the outcome of therapy for neuroblastoma. N Engl J Med 325:1608–1614 28. van den Heuvel-Eibrink MM, Sonneveld P, Pieters R (2000) The prognostic significance of membrane transport-associated multidrug resistance (MDR) proteins in leukemia. Int J Clin Pharmacol Ther 38:94–110 29. Pallis M, Russell N (2004) Strategies for overcoming p-glycoprotein-mediated drug resistance in acute myeloblastic leukaemia. Leukemia 18:1927–1930 30. van der Holt B, Lowenberg B, Burnett AK et al (2005) The value of the MDR1 reversal agent PSC-833 in addition to daunorubicin and cytarabine in the treatment of elderly patients with previously untreated acute myeloid leukemia (AML), in relation to MDR1 status at diagnosis. Blood 106:2646–2654 31. Donnenberg VS, Donnenberg AD (2005) Multiple drug resistance in cancer revisited: the cancer stem cell hypothesis. J Clin Pharmacol 45:872–877

425

32. Merino V, Jimenez-Torres NV, MerinoSanjuan M (2004) Relevance of multidrug resistance proteins on the clinical efficacy of cancer therapy. Curr Drug Deliv 1:203–212 33. Ak I, Aslan V, Vardareli E, Gulbas Z (2003) Assessment of the P-glycoprotein expression by 99mTc-MIBI bone marrow imaging in patients with untreated leukaemia. Nucl Med Commun 24:397–402 34. Jekerle V, Wang JH, Scollard DA et al (2006) 99mTc-Sestamibi, a sensitive probe for in vivo imaging of P-glycoprotein inhibition by modulators and mdr1 antisense oligodeoxynucleotides. Mol Imaging Biol 8:333–339 35. Martin C, Berridge G, Mistry P et al (1999) The molecular interaction of the high affinity reversal agent XR9576 with P-glycoprotein. Br J Pharmacol 128:403–411 36. Pusztai L, Wagner P, Ibrahim N et al (2005) Phase II study of tariquidar, a selective P-glycoprotein inhibitor, in patients with chemotherapy-resistant, advanced breast carcinoma. Cancer 104:682–691 37. Walker J, Martin C, Callaghan R (2004) Inhibition of P-glycoprotein function by XR9576 in a solid tumour model can restore anticancer drug efficacy. Eur J Cancer 40:594–605 38. Benson AB III, Trump DL, Koeller JM et al (1985) Phase I study of vinblastine and verapamil given by concurrent iv infusion. Cancer Treat Rep 69:795–799 39. Tsuruo T, Iida H, Tsukagoshi S, Sakurai Y (1981) Overcoming of vincristine resistance in P388 leukemia in  vivo and in  vitro through enhanced cytotoxicity of vincristine and vinblastine by verapamil. Cancer Res 41:1967–1972 40. Tsuruo T, Iida H, Yamashiro M, Tsukagoshi S, Sakurai Y (1982) Enhancement of vincristineand adriamycin-induced cytotoxicity by verapamil in P388 leukemia and its sublines resistant to vincristine and adriamycin. Biochem Pharm­ acol 31:3138–3140 41. Ganapathi R, Grabowski D (1983) Enhancement of sensitivity to adriamycin in resistant P388 leukemia by the calmodulin inhibitor trifluoperazine. Cancer Res 43:3696–3699 42. Goldberg H, Ling V, Wong PY, Skorecki K (1988) Reduced cyclosporin accumulation in multidrug-resistant cells. Biochem Biophys Res Commun 152:552–558 43. Ramu A, Fuks Z, Gatt S, Glaubiger D (1984) Reversal of acquired resistance to doxorubicin in P388 murine leukemia cells by perhexiline maleate. Cancer Res 44:144–148 44. Tsuruo T, Iida H, Kitatani Y et  al (1984) Effects of quinidine and related compounds

426

Crowley, McDevitt, and Callaghan

on cytotoxicity and cellular accumulation of vincristine and adriamycin in drug-resistant tumor cells. Cancer Res 44:4303–4307 45. Bartlett NL, Lum BL, Fisher GA et al (1994) Phase I trial of doxorubicin with cyclosporine as a modulator of multidrug resistance. J Clin Oncol 12:835–842 46. Berman E, McBride M, Lin S, MenedezBotet C, Tong W (1995) Phase I trial of high-dose tamoxifen as a modulator of drug resistance in combination with daunorubicin in patients with relapsed or refractory acute leukemia. Leukemia 9:1631–1637 47. Verweij J, Herweijer H, Oosterom R et  al (1991) A phase II study of epidoxorubicin in colorectal cancer and the use of cyclosporinA in an attempt to reverse multidrug resistance. Br J Cancer 64:361–364 48. Deferme S, Van Gelder J, Augustijns P (2002) Inhibitory effect of fruit extracts on P-glycoprotein-related efflux carriers: an invitro screening. J Pharm Pharmacol 54:1213–1219 49. Ford JM, Prozialeck WC, Hait WN (1989) Structural features determining activity of phenothiazines and related drugs for inhibition of cell growth and reversal of multidrug resistance. Mol Pharmacol 35:105–115 50. Hyafil F, Vergely C, Du Vignaud P, GrandPerret T (1993) In vitro and in vivo reversal of multidrug resistance by GF120918, an acridonecarboxamide derivative. Cancer Res 53:4595–4602 51. Kane GC, Lipsky JJ (2000) Drug-grapefruit juice interactions. Mayo Clin Proc 75:933–942 52. Newman MJ, Rodarte JC, Benbatoul KD et  al (2000) Discovery and characterization of OC144–093, a novel inhibitor of P-glyco­ protein-mediated multidrug resistance. Cancer Res 60:2964–2972 53. Nogae I, Kohno K, Kikuchi J et  al (1989) Analysis of structural features of dihydropyridine analogs needed to reverse multidrug resistance and to inhibit photoaffinity labeling of P-glycoprotein. Biochem Pharmacol 38:519–527 54. Roe M, Folkes A, Ashworth P et al (1999) Reversal of P-glycoprotein mediated multidrug resistance by novel anthranilamide derivatives. Bioorg Med Chem Lett 9: 595–600 55. Sadzuka Y, Sugiyama T, Sonobe T (2000) Efficacies of tea components on doxorubicin induced antitumor activity and reversal of multidrug resistance. Toxicol Lett 114:155–162 56. Twentyman PR, Bleehen NM (1991) Resistance modification by PSC-833, a

novel non-immunosuppressive cyclosporin [corrected]. Eur J Cancer 27:1639–1642 57. Boniface GR, Ferry D, Atsmon J et al (2002) XR9576 (tariquidar), a potent and specific P-glycoprotein inhibitor, has minimal effects on the pharmacokinetics of paclitaxel, doxorubicin, and vinorelbine and can be administered with full-dose chemotherapy in patients with cancer. Proc Am Soc Clin Oncol 39:2173 58. Fox E, Bates SE (2007) Tariquidar (XR9576): a P-glycoprotein drug efflux pump inhibitor. Expert Rev Anticancer Ther 7:447–459 59. Stewart A, Steiner J, Mellows G et al (2000) Phase I trial of XR9576 in healthy volunteers demonstrates modulation of P-glycoprotein in CD56+ lymphocytes after oral and intravenous administration. Clin Cancer Res 6:4186–4191 60. Baer MR, George SL, Dodge RK et al (2002) Phase 3 study of the multidrug resistance modulator PSC-833 in previously untreated patients 60 years of age and older with acute myeloid leukemia: Cancer and Leukemia Group B Study 9720. Blood 100:1224–1232 61. Giaccone G, Linn SC, Welink J et al (1997) A dose-finding and pharmacokinetic study of reversal of multidrug resistance with SDZ PSC 833 in combination with doxorubicin in patients with solid tumors. Clin Cancer Res 3:2005–2015 62. Gruber A, Bjorkholm M, Brinch L et  al (2003) A phase I/II study of the MDR modulator Valspodar (PSC 833) combined with daunorubicin and cytarabine in patients with relapsed and primary refractory acute myeloid leukemia. Leuk Res 27:323–328 63. Kamath AV, Chong S, Chang M, Marathe PH (2005) P-glycoprotein plays a role in the oral absorption of BMS-387032, a potent cyclin-dependent kinase 2 inhibitor, in rats. Cancer Chemother Pharmacol 55:110–116 64. Kitazaki T, Oka M, Nakamura Y et al (2005) Gefitinib, an EGFR tyrosine kinase inhibitor, directly inhibits the function of P-glycoprotein in multidrug resistant cancer cells. Lung Cancer 49:337–343 65. Dawson RJ, Locher KP (2006) Structure of a bacterial multidrug ABC transporter. Nature 443:180–185 66. Fischer E (1894) Einfluss der Configuration auf die Wirkung der Enzyme. Ber Dtsch Chem Ges 27:2985–2993 67. McFarland BJ, Strong RK (2003) Thermodynamic analysis of degenerate recognition by the NKG2D immunoreceptor: not induced fit but rigid adaptation. Immunity 19:803–812

Generating Inhibitors of P-Glycoprotein: Where to, Now? 68. Koshland DE (1958) Application of a theory of enzyme specificity to protein synthesis. Proc Natl Acad Sci USA 44:98–104 69. Sethi DK, Agarwal A, Manivel V, Rao KV, Salunke DM (2006) Differential epitope positioning within the germline antibody paratope enhances promiscuity in the primary immune response. Immunity 24:429–438 70. Tamai I, Safa AR (1991) Azidopine noncompetitively interacts with vinblastine and cyclosporin A binding to P-glycoprotein in multidrug resistant cells. J Biol Chem 266:16796–16800 71. Dey S, Ramachandra M, Pastan I, Gottesman MM, Ambudkar SV (1997) Evidence for two nonidentical drug-interaction sites in the human P-glycoprotein. Proc Natl Acad Sci USA 94:10594–10599 72. Ferry DR, Russell MA, Cullen MH (1992) P-glycoprotein possesses a 1, 4-dihydropyridine selective drug acceptor site which is allosterically coupled to a vinca alkaloid selective binding site. Biochem Biophys Res Commun 188:440–445 73. Martin C, Berridge G, Higgins CF et  al (2000) Communication between multiple drug binding sites on P-glycoprotein. Mol Pharmacol 58:624–632 74. Orlowski S, Mir LM, Belehradek J, Garrigos M (1996) Effects of steroids and verapamil on P-glycoprotein ATPase activity: progesterone, desoxycorticosterone and verapamil are mutually non-exclusive modulators. Biochem J 317:515–522 75. Pascaud C, Garrigos M, Orlowski S (1998) Multidrug resistance transporter P-glyco­ protein has distinct but interacting binding sites for cytotoxic drugs and reversing agents. Biochem J 333:351–358 76. Scala S, Akhmed N, Rao US et  al (1997) P-glycoprotein substrates and antagonists cluster into two distinct groups. Mol Pharmacol 51:1024–1033 77. Shapiro AB, Ling V (1997) Positively cooperative sites for drug transport by P-glycoprotein with distinct drug specificities. Eur J Biochem 250:130–137 78. Lu L, Leonessa F, Clarke R, Wainer IW (2001) Competitive and allosteric interactions in ligand binding to P-glycoprotein as observed on an immobilized P-glycoprotein liquid chromatographic stationary phase. Mol Pharmacol 59:62–68 79. Martin C, Berridge G, Mistry P et al (2000) Drug binding sites on P-glycoprotein are altered by ATP binding prior to nucleotide hydrolysis. Biochemistry 39:11901–11906

427

80. Martin C, Higgins CF, Callaghan R (2001) The vinblastine binding site adopts high- and low-affinity conformations during a transport cycle of P-glycoprotein. Biochemistry 40: 15733–15742 81. Wang EJ, Casciano CN, Clement RP, Johnson WW (2000) Two transport binding sites of P-glycoprotein are unequal yet contingent: initial rate kinetic analysis by ATP hydrolysis demonstrates intersite dependence. Biochim Biophys Acta 1481:63–74 82. Bruggemann EP, Germann UA, Gottesman MM, Pastan I (1989) Two different regions of P-glycoprotein [corrected] are photoaffinity-labeled by azidopine. J Biol Chem 264:15483–15488 83. Loo TW, Clarke DM (2002) Location of the rhodamine-binding site in the human multidrug resistance P-glycoprotein. J Biol Chem 277:44332–44338 84. Pleban K, Kopp S, Csaszar E et  al (2005) P-glycoprotein substrate binding domains are located at the transmembrane domain/ transmembrane domain interfaces: a combined photoaffinity labeling-protein homology modeling approach. Mol Pharmacol 67:365–374 85. Juliano RL, Ling V (1976) A surface glycoprotein modulating drug permeability in Chinese hamster ovary cell mutants. Biochim Biophys Acta 455:152–162 86. Stouch TR, Gudmundsson O (2002) Progress in understanding the structureactivity relationships of P-glycoprotein. Adv Drug Deliv Rev 54:315–328 87. Raub TJ (2006) P-glycoprotein recognition of substrates and circumvention through rational drug design. Mol Pharmacol 3:3–25 88. Cramer J, Kopp S, Bates SE, Chiba P, Ecker GF (2007) Multispecificity of drug transporters: probing inhibitor selectivity for the human drug efflux transporters ABCB1 and ABCG2. ChemMedChem 2:1783–1788 89. Zamora JM, Pearce HL, Beck WT (1988) Physical-chemical properties shared by compounds that modulate multidrug resistance in human leukemic cells. Mol Pharmacol 33:454–462 90. Pearce HL, Safa AR, Bach NJ et  al (1989) Essential features of the P-glycoprotein pharmacophore as defined by a series of reserpine analogs that modulate multidrug resistance. Proc Natl Acad Sci USA 86:5128–5132 91. Ueda K, Okamura N, Hirai M et  al (1992) Human P-glycoprotein transports cortisol, aldosterone, and dexamethasone, but not progesterone. J Biol Chem 267:24248–24252

428

Crowley, McDevitt, and Callaghan

92. Schinkel AH, Wagenaar E, Mol CA, van Deemter L (1996) P-glycoprotein in the blood–brain barrier of mice influences the brain penetration and pharmacological activity of many drugs. J Clin Invest 97: 2517–2524 93. Gottesman MM, Pastan I (1993) Bio­ chemistry of multidrug resistance mediated by the multidrug transporter. Annu Rev Biochem 62:385–427 94. Bain LJ, McLachlan JB, LeBlanc GA (1997) Structure-activity relationships for xenobiotic transport substrates and inhibitory ligands of P-glycoprotein. Environ Health Perspect 105:812–818 95. Seelig A (1998) A general pattern for substrate recognition by P-glycoprotein. Eur J Biochem 251:252–261 96. Ecker G, Huber M, Schmid D, Chiba P (1999) The importance of a nitrogen atom in modulators of multidrug resistance. Mol Pharmacol 56:791–796 97. Martin YC, Kofron JL, Traphagen LM (2002) Do structurally similar molecules have similar biological activity? J Med Chem 45:4350–4358 98. Cramer RD III, Patterson DE, Bunce JD (1988) Comparative molecular field analysis (CoMFA). 1. Effect of shape on binding of steroids to carrier proteins. J Am Chem Soc 110:5959–5967 99. Klebe G, Abraham U, Mietzner T (1994) Molecular similarity indices in a comparative analysis (CoMSIA) of drug molecules to correlate and predict their biological activity. J Med Chem 37:4130–4146 100. Baroni M, Costantino G, Cruciani G et  al (1993) Generating optimal linear PLS estimations (GOLPE): an advanced chemometric tool for handling 3D-QSAR problems. Quant Struct Act Relat 12:9–20 101. Van Drie JH (2004) Pharmacophore discovery: a critical review. In: Bultinck P (ed) Computational medicinal chemistry for drug discovery. Marcel Dekker, New York, pp 437–460 102. Chang C, Swaan PW (2006) Computational approaches to modeling drug transporters. Eur J Pharm Sci 27:411–424 103. Ekins S, Mestres J, Testa B (2007) In silico pharmacology for drug discovery: applications to targets and beyond. Br J Pharmacol 152:21–37 104. Bhat UG, Winter MA, Pearce HL, Beck WT (1995) A structure-function relationship among reserpine and yohimbine analogues in their ability to increase expression of

mdr1 and P-glycoprotein in a human colon carcinoma cell line. Mol Pharmacol 48:682–689 105. Dhainaut A, Regnier G, Atassi G et al (1992) New triazine derivatives as potent modulators of multidrug resistance. J Med Chem 35:2481–2496 106. Boote DJ, Dennis IF, Twentyman PR et  al (1996) Phase I study of etoposide with SDZ PSC 833 as a modulator of multidrug resistance in patients with cancer. J Clin Oncol 14:610–618 107. Rowinsky EK, Smith L, Wang YM et  al (1998) Phase I and pharmacokinetic study of paclitaxel in combination with biricodar, a novel agent that reverses multidrug resistance conferred by overexpression of both MDR1 and MRP. J Clin Oncol 16:2964–2976 108. Stupp R, Bauer J, Pagani O et  al (1998) Ventricular arrhythmia and torsade de pointe: dose limiting toxicities of the MDRmodulator S9788 in a phase I trial. Ann Oncol 9:1233–1242 109. Kim RB, Wandel C, Leake B et  al (1999) Interrelationship between substrates and inhi­ bitors of human CYP3A and P-glycoprotein. Pharm Res 16:408–414 110. Wacher VJ, Wu CY, Benet LZ (1995) Overlapping substrate specificities and tissue distribution of cytochrome P450 3A and P-glycoprotein: implications for drug delivery and activity in cancer chemotherapy. Mol Carcinog 13:129–134 111. Erhlich P (1909) Current status of Chemotherapy Dtsch Chem Ges 42:17 112. van Drie JH (2003) Pharmacophore disco­ very – lessons learned. Curr Pharm Des 9: 1649–1664 113. Martin YC, Bures MG, Danaher EA et  al (1993) A fast new approach to pharmacophore mapping and its application to dopaminergic and benzodiazepine agonists. J Comput Aided Mol Des 7:83–102 114. Jones G, Willett P, Glen RC (1995) A genetic algorithm for flexible molecular overlay and pharmacophore elucidation. J Comput Aided Mol Des 9:532–549 115. Clement OO, Mehl AT (2000) HipHop: pharmacophore based on multiple commonfeature alignments. In: Guner OF (ed) Pharmacophore Perception, Development, and Use in Drug Design. IUL, San Diego, pp 69–84 116. Li H, Sutter J, Hoffman R (2000) HypoGen: an automated system for generating 3D predictive pharmacophore models. In: Guner OF (ed) Pharmacophore perception,

Generating Inhibitors of P-Glycoprotein: Where to, Now? development, and use in drug design. IUL, San Diego, pp 173–189 117. Ekins S, Kim RB, Leake BF et  al (2002) Application of three-dimensional quantitative structure-activity relationships of P-glycoprotein inhibitors and substrates. Mol Pharmacol 61:974–981 118. Penzotti JE, Lamb ML, Evensen E, Grootenhuis PD (2002) A computational ensemble pharmacophore model for identifying substrates of P-glycoprotein. J Med Chem 45:1737–1740 119. Rebitzer S, Annibali D, Kopp S et al (2003) In silico screening with benzofurane- and benzopyrane-type MDR-modulators. Farmaco 58:185–191 120. Langer T, Eder M, Hoffmann RD, Chiba P, Ecker GF (2004) Lead identification for modulators of multidrug resistance based on in silico screening with a pharmacophoric feature model. Arch Pharm (Weinheim) 337:317–327 121. Pleban K, Kaiser D, Kopp S et  al (2005) Targeting drug-efflux pumps – a pharmacoinformatic approach. Acta Biochim Pol 52:737–740 122. Bednarczyk D, Ekins S, Wikel JH, Wright SH (2003) Influence of molecular structure on substrate binding to the human organic cation transporter, hOCT1. Mol Pharmacol 63:489–498 123. Chang C, Swaan PW, Ngo LY et  al (2004) Molecular requirements of the human nucleoside transporters hCNT1, hCNT2, and hENT1. Mol Pharmacol 65:558–570 124. Geldenhuys WJ, Malan SF, Murugesan T, Van der Schyf CJ, Bloomquist JR (2004) Synthesis and biological evaluation of pentacyclo[5.4.0.0(2, 6).0(3, 10).0(5, 9)]undecane derivatives as potential therapeutic agents in Parkinson’s disease. Bioorg Med Chem 12:1799–1806 125. Yates CR, Chang C, Kearbey JD et al (2003) Structural determinants of P-glycoproteinmediated transport of glucocorticoids. Pharm Res 20:1794–1803 126. Pajeva IK, Wiese M (2002) Pharmacophore model of drugs involved in P-glycoprotein multidrug resistance: explanation of structural variety (hypothesis). J Med Chem 45:5671–5686 127. Garrigues A, Loiseau N, Delaforge M et  al (2002) Characterization of two pharmacophores on the multidrug transporter P-glyco­ protein. Mol Pharmacol 62: 1288–1298 128. Ekins S, Crumb WJ, Sarazan RD, Wikel JH, Wrighton SA (2002) Three-dimensional

429

quantitative structure-activity relationship for inhibition of human ether-a-go-go-related gene potassium channel. J Pharmacol Exp Ther 301:427–434 129. Chang C, Bahadduri PM, Polli JE, Swaan PW, Ekins S (2006) Rapid identification of P-glycoprotein substrates and inhibitors. Drug Metab Dispos 34:1976–1984 130. Ecker GF, Stockner T, Chiba P (2008) Computational models for prediction of interactions with ABC-transporters. Drug Discov Today 13:311–317 131. Slate DL, Bruno NA, Casey SM et al (1995) RS-33295-198: a novel, potent modulator of P-glycoprotein-mediated multidrug resistance. Anticancer Res 15:811–814 132. Kohler S, Stein WD (2003) Optimizing chemotherapy by measuring reversal of P-glycoprotein activity in plasma membrane vesicles. Biotechnol Bioeng 81:507–517 133. Mistry P, Stewart AJ, Dangerfield W et  al (2001) In vitro and in  vivo reversal of P-glycoprotein-mediated multidrug resistance by a novel potent modulator, XR9576. Cancer Res 61:749–758 134. van Zuylen L, Sparreboom A, van der Gaast A et  al (2000) The orally administered P-glycoprotein inhibitor R101933 does not alter the plasma pharmacoki-netics of docetaxel. Clin Cancer Res 6:1365–1371 135. Fracasso PM, Goldstein LJ, de Alwis DP et al (2004) Phase I study of docetaxel in combination with the P-glycoprotein inhibitor, zosuquidar, in resistant malignancies. Clin Cancer Res 10:7220–7228 136. Le LH, Moore MJ, Siu LL et al (2005) Phase I study of the multidrug resistance inhibitor zosuquidar administered in combination with vinorelbine in patients with advanced solid tumours. Cancer Chemother Pharmacol 56:154–160 137. Chi KN, Chia SK, Dixon R et  al (2005) A phase I pharmacokinetic study of the P-glycoprotein inhibitor, ONT-093, in combination with paclitaxel in patients with advanced cancer. Invest New Drugs 23:311–315 138. Higgins CF, Linton KJ (2004) The ATP switch model for ABC transporters. Nat Struct Mol Biol 11:918–926 139. Sauna ZE, Ambudkar SV (2007) About a switch: how P-glycoprotein (ABCB1) harnesses the energy of ATP binding and hydrolysis to do mechanical work. Mol Cancer Ther 6:13–23 140. Senior AE, Gadsby DC (1997) ATP hydrolysis cycles and mechanism in P-glycoprotein and CFTR. Semin Cancer Biol 8:143–150

430

Crowley, McDevitt, and Callaghan

141. Boumendjel A, Di Pietro A, Dumontet C, Barron D (2002) Recent advances in the discovery of flavonoids and analogs with highaffinity binding to P-glycoprotein responsible for cancer cell multidrug resistance. Med Res Rev 22:512–529 142. Conseil G, Baubichon-Cortay H, Dayan G et al (1998) Flavonoids: a class of modulators with bifunctional interactions at vicinal ATPand steroid-binding sites on mouse P-glycoprotein. Proc Natl Acad Sci USA 95:9831–9836 143. Morris ME, Zhang S (2006) Flavonoid–drug interactions: effects of flavonoids on ABC transporters. Life Sci 78:2116–2130 144. Harborne JB, Williams CA (2000) Advances in flavonoid research since 1992. Phyto­ chemistry 55:481–504 145. Castro AF, Altenberg GA (1997) Inhibition of drug transport by genistein in multidrugresistant cells expressing P-glycoprotein. Biochem Pharmacol 53:89–93 146. Jodoin J, Demeule M, Beliveau R (2002) Inhibition of the multidrug resistance P-glycoprotein activity by green tea polyphenols. Biochim Biophys Acta 1542:149–159 147. Zhang S, Morris ME (2003) Effects of the flavonoids biochanin A, morin, phloretin, and silymarin on P-glycoprotein-mediated transport. J Pharmacol Exp Ther 304:1258–1267 148. Cermak R (2008) Effect of dietary flavonoids on pathways involved in drug metabolism. Expert Opin Drug Metab Toxicol 4:17–35 149. Moon YJ, Wang X, Morris ME (2006) Dietary flavonoids: effects on xenobiotic and carcinogen metabolism. Toxicol In Vitro 20:187–210 150. von Itzstein M (2007) The war against influenza: discovery and development of sialidase inhibitors. Nat Rev Drug Discov 6:967–974 151. Colman PM, Varghese JN, Laver WG (1983) Structure of the catalytic and antigenic sites in influenza virus neuraminidase. Nature 303:41–44 152. Varghese JN, Laver WG, Colman PM (1983) Structure of the influenza virus glycoprotein antigen neuraminidase at 2.9 A resolution. Nature 303:35–40 153. Varghese JN, McKimm-Breschkin JL, Caldwell JB, Kortt AA, Colman PM (1992) The structure of the complex between influenza virus neuraminidase and sialic acid, the viral receptor. Proteins 14:327–332 154. Goodford PJ (1985) A computational procedure for determining energetically favorable binding sites on biologically important macromolecules. J Med Chem 28:849–857

155. Gubareva LV, Webster RG, Hayden FG (2001) Comparison of the activities of zanamivir, oseltamivir, and RWJ-270201 against clinical isolates of influenza virus and neuraminidase inhibitor-resistant variants. Antimicrob Agents Chemother 45:3403–3408 156. Smith BJ, McKimm-Breshkin JL, McDonald M et al (2002) Structural studies of the resistance of influenza virus neuramindase to inhibitors. J Med Chem 45:2207–2212 157. Malaisree M, Rungrotmongkol T, Decha P et al (2008) Understanding of known drug– target interactions in the catalytic pocket of neuraminidase subtype N1. Proteins 71:1908–1918 158. Cooper NJ, Sutton AJ, Abrams KR et al (2003) Effectiveness of neuraminidase inhibitors in treatment and prevention of influenza A and B: systematic review and meta-analyses of randomised controlled trials. BMJ 326:1235 159. Eiland LS, Eiland EH (2007) Zanamivir for the prevention of influenza in adults and children age 5  years and older. Ther Clin Risk Manag 3:461–465 160. McKimm-Breschkin JL, Sahasrabudhe A, Blick TJ et  al (1998) Mutations in a conserved residue in the influenza virus neuraminidase active site decreases sensitivity to Neu5Ac2en-derived inhibitors. J Virol 72:2456–2462 161. Mendel DB, Tai CY, Escarpe PA et al (1998) Oral administration of a prodrug of the influenza virus neuraminidase inhibitor GS 4071 protects mice and ferrets against influenza infection. Antimicrob Agents Chemother 42:640–646 162. Lee J-Y, Urbatsch IL, Senior AE, Wilkens S (2002) Projection structure of P-glycoprotein by electron microscopy. Evidence for a closed conformation of the nucleotide binding domains. J Biol Chem 277:40125–40131 163. Lee J-Y, Urbatsch IL, Senior AE, Wilkens S (2008) Nucleotide-induced structural changes in p-glycoprotein observed by electron microscopy. J Biol Chem 283:5769–5779 164. Rosenberg MF, Callaghan R, Ford RC, Higgins CF (1997) Structure of the multidrug resistance P-glycoprotein to 2.5  nm resolution determined by electron microscopy and image analysis. J Biol Chem 272:10685–10694 165. Rosenberg MF, Callaghan R, Modok S, Higgins CF, Ford RC (2005) Threedimensional structure of P-glycoprotein: the transmembrane regions adopt an asymmetric configuration in the nucleotide-bound state. J Biol Chem 280:2857–2862

Generating Inhibitors of P-Glycoprotein: Where to, Now? 166. Rosenberg MF, Kamis AB, Callaghan R, Higgins CF, Ford RC (2003) Threedimensional structures of the mammalian multidrug resistance P-glycoprotein demonstrate major conformational changes in the transmembrane domains upon nucleotide binding. J Biol Chem 278:8294–8299 167. Rosenberg MF, Velarde G, Ford RC et  al (2001) Repacking of the transmembrane domains of P-glycoprotein during the transport ATPase cycle. EMBO J 20:5615–5625 168. Locher KP, Lee AT, Rees DC (2002) The E. coli BtuCD structure: a framework for ABC transporter architecture and mechanism. Science 296:1091–1098 169. Pinkett HW, Lee AT, Lum P, Locher KP, Rees DC (2007) An inward-facing conformation of a putative metal-chelate-type ABC transporter. Science 315:373–377 170. Hvorup RN, Goetz BA, Niederer M et  al (2007) Asymmetry in the structure of the ABC transporter-binding protein complex BtuCD-BtuF. Science 317:1387–1390 171. Hollenstein K, Dawson RJ, Locher KP (2007) Structure and mechanism of ABC transporter proteins. Curr Opin Struct Biol 17:412–418 172. Oldham ML, Khare D, Quiocho FA, Davidson AL, Chen J (2007) Crystal structure of a catalytic intermediate of the maltose transporter. Nature 450:515–521 173. Globisch C, Pajeva IK, Wiese M (2008) Identification of putative binding sites of P-glycoprotein based on its homology model. ChemMedChem 3:280–295 174. Jones PM, George AM (2002) Mechanism of ABC transporters: a molecular dynamics simulation of a well characterized nucleotidebinding subunit. Proc Natl Acad Sci USA 99:12639–12644 175. O’Mara ML, Tieleman DP (2007) P-glycoprotein models of the apo and ATPbound states based on homology with Sav1866 and MalK. FEBS Lett 581:4217–4222 176. Ravna AW, Sylte I, Sager G (2007) Molecular model of the outward facing state of the human P-glycoprotein (ABCB1), and comparison to a model of the human MRP5 (ABCC5). Theor Biol Med Model 4:33 177. Stenham DR, Campbell JD, Sansom MS et al (2003) An atomic detail model for the human ATP binding cassette transporter P-glycoprotein derived from disulfide crosslinking and homology modeling. FASEB J 17:2287–2289 178. Dawson RJ, Locher KP (2007) Structure of the multidrug ABC transporter Sav1866

431

from Staphylococcus aureus in complex with AMP-PNP. FEBS Lett 581:935–938 179. Vandevuer S, Van Bambeke F, Tulkens PM, Prevost M (2006) Predicting the threedimensional structure of human P-glycoprotein in absence of ATP by computational techniques embodying crosslinking data: insight into the mechanism of ligand migration and binding sites. Proteins 63:466–478 180. Zolnerciks JK, Wooding C, Linton KJ (2007) Evidence for a Sav1866-like architecture for the human multidrug transporter P-glyco­ protein. FASEB J 21:3937–3948 181. Greenberger LM (1993) Major photoaffinity drug labeling sites for iodoaryl azidoprazosin in P-glycoprotein are within, or immediately C-terminal to, transmembrane domains 6 and 12. J Biol Chem 268:11417–11425 182. Greenberger LM (1998) Identification of drug interaction sites in P-glycoprotein. Methods Enzymol 292:307–317 183. Greenberger LM, Lisanti CJ, Silva JT, Horwitz SB (1991) Domain mapping of the photoaffinity drug-binding sites in P-glycoprotein encoded by mouse mdr1b. J Biol Chem 266:20744–20751 184. Taylor AM, Storm J, Soceneantu L et  al (2001) Detailed characterization of cysteineless P-glycoprotein reveals subtle pharmacological differences in function from wild-type protein. Br J Pharmacol 134:1609–1618 185. Loo TW, Bartlett MC, Clarke DM (2006) Transmembrane segment 7 of human P-glycoprotein forms part of the drug-binding pocket. Biochem J 399:351–359 186. Loo TW, Bartlett MC, Clarke DM (2006) Transmembrane segment 1 of human P-glyco­ protein contributes to the drug-binding pocket. Biochem J 396:537–545 187. Loo TW, Clarke DM (1997) Identification of residues in the drug-binding site of human P-glycoprotein using a thiol-reactive substrate. J Biol Chem 272:31945–31948 188. Loo TW, Clarke DM (1999) Identification of residues in the drug-binding domain of human P-glycoprotein. Analysis of transmembrane segment 11 by cysteine-scanning mutagenesis and inhibition by dibromobimane. J Biol Chem 274:35388–35392 189. Loo TW, Clarke DM (2000) Identification of residues within the drug-binding domain of the human multidrug resistance P-glycoprotein by cysteine-scanning mutagenesis and reaction with dibromobimane. J Biol Chem 275:39272–39278 190. Loo TW, Clarke DM (2001) Defining the drug-binding site in the human multidrug

432

Crowley, McDevitt, and Callaghan

resistance P-glycoprotein using a methanethiosulfonate analog of verapamil, MTS-verapamil. J Biol Chem 276:14972–14979 191. Ecker GF, Csaszar E, Kopp S et  al (2002) Identification of ligand-binding regions of P-glycoprotein by activated-pharmacophore photoaffinity labeling and matrix-assisted laser desorption/ionization-time-of-flight mass spectrometry. Mol Pharmacol 61: 637–648 192. Loo TW, Clarke DM (2005) Recent progress in understanding the mechanism of P-glycoprotein-mediated drug efflux. J Membr Biol 206:173–185 193. Loo TW, Clarke DM (1997) Drug-stimulated ATPase activity of human P-glycoprotein requires movement between transmembrane segments 6 and 12. J Biol Chem 272: 20986–20989 194. Loo TW, Clarke DM (2000) The packing of the transmembrane segments of human multidrug resistance P-glycoprotein is revealed

by disulfide cross-linking analysis. J Biol Chem 275:5253–5256 195. Maki N, Hafkemeyer P, Dey S (2003) Allosteric modulation of human P-glyco­ protein. Inhibition of transport by preventing substrate translocation and dissociation. J Biol Chem 278:18132–18139 196. Maki N, Dey S (2006) Biochemical and pharmacological properties of an allosteric modulator site of the human P-glycoprotein (ABCB1). Biochem Pharmacol 72:145–155 197. Maki N, Moitra K, Ghosh P, Dey S (2006) Allosteric modulation bypasses the requirement for ATP hydrolysis in regenerating low affinity transition state conformation of human P-glycoprotein. J Biol Chem 281: 10769–10777 198. Shapiro AB, Fox K, Lam P, Ling V (1999) Stimulation of P-glycoprotein-mediated drug transport by prazosin and progesterone. Evidence for a third drug-binding site. Eur J Biochem 259:841–850

Chapter 19 Immunosuppressors as Multidrug Resistance Reversal Agents Hamid Morjani and Claudie Madoulet Abstract Multidrug-resistance (MDR) is the major reason for failure of cancer therapy. ATP-binding cassette (ABC) transporters contribute to drug resistance via ATP-dependent drug efflux. P-glycoprotein (Pgp), which is encoded by MDR1 gene, confers resistance to certain anticancer agents. The development of agents able to modulate MDR mediated by Pgp and other ABC transporters remained a major goal for the past 20 years. The calcium blocker verapamil was the first drug shown to be a modulator of Pgp, and since many different chemical compounds have been shown to exert the same effect in vitro by blocking Pgp activity. These included particularly immunosuppressors. Cyclosporin A (CSA) was the first immunosuppressor that have been shown to modulate Pgp activity in laboratory models and entered very early into clinical trials for reversal of MDR. The proof of reversing activity of CSA was found in phase II studies with myeloma and acute leukemia. In phase III studies, the results were less convincing regarding the response rate, progression-free survival, and overall survival, which were detected in advanced refractory myeloma. The non-immunosuppressive derivative PSC833 (valspodar) was subsequently developed. This compound showed tenfold higher potency in reversal of MDR mediated by Pgp. However, pharmacokinetic interactions required reductions in the dose of the concurrently administered anticancer agents. The pharmacokinetic interactions were likely because of decreased clearance of the anticancer agents, possibly as a result of Pgp inhibition in organs such as the gastrointestinal tract and kidney, as well as inhibition of cytochrome P450. Finally, CSA and PSC833 have been shown also to modulate the ceramide metabolism which stands as second messenger of anticancer agent-induced apoptosis. In fact, CSA and PSC833 are also able to respectively inhibit ceramide glycosylation and stimulate de novo ceramide synthesis. This could enhance the cellular level of ceramide and potentiate apoptosis induced by some anticancer agents. Key words: Multidrug resistance, P-glycoprotein, Ceramide pathways, Immunosuppressors, Reversal

1. Introduction Drug resistance is the major reason for failure of cancer therapy. When one drug elicits a response in tumor cells resulting in resistance to a large variety of – chemically unrelated – drugs, this is called multidrug resistance (MDR). ATP-binding cassette (ABC) J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_19, © Humana Press, a part of Springer Science + Business Media, LLC 2010

433

434

Morjani and Madoulet

transporters contribute to drug resistance via ATP-dependent drug efflux (1). A diverse array of drugs has been identified that sensitize multidrug-resistant cells to chemotherapy. The drugs selected for the initial clinical studies were ones already approved for clinical use, and modulators such as verapamil (2), calmodulin inhibitors (3), cyclosporin A (CSA) (4–7), and quinolines (8) were the agents most frequently evaluated. Plasma concentrations equivalent to the concentrations necessary to inhibit drug efflux in vitro were difficult or impossible to achieve with these drugs because of toxicity. Early on, it became clear that reversing the resistance of malignancies like renal cell cancer and colorectal cancer would not be possible, despite high levels of expression of ABC transporters in these tumors. It also became clear that modulators block normal excretory function and delay clearance of chemotherapy. The limitation of the potency of the modulators has been addressed by the development of compounds that are less toxic and more effective as inhibitors. These second generation modulators include CSA (9–12) and the cyclosporin D analog valspodar (PSC833) (13–16). Results from clinical trials with PSC833 have clearly demonstrated a need to reduce the dose of anticancer agents used in combination with it (17, 18), probably because of decreased clearance of the anticancer agents and inhibition of cytochrome P450. Recent clinical studies on the third-generation inhibitors have shown no significant drug interactions with common chemotherapy agents (19). In this issue, we propose to first review the interaction of immunosuppressors with P-glycoprotein (Pgp)-mediated drug efflux. As CSA and PSC833 have been shown also to modulate the metabolism of ceramide which stands as second messenger of anticancer agent-induced apoptosis, we propose to review also the effects of these two compounds on respectively ceramide glycosylation and de novo ceramide synthesis. These effects are able to enhance the cellular level of ceramide and consequently potentiate apoptosis induced by some anticancer agents.

2. Drugs Extruded by Pgp and Common Substrates with Other Pumps 2.1. Drugs Extruded by Pgp

Drug resistance was first documented experimentally in mouse leukemia cells that acquired resistance to 4-amino-N10-methylpteroylglutamic acid (20). In 1973, Dano discovered active

Immunosuppressors as Multidrug Resistance Reversal Agents

435

outward transport of daunorubicin by drug-resistant cells that were cross-resistant to other chemotherapeutic agents, such as vinca-alkaloids (vincristine, vinblastine) and other anthracyclines (doxorubicin) (21). Other authors noted a constant over-expression of a 170 kDa membrane protein termed Pgp (22). The gene encoding Pgp was cloned and identified as MDR1 (23). Riordan and Ling also purified this protein from plasma membrane vesicles of Chinese hamster ovary cell mutants with reduced colchicine permeability (24). More later, taxotere and taxol and other molecules from the cytoskeleton poison group have been identified as substrates of Pgp (25, 26). 2.2. Common Substrates with Other Pumps

Pgp is highly complex transporter and has the ability of recognizing and transporting a large number of structurally diverse, mainly hydrophobic compounds. In addition to its overlapping substrate specificity with other transporters such as multidrug resistance-associated protein (MRP) family and breast cancer resistance protein (BCRP), Pgp can handle unique compounds. Pgp is a transporter for large hydrophobic, either uncharged or slightly positively charged compounds while the MRP family primarily transports hydrophobic anionic conjugates and extrudes hydrophobic uncharged drugs. The MRP1-related uncharged drug transport is linked to the transport of cellular free reduced glutathione (27). The exact spectrum of the BCRP-transported substrates has not yet been explored in detail, and these studies are complicated by the variable substratemutants of BCRP observed in the most recent studies (28). Some of the key molecules are presented in Table 19.1 and are unfortunately also MDR substrates for the patients.

Table 19.1 Summary of the key molecules transported by Pgp and/or other ABC proteins Compound

Type of compound

Specificity

Doxorubicin Idarubicin Daunorubicin

Anthracyclines

Pgp, MRP1, BCRP

Mitoxantrone

Anthraquinone

Pgp, BCRP

Methotrexate

Antimetabolite

MRP1, MRP3, BCRP

Topotecan and SN38

Camptothecins

BCRP

Vinblastine Navelbine Vincristine

Vinca alkaloids

Pgp Pgp Pgp, MRP1

Etoposide (VP16)

Epipodophyllotoxin

Pgp, MRP1

436

Morjani and Madoulet

3. Agents that Modulate Pgp Activity

The clinical importance of Pgp might also be determined through trials designed to abrogate Pgp function. Toward this end, less than 10 years after the discovery of Pgp-mediated MDR, the first Phase I and II clinical trials began to test the clinical potential of Pgp inhibitors. Initial trials used “first-generation” Pgp inhibitors, including verapamil, quinine, and CSA. A randomized Phase III clinical trial showed the benefit of addition of CSA to treatment with cytarabine and daunorubicin in patients with poor-risk AML (12). Similarly, quinine was shown to increase the complete remission rate as well as survival in Pgp-positive myelodysplastic syndrome cases treated with intensive chemotherapy (29), suggesting that successful Pgp modulation was feasible. However, several other trials failed to show improvement of the outcome (30). The second generation of inhibitors was devoid of side effects related to the primary toxicity of the compounds (R-enantiomer of verapamil, PSC-833 “valspodar,” and VX-710 “biricodar”). The two first molecules were able to inhibit Pgp without blocking calcium channels or immunosuppressive effects, respectively (31). Third-generation inhibitors are designed specifically for high transporter affinity and low pharmacokinetic interaction. Table 19.2 summarizes inhibitors of Pgp (32, 33).

Table 19.2 Summary of molecules that are able to inhibit P-glycoprotein Generation

Pgp inhibitors

Reference

First

Amiodarone Cyclosporin A (CSA) Quinidine Quinine Verapamil Nifedipine Dexniguldipine

(34) (35) (32, 36) (29, 37) (2) (32) (38)

Second

PSC833 VX-710 (Biricodar) GG918

(35) (39) (40, 41)

Third

LY475776 LY335979 (Zosuquidar) XR-9576 (Tariquidar) V-104 R101933 (Laniquidar) S9788

(42) (43, 44) (45, 46) (47) (48, 49) (50, 51)

Other

Disulfiram Pluronic L61

(52) (53)

Immunosuppressors as Multidrug Resistance Reversal Agents

4. Modulation of Pgp Activity by CSA and PSC833

437

CSA and PSC833 (Fig.  19.1) belong to the group of MDR modulators that were first described to inhibit the Pgp-associated ATPase activity. The best characterized is PSC833, which at nanomolar concentrations inhibits the ATPase and transport function of Pgp. Concerning the transport of CSA and PSC833, data are not clear. Several works are in favor of the view that CSA and PSC833 are substrates for Pgp (54–56) and other data are not (57). The last report considers that PSC833 is indeed a substrate of Pgp, but a slow one. That is, when PSC833 competes with a substrate, it will win out due to its higher affinity (a larger interaction surface) (58). But since its transport rate is slower, it will slow down the turnover rate, which will then be reflected in the decrease in ATPase activity. So PSC833 seems to be a partial antagonist, since it does not completely block Pgp function, but just slows it down due to the bulkiness of this molecule acting as an “obstructive” substrate, slowing down the Pgp machinery. As MDR modulators from the first generation and at the opposite of some others from the second generation (VX-710 and GG918), CSA and PSC833 inhibit exclusively Pgp (Table 19.3). Studies performed in vitro on samples from leukemia patients showed that CSA and PSC833 were appropriate inhibitors that can be used from MDR diagnosis. In fact, Merlin and co-workers have shown that in the Pgp-positive samples, cellular daunorubicin uptake was increased in the presence of CSA (59). More later, the same group demonstrated the influence of PSC833 on cellular daunorubicin uptake in bone marrow specimens from patients with acute myeloid leukemia (60). Legrand and co-workers have demonstrated also that there was a good correlation between Pgp expression and the in vitro modulatory effect of CSA on cellular calcein-AM uptake in samples from patients with acute myeloid leukemia (61). A functional study of calcein uptake and efflux in the presence of respectively CSA and probenecid (a specific

Fig. 19.1. Structure of cyclosporin A and PSC833 (valspodar).

438

Morjani and Madoulet

Table 19.3 Specificity of molecules that are able to modulate resistance to anticancer drugs Compound

P-gp

MRP1

BCRP

Verapamil

+

−

−

Quinine, quinidine

+

−

−

Cyclosporin A (CSA) PSC833 (Valspodar)

+

−

−

Biricodar (VX-710)

+

+

−

S9788

+

−

−

Elacridar (GG918)

+

−

+

Zosuquidar (LY335979)

+

−

−

Tariquidar (XR9576)

+

−

−

i­nhibitor of MRP1) has shown that it was possible to discriminate the Pgp and MRP1 transport activities (62). CSA entered early into trials of reversal of MDR. The proof of reversing activity of CSA was found in phase II studies with myeloma (9) and acute leukemia (10). Phase III studies were conducted in hematological malignancies and no effect of CSA on the overall response rate and progression-free survival in myeloma patients was observed (11), whereas among several studies only one showed a positive effect of CSA in acute myeloblastic leukemia (12). The widely tested second-generation compound was PSC833 (valsopodar). PSC833 was 10 times more potent than CSA (35) and was the first molecule from cyclosporins group without immunosuppressive properties (6). PSC833 showed modulation of MDR in vivo with a lower renal toxicity when compared with CSA (7). During phase I studies, an important effect of this compound on the pharmacokinetics of the associated drugs was shown (13–16). In most cases, either a doubling of the time-plasma concentration area under the curve or an important increase in elimination half-life was found. Therefore, it was not possible to separate the pharmacokinetic effects of PSC833 from its pharmacodynamic effects. Those results were in general disappointing, particularly in acute myeloblastic leukemia trials and then suggested that it was needed to reduce the dose of anticancer agents used in combination with it and dose reduction ranged from 25% for etoposide to 66% for taxol (63, 64). Those dose reductions were required to prevent toxicities of the anticancer agent in

Immunosuppressors as Multidrug Resistance Reversal Agents

439

combined therapy, and have compromised drug concentration in the tumor even with complete inhibition of Pgp. One treatment that has not been fully explored is that of ­prevention of the emergence of resistance through the use of Pgp inhibitors. In the laboratory, PSC833 reduced the mutation rate for doxorubicin-selected resistance in sarcoma cells by tenfold, thus reducing the development of resistant clones via the MDR mechanism. In those sarcoma cells, resistance was mediated by an alternative pathway with reduced expression of topoisomérase II, the target enzyme for doxorubicin (65). Another study examined six agents for their ability to prevent vincristine resistance in a rhabdomyosarcoma cell line (66). MDR modulators and particularly PSC833 prevented the development of resistance, suggesting the role of the use of Pgp inhibitors prior to cytotoxic therapy.

5. Modulation of Ceramide Metabolism by Immuno sup-pressors

5.1. Modulation of De Novo Synthesis of Ceramide

Sphingolipid are a large family of lipids that are implicated in ­signal transduction process, and ceramide is the most studied (67–69). In fact, ceramide stands as second messenger of anticancer agent-induced apoptosis. CSA and PSC833 have been shown to modulate the ceramide metabolism (de novo synthesis and glycosylation). These effects are able to enhance the cellular level of ceramide and consequently ­potentiate apoptosis induced by some anticancer agents. Ceramide synthesis results from sphingomyelin hydrolysis (67) or by de novo synthesis. De novo synthesis of ceramide is initiated at the cytosolic surface of the endoplasmic reticulum (RE) membrane by condensation of l-serine and palmitoyl coenzyme A (69–71) (Fig. 19.2). This NADPH-dependent reaction gives

Fig. 19.2. Ceramide de novo synthesis and glycosylation pathways.

440

Morjani and Madoulet

3-ketosphinganine (72), which is reduced to dihydrosphingosine. The next step is an acylation reaction. The enzyme acyltransferase transfers a long-chain fatty acid to the amino group of the molecule and generates dihydroceramide. The introduction of a trans –4,5 double bond by a desaturase converts the dihydroceramide to ceramide (Fig. 19.2) (73). A number of cytotoxic agents have been shown to activate de novo synthesis of ceramide. Daunorubicin is able to promote ceramide formation and apoptosis by ceramide synthase activation in mouse and human leukemia cells (74, 75). Vincristine activates also de novo synthesis of ceramide and subsequently induces apoptosis in several cell lines (76–78). Similar data have been observed when taxol was used to induce apoptosis in leukemia and breast carcinoma (79, 80). This is a short list of compounds that induce de  novo synthesis of ceramide and apoptosis. Several other anti­ cancer drug are able to induce the same effects (etoposide, camptothecins, retinoides, N-(4-hydroxyphenyl)retinamide, gemcitabine). PSC833 is able to increase the cellular level of ceramide and this effect is a Pgp-independent mechanism (81, 82). In KB-V-1 human epidermoid carcinoma cells, PSC833 activates ceramide synthesis and increases the cytotoxic effects of vinblastine (78). Similar data have been reported on human ovarian carcinoma cells SKOV-3 (83). Cells, whose ceramide glycosylation is enhanced, are generally resistant to PSC833 effect (84). This can be explained by the fact that ceramide generated by PSC833 is converted to glycosylceramides (85) (Fig. 19.2). Wang and co-workers demonstrated that generation of ceramide by PCS833 results from activation of serine palmitoyltransferase enzyme (86). Additionally, PSC833 has been reported to restore sphingomyelin-ceramide pathway stimulation. In fact, Bezombes and co-workers have reported that KG1a cells, which are inherently resistant to TNFa and do not produce ceramide via sphingomyelin hydrolysis upon cytokine treatment, can be sensitized by PSC833 (87). The authors suggested that resistance to TNFa-mediated apoptosis of these cells was linked to the disposability of the sphingomyelin pool, and a role for Pgp in sphingomyelin transverse plasma membrane asymmetry which can be affected in this case by PCS833. 5.2. Modulation of Ceramide Glycosylation

Once generated ceramide can accumulate in the cell or may be converted into a variety of metabolites. In fact, ceramide is a precursor of glucosylceramide and galactosylceramide (Fig.  19.2). The glucosylceramide (GlcCer) synthesis results from transfer of glucose of UDP-glucose to ceramide by a glucosyltransferase. The GlcCer is a metabolic precursor of lactosylceramide (LacCer) and higher order glycosylceramides. Lavie and co-workers have shown that MDR cells, which over-express Pgp, display an elevation of GlcCer when compared

Immunosuppressors as Multidrug Resistance Reversal Agents

441

with their drug-sensitive counterparts (88–90). Kok and co-workers have also reported that colchicine-resistant colon carcinoma cells over-expressing MRP1 have elevated GlcCer level (91). It has then been suggested to consider GlcCer level as a diagnosis marker for drug resistance in tumors (92). The increase in GlcCer level has been explained by a higher activity of GlcCer synthase (GCS) (89) and/or uncoupling of GlcCer conversion to LacCer (93). The use of specific glucosylceramide synthase inhibitors like 1-phenyl-2-palmitoylamino-3-morpholino-1-propanolol (PPMP) decreases the GlcCer level in resistant cells and partially restores sensitivity to chemotherapeutic agents (94–96). Similarly, tamoxifen is also able to the same effect (97, 98). CSA is able to block the glucosylation of ceramide in MDR cells (84) and this is accompanied by a sensitization of MDR cells to the cytotoxic effects of anticancer agents (90, 94). However, the mechanism by which ceramide glycosylation is inhibited is not clear.

6. Conclusion A number of lessons have been learned from the evolution of the field of MDR. However, the reversal of MDR has not yet reached the level of routine clinical applications. The future of the potential therapeutic area remains uncertain. This is not for lack of molecules, since hundreds of compounds have been selected or designed with comprehensive studies on structure–activity relationships in several chemical families. Rather, the reason for this failure originates from the inadequate design of clinical trials. The pharmacological and toxicological properties of MDR modulators should have been taken into consideration with those of anticancer drugs in terms of the benefit/risk ratio of combination. The pharmacokinetic interaction of CSA or PSC833 with cytotoxic drugs has considerably rendered difficult the clinical development of these drugs as MDR modulators. In fact, CSA and PSC833 appear to be the only ones to present this interaction that might well be related to the specific inhibition of another ABC protein that Pgp involved in the biliary elimination of drugs. The pharmacokinetic studies of the third-generation inhibitors (tariquidar XR9576, zosuquidar LY335979, and laniquidar R101933) have shown no appreciable impact on cytochrome P450 drug metabolism and no clinically significant drug interactions with common chemotherapy agents. Finally, the emergence of modulators with several ABC protein targets will be of clinical interest in malignancies often expressing several efflux pumps simultaneously (Table 19.3).

442

Morjani and Madoulet

References 1. Gottesman MM (1993) How cancer cells evade chemotherapy. Cancer Res 53:747–754 2. Tsuruo T, Iida H, Tsukagoshi S, Sakurai Y (1981) Overcoming of vincristine resistance in P388 leukemia in vivo and in vitro through enhanced cytotoxicity of vincristine and vinblastine by verapamil. Cancer Res 41: 1967–1972 3. Cano-Gauci DF, Riordan JR (1987) Action of calcium antagonists on multidrug resistant cells. Specific cytotoxicity independent of increased cancer drug accumulation. Biochem Pharmacol 36:2115–2123 4. Silbermann MH, Boersma AW, Janssen AL et al (1989) Effects of cyclosporin A and verapamil on the intracellular daunorubicin accumulation in Chinese hamster ovary cells with increasing levels of drug-resistance. Int J Cancer 44:722–726 5. Hait WN, Stein JM, Koletsky AJ, Harding MW, Handschumacher RE (1989) Activity of cyclosporin A and a non-immunosuppressive cyclosporin against multidrug resistant leukemic cell lines. Cancer Commun 1:35–43 6. Twentyman PR, Bleehen NM (1991) Resistance modification by PSC-833, a novel non-immunosuppressive cyclosporin. Eur J Cancer 27:1639–1642 7. Boesch D, Gaveriaux C, Jachez B, PourtierManzanedo A, Bollinger P, Loor F et al (1991) In vivo circumvention of P-glycoproteinmediated multidrug resistance of tumor cells with SDZ PSC 833. Cancer Res 51:4226–4233 8. Beck WT, Cirtain MC, Glover CJ, Felsted RL, Safa AR (1988) Effects of indole alkaloids on multidrug resistance and labeling of P-glycoprotein by a photoaffinity analog of vinblastine. Biochem Biophys Res Commun 153:959–966 9. Sonneveld P, Durie BG, Lokhorst HM et  al (1992) Modulation of multidrug-resistant multiple myeloma by cyclosporin. The Leukaemia Group of the EORTC and the HOVON. Lancet 340:255–259 10. List AF, Spier C, Greer J et al (1993) Phase I/II trial of cyclosporine as a chemotherapy-resistance modifier in acute leukemia. J Clin Oncol 11:1652–1660 11. Sonneveld P, Suciu S, Weijermans P et  al (2001) Cyclosporin A combined with vincristine, doxorubicin and dexamethasone (VAD) compared with VAD alone in patients with advanced refractory multiple myeloma: an EORTC-HOVON randomized phase III study (06914). Br J Haematol 115:895–902

12. List AF, Kopecky KJ, Willman CL et al (2001) Benefit of cyclosporine modulation of drug resistance in patients with poor-risk acute myeloid leukemia: a Southwest Oncology Group study. Blood 98:3212–3220 13. Boote DJ, Dennis IF, Twentyman PR et  al (1996) Phase I study of etoposide with SDZ PSC 833 as a modulator of multidrug resistance in patients with cancer. J Clin Oncol 14:610–618 14. Giaccone G, Linn SC, Welink J et al (1997) A dose-finding and pharmacokinetic study of reversal of multidrug resistance with SDZ PSC 833 in combination with doxorubicin in patients with solid tumors. Clin Cancer Res 3:2005–2015 15. Advani R, Saba HI, Tallman MS et al (1999) Treatment of refractory and relapsed acute myelogenous leukemia with combination chemotherapy plus the multidrug resistance modulator PSC 833 (Valspodar). Blood 93:787–795 16. Fracasso PM, Westervelt P, Fears CL et  al (2000) Phase I study of paclitaxel in combination with a multidrug resistance modulator, PSC 833 (Valspodar), in refractory malignancies. J Clin Oncol 18:1124–34 17. Baer MR, George SL, Dodge RK et al (2002) Phase 3 study of the multidrug resistance modulator PSC-833 in previously untreated patients 60 years of age and older with acute myeloid leukemia: Cancer and Leukemia Group B Study 9720. Blood 100:1224–1232 18. Bates SE, Bakke S, Kang M et  al (2004) A phase I/II study of infusional vinblastine with the P-glycoprotein antagonist valspodar (PSC 833) in renal cell carcinoma. Clin Cancer Res 10:4724–4733 19. Thomas H, Coley HM (2003) Overcoming multidrug resistance in cancer: an update on the clinical strategy of inhibiting p-glycoprotein. Cancer Control 10:159–165 20. Burchenal JH, Robinson E, Johnston SF, Kushida MN (1950) The induction of resistance to 4-amino-N10-methyl-pteroylglutaminic acid in a strain of transmitted mouse leukemia. Science 111:116–117 21. Dano K (1973) Active outward transport of daunomycin in resistant Ehrlich ascites tumor cells. Biochim Biophys Acta 323:466–483 22. Juliano RL, Ling V (1976) A surface glycoprotein modulating drug permeability in Chinese hamster ovary cell mutants. Biochim Biophys Acta 455:152–162 23. Ueda K, Clark DP, Chen CJ et al (1987) The human multidrug resistance (mdr1) gene.

Immunosuppressors as Multidrug Resistance Reversal Agents

24.

25.

26.

27.

28.

29.

30.

31.

32. 33.

34.

cDNA cloning and transcription initiation. J Biol Chem 262:505–508 Riordan JR, Ling V (1979) Purification of P-glycoprotein from plasma membrane vesicles of Chinese hamster ovary cell mutants with reduced colchicine permeability. J Biol Chem 254:12701–1275 Riou JF, Naudin A, Lavelle F (1992) Effects of Taxotere on murine and human tumor cell lines. Biochem Biophys Res Commun 187:164–170 Greenberger LM, Lothstein L, Williams SS, Horwitz SB (1988) Distinct P-glycoprotein precursors are overproduced in independently isolated drug-resistant cell lines. Proc Natl Acad Sci USA 85:3762–3766 Leier I, Jedlitschky G, Buchholz U et  al (1994) The MRP gene encodes an ATPdependent export pump for leukotriene C4 and structurally related conjugates. J Biol Chem 269:27807–27810 Honjo Y, Hrycyna CA, Yan QW et al (2001) Acquired mutations in the MXR/BCRP/ ABCP gene alter substrate specificity in MXR/ BCRP/ABCP-overexpressing cells. Cancer Res 61:6635–6639 Wattel E, Solary E, Hecquet B et  al (1999) Quinine improves results of intensive chemotherapy (IC) in myelodysplastic syndromes (MDS) expressing P-glycoprotein (PGP). Updated results of a randomized study. Groupe Français des Myélodysplasies (GFM) and Groupe GOELAMS. Adv Exp Med Biol 457:35–46 Daenen S, van der Holt B, Verhoef GE et al (2004) Addition of cyclosporin A to the combination of mitoxantrone and etoposide to overcome resistance to chemotherapy in refractory or relapsing acute myeloid leukaemia: a randomised phase II trial from HOVON, the Dutch-Belgian HaematoOncology Working Group for adults. Leuk Res 28:1057–1067 Höllt V, Kouba M, Dietel M, Vogt G (1992) Stereoisomers of calcium antagonists which differ markedly in their potencies as calcium blockers are equally effective in modulating drug transport by P-glycoprotein. Biochem Pharmacol 43:2601–2608 Robert J, Jarry C (2003) Multidrug resistance reversal agents. J Med Chem 46:4805–4817 Szakács G, Paterson JK, Ludwig JA, BoothGenthe C, Gottesman MM (2006) Targeting multidrug resistance in cancer. Nat Rev Drug Discov 5:219–234 Seelig A, Gatlik-Landwojtowicz E (2005) Inhibitors of multidrug efflux transporters:

35. 36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

443

their membrane and protein interactions. Mini Rev Med Chem 5:135–151 Twentyman PR (1992) Cyclosporins as drug resistance modifiers. Biochem Pharmacol 43:109–117 Tanigawara Y, Okamura N, Hirai M et  al (1992) Transport of digoxin by human P-glycoprotein expressed in a porcine kidney epithelial cell line LLC-PK1. J Pharmacol Exp Ther 263:840–845 Wattel E, Solary E, Hecquet B et  al (1998) Quinine improves the results of intensive chemotherapy in myelodysplastic syndromes expressing P glycoprotein: results of a randomized study. Br J Haematol 102:1015–1024 Hofmann J, Wolf A, Spitaler M et al (1992) Reversal of multidrug resistance by B859–35, a metabolite of B859–35, niguldipine, verapamil and nitrendipine. J Cancer Res Clin Oncol 118:361–366 Minderman H, O’Loughlin KL, Pendyala L, Baer MR (2004) VX-710 (biricodar) increases drug retention and enhances chemosensitivity in resistant cells overexpressing P-glycoprotein, multidrug resistance protein, and breast cancer resistance protein. Clin Cancer Res 10:1826–1834 Hyafil F, Vergely C, Du Vignaud P, GrandPerret T (1993) In vitro and in vivo reversal of multidrug resistance by GF120918, an acridonecarboxamide derivative. Cancer Res 53:4595–4602 Zhou DC, Simonin G, Faussat AM, Zittoun R, Marie JP (1997) Effect of the multidrug inhibitor GG918 on drug sensitivity of human leukemic cells. Leukemia 11:1516–1522 Dantzig AH, Shepard RL, Pratt SE III et  al (2004) Evaluation of the binding of the tricyclic isoxazole photoaffinity label LY475776 to multidrug resistance associated protein 1 (MRP1) orthologs and several ATP-binding cassette (ABC) drug transporters. Biochem Pharmacol 67:1111–1121 Shepard RL, Cao J, Starling JJ, Dantzig AH (2003) Modulation of P-glycoprotein but not MRP1- or BCRP-mediated drug resistance by LY335979. Int J Cancer 103:121–125 Tang R, Faussat AM, Perrot JY et al (2008) Zosuquidar restores drug sensitivity in P-glycoprotein expressing acute myeloid leukemia (AML). BMC Cancer 8:51 Martin C, Berridge G, Mistry P et al (1999) The molecular interaction of the high affinity reversal agent XR9576 with P-glycoprotein. Br J Pharmacol 128:403–411 Walker J, Martin C, Callaghan R (2004) Inhibition of P-glycoprotein function by

444

47.

48.

49.

50.

51.

52. 53.

54.

55.

56.

57. 58.

Morjani and Madoulet XR9576 in a solid tumour model can restore anticancer drug efficacy. Eur J Cancer 40:594–605 Evers R, Kool M, Smith AJ et  al (2000) Inhibitory effect of the reversal agents V-104, GF120918 and pluronic L61 on MDR1 Pgp-, MRP1- and MRP2-mediated transport. Br J Cancer 83:366–374 van Zuylen L, Sparreboom A, van der Gaast A et  al (2000) The orally administered P-glycoprotein inhibitor R101933 does not alter the plasma pharmacokinetics of docetaxel. Clin Cancer Res 6:1365–1371 van Zuylen L, Sparreboom A, van der Gaast A et  al (2002) Disposition of docetaxel in the presence of P-glycoprotein inhibition by intravenous administration of R101933. Eur J Cancer 38:1090–1099 Pierré A, Dunn TA, Kraus-Berthier L et  al (1992) In vitro and in vivo circumvention of multidrug resistance by Servier 9788, a novel triazinoaminopiperidine derivative. Invest New Drugs 10:137–148 Punt CJ, Voest EE, Tueni E et al (1997) Phase IB study of doxorubicin in combination with the multidrug resistance reversing agent S9788 in advanced colorectal and renal cell cancer. Br J Cancer 76:1376–1381 Vastag B (2000) Almost serendipity: alcoholism drug reverses drug resistance in vitro. J Natl Cancer Inst 92:864–865 Venne A, Li S, Mandeville R, Kabanov A, Alakhov V (1996) Hypersensitizing effect of pluronic L61 on cytotoxic activity, transport, and subcellular distribution of doxorubicin in multiple drug-resistant cells. Cancer Res 56:3626–3629 Saeki T, Ueda K, Tanigawara Y, Hori R, Komano T (1993) Human P-glycoprotein transports cyclosporin A and FK506. J Biol Chem 268:6077–6080 Shirai A, Naito M, Tatsuta T et  al (1994) Transport of cyclosporin A across the brain capillary endothelial cell monolayer by P-glycoprotein. Biochim Biophys Acta 1222:400–404 Schramm U, Fricker G, Wenger R, Miller DS (1995) P-glycoprotein-mediated secretion of a fluorescent cyclosporin analogue by teleost renal proximal tubules. Am J Physiol 268:F46–F52 Stein WD (1997) Kinetics of the multidrug transporter (P-glycoprotein) and its reversal. Physiol Rev 77:545–590 Litman T, Zeuthen T, Skovsgaard T, Stein WD (1997) Structure-activity relationships of P-glycoprotein interacting drugs: kinetic char-

59.

60.

61.

62.

63. 64.

65.

66.

67. 68. 69. 70.

acterization of their effects on ATPase activity. Biochim Biophys Acta 1361:159–168 Merlin JL, Guerci A, Marchal S et al (1994) Comparative evaluation of S9788, verapamil, and cyclosporine A in K562 human leukemia cell lines and in P-glycoprotein-expressing samples from patients with hematologic malignancies. Blood 84:262–269 Merlin JL, Guerci AP, Marchal S et al (1998) Influence of SDZ-PSC833 on daunorubicin intracellular accumulation in bone marrow specimens from patients with acute myeloid leukaemia. Br J Haematol 103:480–487 Legrand O, Simonin G, Perrot JY, Zittoun R, Marie JP (1998) Pgp and MRP activities using calcein-AM are prognostic factors in adult acute myeloid leukemia patients. Blood 91:4480–4488 Legrand O, Simonin G, Beauchamp-Nicoud A, Zittoun R, Marie JP (1999) Simultaneous activity of MRP1 and Pgp is correlated with in  vitro resistance to daunorubicin and with in vivo resistance in adult acute myeloid leukemia. Blood 94:1046–1056 Leonard GD, Fojo T, Bates SE (2003) The role of ABC transporters in clinical practice. Oncologist 8:411–424 Bates SE (2002) Solving the problems of multidrug-resistance: ABC transporters in clinical oncology. In: Holland IB, Cole SP, Kuchler K et  al (eds) ABC proteins: from bacteria to man. Elsevier Science, London, pp 359–391 Beketic-Oreskovic L, Duran GE et al (1995) Decreased mutation rate for cellular resistance to doxorubicin and suppression of mdr1 gene activation by the cyclosporin PSC 833. J Natl Cancer Inst 87:1593–1602 Cocker HA, Tiffin N, Pritchard-Jones K, Pinkerton CR, Kelland LR (2001) In vitro prevention of the emergence of multidrug resistance in a pediatric rhabdomyosarcoma cell line. Clin Cancer Res 7:3193–3198 Levade T, Jaffrezou JP (1999) Signalling sphingomyelinases: which, where, how and why ? Biochim Biophys Acta 1438:1–17 Sietsma H, Veldman RJ, Kok JW (2001) The involvement of sphingolipids in multidrug resistance. J Membr Biol 181:153–162 Perry DK (2000) The role of de  novo ceramide synthesis in chemotherapy-induced apoptosis. Ann N Y Acad Sci 905:91–96 Mandon EC, Ehses I, Rother J, van Echten G, Sandhoff K (1992) Subcellular localization and membrane topology of serine palmitoyltransferase, 3-dehydrosphinganine reductase, and sphinganine N-acyltransferase in mouse liver. J Biol Chem 267:11144–11148

Immunosuppressors as Multidrug Resistance Reversal Agents 71. Michel C, van Echten-Deckert G (1997) Conversion of dihydroceramide to ceramide occurs at the cytosolic face of the endoplasmic reticulum. FEBS Lett 416:153–155 72. Hannun YA, Luberto C, Argraves KM (2001) Enzymes of sphingolipid metabolism: from modular to integrative signaling. Biochemistry 40:4893–4903 73. Michel C, van Echten-Deckert G, Rother J et al (1997) Characterization of ceramide synthesis. A dihydroceramide desaturase introduces the 4, 5-trans-double bond of sphingosine at the level of dihydroceramide. J Biol Chem 272:22432–22437 74. Bose R, Verheij M, Haimovitz-Friedman A et  al (1995) Ceramide synthase mediates daunorubicin-induced apoptosis: an alternative mechanism for generating death signals. Cell 82:405–414 75. Turnbull KJ, Brown BL, Dobson PR (1999) Caspase-3-like activity is necessary but not sufficient for daunorubicin-induced apoptosis in Jurkat human lymphoblastic leukemia cells. Leukemia 13:1056–1061 76. Zhang J, Alter N, Reed JC et  al (1996) Bcl-2 interrupts the ceramide-mediated pathway of cell death. Proc Natl Acad Sci USA 93:5325–5328 77. Olshefski RS, Ladisch S (2001) Glucosylceramide synthase inhibition enhances vincristine-induced cytotoxicity. Int J Cancer 93:131–138 78. Cabot MC, Giuliano AE, Han TY, Liu YY (1999) SDZ PSC 833, the cyclosporine A analogue and multidrug resistance modulator, activates ceramide synthesis and increases vinblastine sensitivity in drug-sensitive and drugresistant cancer cells. Cancer Res 59:880–885 79. Myrick D, Blackinton D, Klostergaard J et al (1999) Paclitaxel-induced apoptosis in Jurkat, a leukemic T cell line, is enhanced by ceramide. Leuk Res 23:569–278 80. Mehta S, Blackinton D, Omar I et al (2000) Combined cytotoxic action of paclitaxel and ceramide against the human Tu138 head and neck squamous carcinoma cell line. Cancer Chemother Pharmacol 46:85–92 81. Cabot MC, Han TY, Giuliano AE (1998) The multidrug resistance modulator SDZ PSC 833 is a potent activator of cellular ceramide formation. FEBS Lett 431:185–188 82. Goulding CW, Giuliano AE, Cabot MC (2000) SDZ PSC 833 the drug resistance modulator activates cellular ceramide formation by a pathway independent of P-glycoprotein. Cancer Lett 149:143–151 83. Senchenkov A, Litvak DA, Cabot MC (2001) Targeting ceramide metabolism – a strategy

445

for overcoming drug resistance. J Natl Cancer Inst 93:347–357 84. Lucci A, Han TY, Liu YY, Giuliano AE, Cabot MC (1999) Multidrug resistance modulators and doxorubicin synergize to elevate ceramide levels and elicit apoptosis in drug-resistant cancer cells. Cancer 86:300–311 85. Lucci A, Han TY, Liu YY, Giuliano AE, Cabot MC (1999) Modification of ceramide metabolism increases cancer cell sensitivity to cytotoxics. Int J Oncol 15:541–546 86. Wang H, Giuliano AE, Cabot MC (2002) Enhanced de  novo ceramide generation through activation of serine palmitoyltransferase by the P-glycoprotein antagonist SDZ PSC 833 in breast cancer cells. Mol Cancer Ther 1:719–726 87. Bezombes C, Maestre N, Laurent G et  al (1998) Restoration of TNF-alpha-induced ceramide generation and apoptosis in resistant human leukemia KG1a cells by the P-glycoprotein blocker PSC833. FASEB J 12:101–109 88. Lavie Y, Cao H, Bursten SL, Giuliano AE, Cabot MC (1996) Accumulation of glucosylceramides in multidrug-resistant cancer cells. J Biol Chem 271:19530–19536 89. Liu YY, Han TY, Giuliano AE, Cabot MC (1999) Expression of glucosylceramide synthase, converting ceramide to glucosylceramide, confers adriamycin resistance in human breast cancer cells. J Biol Chem 274:1140–1146 90. Morjani H, Aouali N, Belhoussine R et  al (2001) Elevation of glucosylceramide in multidrug-resistant cancer cells and accumulation in cytoplasmic droplets. Int J Cancer 94:157–165 91. Kok JW, Veldman RJ, Klappe K et al (2000) Differential expression of sphingolipids in MRP1 overexpressing HT29 cells. Int J Cancer 87:172–178 92. Lucci A, Cho WI, Han TY et  al (1998) Glucosylceramide: a marker for multipledrug resistant cancers. Anticancer Res 18:475–480 93. Veldman RJ, Klappe K, Hinrichs J et  al (2002) Altered sphingolipid metabolism in multidrug-resistant ovarian cancer cells is due to uncoupling of glycolipid biosynthesis in the Golgi apparatus. FASEB J 16: 1111–1113 94. Lavie Y, Cao H, Volner A et al (1997) Agents that reverse multidrug resistance, tamoxifen, verapamil, and cyclosporin A, block glycosphingolipid metabolism by inhibiting ceramide glycosylation in human cancer cells. J Biol Chem 272:1682–1687

446

Morjani and Madoulet

95. Lucci A, Giuliano AE, Han TY et  al (1999) Ceramide toxicity and metabolism differ in wild-type and multidrug-resistant cancer cells. Int J Oncol 15:535–540 96. Nicholson KM, Quinn DM, Kellett GL, Warr JR (1999) Preferential killing of multidrugresistant KB cells by inhibitors of glucosylceramide synthase. Br J Cancer 81:423–430

97. Cabot MC, Giuliano AE, Volner A, Han TY (1996) Tamoxifen retards glycosphingolipid metabolism in human cancer cells. FEBS Lett 394:129–131 98. Pommerenke E, Mattern J, Volm M (1994) Modulation of doxorubicin-toxicity by tamoxifen in multidrug-resistant tumor cells in  vitro and in vivo. J Cancer Res Clin Oncol 120:422–426

Chapter 20 Overcoming Multidrug Resistance by RNA Interference Alexandra Stege, Andrea Krühn, and Hermann Lage Abstract The ATP-binding cassette (ABC)-transporter P-glycoprotein (Pgp, also known as ABCB1) is the best characterized factor involved in multidrug resistance (MDR) of cancer cells. Pgp, which is encoded by the MDR1 gene, acts as a membrane-embedded drug extrusion pump for multiple structurally unrelated cytotoxic drugs. Inhibition of the pump activity of Pgp by low-molecular weight pharmacologically active compounds as a method to reverse MDR in cancer patients has been studied extensively, but so far clinical trials have generally been disappointing. Thus, experimental strategies for overcoming MDR are under investigation. These approaches include the application of the RNA interference (RNAi) technology. RNAi is a physiological mechanism triggered by small double-stranded RNA molecules resulting in a sequence-specific gene-silencing. Besides its potential for development of novel therapeutics, RNAi also offers the possibility for specific inhibition of cellular targets in functional investigations. For specific inhibition of Pgp by triggering the RNAi pathway, transient gene-silencing by application of small interfering RNA (siRNA), and stable inhibition by transfection of MDR cancer cells with short hairpin RNA (shRNA) encoding expression cassettes encoded on plasmid DNA are described. Efficacy of RNAi on MDR1 mRNA expression level is determined by quantitative real-time RT-PCR and Northern blot. The consequences of RNAi on protein expression level are measured by Western blot and immunohistochemistry. The effects on the drug extrusion activity are measured by a drug accumulation assay based on flow cytometry, and reversal of the drug-resistant phenotype by assessment of drug-specific IC50-values by a cell proliferation assay based on colorimetry. Key words: Multidrug resistance, Cancer, RNAi, siRNA, shRNA, Gene therapy

1. Introduction A decade ago, the RNA interference (RNAi) mechanism was initially characterized in a model organism, the nematode Caenorhabditis elegans (1). In particular, following the discovery that the RNAi pathway can be triggered in mammalian cells in response to double-stranded small interfering RNA (siRNA) of ~21  nt in length (2), RNAi technology entranced to many bio-medical research laboratories worldwide, and started to J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_20, © Humana Press, a part of Springer Science + Business Media, LLC 2010

447

448

Stege, Krühn, and Lage

replace alternative gene-silencing techniques such as antisense- and ribozyme methodologies. Consequently, the RNAi platform is now a widely distributed, well established technology for highthroughput analyses as well as for functional studies in  vitro. Above and beyond its potential as powerful laboratory tool, RNAi technology also offers the possibility for design of novel therapeutics targeting any gene product of interest. Cancer is a complex disorder where various cellular pathways composed of multiple interacting factors are altered in normal function. Thus, above all the utilization of the RNAi pathway provides great opportunities for cancer research as well as for the development of novel targeted anticancer agents (3). Besides factors involved in oncogenesis and apoptosis-regulating pathways, regulation of cell cycle, cell senescence, and in tumor–host interactions, also proteins associated with resistance against antineoplastic agents such as ABC-transporters like MDR1/Pgp are specially suited as target using RNAi technology. 1.1. RNAi Pathway

Naturally, RNAi is a posttranscriptional gene silencing mechanism. Although homologous proteins of the RNAi machinery were also identified in prokaryotic organisms, RNAi appears to be restricted to eukaryotae. The RNAi pathway (detailed overviews of the biochemical features of RNAi in (4–6)) is initiated by processing of long regulatory double-stranded RNA (dsRNA) into approximately 21–28  nt siRNAs by Dicer, a endoribonuclease-III-like enzyme. Dicer’s cleavage activity is assisted by the dsRNA binding protein TRBP. These physiologically produced siRNAs have symmetric 2–3 nt 3¢-overhangs, 3¢-hydroxyl, and 5¢-phosphate residues. In the case of experimental or potential therapeutic utilization of the RNAi pathway, endoribonucleolytic digestion by Dicer is merely necessary for the intracellular generation of biologic active siRNA from short hairpin RNA (shRNA) encoded by DNA expression cassettes (Fig. 20.1). In a concerted reaction, siRNAs are loaded into the RNAinduced silencing complex (RISC). Prerequisite for integration of siRNAs into RISC is the phosphorylation at the 5¢-end of the RNA molecules. Unwinding of the siRNA duplex activates RISC. The siRNA’s antisense strand guides the complex to the homologous target mRNA where the endoribonucleolytic cleavage by Argonaute-2 (Slicer), a nuclease residing within the RISC complex, is catalyzed. Cleavage of the target mRNA occurs at a defined site 10 nt from the 5¢-phosphate of the antisense strand within the center of the duplex region between the target molecule and the siRNA strand. The cleaved target mRNA is no longer protected against endogenous RNases and can be degraded due to the loss of the 7-methylguanine cap structure of the 3¢-cleavage product and the loss of the poly(A)-tail of the 5¢-product. Besides triggering the RNAi pathway, siRNAs can

Overcoming Multidrug Resistance by RNA Interference

449

Fig.  20.1. Construction of an anti-MDR1/Pgp shRNA expression system. The shRNA-encoding DNA consists of BbsIspecific 5¢- and 3¢-assymetric cohesive overhangs that are not compatible, of the target sense and antisense sequence separated by a loop structure, and of the termination site composed by five thymidines (5 T). The chemically synthesized shRNA-encoding DNA oligonucleotide is cloned into the shRNA expression vector psiRNA-hH1zeo. Digestion of psiRNAhH1zeo with the restriction endonuclease BbsI results in lost of a region encoding an active ß-galactosidase (LacZ) producing EM7-lacZ a-peptide. Finally, the vector drives the synthesis of a shRNA which is intracellular processed into the corresponding siRNA.

also be involved in inhibition of mRNA translation and were demonstrated to induce transcriptional repression through RNA-directed DNA methylation (6). 1.2. Application of the RNAi Pathway as Laboratory or Therapy Tool

RNAi-mediated gene-silencing in mammalian cells including human cancer cells can be achieved by transfection of chemically or enzymatically synthesized siRNA molecules with liposomes or electroporation. On the other hand, RNAi can be triggered by gene transfer using DNA vectors containing expression cassettes encoding siRNA single strands, or shRNAs which will be intracellularly processed into the corresponding siRNAs. Commonly RNA polymerase III-specific promoters are used, i.e., H1-RNA

450

Stege, Krühn, and Lage

promoter, U6-RNA promoter, or tRNA promoter tRNAVAL. The RNA-polymerase III-depending promoters have a defined start of transcription and a termination signal consisting of five consecutive thymidine residues (T5). Therewith, these promoters can be used to direct the synthesis of small RNA molecules of interest lacking a poly-adenosin tail. Cleavage of the RNA transcript at the termination site is after the second uridine. Thus, RNA-polymerase III promoters direct the synthesis of small RNAs that are similar to the ends of chemically synthesized siRNAs containing two 3¢-overhanging thymidines or uridines. In shRNAs, the sequence of interest consists of a 19-nt sequence homologous to the target mRNA, linked with a 5–11-nt spacer sequence to the reverse complement of the same 19-nt targetspecific sequence. The synthesized RNA transcript folds back to its complementary strand to form a 19-base pair shRNA molecule, which is then processed by Dicer to a corresponding siRNA and passed into the RNAi pathway. 1.3. Reversal of MDR by RNAi

For overcoming cancer multidrug resistance (MDR) various low-molecular weight pharmacologically active compounds designated as chemosensitizers or MDR modulators have been identified. The expectation was that these compounds inhibit cellular pathways involved in MDR and, therewith, guarantee the ongoing efficacy of administrated anticancer drugs. Predominantly, chemosensitizers were developed as inhibitors for ABC transporters, in particular for targeting Pgp (7). Although different generations of ABC transporter inhibitors were highly effective in cell culture models, so far all of them failed in clinical trials. Reasons for the absence of clinical benefit included the lack of pretherapeutic diagnostics of Pgp expression, no consideration of the activity of other mechanisms involved in MDR, the necessity of high inhibitor concentrations including unwanted side effects, and unpredictable pharmacokinetic interactions with the therapeutic anticancer agents. Consequently, efforts were made to develop alternative, less toxic, and more efficient strategies to overcome MDR. These endeavors included the development of novel therapeutic strategies based on RNAi technology (8). The first studies demonstrating the proof of principle that cancer MDR can be modulated by RNAi were simultaneously published in 2003 (9, 10). In both investigations, chemically synthesized siRNAs were applied for MDR modulation by transient down-regulation of Pgp expression. Following these original experiments, in many other studies RNAi-depended downregulation of Pgp or alternative ABC transporters involved in MDR were reported from various cell models of different cancer entities. In those studies, siRNAs as well as RNAi effectors-encoding DNA expression cassettes based on plasmids, viruses, or transposons were applied for in vitro as well as for in vivo studies (8).

Overcoming Multidrug Resistance by RNA Interference

451

2. Materials 2.1. Human Carcinoma Cell Lines and Cell Culture

1. Choose a suited MDR1/Pgp-positive cell line (see Note 1). 2. Modified L-15 medium: Leibovitz L-15 medium without glutamine (Lonza BioWhittaker) supplemented with 7.5 mg/l fetuin, 10% fetal calf serum, 0.05% (w/v) d(+)-glucose (45%), 1.0  mM l-glutamine, 80  IE/l H-insulin, 0.1125% (w/v) NaHCO3, 1× MEM-vitamins, and 2.5 mg/l transferrin. 3. Daunorubicin hydrochloride (Daunoblastin®; Pfizer) is dissolved in tissue-culture water at 2 mg/ml, stored in aliquots at −20°C, and then added to tissue culture flasks as required. 4. 10× Phosphate-Buffered Saline (PBS), pH 7.4 (Gibco). 5. Solution of trypsin (0.5%)/ethylene diamine tetraacetic acid (EDTA; 0.2%) in PBS, 10× (Biochrom).

2.2. Transient MDR1/ Pgp Silencing by siRNAs 2.2.1. Annealing of Sense and Antisense siRNA Oligo Pairs 2.2.2 Transient Transfection of MDR Cancer Cells with siRNAs

1. Nuclease-free water. 2. 5× annealing buffer (Final buffer concentration is: 100  mM potassium acetate, 30  mM HEPES-KOH pH 7.4, 2  mM magnesium acetate).

1. Suited cell culture medium. For EPG85-257RDB cells: modified L-15 medium (see Subheading 2.1). 2. Oligofectamine™ Reagent (Invitrogen). 3. Serum-free medium: Opti-MEM® with GlutaMax™ I (Invitrogen). 4. 10× PBS pH 7.4 (Gibco).

2.3. Stable MDR1/Pgp Silencing by shRNAEncoding Plasmids 2.3.1. Construction of Plasmid Expression Vectors Encoding shRNAs 2.3.2. Stable Transfection of MDR Cancer Cells with shRNA Vectors

1. Plasmid: psiRNA-hH1zeo G2 (InvivoGen). 2. Bacteria: GT116; genotype: F¯ mcrA D(mrr-hsdRMS-mcrBC) F80lacZDM15 DlacX74 recA1 endA1 Ddcm DsbcC-sbcD (InvivoGen). 3. Media: Fast-Media Zeo X-Gal agar and Fast-Media Zeo TB (InvivoGen). 1. Suited cell culture medium. For EPG85-257RDB cells: modified L-15 medium (see Subheading 2.1). 2. Serum-free medium: Opti-MEM® with GlutaMax™ I (Invitrogen). 3. 10× PBS, pH 7.4 (Gibco). 4. SuperFect® Transfection Reagent (Qiagen). 5. 10  mM Tris(hydroxymethyl)aminomethane hydrochloride (Tris–HCl).

452

Stege, Krühn, and Lage

2.4. Quantitative Real-Time RT-PCR

1. LightCycler FastStart DNA Master SYBR Green I (Roche Diagnostics). 2. Light cycler capillaries (Roche Diagnostics). 3. Light cycler instrument (Roche Diagnostics). 4. Oligonucleotide primer (see Note 2). 5. RelQuant Software (Roche Diagnostics).

2.5. Northern Blot

1. BioMax MR-films (Eastman Kodak). 2. Exposure cassette. 3. Hybridization flasks. 4. Megaprime™ DNA Labelling System (Amersham Biosciences). 5. 25× MOPS buffer: 5 M 3-(N-morpholino)propanesulfonic acid (MOPS), 12.5 M sodium acetate, 0.25 M EDTA in ddH2O. 6. Nylontransfermembran Hybond–N+ (Amersham Biosciences). 7. PBS pH 7.4. 8. a-(32P)-dCTPs (see Note 3). 9. 2× RNA Loading Dye Solution (Fermentas). 10. RNeasy® Mini Kit (Qiagen). 11. 20× SSC, pH 7.0; 3 M NaCl, 0.3 M Na3-citrate in ddH2O. 12. Trans-Blot BD Semi-Dry Transfer Cell (Bio-Rad Laboratories). 13. UV-Crosslinker UVC1000 (Hofer). 14. Wash buffer I: 2× SSC, 0.1% SDS in ddH2O. 15. Wash buffer II: 0.1× SSC, 0.1% SDS in ddH2O. 16. Whatman paper.

2.6. Western Blot 2.6.1. Membrane Protein Isolation

1. 10 mM Tris–HCl, pH 8.8. 2. Proteinase inhibitor cocktail tablets (Roche Diagnostics). 3. Phosphate buffered saline (1× PBS). 4. 4× sample buffer: 60 mM 1 M Tris–HCl pH 6.8, 100 mM dithiothreitol (DTT), 2 ml 20% SDS, 10% glycerol, crumb of brome phenol blue.

2.6.2. SDS-PAGE

1. Acrylamid-Bis-Acrylamid (19:1) (Qbiogene). 2. Ammonium peroxodisulfate, APS. 3. Color marker (high range, 29–205 kDa; Sigma-Aldrich). 4. 2-propanol. 5. 10× running buffer: 144 g glycine, 10 g SDS, 30 g Tris-base in ddH2O. 6. 4× sample buffer: 160 mM Tris–HCl (pH 6.8), 100 mM DTT, 10% (v/v) SDS, 2% (v/v) glycine and brome phenol blue.

Overcoming Multidrug Resistance by RNA Interference

453

7. Sodium dodecyl sulfate, SDS. 8. N,N,N¢,N¢-Tetramethylethylendiamin, TEMED. 9. 1× Transfer buffer: 150  mM glycin, 20% (v/v) methanol, 25 mM Tris-base. 2.6.3. Blotting

1. Color marker high range 29–205 kDa (Sigma-Aldrich). 2. ECL™ Western Blotting Analysis System (Amersham Biosciences). 3. Hypolysis buffer: 1 M Tris–HCl, pH 8.0, proteinase inhibitor cocktail tablets (Roche Diagnostics). 4. ImmunoPure® goat anti-Mouse IgG, (H + L), peroxidise conjugated (Perbio Science). 5. Monoclonal antibody against MDR1/Pgp, C219 (Alexis Biochemicals). 6. Mouse anti-actin monoclonal antibody C4 (Chemicon). 7. Skim milk (Difco Laboratories). 8. Stripping buffer: 100 mM trisodium citrate pH 2.2. 9. SuperSignal® West Pico chemiluminescent substrate (Perbio Science). 10. 1× TBS, pH 7.4–7.6: 150  mM NaCl, 7  mM Tris-base, 43 mM Tris–HCl in ddH2O. 11. 1× TBST, pH 7.4–7.6: 1× TBS with 0.05% (v/v) Tween 20. 12. Tween® 20, Polysorbate 20.

2.7. Immunohisto­ chemistry

1. Methanol/Acetone mixture 1:1. 2. DakoCytomation LSAB+ System-HRP (Dako). 3. Hydrogen peroxide (H2O2). 4. Mayer’s Hematoxylin (Dako). 5. Monoclonal antibody against MDR1/Pgp, C219 (Alexis Biochemicals). 6. Mouse anti-actin monoclonal antibody C4 (Chemicon® International, Inc.). 7. NovaRed (Vector Laboratories). 8. 10× PBS pH 7.4 (Gibco). 9. Primary Mouse Negative Control (Dako).

2.8. Drug Accumulation Assay 2.9. Cell Proliferation Assay

1. Daunorubicin hydrochloride (Daunoblastin®; Pfizer). 2. 10× PBS, pH 7.4 (Gibco). 1. Acetic acid. 2. 10× PBS, pH 7.4 (Gibco).

454

Stege, Krühn, and Lage

3. Sulforhodamine B (SRB) (Sigma-Aldrich). 4. Tris(hydroxymethyl)aminomethane hydrochloride (Tris–HCl). 5. Daunorubicin hydrochloride (Daunoblastin®; Pfizer).

3. Methods In this section, the experimental strategy for silencing of MDR1/ Pgp is described. Several cancer cell lines with over-expression of MDR1/Pgp or alternative ABC transporters are available. Following treatment of these cell models with RNAi-mediating agents, the biological effects can be analyzed on different levels. RNAi effects on mRNA expression can be determined by quantitative real-time RT-PCR or Northern blot, effects on protein expression by Western blot, or immunohistochemistry. On functional level, the inhibition of ABC transporters can easily be measured by a flow cytometry-based drug accumulation assay. The consequences on the drug resistance phenotype can be measured by a colorimetric cell proliferation assay. For synthesis of siRNAs, many suppliers offer chemically synthesized siRNAs of good quality. Examples of suppliers are Thermo Fisher Scientific, Waltham, MA USA; Invitrogen, Carlsbad CA, USA; Eurogentec, Seraing, Belgium. 3.1. Transient MDR1/ Pgp Silencing by siRNAs 3.1.1. Selection of siRNA Sequences 3.1.2. Annealing of Sense and Antisense siRNA Strands

The first step in preparation of RNAi-based experiments is the design of the RNAi effectors, i.e., siRNAs (see Note 4). 1. Dilute each RNA strand in RNase-free water to a concentration of 50 mM. 2. Take 30 ml of each RNA strand and combine it with 15 ml of 5× annealing buffer. 3. Heat the solution for 1 min at 90°C; centrifuge the tube for 15 s. 4. Incubate at 37°C for 1 h. 5. The final concentration of the siRNA duplexes is now 20 mM. 6. Store aliquots at −20°C. Multiple freeze-thawed cycles should be avoided.

3.1.3. Transient Transfection of MDR Cancer Cells with siRNAs

1. Seed 5 × 105 exponential growing cells per well in 6-well plates out, let them grow for 24 h (see Note 5). 2. The next day dilute for each well 10 ml of the siRNA duplexes (20 mM) in 175  ml medium without serum (Opti-MEM) ⇒ Solution 1.

Overcoming Multidrug Resistance by RNA Interference

455

3. For each well in a 6-well plate, dilute 3 ml of Oligofectamine reagent in 12 ml Opti-MEM. Allow diluted reagent to sit for 5–10 min ⇒ Solution 2. 4. Combine Solution 1 and 2, mix gently, and incubate at room temperature for 20 min ⇒ siRNA Complex Solution. 5. Wash cells once with 1× PBS. Add 800 ml Opti-MEM to each well containing cells. 6. Mix gently and add 200 ml of the siRNA Complex Solution onto the cells. The final siRNA duplex concentration in each well will be 200 nM. 7. Incubate the cells for 4 h at 37°C in a CO2 incubator. 8. After incubation, add growth medium containing three times the normal concentration of serum without removing the transfection mixture. 9. Examine the silencing at different time points, e.g., 1–7 days, after transfection (see Note 6). An example of the results produced is shown in Fig. 20.2. 3.2. Stable MDR1/Pgp Silencing by shRNAEncoding Plasmid DNA 3.2.1. Construction of Plasmid Expression Vectors Encoding shRNAs

1. Select the target sequences for shRNA as described in Subheading  3.1.1 or construct shRNA expression vector on the basis of proven chemically synthesized siRNAs (see Note 7) (Fig. 20.1). 2. Two suited homologous single-stranded DNA (ssDNA) molecules were chemically synthesized. 3. Anneal the ssDNA molecules by incubation of 2 ml of each complementary ssDNA oligonucleotide (25 mM) in 6 ml 0.5 M NaCl in a total volume of 20 ml. 4. Incubate the annealing mixture at 80°C for 2 min followed by cooling to 35°C. 5. Clone the annealed dsDNA, consisting of the anti-MDR1 sense sequence, a 5-nt 3¢-CCACC-5¢ spacer sequence, the anti-MDR1 antisense sequence, and BbsI-specific 5¢-overhangs, into the BbsI restriction site of the expression vector psiRNA-hH1zeo (InvivoGen, San Diego, CA). 6. Transform the shRNA encoding expression vector in the E. coli strain GT116 (see Note 8). 7. Confirm the correct insertion of the specific shRNA encoding DNA molecules by restriction and sequencing.

3.2.2. Stable Transfection of MDR Cancer Cells with shRNA-Encoding Plasmid Vectors

1. Perform the experiment in 50–60% confluent 6-well plates. 2. For each well mix 2 mg of plasmid DNA (including controls, see Note 9) with 65 ml serum-free medium Opti-MEM and adjust the solution with 10 mM Tri–HCl (sterile) to a volume of 75 ml.

456

Stege, Krühn, and Lage

Fig. 20.2. (a) Time kinetics of siRNA-mediated decrease of MDR1 mRNA expression by Northern blot and cellular MDR1/ Pgp content by Western blot in human gastric carcinoma cells. As controls, the multidrug-resistant cell line EPG85257RDB was treated with single sense and single antisense siRNA strands. Furthermore, it has to be demonstrated that the transfection procedure without RNA molecules has no effects on the gene expression level (not shown in the figure). For comparison, the parental cell variant EPG85-257P is included in the experiments. (b) Growth curves for assessment of IC50-values to daunorubicin following treatment of EPG85-257RDB cells with anti-MDR1 siRNAs, single sense and single antisense siRNA strands in comparison to the parental cell line EPG85-257P. It has to be noticed that the x-axis shows a logarithmic scale. Accordingly, the reversal of drug resistance by anti-MDR1 siRNAs is more than 50%.

3. Add 10 ml of the transfection reagent SuperFect, pipette five times up and down and incubate for 10  min at room temperature. 4. In the meantime wash the cells with 1× PBS. 5. After incubation add 500 ml Opti-MEM and transfer the transfection complex to each well. One well, which serves as

Overcoming Multidrug Resistance by RNA Interference

457

a transfection control, has to be incubated with the transfection complex without plasmid DNA. 6. Incubate the cells for 3 h at 37°C in a CO2 incubator. 7. Control the cells in the microscope and wash the them twice with 1× PBS and let them grow in suited serum-containing cell culture medium (for EPG85257RDB cells modified L-15 medium) without any further antibiotics or additives. 8. 24 h later change the medium into cell culture medium containing selection agents (see Note 10). 9. When the cells in the control well have been died, expand the cells from each well to multiple 6 cm2 dishes. 10. After 3 weeks, pick visible clones in 12 wells, and finally transfer them to regular cell culture flasks. 11. Prescreening of the clones for gene-silencing activity by Northern blot analyses or real-time RT-PCR. 3.3. Real-Time RT-PCR

For quantitative mRNA expression analysis, a real-time RT-PCR protocol is suited. There are different well established real-time RT-PCR methodologies available. Here, the application of a LightCycler instrument and PCR product detection by SYBR Green I fluorescent dye is depicted. The reaction mixtures were composed as described in manufacturer’s manual (Roche Diagnostics). 1. Isolate total cellular RNA by using a suitable RNA Isolation Kit (e.g., RNeasy Mini Kit, Qiagen). Confluence of the cells should be 50–70%. 2. For cDNA synthesis, perform reverse transcription of RNA by using the SuperScript First-Strand Synthesis System for RT-PCR (Invitrogen). 3. Produce an appropriate amount cDNA of any mRNA to last for all following real-time PCRs (5–10 mg) as a calibrator. 4. Perform a dilution series (107–100) of any cDNA containing the target gene sequence of MDR1/Pgp and the reference gene sequence of aldolase (see Note 11). 5. Establish standard curves for the target gene and the reference gene by performing a real-time PCR. Utilize each dilution of the dilution series, the calibrator, and a nontemplate control as duplicates. 6. Perform a real-time PCR to quantify the cDNA of interest in the samples using target gene primers for one run and reference gene primers for a second run to normalize the samples (see Note 2). 7. Use the RelQuant Software to evaluate the results of the realtime PCR runs.

458

Stege, Krühn, and Lage

3.4. Northern Blot

1. Isolate total cellular RNA by using a suitable RNA isolation kit (e.g., RNeasy Mini Kit, Qiagen). 2. Separate 10 mg of total cellular RNA on a 1% agarose gel containing 6% formaldehyde. 3. Transfer RNA to a nylon-membrane using reverse capillary blotting as described by Zhou et al. (11). 4. Perform a standard RT-PCR using cDNA from a cancer cell line expressing MDR1/Pgp using MDR1-fw and MDR-1 rev primers (see Note 2) to amplify the target sequence. 5. Label PCR-generated MDR1-specific cDNA probe with (32P) using a MegaprimeTM DNA Labelling System according to manufacturer’s instructions. 6. Prehybridize blot-membrane for 30 min at 58°C. 7. Hybridize (32P) dCTP-labeled MDR1/Pgp-specific and (32P) dCTP-labeled aldolase (loading control) cDNA fragments to blot-membrane overnight at 58°C. 8. Wash blot-membrane twice with wash buffer I for 30 min at 21°C and twice with wash buffer II for 30 min at 55°C. 9. Develop BioMax MR-film over night at −80°C.

3.5. Western Blot 3.5.1. Membrane Protein Isolation

Isolate membrane proteins by the method described by Dieckmann-Schuppert and Schnittler (11): 1. Cultivate cancer cells in 10-cm petri dishes to a confluence of 75%. All following steps on ice. 2. Dissolve 1 Protease inhibitor cocktail tablet in 2 ml ddH2O (=proteinase inhibitor solution). 3. Mix 19.2 ml 10 mM Tris–HCl and 0.8 ml proteinase inhibitor solution (=hypolysis buffer). 4. Remove cell culture medium of the cultivated cancer cells. 5. Wash cancer cells twice with 1× PBS. 6. Add 3 ml hypolysis buffer to cancer cells and incubate 5 min. 7. Scrape off cancer cells from petri dish with cell scraper. 8. Homogenize cancer cells in a homogenizer by moving the pistil up and down 8–12 times. 9. Transfer cancer cell lysate into centrifuge tube. 10. Centrifuge cancer cell lysate at 1,000 g and 4°C for 10 min. 11. Transfer supernatant into an ultracentrifugate tube. 12. Centrifuge supertanant at 150,000 g and 4°C for 30 min. 13. Dissolve pellet in 50–100 ml 2× sample buffer. 14. Transfer protein suspension into new Eppendorf tube. 15. Incubate protein suspension at 95°C for 10 min.

Overcoming Multidrug Resistance by RNA Interference

459

16. Centrifuge protein suspension at 16,000 g for 5 min. 17. Transfer supertanant containing membrane proteins into new Eppendorf tube. 18. Determine protein concentration. 19. Store isolated membrane proteins at −80°C. 3.5.2. SDS-PAGE

1. The following instructions assume to the use of the electrophoresis chamber Mini Protean II (BioRad). 2. Clean glass plates with ddH2O and ethanol. 3. Assemble glass plates in casting stand. 4. Prepare the 7.5% Separating-solution (3.75  ml acrylamide/ bis-acrylamide (19:1), 200 ml 10% SDS, 7.5 ml 1 M Tris–HCl, pH 8.8). 5. Add 150 ml 10% APS and 15 ml TEMED and immediately pipette into gel to about 2.0 cm below the front of the small glass plate. 6. To remove air bubbles put a thin layer of 2-propanol on the gel. 7. After polymerization (~20 min), pour off 2-propanol. 8. For the 4% stacking gel mix 1 ml acrylamide/bis-acrylamide (19:1), 100 ml 10% SDS, 2.5  ml 0.5  M Tris–HCl, pH 6.8, 75 ml 10% APS, and 15 ml TEMED. 9. Immediately pipette the stacking solution onto gel until flash with top edge of small glass plate. 10. Add the comb and make sure that it’s centered. Allow stacking gel to polymerize. 11. After polymerization, remove comb, fill syringe and needle with 1× running buffer. 12. Meanwhile, add 2× sample buffer to protein samples, mix, boil at 95°C for 5 min. 13. Load molecular weight marker and the samples onto the wells. 14. Place gel in electrode assembly; place gel into tank and fill with 1× running buffer. 15. Attach to electrode assembly/tank to power supply. Run at 63 V for 30 min and thereafter 95 V until you can see loading buffer reach bottom edge of separating gel. 16. Upon completion of gel run, disassemble. Carefully remove gel from between the glass plate. 17. Remove the stacking gel and equilibrate the gel in 1× transfer buffer.

460

Stege, Krühn, and Lage

3.5.3. Blotting

1. After separating 20 mg of membrane proteins by SDS-PAGE, transfer proteins to a 0.2 mm cellulose nitrate membrane by a semi-dry electro-blotting procedure. 2. For blocking incubate the blot-membrane in 5% skim milk in 1× TBS over night at 4°C. 3. Incubate blot-membrane with mouse mAb C219 directed against MDR1/Pgp (1:100) and with mouse mAb C4 directed against actin (1:5,000) as loading control in 5% skim milk in 1× TBST for 2 h (see Note 12). 4. Incubate blot-membrane with peroxidase-conjugated mouse anti-rabbit IgG (1:10,000) in 1× TBST for 1 h. 5. Wash blot-membrane four times with 1× TBST (10 min) and twice with 1× TBS (5 min). 6. Use SuperSignal West Pico Chemiluminescent Substrate (Perbio Science). 7. Expose nitrate membrane to Hyperfilm ECL (Amersham) according to the manufacturer’s instructions.

3.6. Immunohisto­ chemistry

1. Seed MDR cancer cells on cell culture dishes containing slides and culture them until they reached ca. 70% confluence (2–3 days). 2. Wash the slides once with 1× PBS and fix the cells with a methanol/acetone (1:1) mixture for 10 min at −20°C. 3. For blocking the endogenous peroxidise activity, incubate the air-dried slides in 3% H2O2 for 5 min. 4. Perform the immunological staining reaction by using the mouse mAb C219 directed against MDR1/Pgp in a 1:100 dilution for 18 h at 4°C (see Note 13). 5. Perform in parallel control reactions by the substitution of specific antibody with a primary mouse negative control antibody. 6. Subsequently, incubate the slides consecutively in biotinylated antibodies (15 min, at room temperature), streptavidin-biotinylated peroxidase complex (15  min, at room temperature), LSAB+ and HRP. 7. Use as chromogen, e.g., NovaRed (10  min, at room temperature). 8. Counterstain and dehydrate the preparations with Mayer’s haematoxylin. 9. Evaluate the slides in a microscope.

3.7. Cell Proliferation Assay

For assessment of the level of drug resistance determine IC50 values in a cell proliferation assay based on SRB (13), a dye staining the cellular proteins which are corresponding to the numbers of cells (see Note 14).

Overcoming Multidrug Resistance by RNA Interference

461

1. Seed 500–750 MDR cancer cells in 100 ml cell culture medium per well in 96-well plates (see Note 15). 2. Cultivate seeded cells for 48 h in a CO2 incubator. 3. Add different concentrations of daunorubicin in a dilution series. Each concentration should be represented in triplicates (see Note 16). 4. Cultivate daunorubicin-treated cells for 5 days. 5. Fix cancer cells using 200 ml of 10% trichloroacetic acid for 2 h at 4°C. 6. Wash cancer cells five times with tap water. 7. Stain cellular proteins using 0.4% (w/v) SRB in 1% acetic acid at room temperature for 10 min. 8. Wash stained cells five times with 1% acetic acid. 9. Dry 96-well plates over night. 10. Add 300 ml 20 mM Tris-Base pH 10.0 to elute protein-SRB complexes. 11. Perform a photometric determination of the degree of staining intensity (see Note 17). 12. To calculate the IC50 values set the absorbance value of untreated cells of the same 96-well plate as the treated cells under investigation to 100% (see Note 18). 3.8. Drug Accumulation Assay of MDR Cancer Cells

1. Treat exponentially growing cancer cells with 10 mM (5.8 mg/ ml) of daunorubicin for 1 h (see Note 19). 2. Wash the cells with ice-cold 1× PBS, trypsinize and harvest them by centrifugation. 3. Resuspend the cell pellet in 500 ml 1× PBS. 4. Determine the intracellular daunorubicin content by flow cytometry, e.g., using a FACScan (Calibur 750; BectonDickinson, San Jose, CA) (see Note 20). 5. Analyze a minimum of 104 cells for each sample. 6. Use at least three independent experiments in duplicate to assess a geometric mean (see Note 18).

4. Notes 1. In the past, many MDR1/Pgp-positive cell models were established from drug-sensitive cancer cells by in  vitro exposure to drugs that are substrates of this pump protein, e.g., anthracyclines. The advantage of these models is that the

462

Stege, Krühn, and Lage

degree of drug resistance and its modulation by RNAi can be directly compared with the corresponding parental cells as a control. Examples for well-known MDR1/Pgp-positive cell lines established by this procedure are the CML cell line K562/A02, the gastric carcinoma cell line EPG85-257RDB, the pancreatic carcinoma cell line EPP85-181RDB, or MCF-7/ AdrR cells originally described to be derived from the breast cancer cell line MCF-7, but meanwhile identified as derived from OVCAR-8 ovarian adenocarcinoma cells. Thus, these cells were later redesignated as NCI/ADR-RES (14). To maintain the drug-resistant phenotype, cell culture medium for the MDR1/Pgp-expressing cell lines should be supplemented with appropriate concentration of the selection agent. For the experiments described here, the human gastric carcinoma cell line EPG85-257P and its multidrug-resistant subline EPG85257RDB (15) overexpressing the ABC-transporter MDR1/ Pgp were used. Both cell variants were cultured in modified Leibovitz L-15 medium, whereby EPG85-257RDB cells were supplemented with 2.5 mg/ml daunorubicin. 2. Oligodeoxynucleotide primers used for amplification of the MDR1-specific template were MDR-fw: 5¢-GCC CTT GGA ATT ATT TCT TT-3¢ and MDR-rev: 5¢-TGG GTG AAG GAA AAT GTA AT-3¢; as control oligodeoxynucleotide primers specific for aldolase Ald-fw: 5¢-ATC CTG GCT GCA GAT GAG TC-3¢, and Ald-rev: 5¢-GCC CTT GTC TAC CTT GAT GC-3¢ are useful. Reverse transcription can be performed with SuperScript II enzyme (Gibco BRL) using arbitrary hexamers. Cycling conditions for MDR1 (aldolase) are as follows: initial enzyme activation at 95°C for 10 min, followed by 45 cycles at 94°C (95°C) for 15 s, 50°C (54°C) for 5  s, and 72°C for 10  s. All cycling reactions should be performed in the presence of 4  mM MgCl2. Gene-specific fluorescence should be measured at 84°C (86°C) and confirmed by melting curve analysis. 3. a-(32P)-dCTP, 10 mCi/ml. 4. A nice overview of history, mechanism, and current recommendations for the selection of siRNA target sites is available at the homepage of Thomas Tuschl, a pioneer in this field (http://www.rockefeller.edu/labheads/tuschl/). Further information and commercial providers of siRNAs are listed below: General Information: http://www.rnaiweb.com/ RNAi/siRNA_Design/; Design Tools: (1) http://www1. qiagen.com/Products/GeneSilencing/CustomSiRna/ SiRnaDesigner.aspx; (2) http://www.invitrogen.com/site/ us/en/home/Products-and-Services/Applications/RNAiEpigenetics-and-Gene-Regulation/RNAi.html?cid=invggl 123000000007882s; (3) http://www.dharmacon. com/DesignCenter/DesignCenterPage.aspx. For transient

Overcoming Multidrug Resistance by RNA Interference

463

gene-silencing experiments described here, siRNAs synthesized by Dharmacon (now Thermo Fisher Scientific, Waltham, MA USA) were used. A highly effective siRNA sequence was 5¢-AAG AAG GAA AAG AAA CCA ACU-3¢, homologous to nt 503–523 of the MDR1 mRNA consensus sequence NM_000927. The desalted oligonucleotides contained 3¢dTdT extensions and have been converted to the 2¢-hydroxyl form (2¢-OH deprotection) (see Note 4). A detailed discussion of these experiments was published by Nieth et al. (9). An overview of various siRNA sequences used for silencing of the MDR1 gene was published by Lage (8). 5. The exact cell number depends on the cell model, on the duplication rate, and the transfection efficiency. The instructions depicted here concern to the cell line EPG85-257RDB. Thus, experiments with different dilution series should be performed in advance. 6. Down regulation of the ABC-transporter protein depends on the biological half-life time. The MDR1 mRNA was found to have a biological half-life of approximately 4 h, and the corresponding ABC-transporter protein MDR1/Pgp exhibits a half-life of approximately 16 h (16). 7. Own experience and experience by others working in the field show that high effective siRNA target sites are in 50–75%, which are also highly effective for shRNAs. A detailed discussion of these experiments was published by Stege et al. (17). 8. In the experiments described here, the plasmid psiRNAhH1zeo (InvivoGen) (Fig. 20.1) was used for the construction of the anti-MDR1 shRNA-encoding expression vector. There are also well established shRNA expression systems available from alternative suppliers like Ambion, Clontech, or Invitrogen. 9. Before starting the experiment, the optimal amount of plasmid DNA for the best transfection efficiency in a given MDR cancer cell line has to be determined empirically. In the experiments described here, the “classical” multidrugresistant gastric carcinoma cell line EPG85-257RDB was transfected with 2 mg of the anti-MDR1 shRNA-encoding plasmid, or with 2 mg control plasmid DNA, i.e., with an antiluciferase shRNA-encoding vector and with the original psiRNA-hH1zeo plasmid containing an active ß-galactosidase (LacZ) producing EM7-lacZ a-peptide cassette instead of a shRNA-specific sequence. 10. The selection agent depends on the resistance marker expressed by the used expression vector. Furthermore, it should be taken into consideration, that the selection agent is not a substrate transported by MDR1/Pgp or an alternative ABC-transporter. In the experiments described here,

464

Stege, Krühn, and Lage

the cells were incubated with modified Leibovitz L-15 medium containing 400 mg/ml zeocin. 11. The dilution series is performed using cDNA from the gastric carcinoma cell line EPG85-257RDB expressing MDR1/ Pgp. The target sequence is homologous to nt 3,052– 3,072 of MDR1 mRNA consensus sequence NM_000927. Oligonucleotide primer sequences are mentioned in Note 2. 12. Best results are achieved by incubation of the blot membrane with each primary antibody separately. After detecting MDR1/ Pgp with the primary antibody mAb C219 and the secondary peroxidase-conjugated mouse anti-rabbit IgG antibody, strip the blot membrane by incubation with stripping buffer twice for 1  h at 4°C and once for 15  min at room temperature. Wash 4× 10 min with TBS and block the blot membrane with 5% skim milk in TBS over night again before incubation with mAb C4 directed against actin. 13. To avoid high, nonspecific background staining dilute the antibody in Antibody Diluent Background Reducing Solution (Dako). 14. The IC50-value is the drug concentration inhibiting cell growth to 50% of those of the control without drug exposure. 15. Depending on the cell line the number of seeded cells may differ. The instructions depicted here concern to the cell line EPG85-257RDB. It is important to fill the outer wells as well with fluid to avoid evaporation of cell culture medium of cell-containing wells. 16. Depending on the level of resistance of the cancer cells against daunorubicin, an appropriate range of concentrations should be applied. Parental gastric carcinoma cell line EPG85-257P was treated with 0.01–35  nM and the multidrug-resistant gastric carcinoma cell line EPG85-257RDB was treated with 35–3,500 nM of daunorubicin. The range of concentrations used may differ using other cell lines. 17. Wavelength for detection of SRB is 564 nm. 18. In cell proliferation assays and anthracycline accumulation studies, levels of statistical significance can be evaluated by calculation of the two-tail P-values by performing the unpaired t-test using the Prism software (GraphPad Software). 19. In the experiments described here, 2.5 × 105 cells of the cancer cell lines EPG85-257RDB or EPG85-257P were seeded in 6-well plates. After 48 h, the cells were incubated with 1 mM (580 ng/ml) daunorubicin for 60 min at 37°C. 20. Wavelength for detection of daunorubicin is 480  nm for excitation and 550 nm for emission.

Overcoming Multidrug Resistance by RNA Interference

465

Acknowledgments Own experiments for overcoming cancer MDR by RNAi were supported by grants LA 1039/2-1, LA 1039/2-3 and LA 1039/5-1 of the “Deutsche Forschungsgemeinschaft” (DFG), and by the “RNA-network” funded by the “Bundesministerium für Bildung und Forschung” (BMBF) and Berlin as well as by grant no. 01GU0615 of the BMBF. References 1. Fire A, Xu S, Montgomery MK et al (1998) Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391:806–811 2. Elbashir SM, Harborth J, Lendeckel W et al (2001) Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 411:494–498 3. Lage H (2005) Potential applications of RNA interference technology in the treatment of cancer. Fut Oncol 1:103–113 4. Hannon GJ (2002) RNA interference. Nature 418:244–251 5. Dorsett Y, Tuschl T (2004) siRNAs: applications in functional genomics and potential as therapeutics. Nat Rev Drug Discov 3: 318–329 6. Martin SE, Caplen NJ (2007) Application of RNA interference in mammalian systems. Annu Rev Genomics Hum Genet 8:81–108 7. Szakács G, Paterson JK, Ludwig JA, BoothGenthe C, Gottesman MM (2006) Targeting multidrug resistance in cancer. Nat Rev Drug Discov 5:219–234 8. Lage H (2006) MDR1/P-glycoprotein (ABCB1) as target for RNA interferencemediated reversal of multidrug resistance. Curr Drug Targets 7:813–821 9. Nieth C, Priebsch A, Stege A, Lage H (2003) Modulation of the classical multidrug resistance (MDR) phenotype by RNA interference (RNAi). FEBS Lett 545:144–150 10. Wu H, Hait WN, Yang JM (2003) Small interfering RNA-induced suppression of

MDR1 (P-glycoprotein) restores sensitivity to multidrug-resistant cancer cells. Cancer Res 63:1515–1519 11. Zhou DC, Marie JP, Suberville AM, Zittoun R (1992) Relevance of mdr1 gene expression in acute myeloid leukemia and comparison of different diagnostic methods. Leukemia 6:879–885 12. Dieckmann-Schuppert A, Schnittler H (1996) A simple assay for quantification of protein in tissue sections, cell cultures, and cell homogenates, and of protein immobilized on solid surfaces. Cell Tissue Res 288:119–126 13. Skehan P, Storeng R, Scudiero D et al (1990) New colorimetric cytotoxicity assay for anticancer-drug screening. J Natl Cancer Inst 82:1107–1112 14. Liscovitch M, Ravid D (2007) A case study in misidentification of cancer cell lines: MCF-7/ AdrR cells (re-designated NCI/ADR-RES) are derived from OVCAR-8 human ovarian carcinoma cells. Cancer Lett 245:350–352 15. Lage H (2003) Molecular analysis of therapy resistance in gastric cancer. Dig Dis 21:326–338 16. Alemán C, Annereau JP, Linag XJ et al (2003) P-glycoprotein, expressed in multidrug resistant cells, is not responsible for alterations in membrane fluidity or membrane potential. Cancer Res 63:3084–3091 17. Stege A, Priebsch A, Nieth C, Lage H (2004) Stable and complete overcoming of MDR1/ P-glycoprotein-mediated multidrug resistance in human gastric carcinoma cells by RNA interference. Cancer Gene Ther 11:699–706

sdfsdf

Chapter 21 Circumventing Tumor Resistance to Chemotherapy by Nanotechnology Xing-Jie Liang, Chunying Chen, Yuliang Zhao, and Paul C. Wang Abstract Patient relapse and metastasis of malignant cells is very common after standard cancer treatment with surgery, radiation, and/or chemotherapy. Chemotherapy, a cornerstone in the development of present day cancer therapy, is one of the most effective and potent strategies to treat malignant tumors. However, the resistance of cancer cells to the drugs remains a significant impediment to successful chemotherapy. An additional obstacle is the inability of chemotherapeutic drugs to selectively target tumor cells. Almost all the anticancer agents have severe side effects on normal tissues and organs. The toxicity of currently available anticancer drugs and the inefficiency of chemotherapeutic treatments, especially for advanced stages of the disease, have limited the optimization of clinical drug combinations and effective chemotherapeutic protocols. Nanomedicine allows the release of drugs by biodegradation and self-regulation of nanomaterials in vitro and in vivo. Nanotechnologies are characterized by effective drug encapsulation, controllable self-assembly, specificity and biocompatibility as a result of their own material properties. Nanotechnology has the potential to overcome current chemotherapeutic barriers in cancer treatment, because of the unique nanoscale size and distinctive bioeffects of nanomaterials. Nanotechnology may help to solve the problems associated with traditional chemotherapy and multidrug resistance. Key words: Cancer chemotherapy, Drug resistance, Nanomedicine, Nanotechnology

1. Obstacles to Cancer Treatment and the Potential of Nanotechnology

Cancer remains one of the main causes of death in humans and thus great efforts have been undertaken to develop cancer treatments (1). Cancer cells are notorious in their resistance to chemotherapy in the clinic. In fact, an enormous body of research strongly suggests that drug-resistant cancer cells that remain alive after chemotherapy are responsible for the reappearance of tumors and the poor prognosis for patients. The occurrence of drug resistance leads to the failure of tumor treatment. This is a difficult obstacle to overcome, as tumor resistance mechanisms have various origins

J. Zhou (ed.), Multi-Drug Resistance in Cancer, Methods in Molecular Biology, vol. 596, DOI 10.1007/978-1-60761-416-6_21, © Humana Press, a part of Springer Science + Business Media, LLC 2010

467

468

Liang et al.

Fig. 21.1. Mechanisms of clinical multidrug resistance during anticancer therapy. A number of mechanisms might be responsible for pleiotropic drug resistance in the clinic. Cellular resistance to chemotherapy is associated with overexpression of ABC membrane transporters such as Pgp, Mrp or BRCP that are responsible for “pumping” drugs out of cancer cell. Cytoskeletal disruption and other alterations prevent the correct localization of membrane proteins, and disrupted cell signaling in cancer cells may lead to drug resistance. Abnormal vasculature reduces the efficient biodistribution of anticancer drugs, resulting in less drug accumulation in the tumor.

(Fig. 21.1). It is known that several members of the ATP-binding cassette (ABC) transporter family play an important role in cancer cell with resistance to different drugs (2). Studies have demonstrated that ABCB1, ABCC1, ABCG2, and other members of the ABC family are expressed in different types of human cancers, and their expression is related to the outcome of chemotherapy: the higher their expression in a tumor, the more resistant cancer cells are to chemotherapy. For example, a poor survival rate characterizes gliomas and tumor-derived endothelial cells that express ABCB1 and ABCC1 subfamily members (3). In the tumor tissue, tumor resistance can be connected to the physiology of the tumor tissue, including a poor vasculature and unsuitable physicochemical conditions. To overcome drug resistance, many attempts have been made using strategies that consider the different chemotherapeutic mechanisms either at the cellular level or at the tissue level. In the clinic, multidrug resistance (MDR) occurs in over 50% of patients whose cancer relapses, accounting in large part for the high mortality associated with cancer. Tumor resistance to chemotherapy in the clinic can be due to the inefficient distribution of drug relative to its targeted tumor tissue. MDR may become evidence either as a lack of tumor size reduction or as a clinical relapse after an initial positive response to antitumor treatment (4). In clinical practice, drug resistance constitutes the failure of a patient to achieve a complete or partial response to therapy. In the laboratory, however, drug resistance is a cellular phenomenon and reflects the inability to demonstrate cytotoxicity at physiologically achievable drug concentrations in cancer cells. Drug resistance may be considered to be either intrinsic or acquired. Frequently, resistance is intrinsic to the cancer at the beginning, but as therapy becomes more and more effective, acquired ­resistance also becomes common. Intrinsic resistance occurs when

Circumventing Tumor Resistance to Chemotherapy by Nanotechnology

469

tumor cells are capable of escaping exposure or repairing damage induced by the cytotoxic effects of chemotherapy at initial exposure. Finally, acquired resistance dominates when resistant cells survive from a population that was initially sensitive to chemotherapy. Both intrinsic and acquired resistance may operate along several different pathways, including decreased drug accumulation, decreased drug activation, increased repair of drug-induced damage, altered drug targets, altered gene expression and drug barriers (2). The development of resistance to chemotherapy is frequently associated with broad cross resistance even to structurally dissimilar drugs, suggesting the existence of more than one potential mechanism of resistance. Multiple changes often appear simultaneously in highly resistant tumor cell lines. This observation has led to the widely accepted hypothesis that tumor resistance to chemotherapy is usually multifactorial. Nanotechnology has the potential to overcome current obstacles to chemotherapy, because of the unique properties of nanoparticles (1–100  nm) (5). For example, solid tumors have unique features, such as leaky tumor blood vessels and defective lymphatic drainage, that promote the delivery and retention of macromolecules or nanoscale particles, a phenomenon recog nized as the enhanced permeability and retention (EPR) effect. Nanoparticles can be constructed at a certain size for effective biodistribution and accumulation in the tumor. Nanoparticles are characterized by self-assembly, stability, specificity, drug encapsulation and biocompatibility as a result of their material composition (6). Many researchers now make investigation to find out how to employ nanotechnology to overcome tumor multidrug resistance in vivo and in vitro.

2. Nanotechnology Allows Specific Targeting of Tumor

A major problem limiting the success of many anticancer agents lies in their inability to target tumor cells and tissues selectively. Therefore, almost all anticancer agents result in severe side effects to normal tissues and organs. In chemotherapy, pharmacologically active concentrations of an anticancer drug in the tumor tissue are often reached at the expense of massive contamination of the rest of the body. This poor specificity creates a toxicological problem that represents a serious obstacle to effective antitumor therapy. Recently, progress has been made on the design of nanoparticles with surface properties that allow better accumulation in tumor tissue after systemic administration. To improve the specificity of nanoparticles, a molecular recognition moiety is connected to the surface of the nanoparticles to target cancer cells in tumor tissue after intravenous administration.

470

Liang et al.

For example, folic acid has been used to be conjugated to surface of nanoparticles (7). The rationale behind the choice of folic acid as a targeting moiety is that folic acid binding proteins are frequently overexpressed on the surface of human cancer cells. The folate-coated nanoparticles showed a tenfold higher apparent affinity for the folate binding protein than the free folate, as measured by surface plasma resonance. This increased apparent affinity was attributed to the fact that the particles represent a multivalent form of the folic acid ligand and can therefore display stronger interactions with the folate receptor. Thus, it could be expected that the folate-grafted nanoparticles would also strongly interact with the surface of malignant cells on which the folate binding protein can be overexpressed; such binding can eventually promote endocytosis of the nanoparticles mediated by folate binding protein. Indeed, only the cancer cells overexpressing the folate binding protein showed intensive uptake of the folate-decorated nanoparticles. The cancer cells that did not express the folate binding protein on the cell surface did not show any uptake of those nanoparticles. In addition, none of the cells was able to internalize PEG-coated nanoparticles without folate coating. The development of various nanoparticles with different ligands now offers a choice for targeted tumors with drug resistance. The suspension of nanoparticles is very stable, as evaluated by size measurements, and can be lyophilized. The surface properties, including the zeta potential, complement activation and protein adsorption pattern, are defined by the nature of the materials used to synthesize nanoparticles. Indeed, the biological activity of heparin grafted on the surface of nanoparticles was preserved at a level of 70% when compared to the activity measured for a heparin solution. The variety of biomolecules that can be conjugated to nanoparticles offers many possibilities for the design of targeted nanoparticles using a biomimetic approach. Chemotherapies should ensure a specific toxic effect against the targeted tumor cells, even if the increased complexity of the outer surface obstructs their diffusion into tumor. Utilizing a ligand that binds specifically to its receptor on a malignant cell may help to reduce the dose-limiting cytotoxicity of the drug and also enable the drug to bypass the drug resistance mechanism especially caused by P-glycoprotein (Pgp) overexpression, via internalization through receptor-mediated endocytosis. This strategy not only targets the malignant cells directly, but also aims at destroying nonmalignant tumor components that are crucial for tumor survival and development. Heparin-paclitaxel-Folic Acid with its highly specific tumor uptake and potent antitumor properties fits the profile of this strategic requirement very well. More recently, liposomes have been modified by conjugating them with monoclonal antibodies directed against tumor antigens (8). Furthermore, copolymer nanoparticles can form a shell

Circumventing Tumor Resistance to Chemotherapy by Nanotechnology

471

with a hydrophobic inner part that contains the drug. They have a kinetic behavior similar to that of liposomes. These vectors may carry drugs, radioisotopes and/or labeling agents, and are directed against the specific surface structure of tumor cells, hence increasing the specific distribution and accumulation of drugs within tumors.

3. Nanotechnology Can Overcome Drug Resistance Due to Different Mechanisms

Drug resistance is known to develop through a variety of molecular mechanisms within the tumor (Fig.  21.1), and various approaches overcoming tumor resistance to chemotherapy are based on various pathways (9). For example, the enzyme glucosylceramide synthase (GCS), responsible for bioactivation of the proapoptotic mediator ceramide to a nonfunctional moiety glucosylceramide, is found to be overexpressed in many multidrugresistant tumor types and has been implicated in cell survival in the presence of chemotherapy. In an attempt to circumvent the mechanisms that cancer cells use to avoid cell death following chemotherapy, a polymeric nanoparticle was created to deliver ceramide, which triggers resistant cells to apoptosis under paclitaxel treatment. Treatment with the multifunctional nanoparticle produced 100% mortality among cultured cells. To overcome MDR in a human ovarian cancer cell line, modified poly(epsiloncaprolactone) (PEO-PCL) nanoparticles were used to encapsulate and deliver therapeutic agents for enhanced efficacy (9). With nanoparticle drug delivery, the resistant cells can be sensitized to paclitaxel near the IC50 concentration of sensitive cells. Chemotherapy enhanced via nanoparticle delivery has a promising potential as a strategy to overcome MDR. Tumor cells can develop simultaneous resistance to multiple anticancer drugs (2). An alternate strategy suggested for overcoming MDR is association of the drug with nanoparticles (10). The rationale behind this strategy is to increase the intracellular concentration of the drug and other agents using endocytosis. Doxorubicin, an anticancer drug widely used in cancer therapy and a known substrate of Pgp, was encapsulated in various types of nanoparticles. The sensitivity of resistant cells to the doxorubicinloaded nanoparticles was then evaluated by measuring the cytotoxic effect produced by increasing the concentration of the doxorubicin-loaded nanoparticles. Resistant cells treated with alginate or lactide-co-glycolide modified nanoparticles showed the same sensitivity to the treatment as the free drug (11). In contrast, resistant cells treated with doxorubicin-loaded poly(alkyl cyanoacrylate) (PACA) nanoparticles showed a much higher sensitivity to the drug, relative to the free drug. The sensitivity of

472

Liang et al.

the resistant cells even reached the level of sensitivity of the parent sensitive cells, suggesting that the PACA nanoparticles can totally overcome the tumor resistance to doxorubicin (12). To explain the ability of nanoparticles to overcome doxorubicin resistance, researchers have proposed a mechanism based on the adhesion of the nanoparticles to the surface of resistant cancer cells. Adhesion is followed by the simultaneous release of the drug and nanoparticle degradation products that pass through the cell membrane without being recognized by Pgp. To circumvent MDR, some proposed the use of competitive inhibitors such as verapamil. However, the clinical use of verapamil to overcome MDR is limited due to the serious adverse effects of this compound. More recent studies that have been designed to further improve the efficacy of nanoparticles in overcoming MDR have been based on limiting the activity of Pgp. This strategy is also an interesting alternative to promote the efficacy of doxorubicin-loaded nanoparticles. Soma et al. suggested co-encapsulating doxorubicin and cyclosporin A within the nanoparticles (12). Cyclosporin A is a chemosensitizing compound that can bind to Pgp and inhibit the pump efflux activity. Doxorubicin was incorporated within the core of the nanoparticles while cyclosporin A was located at the nanoparticle surface. Using different formulations of the drug-loaded nanoparticles, it was shown that the association of both doxorubicin and cyclosporin A within a single nanoparticle led to the most effective growth rate inhibition of the resistant cells. The association of cyclosporin A with doxorubicin nanospheres would also ensure that cyclosporin A reaches the same sites with the anticancer drug and also reduces its toxic side-effects. Other strategies proposed to regulate the expression of the Pgp have involved using siRNA (13, 14). However, the results obtained were disappointing because of the long half-life of Pgp, making its down-regulation difficult (14).

4. Engineered Nanoparticles Facilitates Targeting of Tumors

Nanoparticles have the potential to enhance the protection of drugs against biotransformation and rapid clearance in vivo (15). In order to do so, they must have long-circulating properties to reach the tumor tissue. In addition, they should have the proper biodistribution to target the tumor. With these objectives, studies have focused on customization of the surface properties of nanoparticles. Researchers have sought to modify the nanoparticle biodistribution to target tumors using poly(ethylene glycol) (PEG) as a coating material at the nanoparticle surface in order to reduce protein adsorption and complement activation (16). PEG-coated

Circumventing Tumor Resistance to Chemotherapy by Nanotechnology

473

nanoparticles were prepared from a poly(PEG cyanoacrylatecohexadecyl cyanoacrylate) copolymer (17). These nanoparticles circulated longer in the blood stream, while their uptake by the liver was reduced (18). They were found to accumulate in the brain to a larger extent than other formulations, including the non-PEGcoated nanoparticles (19, 20). The concentration of PEG-coated nanoparticles in the central nervous system was shown to be greatly increased, especially in the white matter when compared to conventional nanoparticles. Recently, these nanoparticles were shown to accumulate specifically in a glioma implanted into a rat brain. The accumulation was found to occur mainly in the tumoral tissue, while the amount of nanoparticles found in the adjacent healthy tissue and in the control hemisphere was much lower (21, 22). The comparable distribution in tumor and normal tissue was attributed to the difference in the microvascular permeability between healthy and tumor tissue, combined with an increased circulation time in the blood stream. Maeda et  al. found that Evans blue dye, which binds with plasma albumin, concentrated selectively in tumor tissues following intravenous (i.v.) injection (23). The same behavior was also noticed with radiolabeled plasma proteins, including transferrin (90  kDa) and IgG (160  kDa), whereas smaller proteins such as neocarzinostatin (12  kDa) did not accumulate in tumors (24). The tumor accumulation reaches up to several fold higher than that of the plasma due to lack of efficient lymphatic drainage in the solid tumor; this provides an ideal application for EPR-based selective anticancer drug delivery and distribution in a tumor. Tumor blood vessels are thought to have relatively large pore structures and poorly aligned defective endothelial cells lacking a smooth muscle layer (25). Extensive production of vascular permeability enhancing factors, such as nitric oxide (NO), lead to highly abnormal transport dynamics across tumor capillaries, especially for nanosized macromolecular drugs. Thus, it becomes possible for anticancer nanomedicines of certain sizes to cross selectively into tumor tissues (26). Furthermore, tumor tissues usually lack effective lymphatic drainage (27, 28), which leads to prolonged retention of nanoparticles. Due to their size, nanoscale particles containing anticancer drugs administered intravenously (i.v.) can escape renal clearance. Often they cannot penetrate the tight endothelial junctions of normal blood vessels, but can extravasate in tumor vasculature and become trapped in the tumor vicinity. Establishment of this principle hastened the development of various multifunctional nanoparticles for targeted cancer chemotherapy. Indeed, this highly selective local distribution of nanoparticles in tumor tissues has proven superior in therapeutic effect with minimal side effects in both preclinical and clinical settings. Gabizon et  al. found that 100  nm nanoparticles can passively enter tumor tissues, thereby, increasing selectivity of anticancer

474

Liang et al.

drug delivery at the tumor site, while markedly reducing drug accumulation and toxicity in many susceptible healthy tissues (29). If the level of drug resistance is comparable to the drug levels in tumor, MDR may be overcome by increasing delivery of anticancer drugs based only on mass action (30). Biocompatible and sterically stabilized micelles (SSMs) have been used as nanocarriers for chemotherapeutic agents. Drug solubilization in SSMs is reproducible and is attributed to the avoidance of drug aggregate formation. Furthermore, SSMs composed of polyethylene glycol (PEGylated) phospholipids are attractive nanocarriers for drug delivery because they are sufficiently small (14 nm) to cross through the leaky microvasculature of tumors and penetrate tissues for passive targeting of solid cancers in vivo, resulting in high drug concentration in tumors and reduced drug toxicity to the normal tissues (31).

5. Nanoparticle Properties Improve Drug Accumulation in Tumors

During the past few years, several strategies have been ­investigated to improve the clinical effectiveness of chemotherapy; in particular, attention has been focused on drugs and their pharmaceutical formulations. A promising strategy is the use of carriers to transport drugs that are already employed in the clinic such as platinum complexes. Nanoparticles and liposomes represent two major formulations that are in active clinical evaluation (32). The EPR effect allows significant increase in drug concentration; therefore, the cytotoxicity against tumor cells is increased, while normal tissues are spared from the drug-induced damage. However, it is evident that the incomplete and immature vasculature within the tumors plays a fundamental role in drug resistance. The immature vasculature leads to reduced oxygenation and nourishment of cancer cells, and cancer cells adapt to grow in these critical conditions. The adaptation leads to changes in gene expression and metabolic pathways, which contributes to diminishing pH values in the tumor until an acidic pH is achieved and maintained (33). In these conditions, drug resistance phenomena may begin to occur because many drugs become ionized due to their pKa values within the range from 5.8 to 8.5 (34). Weak basic drugs, such as anthracyclines and vinca alkaloids, diffuse poorly in an acidic extracellular milieu because their ionized status obstructs their passage through cell membranes. Low pH may cause tumor resistance to mitoxantrone, a weak basic drug. Conversely, weak acid agents, such as chlorambucil and ­5-fluorouracil, have an advantage in terms of distribution within the tumor and cytoplasmic sequestration because of the ­neutral-to-alkaline pHi.

Circumventing Tumor Resistance to Chemotherapy by Nanotechnology

475

Polymer micelles as powerful chemotherapeutic nanoparticles afford several advantages for targeted drug delivery in cancer, including increased drug solubility, prolonged circulation halflife, selective accumulation at tumor sites, and a decrease in toxicity. However, the technology still lacks tumor specificity and controlled release of the entrapped agents. Therefore, the focus has gradually shifted from passive targeting micelles to active targeting and responsive systems that carry additional mechanisms for site-specific release. pH-sensitive formulations are examples of how the versatility of micelles can lead to a fusion of chemical customization with biological insight to achieve active drug delivery. In addition, hyperthermia may also increase drug accumulation within a tumor and has been evaluated in association with liposomes. Hypoxia may also exert a significant influence on drug sensitivity through the modulation of mRNA levels of several genes (35). For example, the chronic influence of hypoxia may lead to etoposide and vincristine resistance by modifying gene expression of HIF-1a (36). Furthermore, hypoxic conditions increase the heme content and induce the expression of ABCG2 protein in stem cells. This induction allows cellular survival by removing heme from the cytoplasm and thus diminishing the formation of reactive oxygen species (ROS) (37). Therefore, the hypoxia-induced expression of ABCG2 seems to give a double advantage to cancer cells, allowing survival in critical conditions and making the tumor resistant to drugs. Modification of extracellular (pHe) and intracellular (pHi) could help to reverse drug resistance in tumors. Recent studies have demonstrated that some drugs exert their cytotoxic effects by altering the regulation of pHi through production of H2O2 in the mitochondria (38). Furthermore, an acidic pHi increases tumor sensitivity toward several drugs. Recent data support the use of proton-pump inhibitors (PPIs) to increase pHe and the pH of lysosomal organelles. Pretreatment with PPIs may reverse the MDR tumor phenotype, likely through the inhibition of drug excretion by ABC family members (i.e., ABCB1 or ABCG2). In some cases, it has been hypothesized that PPIs could induce drug accumulation within vesicle-like structures which cannot be excreted. Bioreductive drugs represent a logical consequence in the drug development process based on the knowledge of biologic characteristics of tumors. In contrast with other bioreductive drugs, tirapazamine is active at intermediate oxygen concentrations, which acts synergistically with several antineoplastic agents such as cisplatin. Block-copolymer micelles are spherical supramolecular assemblies of amphiphilic copolymers that have core–shell architecture. The core is a loading space that can accommodate hydrophobic drugs, and the shell is a hydrophilic brush-like corona that makes

476

Liang et al.

the micelle water soluble. Block-copolymer micelles allow ­delivery of poorly soluble contents and avoid the pHe and pHi limitation. The micelles localize in several cytoplasmic organelles, including the mitochondria, but not the nucleus (39). Administering immunomicelles loaded with the anticancer drug taxol to mice with lung carcinoma resulted in increased accumulation of taxol in the tumor. Furthermore, nanoparticle shape may be important in designing better nanotechnology-based drug delivery vehicles. Filomicelles are about ten times longer than their spherical counterparts and are more persistent than any known synthetic nanoparticle (40). Preliminary results further demonstrate that filomicelles can effectively deliver the anticancer drug paclitaxel and shrink humanderived tumors in mice. Although these findings show that long-circulating vehicles need not be nanospheres, they lend insight into possible shape effects on nanoparticle function.

6. Nanoparticles Used for Tumor Treatment

Because cells will typically internalize nanomaterials below 100 nm, nanostructures have the ability to enter the cells due to their nanoscale size. Some of the leading nanostructures being used for this purpose include fullerenes, dendrimers, and nanoshells (Table 21.1). Fullerenes (or Buckyballs) are natural hollow spheres, 1 nm in diameter, made with carbon atoms. Fullerenes create a unique drug delivery platform that allows active pharmacophores to be conjugated to their surface in three-dimensional orientations for precise control in matching fullerene compounds to biological targets, in entrapping atoms within the fullerene cage, and for attaching fullerene derivatives to targeting agents. One of these fullerenes investigated by our group is [Gd@ C82(OH)22]n, which is a water-soluble hydroxyl modified metalfullerene. This nanoparticle has a strong capacity to enhance immunity and protect the normal tissues from tumor invasion, with almost no toxicity in vivo and/or in vitro (41). In comparison with conventional antitumor chemicals such as cisplatin and cyclophosphamide, this nanoparticle is highly efficient at suppressing tumor growth. Its action is not due to toxic effects on tumor cells because it does not affect tumor cell proliferation directly under the administrated concentration. The distribution in the tissues is mainly in bones (about 1% of administration), then the pancreas, kidney and spleen, in that order. About 50% of [Gd@C82(OH)22]n are excreted in the urine and 35% in the feces, which suggests that this nanoparticle reaches tissues and organs through blood circulation and does not remain in the blood after 24 h of administration. This could be improved by appropriate modification of this nanoparticle.

Circumventing Tumor Resistance to Chemotherapy by Nanotechnology

477

Table 21.1 Nanomaterials potential for nanomedical application Subjects

Properties

Nanomaterials

Nanocrystals

Materials with nanocrystalline structure are different in their atomic structure, crystallographic orientation, or chemical composition

Ceramic, metal (quantum dots, nanogold, nanosilver, etc.) and metal oxide nanoparticles (CuO, ZnO, TiO2, SiO2, etc.)

Carbon nanotubes/ fullerenes

Carbon-based nanomaterials are composed mostly of carbon in the form of a hollow spheres, ellipsoids, or tubes etc. Fullerenes are characterized with an elongated sphere of carbon atoms formed by interconnecting six-member rings and 12 isolated five-member rings forming hexagonal and pentagonal faces

Hollow cylinders of carbon atoms such as: Carbon nanotubes (CNTs) Fullerences including C60, C70, C80, Gd@C82, C84, etc.

Organic nanoparticles

Nanomaterials with three components: a central core, an interior dendritic structure (the branches), and an exterior surface (the end groups)

Polymers built from branched units with numerous chain ends on the surface such as various dendrimers

Inorganic–organic hybrid nanoparticles

Hybrid nanomaterials consist of one material as matrix filled with another material

Nanoparticles or nanofibers with at least two different materials such as polyhedral siliconcontaining organic polymers

Other nanomaterials without clear definition, classification

Another nanomaterial used for cancer treatment is the ­ endrimer, which was used to treat tumor cells without triggering d an immune response. This is due to the dendrimer’s small size and branched structure. Dendrimers can be designed to release attached compounds in response to a specific molecule or chemical reaction. In addition, a layered sphere called nanoshell is being developed for cancer therapy. The nanoshell has a gold exterior layer which covers interior layers of silica and drugs. Nanoshells can be made to absorb light energy and then convert it to heat. As a result, when nanoshells accumulate next to a target area such as a tumor cell, they can release tumor-specific antibodies when infrared light is administered. Successful design of nanoparticles to treat tumors effectively requires assembly of the appropriate targeting ligands on nanocarriers and long-circulating nanosystems with appropriate surface modification and the capability to control particle stability, aggregation, receptor binding and subsequent biochemical cascades and signaling processes. The size of the particles must be large enough (30–100 nm) to avoid leakage into blood capillaries but not so large (>100  nm) that they become susceptible to

478

Liang et al.

­ acrophage-based clearance. Surface manipulation can control m the particle aggregation at interstitial sites, optimizing nanoparticle retention at lymph nodes. Very small particles (1–20  nm) with long circulatory residence times could slowly penetrate the vasculature into the interstitial spaces, and be transported by lymphatic vessels to lymph nodes. This phenomenon is quite important when designing nanoparticles to allow differential leakage from the blood circulation system through the permeable endothelium in lymph nodes. To date, many different nanoparticles have been synthesized and developed for effective treatment of tumors (Table  21.1). Recently, a multistage nanoparticle system has been employed. This multistage system consists of mesoporous material and nanoparticles as two major components. Modified mesoporous material is about to carry nanoparticles to their designated targeting site; it then degrades to release the nanoparticles into the targeted tissue. The released nanoparticles consequently merge into cells for efficient treatment.

7. Nanotechnology Can Improve the Bioavailability of Poorly Soluble Anticancer Drugs

Nanotechnology has been successfully utilized to create a new drug delivery system that can solve the problem of poor water solubility common to many promising and currently available anticancer drugs and thereby increase their effectiveness. The poorly soluble anticancer drugs require the addition of solvents in order for them to be easily absorbed into cancer cells. Unfortunately, these solvents not only dilute the potency of the drugs but increase toxicity as well. Silica-based nanoparticles are used to deliver hydrophobic anticancer drugs and other water-insoluble drugs to human cancer cells (42). The experimental results suggest that mesoporous silica nanoparticles might be used as a vehicle to overcome the insolubility problem of many anticancer drugs. Paclitaxel is widely used to treat multiple types of solid tumors. The commercially available paclitaxel formulation uses cremophor/ethanol (C/E) as solubilizers. Other formulations including nanoparticles have also been introduced. The nanoparticle and C/E formulations showed significant differences when compared to paclitaxel itself. Tissue specificity of the two formulations was different too. The nanoparticles showed longer retention and higher accumulation in organs and tissues. The most striking difference was an eightfold greater drug accumulation and sustained retention in the kidney. These data indicate that the nanoparticulate formulation of paclitaxel affects its clearance as well as distribution in tissues with preferential accumulation in the liver, spleen, small intestine, and kidney.

Circumventing Tumor Resistance to Chemotherapy by Nanotechnology

479

As mentioned above, block-copolymer micelles with core– shell architecture provide a loading space that can accommodate hydrophobic drugs, and the shell is a hydrophilic brush-like corona that makes the micelle water soluble, thereby allowing delivery of the poorly soluble contents and accumulated in tumor. Other nanoparticles consisting of human serum albumin (HSA) and containing different antisense ODNs (ASOs) have also been used for drug delivery insoluble drugs (43). The preparation process was optimized regarding the amount of desolvating agent, stabilization conditions, as well as nanoparticle purification. Wartlick et al. found that the glutaraldehyde cross-linking procedure of the particle matrix was a crucial parameter for biodegradability and drug release of the nanoparticles (43). The drug loading efficiency increased with longer chain length and employment of a phosphorothioate backbone. It indicated that there was no cytotoxic effect observed under nanoparticle concentrations up to 5,000 mg/ml in different tumor cells. In this study, the entrapment of a fluorescent labeled oligonucleotide within the particle matrix was used to detect intracellular drug release of the carrier systems. It was revealed under confocal laser scanning microscopy that nanoparticles cross-linked with low amounts of glutaraldehyde could rapidly be degraded intracellularly and could lead to a significant accumulation of the ASO in cytosolic compartments of the tumor cells.

8. Resistance to Cisplatin: A Broadly Used Anticancer Drug

The platinum coordination complex known as Peyrone’s chloride was firstly synthesized and described by M. Peyrone in 1845; these findings were published in 1965 (44). In the 1960s, Barnett Rosenberg serendipitously discovered its chemotherapeutic cancer activity (45, 46). In 1968, following further tests against various bacteria, cisplatin was administered intraperitoneally to mice at the nonlethal dose of 8 mg/kg, and was shown to cause marked tumor regression (47). The patient was first treated with confirmatory in  vivo tests performed by clinical testing in 1971. Cisplatin was approved by the US Food and Drug Administration (FDA) for clinical application in 1978. Since the biological properties of cisplatin as a anticancer drug were accidentally discovered over 40 years ago, it has had a major impact on the chemotherapeutic treatment of various cancers and is still widely used today. Cisplatin is one of the most widely used and most effective cytotoxic agents, and is broadly employed in the treatment of epithelial malignancies such as lung, head and neck, ovarian, bladder and testicular cancer (48). The action mechanism of cisplatin involves covalent binding to purine DNA bases, which

480

Liang et al.

primarily leads to cellular apoptosis. However, its continued clinical use is impeded by its severe adverse reactions including renal toxicity from renal tubular damage, gastrointestinal toxicity, peripheral neuropathy, asthenia, and ototoxicity (45, 46). The major limitation in the clinical applications of cisplatin is the development of cisplatin resistance by tumors. This arises either by clonal expansion of tumor cells in the heterogeneous tumor cell population with inherent resistance to cisplatin (with mutations in specific genes that confer resistance), or by acquired resistance by some cells in the tumor during treatment and their clonal expansion after killing of the sensitive cells by the drug. Tumor proliferation could be mainly conferred by limited uptake of the cisplatin by drug-resistant cells. Much is currently understood about how tumors commonly exhibit resistance to cisplatin, either intrinsically or as acquired during the courses of therapy. Mechanisms explaining cisplatin resistance include the reduction in cisplatin accumulation inside cancer cells because of barriers across the cell membrane, the faster repair of cisplatin adducts, increased cytoplasmic detoxification and tolerance to DNA damage, the modulation of apoptotic pathways in various cells, the mislocalization of functional membrane protein and a higher concentration of glutathione and metallothioneins in some types of tumors (49). A number of experimental strategies to overcome cisplatin resistance are at the preclinical or clinical stages.

9. Increased Intracellular Cisplatin Accumulation to Reverse Tumor Resistance

Reduced cisplatin intracellular accumulation is the common result in different types of cisplatin-resistant cell lines. Cisplatin is highly polar and enters cells relatively slow in comparison to other classes of small-molecule cancer drugs. The uptake of cisplatin is influenced by factors such as sodium and potassium ion concentrations, pH, and the presence of reducing agents. The role of transporters or gated channels has been postulated in addition to passive diffusion (50). So far, copper transporter-1 (CTR1) is considered to have a substantial role in cisplatin influx (51, 52). Loss of CTR1 was found to lead to a two- to threefold increase in drug resistance (53). In contrast to the mechanism of MDR, which is caused by the overexpression of ABC transporters, it is generally decreased uptake rather than increased efflux that predominates in cisplatin-resistant cells. The efflux proteins such as multidrug resistance protein-1 (MRP1, also known as ABCC1), MRP2 (also known as CMOAT or ABCC2) was reported to be partially associated with cisplatin resistance. The conjugation of cisplatin with glutathione was more readily exported from cells by the ATP-dependent glutathione S-conjugate export (GS-X) pump

Circumventing Tumor Resistance to Chemotherapy by Nanotechnology

481

(that is, MRP1 or MRP2) (54). Other studies also support a role for the glutathione metabolic pathway in acquired and inherited drug resistance to cisplatin. Maintenance of cisplatin levels in tumors for prolonged periods is expected to eradicate cisplatin sensitive cells without offering them a chance to develop resistance. Hurdles are the side effects of cisplatin and the toxicity from the cumulative dose. Several mechanisms can contribute to cisplatin resistance. A common observation, repeatedly reported over many years in many tumor cells with acquired resistance to cisplatin, is that of reduced platinum accumulation in comparison to the parental cells (55). The reduction in cisplatin accumulation inside cancer cells because of the cell membrane barrier is currently considered a major mechanism of acquired cisplatin resistance (56). The copper transporter CTR1 appears to control the accumulation of cisplatin in Saccharomyces cerevisiae. CTR1-deficient cells have reduced the uptake of cisplatin, and are 1.9-fold more resistant to the cytotoxic effect of cisplatin (57). However, until recently, the underlying complex molecular mechanism by which cisplatin enters cells still remained poorly defined. Drug delivery in cancer is important for optimizing the effect of drugs and reducing toxic side effects. Several nanobiotechnologies, mostly based on nanoparticles, have been used to facilitate drug delivery in cancer. The development of less toxic, nanoscale liposomal formulations of cisplatin has been hampered by the low water solubility and low lipophilicity of cisplatin, resulting in very low encapsulation efficiencies. Burger et  al. reported a novel method to efficiently encapsulate cisplatin in a lipid formulation by repeated freezing and thawing (58). The method is unique in that it generates nanocapsules, which are small aggregates of cisplatin covered by a single lipid bilayer. The nanoparticles have an unprecedented drug-to-lipid ratio and an in  vitro cytotoxicity higher when compared to free cisplatin. It suggests that the nanoscale encapsulation may also be generalized to other drugs with low water solubility and lipophilicity. A polymer–metal complex formation between cisplatin and PEG-poly(glutamic acid) block copolymers were prepared before by Nishiyama et al. (59), and their utility was also investigated as a tumor-targeted drug delivery system. Cisplatin-incorporated micelles with a 28 nm size exhibited a sustained drug release and the decay of the carrier itself in physiological saline. These nanoscale micelles showed a remarkably prolonged blood circulation and effectively accumulated in solid tumor sites (59). These data suggest that micelles with cisplatin could be a promising formulation for the targeted therapy of solid tumors. Micelles with a hydrophobic inner core and hydrophilic outer shell allow the chemical entrapment of cisplatin into the micelles; cisplatin is then released slowly into the target organism. As the extracellular

482

Liang et al.

pH of solid tumors has often been shown to be more acidic than normal tissues, this might also explain, in part, the increased tumor delivery of micelles with cisplatin. Overall, numerous mechanisms seem to be involved in tumor resistance to cisplatin studied in vitro. Studies have provided several rational approaches to circumventing clinical cisplatin resistance in patients. The strategy of using delivery vehicles to selectively transport more of a tumor-killing agent to tumors is attractive, and has now been clinically validated with the cytotoxics doxorubicin (liposomal doxorubicin) (60) and paclitaxel (nanoparticle albumin-bound paclitaxel) (61). To exploit the EPR effect of cisplatin in tumors, it has been linked to water-soluble, biocompatible nanomaterials. Trials are continuing with a reformulated cisplatin in an attempt to improve its antitumor activity.

10. Circumventing Cisplatin Resistance by NanotechnologyBased Delivery

Chemotherapy patients can be classified as either ­platinum-sensitive or platinum-resistant, depending on whether they have relapsed or progressed within 26 weeks of completing first-line platinum based chemotherapy (62). Expression of the mitogen-activated protein kinase phosphatase-1 (MKP-1) was a prognostic marker for patients with invasive ovarian carcinomas. The MKP-1 mRNA levels were strongly inducible upon treatment of OVCAR-3 cells with cisplatin. MKP-1 expression is a clinically useful marker to estimate patient prognosis as well as response to cisplatin chemotherapy. Nanotechnology can be applied to encapsulate and protect drugs during transit in  vivo. Drug encapsulation materials include liposomes and polymers (i.e., Polylactide (PLA) and ­Lactide-co-Glycolide (PLGA)). In addition to liposomes and polymers, other types of nanoparticles are also available for encapsulation. Materials such as silica and calcium phosphate (hydroxyapatite) have demonstrated superior properties at nanoscale rather than microscale, and can potentially be better suited for cisplatin delivery challenges. The materials form capsules around cisplatin and permit timed drug release to occur as the drug diffuses through the encapsulation material. Lipoplatin is a liposomal cisplatin formulation currently under clinical trials. The advantage of lipoplatin appears to arise from its 2- to 50-fold higher concentration in human tumors when compared to normal human tissues in biopsies, measured as total platinum with atomic absorption. The lipoplatin formulation can attain a higher concentration in tumors via its preferential extravasation through altered and compromised tumor vasculature. In order to achieve this property, liposomes that enclose ­chemotherapy

Circumventing Tumor Resistance to Chemotherapy by Nanotechnology

483

drugs must have a diameter below 130  nm, long-circulation properties and the ability to escape immune surveillance. The antitumor activity of cisplatin, encapsulated into ­transferrin-conjugated polyethylene glycol liposomes (Tf-PEG liposomes), was studied in nude mice with peritoneal dissemination of human gastric cancer cells. Small unilamellar Tf-PEG, PEG or DSPC/CH liposomes (bare liposomes) encapsulating cisplatin were prepared by reverse-phase evaporation followed by extrusion. The Tf-PEG liposomes were internalized into tumor cells by receptor-mediated endocytosis as shown by electron microscopy. Uptake of Tf-PEG liposomes into the liver and spleen was significantly lower than that of bare liposomes and had antitumor properties in nude mice xenografts that were better than free cisplatin (63). A novel bile acid–cisplatin complex, called as Bamet-R2, with liver vectoriality, was synthesized with the aim of overcoming cisplatin resistance. This complex had increased water solubility by encapsulation into liposomes and enhanced uptake by liver tumor cells. Bamet-R2 was effectively incorporated into liposomes with an increase in the concentration of the drug by more than 6 million fold compared with that in the initial free solution; this is 1,000-fold higher than the encapsulation obtained for cisplatin (64). A lipophilic cisplatin derivative, NDDP, formulated in conventional liposomes was shown not to be a cross-resistant with cisplatin in different in vitro and in vivo systems, and more active than cisplatin against tumor metastasis (65). NDDP was also formulated in liposomes composed of phosphatidylcholine, cholesterol and monosialoganglioside or PEG conjugated to phosphatidylethanolamine with prolonged circulation (66).

11. Nanoparticles Employed for Cisplatin Delivery in Preclinical and Clinical Stage

Local and sustained release of cisplatin near or inside a tumor may have distinct advantages over systemic administration of the drug. Cisplatin formulations in gel-type materials suitable for intratumoral injection have been tested in several laboratories. In general, these methods suffer from inefficient loading of the drug and other hurdles relating to its release mode and overall toxicity of the formulation. Malignant bone tumors are treated with surgical therapy and simultaneous systemic chemotherapy. In order to overcome the toxicity of this approach, the bone-cementing apatite (calcium phosphate) was used for a cisplatin formulation to develop an implant and maintain high concentrations of cisplatin at local sites in animals to counteract local structural weakness after tumor resection and treat residual malignant bone tumors. Approximately 33% of the total bound cisplatin was released after

484

Liang et al.

4.25  days. This approach might be used for the slow and local release of cisplatin in vivo (67, 68). PLGA-mPEG nanoparticles containing cisplatin were prepared by a double emulsion method and characterized with regard to their morphology, size, zeta potential and drug loading. Although intravenous administration of these cisplatin nanoparticles in mice resulted in prolonged cisplatin circulation in the blood, they suffered from loading efficiency for therapeutic applications (69). Degradable starch microspheres in an aqueous crystal suspension were used in clinical trials to achieve intensification of intraarterial chemotherapy of head and neck cancer with high-dose cisplatin (70). Nanoparticle-formulated cisplatin might further broaden its applicability to tumor types such as prostate cancer and small-cell lung cancer. Improved tumor delivery strategies and controlled release of cisplatin with specific modulators of cisplatin-resistance mechanisms might also provide future clinical benefits. Such strategies are sometimes unsuitable for clinical practice because of technical and biologic constraints, but in some cases they represent fruitful efforts to improve cancer chemotherapy. In particular, novel pharmaceutical formulations of cisplatin improve treatment efficacy and tolerability by increasing drug delivery within tumors or in close proximity.

12. Prospective Application of Nanotechnology to Reverse Tumor Resistance

Nanotechnology provides a wide range of new technologies for developing customized solutions that optimize the delivery of pharmaceutical agents. To be therapeutically effective, drugs need to be protected during their transit to the target action site in vivo while maintaining their biological and chemical properties. Some drugs are highly toxic and can cause serious side effects and have reduced therapeutic effect if they decompose during their delivery. Once the drug arrives at its destination, it needs to be released at an appropriate rate so that it can be effective. If the drug is released too rapidly it may not be completely absorbed, or it may cause gastrointestinal irritation and other side effects. The use of nanoparticle for drug delivery could positively impact the rate of absorption, distribution, metabolism, and excretion of the drugs in the body. In addition, nanoparticle delivery can allow the drug to reach its target in a more active form. There are severe restrictions on the nanomaterials and synthesis processes that can be used in drug delivery systems. The drug delivery material must be compatible and has to be easily bound with the drugs; in addition, the nanomaterial has to be easily degraded after use. It can be either metabolized or eliminated via normal excretory routes.

Circumventing Tumor Resistance to Chemotherapy by Nanotechnology

485

Nanotechnology can offer new drug delivery solutions by drug encapsulation. When materials are encapsulated in nanoparticles within the 1–100 nm size range, they have a larger surface area for the same volume, smaller pore size, improved solubility, and different structural properties. This can improve both the diffusion and degradation characteristics of the encapsulated nanomaterial. Nanotechnology involves the creation and use of materials and devices at the atomic and molecular level. Because clinical chemotherapy uses a variety of molecular materials and devices, and nanotechnology has the potential to provide many medical and pharmaceutical insights, such as how molecular materials self-assemble, self-regulate, and self-destroy. The scope of nanotechnology is enormous and it overlaps with the traditional medicine. Although only a subset of nanotechnology is applied to biological processes including medical and pharmaceutical usage, the potential for breakthrough is enormous and is being pursued on multiple fronts.

Acknowledgments We thank Mr. Xu Zhang for research assistance during the preparation of its manuscript. We also appreciate the help of Dr. Michael M. Gottesman for critical reading of the manuscript. Our work is financially supported by the Chinese Academy of Sciences “Hundred Talents Program” (07165111ZX), and the National Basic Research Program of China (2009CB930200). This work was also supported in part by NIH/ NCRR/ RCMI 2G12RR003048, NIH 5U 54CA091431, and USAMRMC W81XWH-05-1-0291 grants. References 1. Leaf C (2004) Why we’re losing the war on cancer (and how to win it). Fortune 149:76–82 2. Gottesman MM (2002) Mechanisms of cancer drug resistance. Annu Rev Med 53:615–627 3. Calatozzolo C, Gelati M, Ciusani E et  al (2005) Expression of drug resistance proteins Pgp, MRP1, MRP3, MRP5 and GST-pi in human glioma. J Neurooncol 74:113–121 4. Links M, Brown R (1999) Clinical relevance of the molecular mechanisms of resistance to anti-cancer drugs. Expert Rev Mol Med 1999:1–21 5. McNeil SE (2005) Nanotechnology for the biologist. J Leukoc Biol 78:585–594

6. Grodzinski P, Silver M, Molnar LK (2006) Nanotechnology for cancer diagnostics: promises and challenges. Expert Rev Mol Diagn 6:307–318 7. Stella B, Arpicco S, Peracchia MT et al (2000) Design of folic acid-conjugated nanoparticles for drug targeting. J Pharm Sci 89:1452–1464 8. Kontermann RE (2006) Immunoliposomes for cancer therapy. Curr Opin Mol Ther 8:39–45 9. van Vlerken LE, Duan ZF, Seiden MV, Amiji MM (2007) Modulation of intracellular ceramide using polymeric nanoparticles to overcome multidrug resistance in cancer. Cancer Res 67:4843–4850

486

Liang et al.

10. Kubiak C, Couvreur P, Manil L, Clausse B (1989) Increased cytotoxicity of nanoparticlecarried Adriamycin in  vitro and potentiation by verapamil and amiodarone. Biomaterials 10:553–556 11. Colin de Verdiere A, Dubernet C, Nemati F et  al (1994) Uptake of doxorubicin from loaded nanoparticles in multidrug-resistant leukemic murine cells. Cancer Chemother Pharmacol 33:504–508 12. Soma CE, Dubernet C, Bentolila D, Benita S, Couvreur P (2000) Reversion of multidrug resistance by co-encapsulation of doxorubicin and cyclosporin A in polyalkylcyanoacrylate nanoparticles. Biomaterials 21:1–7 13. Shoji Y, Nakashima H (2004) Current status of delivery systems to improve target efficacy of oligonucleotides. Curr Pharm Des 10:785–796 14. Brigui I, Djavanbakht-Samani T, Jolles B, Pigaglio S, Laigle A (2003) Minimally modified phosphodiester antisense oligodeoxyribonucleotide directed against the multidrug resistance gene mdr1. Biochem Pharmacol 65:747–754 15. Reddy LH, Sharma RK, Murthy RS (2004) Enhanced tumour uptake of doxorubicin loaded poly(butyl cyanoacrylate) nanoparticles in mice bearing Dalton’s lymphoma tumour. J Drug Target 12:443–451 16. Gref R, Minamitake Y, Peracchia MT et  al (1994) Biodegradable long-circulating polymeric nanospheres. Science 263:1600–1603 17. Stolnik S, Dunn SE, Garnett MC et al (1994) Surface modification of poly(lactide-co-glycolide) nanospheres by biodegradable poly(lactide)-poly(ethylene glycol) copolymers. Pharm Res 11:1800–1808 18. Peracchia MT, Fattal E, Desmaele D et  al (1999) Stealth PEGylated polycyanoacrylate nanoparticles for intravenous administration and splenic targeting. J Control Release 60:121–128 19. Calvo P, Gouritin B, Chacun H et al (2001) Long-circulating PEGylated polycyanoacrylate nanoparticles as new drug carrier for brain delivery. Pharm Res 18:1157–1166 20. Calvo P, Gouritin B, Villarroya H et al (2002) Quantification and localization of PEGylated polycyanoacrylate nanoparticles in brain and spinal cord during experimental allergic encephalomyelitis in the rat. Eur J NeuroSci 15:1317–1326 21. Brigger I, Dubernet C, Couvreur P (2002) Nanoparticles in cancer therapy and diagnosis. Adv Drug Deliv Rev 54:631–651 22. Brigger I, Morizet J, Aubert G et  al (2002) Poly(ethylene glycol)-coated hexadecylcyano-

23.

24.

25.

26.

27.

28. 29.

30.

31.

32. 33. 34.

acrylate nanospheres display a combined effect for brain tumor targeting. J Pharmacol Exp Ther 303:928–936 Maeda H, Greish K, Fang J (2006) The EPR effect and polymeric drugs: a paradigm shift for cancer chemotherapy in the 21st century. In: Satchi-Fainaro R, Duncan R (eds) Polymer therapeutics II: polymers as drugs, conjugates and gene delivery systems. Springer-Verlag, Berlin, Heidelberg, pp 103–121 Maeda H (2001) The enhanced permeability and retention (EPR) effect in tumor vasculature: the key role of tumor-selective macromolecular drug targeting. Adv Enzyme Regul 41:189–207 Yuan F, Dellian M, Fukumura D et al (1995) Vascular permeability in a human tumor xenograft: molecular size dependence and cutoff size. Cancer Res 55:3752–3756 Kopecek J, Kopeckova P, Minko T, Lu ZR, Peterson CM (2001) Water soluble polymers in tumor targeted delivery. J Control Release 74:147–158 Leu AJ, Berk DA, Lymboussaki A, Alitalo K, Jain RK (2000) Absence of functional lymphatics within a murine sarcoma: a molecular and functional evaluation. Cancer Res 60:4324–4327 Jain RK (1994) Barriers to drug delivery in solid tumors. Sci Am 271:58–65 Gabizon A, Shmeeda H, Horowitz AT, Zalipsky S (2004) Tumor cell targeting of liposome-entrapped drugs with phospholipidanchored folic acid-PEG conjugates. Adv Drug Deliv Rev 56:1177–1192 Krishna R, Mayer LD (2000) Multidrug resistance (MDR) in cancer. Mechanisms, reversal using modulators of MDR and the role of MDR modulators in influencing the pharmacokinetics of anticancer drugs. Eur J Pharm Sci 11:265–283 Koo OM, Rubinstein I, Onyuksel H (2006) Camptothecin in sterically stabilized phospholipid nano-micelles: a novel solvent pH change solubilization method. J Nanosci Nanotechnol 6:2996–3000 Yih TC, Al-Fandi M (2006) Engineered nanoparticles as precise drug delivery systems. J Cell Biochem 97:1184–1190 Madshus IH (1988) Regulation of intracellular pH in eukaryotic cells. Biochem J 250:1–8 Mahoney BP, Raghunand N, Baggett B, Gillies RJ (2003) Tumor acidity, ion trapping and chemotherapeutics. I. Acid pH affects the distribution of chemotherapeutic agents in vitro. Biochem Pharmacol 66:1207–1218

Circumventing Tumor Resistance to Chemotherapy by Nanotechnology 35. Acker T, Plate KH (2004) Hypoxia and hypoxia inducible factors (HIF) as important regulators of tumor physiology. Cancer Treat Res 117:219–248 36. Hussein D, Estlin EJ, Dive C, Makin GW (2006) Chronic hypoxia promotes hypoxiainducible factor-1alpha-dependent resistance to etoposide and vincristine in neuroblastoma cells. Mol Cancer Ther 5:2241–2250 37. Krishnamurthy P, Schuetz JD (2006) Role of ABCG2/BCRP in biology and medicine. Annu Rev Pharmacol Toxicol 46:381–410 38. Ahmad KA, Iskandar KB, Hirpara JL, Clement MV, Pervaiz S (2004) Hydrogen peroxidemediated cytosolic acidification is a signal for mitochondrial translocation of Bax during drug-induced apoptosis of tumor cells. Cancer Res 64:7867–7878 39. Savic R, Luo L, Eisenberg A, Maysinger D (2003) Micellar nanocontainers distribute to defined cytoplasmic organelles. Science 300:615–618 40. Geng Y, Dalhaimer P, Cai S et al (2007) Shape effects of filaments versus spherical particles in flow and drug delivery. Nat Nanotechnol 2:249–255 41. Chen C, Xing G, Wang J et  al (2005) Multihydroxylated [Gd@C82(OH)22]n nanoparticles: antineoplastic activity of high efficiency and low toxicity. Nano Lett 5:2050–2057 42. Lu J, Liong M, Zink JI, Tamanoi F (2007) Mesoporous silica nanoparticles as a delivery system for hydrophobic anticancer drugs. Small 3:1341–1346 43. Wartlick H, Spankuch-Schmitt B, Strebhardt K, Kreuter J, Langer K (2004) Tumour cell delivery of antisense oligonuclceotides by human serum albumin nanoparticles. J Control Release 96:483–495 44. Rosenberg B, Vancamp L, Krigas T (1965) Inhibition of cell division in Escherichia coli by electrolysis products from a platinum electrode. Nature 205:698–699 45. Rosenberg B (1977) Noble metal complexes in cancer chemotherapy. Adv Exp Med Biol 91:129–150 46. Hill JM, Speer RJ (1982) Organo-platinum complexes as antitumor agents (review). Anticancer Res 2:173–186 47. Rosenberg B, VanCamp L, Trosko JE, Mansour VH (1969) Platinum compounds: a new class of potent antitumour agents. Nature 222:385–386 48. Oliver T, Mead G (1993) Testicular cancer. Curr Opin Oncol 5:559–567 49. Liang XJ, Shen DW, Gottesman MM (2004) A pleiotropic defect reducing drug accumula-

487

tion in cisplatin-resistant cells. J Inorg Biochem 98:1599–1606 50. Kelland L (2007) The resurgence of platinumbased cancer chemotherapy. Nat Rev Cancer 7:573–584 51. Ishida S, Lee J, Thiele DJ, Herskowitz I (2002) Uptake of the anticancer drug cisplatin mediated by the copper transporter Ctr1 in yeast and mammals. Proc Natl Acad Sci USA 99:14298–14302 52. Katano K, Kondo A, Safaei R et  al (2002) Acquisition of resistance to cisplatin is accompanied by changes in the cellular pharmacology of copper. Cancer Res 62:6559–6565 53. Holzer AK, Howell SB (2006) The internalization and degradation of human copper transporter 1 following cisplatin exposure. Cancer Res 66:10944–10952 54. Ishikawa T (1992) The ATP-dependent glutathione S-conjugate export pump. Trends Biochem Sci 17:463–468 55. Kelland LR, Mistry P, Abel G et  al (1992) Mechanism-related circumvention of acquired cis-diamminedichloroplatinum(II) resistance using two pairs of human ovarian carcinoma cell lines by ammine/amine platinum(IV) dicarboxylates. Cancer Res 52:3857–3864 56. Shen DW, Goldenberg S, Pastan I, Gottesman MM (2000) Decreased accumulation of [14C]carboplatin in human cisplatin-resistant cells results from reduced energy-dependent uptake. J Cell Physiol 183:108–116 57. Lin X, Okuda T, Holzer A, Howell SB (2002) The copper transporter CTR1 regulates cisplatin uptake in Saccharomyces cerevisiae. Mol Pharmacol 62:1154–1159 58. Burger KN, Staffhorst RW, de Vijlder HC et al (2002) Nanocapsules: lipid-coated aggregates of cisplatin with high cytotoxicity. Nat Med 8:81–84 59. Nishiyama N, Okazaki S, Cabral H et  al (2003) Novel cisplatin-incorporated polymeric micelles can eradicate solid tumors in mice. Cancer Res 63:8977–8983 60. Gordon AN, Tonda M, Sun S, Rackoff W (2004) Long-term survival advantage for women treated with pegylated liposomal doxorubicin compared with topotecan in a phase 3 randomized study of recurrent and refractory epithelial ovarian cancer. Gynecol Oncol 95:1–8 61. Gradishar WJ, Tjulandin S, Davidson N et al (2005) Phase III trial of nanoparticle albumin-bound paclitaxel compared with polyethylated castor oil-based paclitaxel in women with breast cancer. J Clin Oncol 23:7794–7803

488

Liang et al.

62. Gore ME, Atkinson RJ, Thomas H et  al (2002) A phase II trial of ZD0473 in platinum-pretreated ovarian cancer. Eur J Cancer 38:2416–2420 63. Iinuma H, Maruyama K, Okinaga K et  al (2002) Intracellular targeting therapy of cisplatin-encapsulated transferrin-polyethylene glycol liposome on peritoneal dissemination of gastric cancer. Int J Cancer 99:130–137 64. Briz O, Serrano MA, Rebollo N et al (2002) Carriers involved in targeting the cytostatic bile acid-cisplatin derivatives cis-diammine-chlorocholylglycinate-platinum(II) and cis-diammine-bisursodeoxycholate-platinum(II) toward liver cells. Mol Pharmacol 61:853–860 65. Li S, Khokhar AR, Perez-Soler R, Huang L (1995) Improved antitumor activity of cisBis-neodecanoato-trans-R, R-1, 2- diaminocyclohexaneplatinum (II) entrapped in long-circulating liposomes. Oncol Res 7:611–617 66. Mori A, Wu SP, Han I et al (1996) In vivo antitumor activity of cis-bis-neodecanoato-trans-R,

67.

68.

69.

70.

R-1, 2-diaminocyclohexane platinum(II) formulated in long-circulating liposomes. Cancer Chemother Pharmacol 37:435–444 Barroug A, Glimcher MJ (2002) Hydroxyapatite crystals as a local delivery system for cisplatin: adsorption and release of cisplatin in vitro. J Orthop Res 20:274–280 Emerich DF, Snodgrass P, Lafreniere D et al (2002) Sustained release chemotherapeutic microspheres provide superior efficacy over systemic therapy and local bolus infusions. Pharm Res 19:1052–1060 Avgoustakis K, Beletsi A, Panagi Z et  al (2002) PLGA-mPEG nanoparticles of cisplatin: in vitro nanoparticle degradation, in vitro drug release and in  vivo drug residence in blood properties. J Control Release 79: 123–135 Kovacs AF, Turowski B (2002) Chemoembolization of oral and oropharyngeal cancer using a high-dose cisplatin crystal suspension and degradable starch microspheres. Oral Oncol 38:87–95

Index A Alzheimer’s disease......................................................... 171 Angiogenesis.............................................................. 24–25 Animal model mouse . .............................20, 33–44, 58, 103, 127, 164, 208, 231, 254, 314, 332, 373, 387, 417, 434, 453 non-human primate.......................................... 389–392 rat................................18, 19, 21, 36, 48, 145, 161, 208, 209, 211–213, 216, 227, 253, 362, 364–367, 387–389, 391, 393, 394, 396–399, 473 Antagonism.....................................................291–319, 413 Anthracyclines daunorubicin..................52, 62, 103, 104, 112, 125, 126, 129–133, 136, 153, 202, 225, 227, 253, 256, 260, 266, 292, 301, 302, 316, 317, 328, 343, 344, 347, 352, 435–437, 440, 451, 453, 454, 456, 461, 462, 464 doxorubicin....................1, 37, 49, 78, 97, 125, 148, 202, 224, 252, 310, 327, 343, 365, 410, 435, 471 epirubicin......................17, 62, 79, 89, 90, 231, 343, 344 mitoxantrone...................52, 59, 62, 79, 82, 84, 88, 111, 132–136, 202, 253, 255, 256, 260, 266, 268, 270, 328, 343, 344, 347, 435, 474 Antisense oligonucleotide (AON)......................... 331–332 Apoptosis. See Cell death ATP-binding cassette (ABC) transporter..............1, 15, 43, 48, 77, 95–115, 142, 199, 223, 252, 325–335, 343, 407, 433, 448, 468 Autophagy. See Cell death

B Biomarker................................................................... 21, 26 Bisulphite sequencing............................................. 183–197 Blood brain barrier (BBB)................. 20, 21, 147, 169–171, 207–209, 368 Breast cancer........................18, 33–44, 57, 78, 96, 123, 145, 199, 227, 251–282, 326, 344, 435, 462 Breast cancer resistance protein (BCRP) expression.......................... 134–136, 254–256, 259–282 polymorphism................................................... 255–259

C Calcein-AM...................... 87, 124–126, 130–133, 225, 437 Cancer chemotherapy

combination chemotherapy.................258, 291–293, 296, 351 prognosis.......................58, 63, 101, 104, 253, 255, 258, 266–268, 273, 280–282, 467, 482 susceptibility..................................................... 105, 258 treatment.....................55, 202, 213, 216, 251–282, 409, 467–469, 477 Cancer stem cell (CSC)............ 2, 6, 7, 42, 60, 89, 275, 282. See also Tumor stem cell Cell death apoptosis...................... 2, 6, 9, 34, 50, 53, 54, 56–57, 59, 65, 66, 143, 171, 261, 292, 342, 433, 434, 439, 440, 448, 471, 480 autophagy................................................................... 56 necrosis.................................... 37, 44n1, 53, 56, 65, 215 Cell proliferation.......................3, 4, 9–10, 25, 59, 453–454, 460–461, 464, 476 Ceramide ������������������������������ 50, 65, 66, 434, 439–441, 471 Chemoresistance. See Drug resistance Chromatin ��������������������������������������������54, 85, 89, 183, 184 Chromatin immunoprecipitation (ChIP)................ 85, 277 Cisplatin............... 5, 17, 50–51, 97, 131, 273, 294, 342, 475 Clinical outcome.................11, 61, 101, 103, 105–106, 108, 110–111, 115, 260, 266, 268, 278, 397. See also Treatment outcome Combination index (CI)......... 105, 253, 303–305, 307–309 Combinatorial chemistry......... 327, 349, 407, 408, 416, 417 CombiPlex.............................................................. 295–318 Comparative genomic hybridization (CGH)................. 63, 82–85, 277 CpG island..................................................183–184, 195n1 Cyclosporin A (CSA)...................... 87, 126, 129, 154, 155, 158, 161, 345, 346, 351, 391, 407, 413, 434, 436–439, 441, 472 Cytochrome P450 enzyme (CYP)........... 51, 52, 65, 348, 387–389, 391–394, 396–399 Cytotoxicity..................... 42, 59, 65, 88, 148, 162, 163, 165, 268, 292, 293, 296, 316, 319, 345, 349, 468, 470, 474, 481

D Daunorubicin. See Anthracyclines DiOC2(3) ��������������������������������������� 125, 126, 129, 133, 134 DNA methylation CpG methylation............................................. 183–197

489

ulti-Drug Resistance in Cancer 490  M Index



 

hypermethylation.........................................50, 183, 188 methyltransferase................................................ 54, 183 DNA repair......................................... 7, 53–55, 65, 66, 342 Docetaxel. See Taxanes Doxorubicin. See Anthracyclines Drug accumulation..................33, 59, 129, 143, 229, 331, 347, 378, 407, 453, 454, 461, 468, 471, 474–476, 478 administration................37, 48, 205, 209, 212, 406, 479 bioavailability.............................. 95, 102–103, 105, 108, 112–115 delivery..................5, 146, 255, 295, 304, 308–313, 316, 334, 335, 368, 379, 471, 473–476, 478, 479, 481, 484, 485 design................................. 156, 296, 334, 346, 409–423 formulation................ 148, 311–318, 472, 474, 481–484 ratio................................................................... 291–319 screening....................................................293, 307, 316 target.........................15–27, 40, 148, 334, 342, 409, 469 Drug absorption, distribution, metabolism, and excretion (ADME) absorption................................................................. 253 distribution............................................................... 253 excretion................................................................... 253 metabolism............................................................... 253 Drug-drug interaction (DDI)................. 98, 203, 209, 212, 215, 217, 292, 362–367, 372, 379, 385–399 Drug resistance acquired resistance...............16, 33–35, 40, 78, 434, 468, 469, 480, 481 intrinsic resistance.................... 3, 8, 16, 33–35, 78, 343, 468, 469, 480 Drug selection multi-step selection.........................................78, 81, 87 single-step selection.............................78–80, 86, 88–90 Drug transport diffusion.........................4, 5, 8–9, 48, 65, 200, 205, 208, 214, 215, 480 efflux..............................................5, 6, 8, 11, 16, 18, 19, 43, 49–50, 78, 130, 143, 147, 199–200, 205, 207–210, 216, 217, 252, 326, 344, 353, 397, 406, 409, 433, 434 entry........................................... 50–52, 65, 96, 169, 229 sequestration...................5, 52–53, 60, 65, 342, 360, 361 uptake................................................ 5, 9, 11, 49, 50, 65

Fluorescence-activated cell sorting (FACS)............. 42–43, 132, 133 Fluorescence in situ hybridization (FISH)........ 82–84, 256 Fumitremorgin C (FTC)............... 124, 135, 136, 254, 270

G Gallium-67/68................................................................ 163 Gene amplification................................................... 82–86, 89 expression.................. 11, 38, 52, 63, 82, 86, 87, 88, 108, 183–197, 204, 262, 282, 331, 399, 456, 469, 474, 475 regulation.........................................90, 145, 204, 462n4 silencing......................... 183–184, 447–449, 457, 462n4 therapy.............................................................. 157, 332 transcription......................24, 25, 83, 84, 145, 146, 184, 203–204, 399 Genetically engineered mouse model (GEMM)....... 33–44 Genetic variability.....................................98, 102, 108, 112 Glutathione (GSH).................... 6, 8, 19, 49, 52, 59, 65, 66, 105, 107, 225, 227–233, 240, 292, 435, 480, 481 Glutathione-S-transferase (GST)...................... 52, 66, 292

H Histone deacetylase (HDAC)............................... 184, 188 Homology modeling...............................409–410, 419–421

I Imaging noninvasive imaging......................................... 141–173 Immune response................................................7, 228, 477 Immunohistochemistry (IHC).................. 16–18, 134, 157, 171, 260, 265, 272–281, 454 Immunosuppressor................................................. 433–441

L

Efflux pump......... 15–21, 135, 137, 345, 406, 408, 420, 472 Epigenetics epigenetic regulation......................................... 183–197 Epirubicin. See Anthracyclines

Leukemia aacute lymphoblastic leukemia (ALL)................ 58, 63, 101, 104, 110, 257, 260, 263, 265, 268–269, 406 achronic myelogenous leukemia (CML)................ 101, 103, 104, 253, 264, 265, 267, 269–271, 462 acute myelogenous leukemia (AML).................. 58, 63, 101, 103, 125, 126, 259–268, 282, 328, 347, 349, 352, 406, 436 Lipid metabolism....................................................... 50, 65 Liposome......... 148, 150, 310–313, 329, 333, 449, 470–471, 474, 475, 482, 483 Lung resistance-related protein (LRP).................... 57, 134, 262, 263, 268, 269, 272, 276–278, 280, 281 Lymphoma.............................. 105, 258, 271–281, 328, 352

F

M

Flow cytometry...................................................... 123–137 Fluorescein (FITC).................... 36, 43, 125–128, 130–134

Macrophage................................7, 10, 11, 57, 212, 228, 478 Maximum tolerable dose (MTD)...... 35, 37, 38, 44, 314, 316

E

Multi-Drug Resistance in Cancer 491 Index    

Median effect analysis..................... 299, 301, 303, 304, 307 Metal complex......................... 157, 158, 160, 163, 167, 312 Microenvironment............................. 2–4, 11, 24, 59, 64, 65 MicroRNA (miRNA)...................................................... 85 Mitomycin C (MMC)................ 50, 59, 150, 151, 343, 344 Mitotic catastrophe.......................................................... 56 Mitoxantrone. See Anthracyclines mRNA degradation...........................................85, 331, 332 Multidrug resistance (MDR) MDR1. See P-glycoprotein modulator................................. 144, 326–330, 333–335, 344–354, 437, 439, 441, 450 reversing (reversal) agent.......... 144, 154, 156, 158, 348, 433–441 Multidrug resistance-associated protein (MRP) expression....................... 17, 18, 25, 129–131, 133–136, 224–227, 229, 252, 266, 268, 269, 271, 274, 278, 280, 281, 332, 406, 441 substrate...........................16, 18, 19, 115, 124, 129–132, 224–233, 236, 240, 435 Myeloma �����������������101, 104, 271, 272, 278, 327, 349, 438

N Nanomedicine................................................................ 473 Nanoparticle.................... 147, 310, 319, 469–479, 481–484 Nanotechnology..................................................... 467–485 Necrosis. See Cell death Neurodegenerative disease...............................170–172, 208 Non-small cell lung cancer (NSCLC)................. 17, 18, 53, 57, 58, 110, 258, 273–275, 277, 279, 294–296, 328 Normal tissue................ 2–4, 6, 9, 10, 18, 97, 132, 199–217, 240, 275, 280, 343, 347, 353, 469, 473, 474, 476, 482 Nucleotide binding domain (NBD)........ 60, 100, 106, 142, 224–230, 233–239, 416–417, 421

O Orthotopic grafting............................. 34, 35, 38, 39, 41, 43

P Paclitaxel. See Taxanes Parkinson’s disease.................................................. 170, 172 Peripheral blood mononuclear cell (PBMC)......... 126, 129, 130, 271 P-glycoprotein (Pgp) activity..........................20, 154, 157, 165–167, 170–173, 208, 211, 212, 333, 349, 353, 361, 372, 436, 437, 472 expression............................17–21, 25, 96, 98, 100, 102, 114, 125, 127–130, 133, 135, 136, 144, 146, 154, 157, 161, 162, 165, 170, 172, 199–217, 252, 326, 331–333, 344, 345, 349, 352, 362, 363, 366, 369, 376, 379, 380, 391, 392, 450, 470, 472

inducer............................... 200–203, 207, 215, 361, 380 inhibitor............. 129, 130, 144, 149, 154–157, 161, 166, 167, 171, 200–203, 206, 207, 209, 211–213, 215–217, 346, 348, 353, 361, 365, 367, 369, 370, 372, 379, 391, 393, 394, 397, 398, 436, 439 localization................................ 200, 201, 208, 210, 368 substrate........................................ 17, 20, 104, 125, 130, 132, 143, 146, 147, 152, 157, 158, 170, 171, 200–203, 205, 206, 208, 211, 213, 215, 349, 370, 371, 378–380, 391, 392, 393, 397, 434–435, 437, 471 Pharmacodynamics....................22, 217, 359, 362, 369, 370, 372, 375, 376, 378, 380 Pharmacogenetics............................................. 95–115, 255 Pharmacokinetic interaction................... 334, 345, 346, 348, 351, 352, 413, 416, 417, 436, 441, 450 Pharmacophore modeling...................................... 414–415 Phenotyping........................................................... 389, 394 Polymerase chain reaction (PCR).................. 157, 184–197, 265, 457, 458 Polymorphism.....................48, 49, 51, 52, 61, 98, 103, 105, 108, 110–112, 114, 170, 203–205, 211, 212, 255–258, 334, 361, 362 Positron emission tomography (PET)........... 16, 18, 22, 23, 26, 157, 162, 167, 168, 170, 173 Prodrug �������������������������������������� 51, 65, 149–152, 173, 419

R Radiopharmaceutical.......................... 17, 19, 146, 157, 158, 160–162, 164, 166–168, 170 Rhodamine............................. 123, 100, 124–126, 129, 130, 132–135, 201, 202, 216, 230, 231, 248, 349, 366 Ribozyme ������������������������������������������������������331, 332, 448 RNA-induced silencing complex (RISC).............. 332, 448 RNA interference (RNAi)...................... 203, 332, 447–464

S Senescence........................................... 7, 34, 53, 56, 65, 448 Short-hairpin RNA (shRNA)....... 332, 448–451, 455–457, 463n7, 463n8, 463n9 Single nucleotide polymorphism (SNP)......................... 61, 63, 98–106, 108, 110, 113–115, 204–206, 211, 255–258, 334, 335 Single-photon emission computed tomography (SPECT)............................... 16, 17, 22, 23, 26, 157–170, 173 Small interfering RNA (siRNA)............... 86, 88, 136, 331, 332, 447–451, 454–456, 462n4, 463n7, 463n9, 472 Solid tumor....................................................18, 26, 33, 58, 59, 63, 104–105, 127, 128, 271–281, 292, 310, 314–316, 326, 327, 406, 469, 473, 478, 481–482 Solute carrier (SLC)................................... 48–50, 108, 223 Sulforhodamine B (SRB)................. 454, 460, 461, 464n17 Synergy....................................................292, 293, 295–313

ulti-Drug Resistance in Cancer 492  M Index



 

T

V

Taxanes docetaxel....................................................153, 343, 344 paclitaxel............................................104, 153, 343, 344 Translational research......................................48, 62–64, 90 Transmembrane domain (TMD)............... 60, 78, 106, 110, 112, 142, 143, 224, 226, 230, 234, 238, 252, 416, 419–422 Transplantation orthotopic transplantation.........................35–36, 39–44 xenotransplantation.............................................. 34, 37 Treatment outcome...................49, 51, 59, 61, 63, 103–104, 114, 251–282, 292. See also Clinical outcome Tumor development.................................... 24, 35, 38, 470, 480 dormancy............................................................ 3, 4, 10 heterogeneity...........................................6–7, 35, 38, 41 Tumor-initiating cell........................................................ 33 Tumor stem cell............ 2–4, 6–11. See also Cancer stem cell

Vascular endothelial growth factor (VEGF)............. 16, 21, 24–27, 332 Vault major vault protein (MVP)........................... 57, 58, 112 Verapamil...............................8, 18, 20, 79, 81, 98, 125, 126, 129, 136, 149, 154, 155, 158, 170–172, 202, 209, 211, 212, 215, 216, 227, 228, 232, 252, 327, 345, 346, 350, 361, 363, 364, 376, 378, 407, 410, 411, 415, 421, 422, 434, 436, 438, 472 Vinca alkaloids vinblastine..........................................153, 343, 344, 435 vincristine..........................................153, 343, 344, 435

X Xenobiotic.............................57, 78, 96, 107, 108, 112, 132, 142–144, 200, 201, 210, 213, 214, 224, 225, 229, 240, 253–255, 362, 368, 398–399, 406