Contributors
Numbers in parentheses indicate the pages on which the authors’ contributions begin.
Saverio Bettuzzi, Dipartimento di Medicina Sperimentale, Sezione di Biochimica, Biochimica Clinica e Biochimica dell’Esercizio Fisico, Universita` di Parma, Via Volturno 39-43100 Parma and Istituto Nazionale Biostrutture e Biosistemi (I.N.B.B.), Rome, Italy (1, 63, 115, 133) Olesya Chayka, Molecular Haematology and Cancer Biology Unit, Institute of Child Health, University College London, United Kingdom (115) Michael Dews, Department of Pathology and Laboratory Medicine, University of Pennsylvania and Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania 19104, USA (115) Julie Y. Djeu, Department of Immunology, H. Lee Moffitt Cancer Center, Tampa, Florida 33612, USA (77) L. M. Fabbri, Department of Oncology, Hematology and Respiratory Diseases, Section of Respiratory Diseases, University of Modena and Reggio Emilia, 41100 Modena, Italy (63) F. Luppi, Department of Oncology, Hematology and Respiratory Diseases, Section of Respiratory Diseases, University of Modena and Reggio Emilia, 41100 Modena, Italy (63) P. Mazzarelli, Department of Biopathology, Institute of Anatomic Pathology, University of Rome ‘‘Tor Vergata’’, 00133 Rome, Italy (45, 93) C. Nucci, Department of Biopathology, Section of Ophthalmology, University of Rome ‘‘Tor Vergata’’, 00133 Rome, Italy (93) F. Panico, Department of Oncology, Hematology and Respiratory Diseases, Section of Respiratory Diseases, University of Modena and Reggio Emilia, 41100 Modena, Italy (63) Sabina Pucci, Department of Biopathology, Institute of Anatomic Pathology, University of Rome ‘‘Tor Vergata’’, 00133 Rome, Italy (45, 93, 115) Maximino Redondo, Department of Biochemistry, Hospital Costa del Sol, CIBER Epidemiologı´a y Salud Pu´blica, Universidad de Ma´laga, Marbella, Spain (21) F. Ricci, Department of Biopathology, Section of Ophthalmology, University of Rome ‘‘Tor Vergata’’, 00133 Rome, Italy (93)
ix
x
Contributors
Federica Rizzi, Dipartimento di Medicina Sperimentale, Sezione di Biochimica, Biochimica Clinica e Biochimica dell’Esercizio Fisico, Universita` di Parma, Via Volturno 39-43100 Parma and Istituto Nazionale Biostrutture e Biosistemi (I.N.B.B.), Rome, Italy (1, 63) Marı´a Jose Roldan, Department of Biochemistry, Hospital Costa del Sol, CIBER Epidemiologı´a y Salud Pu´blica, Universidad de Ma´laga, Marbella, Spain (21) Arturo Sala, Molecular Haematology and Cancer Biology Unit, Institute of Child Health, University College London, United Kingdom (115) L. G. Spagnoli, Department of Biopathology, Institute of Anatomic Pathology, University of Rome ‘‘Tor Vergata’’, 00133 Rome, Italy; and IRCCS San Raffaele Pisana, Rome, Italy (45, 93) Teresa Tellez, Department of Biochemistry, Hospital Costa del Sol, CIBER Epidemiologı´a y Salud Pu´blica, Universidad de Ma´laga, Marbella, Spain (21) Andrei Thomas-Tikhonenko, Department of Pathology and Laboratory Medicine, University of Pennsylvania and Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania 19104, USA (115) Sheng Wei, Department of Immunology, H. Lee Moffitt Cancer Center, Tampa, Florida 33612, USA (77)
Academic Press is an imprint of Elsevier 525 B Street, Suite 1900, San Diego, CA 92101-4495, USA 30 Corporate Drive, Suite 400, Burlington, MA 01803, USA 32 Jamestown Road, London, NW1 7BY, UK Linacre House, Jordan Hill, Oxford OX2 8DP, UK Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands
First edition 2009 Copyright # 2009 Elsevier Inc. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the Publisher. Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email:
[email protected]. Alternatively you can submit your request online by visiting the Elsevier website at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material. Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made. ISBN: 978-0-12-381429-6 ISSN: 0065-230X
For information on all Academic Press publications visit our website at www.elsevierdirect.com
Printed and bound in USA 09 10 11 12 10 9 8 7
6 5 4 3
2 1
Clusterin (CLU) and Prostate Cancer Federica Rizzi*,{ and Saverio Bettuzzi*,{ *Dipartimento di Medicina Sperimentale, Sezione di Biochimica, Biochimica Clinica e Biochimica dell’Esercizio Fisico, Universita` di Parma, Via Volturno 39-43100 Parma, Italy { Istituto Nazionale Biostrutture e Biosistemi (I.N.B.B.), Rome, Italy
I. II. III. IV. V. VI. VII. VIII.
Introduction Briefly About Clinically Relevant Human PCa CLU Signaling in Normal and Transformed Cell Lines CLU Expression in Human PCa Castration Resistant and Metastatic PCa: The Challenge of CLU Antisense Therapy Chemoprevention of PCa: A Role for CLU in the Mechanism of Action of GTCs CLU as Tumor Modulator: Is CLU a Novel Tumor Suppressor Gene? Conclusions References
The role of clusterin (CLU) in prostate tumorigenesis is probably the most highly controversial, with evidence that CLU expression is increased or decreased in different cancer models. For example, some studies showed that CLU expression is increased in advanced stages of prostate cancer and that suppression of CLU expression sensitizes prostate cancer cells to chemotherapeutic drugs killing. In contrast with the hypothesis that CLU is a positive modulator of prostate cancer, we and others have observed that CLU is downregulated during human prostate cancer progression. Accordingly, a meta-analysis of available microarray data shows that CLU mRNA is significantly downregulated in prostate cancer tissue compared to normal prostate in 14 out of 15 independent studies. Recently, it was shown that CLU is silenced by promoter methylation in the murine TRAMP-C2 cell line, as well as in the human prostate cancer cell line LNCaP. Consistently, CLU expression was found to be significantly reduced in untreated and hormone-refractory human prostate carcinomas. This data suggest the importance of epigenetic events in the regulation of CLU in prostate cancer, supporting the idea that prostate cell transformation at early stages requires CLU silencing through chromatin remodeling. # 2009 Elsevier Inc.
I. INTRODUCTION Clusterin (CLU) mRNA level was shown to increase dramatically in the rat ventral prostate following castration. CLU was therefore originally thought to be repressed by androgens (Bettuzzi et al., 1989). In the same paper, it was also shown that CLU mRNA is also upregulated in uterus following ovariectomy, thus questioning the fact that CLU is only regulated by androgens. It was later discovered that the increased CLU levels are most Advances in CANCER RESEARCH Copyright 2009, Elsevier Inc. All rights reserved.
1
0065-230X/09 $35.00 DOI: 10.1016/S0065-230X(09)05001-5
2
Federica Rizzi and Saverio Bettuzzi
likely due to castration-induced apoptosis of the prostatic epithelium rather than direct action of the androgen receptor (AR). Consequently, CLU was generally considered a marker of cell death (Lakins et al., 1998). Since the early 1990s, CLU was believed to play important roles in nearly all most important biological phenomena including cell proliferation and apoptosis, as well as in many diseases including cancer (Rosenberg and Silkensen, 1995). More recently, CLU was found disregulated in many types of cancers, including prostate cancer (PCa) (Shannan et al., 2006). It is now evident that changes in CLU expression are important events in cancer development, but the specific role of CLU in tumorigenesis is still a matter of debate. As a matter of fact, the CLU “paradox” actually arises mainly from a bulk of data reported in the literature which are sometimes conflicting. In different types of cancers, CLU has been reported to be up- or downregulated (Chen et al., 2004; Pucci et al., 2004; Redondo et al., 2000; Xie et al., 2002, 2005; Zhang et al., 2003). A substantial amount of knowledge is now available on CLU biology. Important data derived from studies conducted in the prostate. This chapter will be focusing mainly on the most recent findings concerning the important involvement of CLU in cell transformation and PCa progression, with the main ambitious aim to give an answer to the fundamental question: Is CLU a positive or a negative modulator of prostate tumorigenesis? This critical review will try to reconcile as much as possible the available experimental results and contributions from different laboratories, even when apparently contradictory, in order to shift from the CLU “paradox” to the CLU “paradigm.” We will provide the reader with the most possible complete picture concerning the role of CLU in prostate tumorigenesis. Considering that PCa is one of the major threats in veterans’ life, as well as one of the main causes of cancer-related death in western countries, we will also specifically address whether targeting CLU might pave the way for a novel therapeutic intervention against PCa.
II. BRIEFLY ABOUT CLINICALLY RELEVANT HUMAN PCa PCa is the most prevalent male cancer and the second leading cause of cancer death in men (Brawer, 2000; Damber and Aus, 2008; Delongchamps et al., 2006; Inghelmann et al., 2007). Apart from age and ethnic origin, life habits are strongly well-recognized risk factors. Clinically, PCa is diagnosed as local or advanced, and treatments range from “watch and wait” surveillance to radical local treatment or androgen-deprivation treatment (Damber
The Clusterin Paradox in Prostate Cancer
3
and Aus, 2008). Early signs of the disease are often related to lower urinary tract obstruction, but in majorities of cases PCa remains asymptomatic and may have long latency. Despite debate about effectiveness of screening at reducing specific mortality, in many countries prostate-specific antigen (PSA) test and digital rectal examination are part of routine medical checkup (Thompson and Ankerst, 2007). Throughout Canada, the United States and much of Europe, PSA screening for PCa contributed to dramatic increases in detection rates of PCa, although it remains questionably whether PSA screening significantly reduces mortality caused by PCa (Thompson and Ankerst, 2007). The ideal screening strategy should improve the detection of a disease at a stage in its natural history where treatment can be implemented to prevent death or suffering. This is particularly problematic in PCa, because of the variable natural history of the disease. Unfortunately, determination of serum PSA for diagnosis has been found to be characterized by both low specificity and low sensitivity (about 40%). Moreover, it cannot distinguish indolent from aggressive cancer. Therefore, both overdetection of some PCas and failure to diagnose cases of aggressive disease at a stage that is sufficiently early to achieve a cure occur (Thompson et al., 2005). At present, diagnosis (generally performed when PSA value is higher than 4 g/ml) is based on examination of histopathological or cytological specimens obtained from the gland by several systematic transrectal core biopsies. The most commonly used system for grading adenocarcinoma of the prostate is the Gleason score (Bostwick, 1994; Gleason and Mellinger, 1974). The system assignes a score between 2 and 10, with 2 being the least aggressive and 10 the most aggressive. This score is the sum of the two most common patterns (grades 1–5) of tumor growth. Clinical and pathologic staging of PCa involves determination of the anatomic extent and burden of tumor based on the best available data. The TNM system (primary tumor (T), regional lymph node (N), and metastases (M)) is the most widely used system for PCa staging (Bostwick, 1997). Localized PCa is the most commonly diagnosed stage. The choice of treatment (active monitoring, radical prostatectomy, or radiotherapy) is based on factors such as tumor characteristics and the patient’s life expectancy (Damber and Aus, 2008). The first sign of failure after treatment with curative intent is generally the rising of serum PSA concentration after prostatectomy. For most patients who are at high risk of systemic failure, some form of hormonal therapy is recommended and frequently started before the metastatic disease can be detected with available imaging techniques. Since the 1940s, androgen-ablative therapy has been the standard option for management of advanced PCa, but unfortunately the progression of metastatic androgen-independent PCa occurs in 70–80% of patient treated with hormonal therapy. Thus, advanced PCa represents the final stage of this disease constituting a substantial threat of morbidity and mortality.
4
Federica Rizzi and Saverio Bettuzzi
Docetaxel, alone or in combination with Estramustine, improved the survival of men with hormone-refractory PCa in comparison with Mitoxantrone and Corticosteroids (Petrylak et al., 2004; Tannock et al., 2004). These reports only show a minor improvement in survival of patients undergoing systemic treatment for advanced PCa. New biologically active drugs such as inhibitors of angiogenesis and signal transduction, vaccines, and other immunomodulators are currently under investigation.
III. CLU SIGNALING IN NORMAL AND TRANSFORMED CELL LINES CLU is a stress-inducible gene, responding in vitro to many different stimuli such as oxidative stress (Trougakos and Gonos, 2006), ionizing radiation, and heat shock (Klokov et al., 2004). Moreover, CLU expression is regulated by many cis-acting elements and trans-factors which may be responsible for the complex tissue-specific control of the gene (Michel et al., 1997). Trans-factors which have been shown to interact with the CLU promoter and regulate its activity include: Egr-1 (Criswell et al., 2005), members from the AP-1 complex (Jin and Howe, 1999), HSF1/2 (Loison et al., 2006), and B-MYB (Cervellera et al., 2000). CLU expression can be enhanced by diverse growth factors like transforming growth factor- (TGF- ) (Reddy et al., 1996a,b), nerve growth factor (NGF), and epithelial growth factor (EGF) (Gutacker et al., 1999). It has been reported that B-MYB, a transcription factor involved in the regulation of cell survival, proliferation, and differentiation, directly regulate the expression of CLU (Cervellera et al., 2000). Several evidences showed that CLU is downregulated by oncogenes and oncoproteins. CLU is downregulated in Myc-transformed, p53-null epithelial cells, in which the growth rate increases in response to c-MYC activation (ThomasTikhonenko et al., 2004). Several evidences showed that CLU is downregulated in RAS-transformed cell lines (Klock et al., 1998; Tchernitsa et al., 2004). Accordingly with the latter evidence, CLU is upregulated when HRAS is downregulated (Kyprianou et al., 1991). Recently, Chayka et al. (2009) showed that CLU expression is negatively regulated by the protooncogene MYCN via activation of members of the miR-17-92 cluster of microRNAs. CLU can modulate the activity of important signaling molecules involved in the NF-B pathway (Chayka et al., 2009; Santilli et al., 2003; Takase et al., 2008). An important physiological role of CLU recently found is to inhibit NF-B signaling through stabilization of its specific inhibitors IkBs. This activity may result in suppression of tumor cell motility (Santilli et al., 2003).
The Clusterin Paradox in Prostate Cancer
5
CLU is also subjected to epigenetic modifications and its expression can be modulated by histone acetylase and DNA methylase inhibitors in different species (Hellebrekers et al., 2007; Lund et al., 2006; Nuutinen et al., 2005; Rauhala et al., 2008; Suuronen et al., 2007). The human CLU promoter is rich in CpG islands that can be subjected to methylation, resulting in transcriptional silencing (Hellebrekers et al., 2007; Rauhala et al., 2008). The induction of chromatin rearrangements is emerging as a common feature of oncogene-induced transformation. Accordingly, RAS transformation, for example, is accompanied by silencing of CLU through promoter methylation in rat PCa cells (Lund et al., 2006). A plausible hypothesis to explain the fact that CLU is generally downregulated by oncogenes is that suppression of CLU expression is required for oncogene transformation. Further studies revealed that CLU is methylated in the TRAMP-C2 murine PCa cell line, as well as in the human PCa cell line LNCaP. Coherently its expression was found to be significantly reduced in untreated and hormonerefractory human prostate carcinomas with respect to normal tissue (Rauhala et al., 2008). A similar mechanism of epigenetic regulation was also described in tumor conditioned epithelial cells (Hellebrekers et al., 2007) in which CLU is significantly downregulated through histone H3 Lys 4 methylation and histone H3 deacetylation. In this system, CLU expression is inversely related with neoangiogenesis and cells sprouting. These findings, although requiring further studies, may suggest that a similar picture is probably true during induction of vascular tumor endothelium growth in vivo. Recently, regulation of CLU expression through acetylation/ deacetylation of histone H3 within the CLU promoter was also confirmed in hepatocellular carcinoma cells (Liao et al., 2009). All these evidences reiterate the idea that CLU silencing during neoplastic transformation occurs through chromatin remodeling and might control tumor cell proliferation, survival, and metastatic spread.
IV. CLU EXPRESSION IN HUMAN PCa CLU expression is disregulated in many types of cancers, including prostate. The role of CLU in prostate tumorigenesis is probably the most highly controversial. We found that CLU mRNA and protein products are downregulated in both low-grade and high-grade PCa, suggesting that this is an early event in PCa onset (Bettuzzi et al., 2000; Rizzi et al., 2008). Scaltriti et al. (2004a) have studied the role of CLU in CaP development and progression by Northern and Western blot analysis, immunohistochemistry, and in situ hybridization. The study was conducted in surgical specimens from the prostate of patients affected by prostatic adenocarcinoma graded from
6
Federica Rizzi and Saverio Bettuzzi
1 to 5. CLU was found downregulated in tumor samples in comparison with benign matched tissues. Only rare epithelial cells are CLU positive. Interestingly, stromal cells express CLU and the staining is much stronger in the stromal compartment. In low-grade PCa, CLU colocalized with GAS-1 (a marker of cell quiescence) to the stromal compartment and accumulated in the basal lamina. In high-grade tumors, CLU stains the remnants of stromal matrix, while epithelial cancer cells were found rarely positive. In these cells, CLU staining is confined to the cytoplasm. It might be speculated that extracellular CLU protein (sCLU) secreted by stromal cells is involved in tissue remodeling processes during stromal compartment involution due to cancer progression. Other authors later confirmed the intense staining for CLU in stromal cells, and found that only CLU staining of stroma was associated with PCa recurrence (Pins et al., 2004). These results showing high CLU distribution in the stromal compartment associated to downregulation in the tumor cells apparently disagree with data reported by July et al. (2002). In their work, authors show that CLU expression is limited to the epithelial compartment, being significantly higher in PCa of patients who received neoadjuvant hormone therapy and suggesting that CLU could play an important role in the onset of hormone refractory disease. A possible explanation of this discrepancy may rest on the existence of different CLU forms. It has to be taken into consideration the possibility that they may undergo specific changes during the different phases of neoplastic transformation. In addition, subcellular location of CLU might also be an important key for the right interpretation of its biological functions. These results also indicate that the role of CLU in PCa is very complex. At the moment, we cannot rule out the possibility that CLU might act as a negative tumor modulator in early stages of PCa, while being recruited as a positive tumor growth modulator in more advanced stages. This may very likely happen during the onset of androgen and chemotherapy resistance. Oncomine is a publicly available database that contains a large collection of gene expression experiments on human cancer (Rhodes et al., 2004). We have interrogated the database to understand whether CLU is up- or downregulated in PCa. The meta-analysis of available data showed that CLU mRNA is differentially expressed in cancer tissue compared to normal prostate: in 14 out of 15 independent studies comparing benign tissue to PCa CLU was found significantly downregulated. Interestingly, in their investigation Tomlins et al. used laser microdissected samples and found that CLU gene is differentially expressed in the stroma and in the epithelium, being more expressed in the stromal compartment (Tomlins et al., 2007). Remarkably, CLU expression is inversely proportional to the grade and/or metastatic stage of PCa in 8 out of 8 studies (to visualize the complete set of experiments please follow the link www.oncomine.org). Moreover, we found that the expression of CLU is significantly lower in PCa with Gleason
The Clusterin Paradox in Prostate Cancer
7
score higher than 7, the most reliable marker of poor prognosis (Rizzi et al., 2008). Consistently with this latter observation it was recently suggested that CLU is epigenetically downregulated in PCa and other neoplasias through promoter hypermethylation (Hellebrekers et al., 2007; Lund et al., 2006; Rauhala et al., 2008; Suuronen et al., 2007).
V. CASTRATION RESISTANT AND METASTATIC PCa: THE CHALLENGE OF CLU ANTISENSE THERAPY Prostate localized disease is generally managed by surgery or local radiation therapy. About one-third of patient treated with conventional clinical protocols will develop metastases and undergo androgen-ablation therapy. Unfortunately, too often the disease progresses toward the resistant phenotype turning to hormone-independent state, also called castration-resistant PCa (CRPC). CRPC is unresponsive to further hormonal therapy and prognosis is very poor (median survival is approximately 1 year). Despite the fact that CLU is downregulated in the majority of naı¨ve cancer cells, its expression was found upregulated in hormone-resistant PCa (July et al., 2002). Several reports describing upregulation of CLU in PCa cells resistant to conventional chemotherapy or hormonal therapy have been published (Miyake et al., 2000a,b). Preclinical data clearly indicate that CLU silencing strategies reduced the IC50 of PCa cell lines treated with conventional chemotherapeutic drugs, but this approach was not able to inhibit the growth of established PC3 tumors (Miyake et al., 2000a) or to induce apoptosis when given alone. Combined treatment of Shionogi cells with 500 nM antisense CLU oligo desossi nucleotide and paclitaxel-induced apoptosis, while either agent alone did not (Miyake et al., 2000b). Silencing CLU expression can enhance the cytotoxicity of chemotherapeutic agents (Trougakos and Gonos, 2002) as well as IR (Zellweger et al., 2002) or androgen ablation therapy (Gleave and Miyake, 2005) in PCa cell lines. More recently it was also proposed that CLU is involved in the mechanism of acquisition of chemoresistance by mediating inhibition of TRAIL-triggered apoptosis in PC3 docetaxel resistant PCa cells (Sallman et al., 2007). Recently, it was shown that repeated exposition of PC-3 cell line to docetaxel chemotherapy in vitro allows to develop a docetaxel-resistant cell subline (PC-3dR). In this system, sCLU levels expression increased 2.5-fold in the newly developed docetaxel-refractory PC-3dR cell line compared with parental PC-3 (Sowery et al., 2008). In vivo, growth of PC-3dR xenografts in nude mice was synergistically inhibited by the use of CLU antisense oligonucleotide (ASO) combined with paclitaxel or mitoxantrone (Sowery et al., 2008).
8
Federica Rizzi and Saverio Bettuzzi
The progression of PCa is now recognized to be associated with multiple changes over time in expression of specific molecule involved in pathways regulating cell proliferation and cell death. Mechanism of acquisition of resistance is highly redundant in cancer cells and this might explain why strategies triggering only one target seem to be effective “in vitro,” but are often overcome in the more complex “in vivo” setting. The rational and effective use of targeted therapies to eradicate resistant populations of tumor cells should be grounded on the premise that PCa is a dynamic disease that evolves as it progresses, and that specific molecular determinants mediating sensitivity and resistance may be relevant only during specific states of the disease. Curative approaches must account for this changing dynamic so that clinical outcomes may be improved. ASO therapy is a molecular strategy to specifically target functionally relevant genes. ASOs are chemically modified stretches of single-strand DNA complementary to mRNA regions of a target gene that inhibit translation by forming RNA/DNA duplexes, thereby reducing mRNA and protein levels of the target. Based on the evidences previously reported, ASO directed against CLU mRNA have been developed and approved for clinical trials. The CLU ASO is called OGX-011: a 21-mer modified ASO directed against the translation start site located in exon 2. Two phase I trials, having a unique pharmacodynamic endpoint, have now been completed. The first phase I trial had the primary objective to determine a biologically effective dose of OGX-011 that inhibited CLU expression in human cancer. It was used together with neoadjuvant therapy: volunteer subjects with localized PCa were treated with OGX-011 and androgen ablation therapy prior to radical prostatectomy. The first study demonstrated that active concentration of OGX-011 reached PCa tissues and inhibited CLU expression (Chi et al., 2005). A recommended phase II dose of OGX-011 in combination with androgen depletion therapy was determined (Chi et al., 2005). The second phase I trial established a recommended phase II dose of OGX-011 in combination with docetaxel. Based on the results of these phase I studies, phase II trials of combined OGX-011 and chemotherapy or hormonal therapy have been started in patients with prostate, breast, and lung cancers. All these studies are now closed, and results for prostate and lung cancers are expected for the end of 2009. Recently, the results of the phase II clinical trial in women with locally advanced or metastatic breast cancer have been published. The primary objective of this phase II trial was to assess both safety and efficacy of the combination of OGX-011 and docetaxel for metastatic breast cancer. The detected response rate was actually equal to that expected from the single agent docetaxel. On the basis of these results, the trial was disappointing because it did not meet the criteria to proceed to the second stage of drug evaluation (Chia et al., 2009).
The Clusterin Paradox in Prostate Cancer
9
The only phase II clinical trial currently ongoing is aimed to test the efficacy of OGX-011 in combination with neoadjuvant hormone therapy in men with high-risk localized PCa. As tumors progress, the number of relevant targets that contribute to the tumor phenotype may increase. For this reason, we should consider aggressive curative options during the early clinical disease states, but we must also realize that therapies that were validated in patients with metastatic disease may not target molecules that are relevant in earlier stages of the disease and vice versa. New combination targeted treatment regimens must be developed with recognition of the selective forces that shape the resistance properties of an evolving tumor and the important subset of survival and apoptotic molecules pathways that are most relevant in each clinical disease state.
VI. CHEMOPREVENTION OF PCa: A ROLE FOR CLU IN THE MECHANISM OF ACTION OF GTCs Chemoprevention is by definition a strategy for pharmacological intervention with natural or synthetic compounds that may prevent, inhibit, delay, or reverse carcinogenesis (Sporn et al., 1976). The expanded definition of cancer chemoprevention is referred to the property of a certain compound to block neoplastic inception as well as reversing the progression of transformed cells before the appearance of clinically relevant malignant lesion (William et al., 2009). Carcinogenesis is a multistep, multipath process involving several genetic and epigenetic alterations that begins with genomic instability and ends with the development of clinically relevant cancer. From the molecular point of view, this process requires activation of oncogenes, functional loss of tumor suppressor genes, altered expression of genes involved in cell-cycle control, proliferation/apoptosis, angiogenesis, and invasiveness. PCa represents an ideal candidate for chemoprevention, because of its high incidence and long latency period before the development of clinically evident disease. Several potential chemopreventive agents have been tested, including COX-2 inhibitors (Basler and Piazza, 2004), 5-reductase inhibitors (Thompson et al., 2003), vitamin D analogues and, among natural compounds, green tea extract rich in catechins (GTCs). COX-2 inhibitors interfere with the action of cycloxygenase-2, an enzyme involved in prostaglandin synthesis, cell proliferation, and angiogenesis. The key role of COX-2 inhibitors in men with PCa has been supported by largescale epidemiological investigations (Basler and Piazza, 2004). Population studies have shown that men who take anti-inflammatory drugs have less than half the risk of developing PCa.
10
Federica Rizzi and Saverio Bettuzzi
Finasteride competitively inhibits the activity of the type II 5--reductase, an enzyme whose expression is increased from high-grade prostate intraepithelial neoplasia (HGPIN, a well-recognized preneoplastic lesion) to PCa, modulating PCa risk. The most compelling study supporting 5--reductase inhibitors for PCa prevention arises from the PCa prevention trial, initiated in 1993 and reported in 2003 (Thompson et al., 2003). Ultimately, a 24.8% reduction in PCa risk was seen, although an alarming increase in the number and proportion of poorly differentiated tumors was observed (Thompson et al., 2003). Although COX-2 inhibitors and finasteride hold promises for chemoprevention, toxicity and undesired adverse effect suggest that new strategies are needed. Epidemiological studies have suggested that increased PCa risk is associated with decreased production of vitamin D. The biologically active form of vitamin D was proven to inhibit proliferation of human PCa cells, through mechanisms that include cell-cycle arrest, induction of apoptosis, and altered activation of growth factor signaling. Unfortunately, population-based study did not provide any data supporting a protective effect of vitamin D in prostate carcinogenesis (Packianathan et al., 2004). Anyhow, inhibition of spontaneous and androgen-induced prostate growth by a nonhypercalcemic calcitriol analogue was demonstrated (Crescioli et al., 2003). These studies supported the development of BXL-628, a calcitriol analogue selected for a phase II clinical trial in patients with benign prostate hyperplasia (Crescioli et al., 2004). In this system, upregulated CLU expression was used as secondary biomarker of apoptotic response. Immunohistochemistry experiments demonstrated that expression of nuclear CLU occurred in TUNEL-positive cells committed to apoptotic death. Epidemiological and case control studies have gained support for the chemopreventive effect of bioactive compounds extracted from green tea such as catechins (Jian et al., 2004). The most biologically effective catechin is EGCG. The possible anticarcinogenic activity of green tea catechins and EGCG may be explained by a number of different mechanisms extensively reviewed by Khan et al. (2006). Although the molecular mechanisms of GTCs action are still unclear, programmed cell death induced by GTCs is not related to altered activity of the members of the BCL-2 family, as EGCG did not alter the expression of BCL-2, BCL-X(L), and BAD in DU145 cells. EGCG has been shown to inhibit angiogenesis, thus causing nutrition deficiency of tumor cells. Green tea catechins have also been found to upregulate the synthesis of some hepatic phase II enzymes that are involved in detoxification of xenobiotics, including chemical carcinogens (Khan et al., 2006). GTCs may have antimetastatic potential acting through the inhibition of the proteolytic enzyme urokinase. EGCG and ECG have been demonstrated to inhibit metalloproteinase-2 (MMP-2, also known as gelatinase A) and metalloproteinase-9 (MMP-9, also known as gelatinase B). Finally, EGCG has been found to downregulate the expression of the AR in human PCa cells
The Clusterin Paradox in Prostate Cancer
11
in culture, thus inhibiting androgen action. EGCG is also an inhibitor of 5- reductase. All together, these data may account for the antiproliferative effect of EGCG on cultured human PCa cells. Interestingly, EGCG potently inhibited the growth of both SV40-immortalized PNT1A and metastatic PC-3 cells, while normal human prostatic epithelial cells were not significantly affected (Caporali et al., 2004). Under these conditions, CLU protein accumulated both in immortalized and cancer cell lines treated with IC50 doses of EGCG but not in benign untreated primary cell cultures, where CLU remained almost undetectable (Caporali et al., 2004). This result is remarkable, in that EGCG and catechins in general have been often found inhibiting gene expression and protein activity, interfering with the transcription process probably by direct binding DNA or target proteins, while CLU seems to be one of the rare genes being upregulated by GTCs. In vivo, oral administration of 0.3% GTCs in drinking water to male TRAMP (TRansgenic Adenocarcinoma of Mouse Prostate) mice, spontaneously developing PCa as a function of age, reduced PCa onset from 100% to 20% without any evidence of adverse events. CLU is expressed at basal level in transgenic young mice, but its expression decreases markedly during PCa onset and progression. At late stage of disease, cancer cells are devoid of CLU expression. A similar pattern of expression is shown by caspase 9. Animals responding to GTCs displayed recovery of CLU expression followed by reactivation of caspase 9 expression, while those refractory to GTCs did not express either CLU or caspase 9 (Caporali et al., 2004; Scaltriti et al., 2006). Based upon preclinical findings, we conducted a Proof-of Principle clinical trial to assess the potential efficacy of GTCs in the prevention of PCa in 60 patients bearing HGPIN, the most likely preinvasive stage of PCa (Bettuzzi et al., 2006). The primary endpoint of the study was to determine the impact of GTCs administration on prevalence/progression of PCa. Following 1 year of treatment, only 3% of patients which received the green tea polyphenols were diagnosed with cancer compared with 30% of the placebo group. In a recent follow-up study (Brausi et al., 2008), we showed that the inhibition of PCa progression achieved in subjects after 1 year of GTCs extract administration remained unchanged 2 years after the suspension of the treatment. Preliminary data from our laboratory are confirming that CLU is upregulated in biopsies from GTCs-treated subjects, while it remains unchanged in placebo-treated specimen (Rizzi et al., 2009).
VII. CLU AS TUMOR MODULATOR: IS CLU A NOVEL TUMOR SUPPRESSOR GENE? The first important in vivo evidence about the possible role of CLU as a tumor suppressor came from the work by Thomas-Tikhonenko et al. (2004). In their paper, the authors show that CLU-null mice are prone to
12
Federica Rizzi and Saverio Bettuzzi
development of skin cancers. The role of CLU as negative tumor modulator has been recently investigated in neuroblastoma (Chayka et al., 2009). The authors demonstrated that CLU is negatively regulated by the protooncogene MYCN through activation of the miR 17-92 microRNA cluster. The penetrance of neuroblastomas arising in MYCN-transgenic mice is significantly increased after deletion of the CLU gene, providing the first evidence that it is bona fide a tumor suppressor gene. Consistently with this finding, CLU expression is low or absent in specimens of metastatic human neuroblastomas. CLU is a known NF-B inhibitor in vivo because it stabilizes IkBs, in turn potent inhibitors of NF-B. Depletion of CLU in MYCNtransgenic mice caused an increase of metastasis, activation of NF-B signaling and epithelial-to-mesenchymal transition. Thus, inhibiting NF-B and maintaining an epithelial phenotype are likely to be critical features of the CLU tumor suppressive functions (Chayka et al., 2009). To further assess whether tumorigenesis is affected by loss of CLU expression, TRAMP mice have been crossed with CLU knockout (CluKO) mice in our laboratory. TRAMP mice express SV40 T/t antigen under the control of the minimal rat probasin promoter, which is prostate specific. The TRAMP mouse model displays in situ and invasive carcinoma of the prostate mimicking the whole spectrum of human cancer disease progression from prostate intraepithelial neoplasia (PIN) to androgen-independent disease (Greenberg et al., 1995; Kaplan-Lefko et al., 2003). CLU expression is downregulated during PCa onset and progression in the TRAMP mouse model, both at protein and mRNA level (Caporali et al., 2004). Similarly to what happens in the neuroblastoma experimental system, preliminary data suggest that inactivation of one or both CLU alleles in TRAMP mice is sufficient to result in substantial progression to more advanced invasive disease (Bettuzzi et al., 2009). Thus, cancer progression is certainly not suppressed or delayed by absence of CLU but, on the contrary, loss of CLU significantly favors the development of tumor growth, also suggesting that CLU possesses antimetastatic activity.
VIII. CONCLUSIONS The role of CLU in prostate tumorigenesis is probably the most highly controversial. Many research teams have produced a critical mass of data on CLU action and tumorigenesis. These data are fundamental: nevertheless, besides the possible biases due to different experimental models or tools, or different ways to interpret the same data, contradictions and alternative hypothesis still exist. We should now be able to take advantage of this massive amount of data for a better understanding of CLU action.
The Clusterin Paradox in Prostate Cancer
13
Therefore: is CLU a positive or a negative modulator of prostate tumorigenesis? Taking normal prostate and PCa as the paradigm, we know that CLU is the most potently overexpressed gene during apoptosis-driven rat prostate regression induced by surgical or pharmacological androgen ablation. Under physiological condition, if CLU would exert a prosurvival action, why apoptosis is induced and prostate gland shrinks concomitantly with CLU overexpression? In the model of rat prostate regression, CLU expression is strictly related to induction of apoptosis. Prostate epithelium is subjected to a strict control of cell proliferation. In the benign prostate tissue, two layers of epithelial cells are well organized around the glandular lumen: the basal cells, adhering to the basal lamina, and the fully differentiated luminal, columnar cells, specialized to produce the prostate fluid. Columnar cells originate from basal cells. In this system, the level of production of CLU is very low and only the secreted form is detectable, mostly in the fluid. After castration, CLU overexpression is clearly evident 4 days after castration (please see chapter “Nuclear CLU (nCLU) and the fate of the cell,” of vol. 104, fig. 1). When apoptosis is reduced back to low levels (16–24 days after castration), also CLU expression is almost negligible inside epithelial cells (see chapter “Nuclear CLU (nCLU) and the fate of the cell,” of vol. 104, fig. 1, 16dC and 24dC). Most of CLU protein is present in the cytoplasm of columnar cells when androgens levels are normal, but nuclear localization of CLU is easily detectable 4 days after castration, when apoptosis induction reaches its maximum level in this system. Interestingly, cells positive for nCLU staining are basal cells, responsible for production of luminal cells. Therefore, inhibition of basal cell proliferation and apoptosis induction in luminal cells by cytoplasm CLU and nCLU overexpression may explain prostate gland atrophy following androgen ablation. A putative prosurvival role for CLU has been proposed. Under physiological condition, if CLU would exert a prosurvival action, why apoptosis is induced and prostate gland shrinks when CLU is overexpressed? The same happens in females, where CLU is overexpressed in the uterus following ovariectomy (Bettuzzi et al., 1989). Uterus is highly dependent by estrogens for epithelial cells survival. Thus, it appears that cell proliferation control operated through proapoptotic nuclear CLU induction is independent from androgens, bearing a more general significance in hormone-dependent tissues. Understanding the role of CLU in tumorigenesis is further complicated by the existence of different protein form of CLU. The Oncomine database provides us the fundamental information that CLU is downregulated in the majority of naı¨ve cancers, noticeably including PCa. Anyhow, upregulation of CLU is possible in some cancer cells, especially after adjuvant hormonal therapy (July et al., 2002; Miyake et al., 2005). Scaltriti et al. (2004b,c) and Moretti et al. (2007) have investigated CLU action by comparing immortalized PNT1A cells (mimicking the early stages
14
Federica Rizzi and Saverio Bettuzzi
of transformation) to metastatic, androgen-independent PC-3 cells. In both cell models, nuclear localization of CLU was found inhibiting cell proliferation, causing cell death and also inhibiting cell migration and invasion by interaction with -actinin. Interestingly, stable cell clones overexpressing CLU (and overcoming clonogenic toxicity which massively happens when CLU is overexpressed) do not show any nuclear localization of CLU. These cells acquire resistance to apoptosis and tolerate high intracellular cytoplasmic levels of CLU. Thus, does acquisition of resistance to apoptosis requires secretion of CLU for survival? How eventually more important is the blockade of CLU entering the nucleus? Despite the fact that many in vitro evidences show that CLU is upregulated upon hormonal withdrawal or administration of cytotoxic agent acting as a prosurvival molecule (Gleave et al., 2003; Miyake et al., 2000a,b, 2003; Sowery et al., 2008), the hypothesis that in vivo CLU may indeed act as a tumor suppressor gene is corroborated by the results of a study recently published. In this work, an increased penetrance of metastatic neuroblastoma in mice was reported when one or both CLU alleles are deleted (Chayka et al., 2009). These results have been confirmed in the TRAMP murine model of PCa (Bettuzzi et al., 2009). In fact, crossing CluKO mice with TRAMP mice resulted in a strong enhancement of metastatic spread. Consistently with the hypothesis that CLU is a negative modulator of tumor growth in mammalians, we found that its expression is downregulated during the early stages of cancer progression (Bettuzzi et al., 2000; Rizzi et al., 2008; Scaltriti et al., 2004a), while its expression is restored in TRAMP mice responding to chemoprevention with GTCs (Caporali et al., 2004; Scaltriti et al., 2006). Since no mutations of CLU have been found in human cancer yet, it is likely that the mechanism of its inactivation is epigenetic. This hypothesis is corroborated by the frequent observation of CpG island methylation or histone deacetylation in the proximity of the CLU gene in different cancers (Hellebrekers et al., 2007; Lund et al., 2006; Suuronen et al., 2007). How could we reconcile this scenario with previous findings suggesting that CLU is a prosurvival oncogene? The answer to this important question probably relies in the fact that the process of tumorigenesis often exploits cellular genes, including tumor suppressor genes, to its own purposes. As many other typical tumor-suppressor genes, early-phase-associated events related to physiological action must be distinguished from late stages-associated ones, when the tumor-suppressor factor is inactivated or acquiring improper activity. For instance, it was recently shown that pRb, a paradigmatic tumor suppressor gene, is amplified and plays a promoting role in colon cancer by suppressing E2F1 and enhancing cell survival by activating the Wnt pathway (Morris et al., 2008). In a similar manner, we hypothesize that CLU may conduct a double life interfering with the NF-B signaling. On one hand, it suppresses
The Clusterin Paradox in Prostate Cancer
15
tumorigenesis and metastatic spread by inhibiting NF-B activity. But, on the other hand, since a proapoptotic NF-B signaling is often involved in replication stress induced by chemotherapeutic drugs, highly malignant cells could reactivate CLU expression to suppress NF-B and survive. The emerging concept that CLU negatively affects cancer progression is highly relevant to the clinic. Women bearing metastatic breast cancer have been treated with CLU antisense in combination with docetaxel without experiencing any improvement in the disease progression. PCa patients are being injected with ASO to downregulate CLU expression in clinical trials (Chi et al., 2005, 2008). Results of the phase II clinical trial are still awaited but, in consideration of the results that we have obtained in vivo in the TRAMP/CLU knockout model, we ask ourselves whether systemic delivery of CLU ASO could be harmful to patients instead of being curative. In spite of many reported in vitro experiments in which abolition of CLU expression in resistant cells produced abolition of drug resistance, we all well know that the real clinical setting is much different. Targeting the early phases of PCa, also taking advantage of long latency, is probably more rewarding and more likely to result in a very important clinical improvement with important social implications.
ACKNOWLEDGMENTS This work was supported by grants from the Association for International Cancer Research (AICR Grant No. 06-711) and Univerisity of Parma, FIL 2008.
REFERENCES Basler, J. W., and Piazza, G. A. (2004). Nonsteroidal anti-inflammatory drugs and cyclooxygenase-2 selective inhibitors for prostate cancer chemoprevention. J. Urol. 171, S59–S62; discussion S62-3. Bettuzzi, S., et al. (1989). Identification of an androgen-repressed mRNA in rat ventral prostate as coding for sulphated glycoprotein 2 by cDNA cloning and sequence analysis. Biochem. J. 257, 293–296. Bettuzzi, S., et al. (2000). Tumor progression is accompanied by significant changes in the levels of expression of polyamine metabolism regulatory genes and clusterin (sulfated glycoprotein 2) in human prostate cancer specimens. Cancer Res. 60, 28–34. Bettuzzi, S., et al. (2006). Chemoprevention of human prostate cancer by oral administration of green tea catechins in volunteers with high-grade prostate intraepithelial neoplasia: A preliminary report from a one-year proof-of-principle study. Cancer Res. 66, 1234–1240. Bettuzzi, S., et al. (2009). Genetic inactivation of APOJ/cluster effects on prostate tumourigenesis and metastatic spread. Oncogene [Epub ahead of print].
16
Federica Rizzi and Saverio Bettuzzi
Bostwick, D. G. (1994). Grading prostate cancer. Am. J. Clin. Pathol. 102, S38–S56. Bostwick, D. G. (1997). Staging prostate cancer—1997: Current methods and limitations. Eur. Urol. 32(Suppl. 3), 2–14. Brausi, M., et al. (2008). Chemoprevention of human prostate cancer by green tea catechins: Two years later. A follow-up update. Eur. Urol. 54, 472–473. Brawer, M. K. (2000). Prostate cancer: Where we have been, where we are, where we are going. World J. Surg. 24, 1200–1204. Caporali, A., et al. (2004). The chemopreventive action of catechins in the TRAMP mouse model of prostate carcinogenesis is accompanied by clusterin over-expression. Carcinogenesis 25, 2217–2224. Cervellera, M., et al. (2000). Direct transactivation of the anti-apoptotic gene apolipoprotein J (clusterin) by B-MYB. J. Biol. Chem. 275, 21055–21060. Chayka, O., et al. (2009). Clusterin, a haploinsufficient tumor suppressor gene in neuroblastomas. J. Natl. Cancer Inst. 101, 663–677. Chen, T., et al. (2004). Clusterin-mediated apoptosis is regulated by adenomatous polyposis coli and is p21 dependent but p53 independent. Cancer Res. 64, 7412–7419. Chi, K. N., et al. (2005). A phase I pharmacokinetic and pharmacodynamic study of OGX-011, a 2’-methoxyethyl antisense oligonucleotide to clusterin, in patients with localized prostate cancer. J. Natl. Cancer Inst. 97, 1287–1296. Chi, K. N., et al. (2008). A phase I study of OGX-011, a 2’-methoxyethyl phosphorothioate antisense to clusterin, in combination with docetaxel in patients with advanced cancer. Clin. Cancer Res. 14, 833–839. Chia, S., et al. (2009). Phase II trial of OGX-011 in combination with docetaxel in metastatic breast cancer. Clin. Cancer Res. 15, 708–713. Crescioli, C., et al. (2003). Inhibition of spontaneous and androgen-induced prostate growth by a nonhypercalcemic calcitriol analog. Endocrinology 144, 3046–3057. Crescioli, C., et al. (2004). Inhibition of prostate cell growth by BXL-628, a calcitriol analogue selected for a phase II clinical trial in patients with benign prostate hyperplasia. Eur. J. Endocrinol. 150, 591–603. Criswell, T., et al. (2005). Delayed activation of insulin-like growth factor-1 receptor/Src/ MAPK/Egr-1 signaling regulates clusterin expression, a pro-survival factor. J. Biol. Chem. 280, 14212–14221. Damber, J. E., and Aus, G. (2008). Prostate cancer. Lancet 371, 1710–1721. Delongchamps, N. B., et al. (2006). The role of prevalence in the diagnosis of prostate cancer. Cancer Control. 13, 158–168. Gleason, D., and Mellinger, G. (1974). Prediction of prognosis for prostatic adenocarcinoma by combined histological grading and clinical staging. J. Urol. 111, 7. Gleave, M., and Miyake, H. (2005). Use of antisense oligonucleotides targeting the cytoprotective gene, clusterin, to enhance androgen- and chemo-sensitivity in prostate cancer. World J. Urol. 23, 38–46. Gleave, M., et al. (2003). Antisense targets to enhance hormone and cytotoxic therapies in advanced prostate cancer. Curr. Drug Targets 4, 209–221. Greenberg, N. M., et al. (1995). Prostate cancer in a transgenic mouse. Proc. Natl. Acad. Sci. USA 92, 3439–3443. Gutacker, C., et al. (1999). Nerve growth factor and epidermal growth factor stimulate clusterin gene expression in PC12 cells. Biochem. J. 339(Pt. 3), 759–766. Hellebrekers, D. M., et al. (2007). Identification of epigenetically silenced genes in tumor endothelial cells. Cancer Res. 67, 4138–4148. Inghelmann, R., et al. (2007). Regional estimates of prostate cancer burden in Italy. Tumori 93, 380–386. Jian, L., et al. (2004). Protective effect of green tea against prostate cancer: A case-control study in southeast China. Int. J. Cancer 108, 130–135.
The Clusterin Paradox in Prostate Cancer
17
Jin, G., and Howe, P. H. (1999). Transforming growth factor beta regulates clusterin gene expression via modulation of transcription factor c-Fos. Eur. J. Biochem. 263, 534–542. July, L. V., et al. (2002). Clusterin expression is significantly enhanced in prostate cancer cells following androgen withdrawal therapy. Prostate 50, 179–188. Kaplan-Lefko, P. J., et al. (2003). Pathobiology of autochthonous prostate cancer in a preclinical transgenic mouse model. Prostate 55, 219–237. Khan, N., et al. (2006). Targeting multiple signaling pathways by green tea polyphenol (-)-epigallocatechin-3-gallate. Cancer Res. 66, 2500–2505. Klock, G., et al. (1998). Differential regulation of the clusterin gene by Ha-ras and c-myc oncogenes and during apoptosis. J. Cell. Physiol. 177, 593–605. Klokov, D., et al. (2004). IR-inducible clusterin gene expression: a protein with potential roles in ionizing radiation-induced adaptive responses, genomic instability, and bystander effects. Mutat. Res. 568, 97–110. Kyprianou, N., et al. (1991). Programmed cell death during regression of the MCF-7 human breast cancer following estrogen ablation. Cancer Res. 51, 162–166. Lakins, J., et al. (1998). Clusterin biogenesis is altered during apoptosis in the regressing rat ventral prostate. J. Biol. Chem. 273, 27887–27895. Liao, F. T., et al. (2009). Hepatitis delta virus epigenetically enhances clusterin expression via histone acetylation in human hepatocellular carcinoma cells. J. Gen. Virol. 90, 1124–1134. Loison, F., et al. (2006). Up-regulation of the clusterin gene after proteotoxic stress: Implication of HSF1–HSF2 heterocomplexes. Biochem. J. 395, 223–231. Lund, P., et al. (2006). Oncogenic HRAS suppresses clusterin expression through promoter hypermethylation. Oncogene 25, 4890–4903. Michel, D., et al. (1997). Stress-induced transcription of the clusterin/apoJ gene. Biochem. J. 328(Pt. 1), 45–50. Miyake, H., et al. (2000a). Antisense TRPM-2 oligodeoxynucleotides chemosensitize human androgen-independent PC-3 prostate cancer cells both in vitro and in vivo. Clin. Cancer Res. 6, 1655–1663. Miyake, H., et al. (2000b). Acquisition of chemoresistant phenotype by overexpression of the antiapoptotic gene testosterone-repressed prostate message-2 in prostate cancer xenograft models. Cancer Res. 60, 2547–2554. Miyake, H., et al. (2003). Resistance to cytotoxic chemotherapy-induced apoptosis in human prostate cancer cells is associated with intracellular clusterin expression. Oncol. Rep. 10, 469–473. Miyake, H., et al. (2005). Enhanced expression of the secreted form of clusterin following neoadjuvant hormonal therapy as a prognostic predictor in patients undergoing radical prostatectomy for prostate cancer. Oncol. Rep. 14, 1371–1375. Moretti, R. M., et al. (2007). Clusterin isoforms differentially affect growth and motility of prostate cells: Possible implications in prostate tumorigenesis. Cancer Res. 67, 10325–10333. Morris, E. J., et al. (2008). E2F1 represses beta-catenin transcription and is antagonized by both pRB and CDK8. Nature 455, 552–556. Nuutinen, T., et al. (2005). Induction of clusterin/apoJ expression by histone deacetylase inhibitors in neural cells. Neurochem. Int. 47, 528–538. Packianathan, S., et al. (2004). Designing a randomized phase I/II prostate cancer chemoprevention trial using 1alpha-hydroxy-24-ethyl-cholecalciferol, an analogue of vitamin D3. Cancer J. 10, 357–367. Petrylak, D. P., et al. (2004). Docetaxel and estramustine compared with mitoxantrone and prednisone for advanced refractory prostate cancer. N. Engl. J. Med. 351, 1513–1520. Pins, M. R., et al. (2004). Clusterin as a possible predictor for biochemical recurrence of prostate cancer following radical prostatectomy with intermediate Gleason scores: A preliminary report. Prostate Cancer Prostatic Dis. 7, 243–248.
18
Federica Rizzi and Saverio Bettuzzi
Pucci, S., et al. (2004). Modulation of different clusterin isoforms in human colon tumorigenesis. Oncogene 23, 2298–2304. Rauhala, H. E., et al. (2008). Clusterin is epigenetically regulated in prostate cancer. Int. J. Cancer 123, 1601–1609. Reddy, K. B., et al. (1996a). Transforming growth factor beta (TGF beta)-induced nuclear localization of apolipoprotein J/clusterin in epithelial cells. Biochemistry 35, 6157–6163. Reddy, K. B., et al. (1996b). Interaction of transforming growth factor beta receptors with apolipoprotein J/clusterin. Biochemistry 35, 309–314. Redondo, M., et al. (2000). Overexpression of clusterin in human breast carcinoma. Am. J. Pathol. 157, 393–399. Rhodes, D. R., et al. (2004). ONCOMINE: A cancer microarray database and integrated datamining platform. Neoplasia 6, 1–6. Rizzi, F., et al. (2008). A novel gene signature for molecular diagnosis of human prostate cancer by RT-qPCR. PLoS ONE 3, e3617. Rizzi, F., et al. (2009). Clusterin gene expression as a novel biomarker of response with GTE in humans. Eur. Urol. Suppl. 8, 345. Rosenberg, M. E., and Silkensen, J. (1995). Clusterin: Physiologic and pathophysiologic considerations. Int. J. Biochem. Cell Biol. 27, 633–645. Sallman, D. A., et al. (2007). Clusterin mediates TRAIL resistance in prostate tumor cells. Mol. Cancer Ther. 6, 2938–2947. Santilli, G., et al. (2003). Essential requirement of apolipoprotein J (clusterin) signaling for IkappaB expression and regulation of NF-kappaB activity. J. Biol. Chem. 278, 38214–38219. Scaltriti, M., et al. (2004a). Clusterin (SGP-2, ApoJ) expression is downregulated in low- and high-grade human prostate cancer. Int. J. Cancer 108, 23–30. Scaltriti, M., et al. (2004b). Clusterin overexpression in both malignant and nonmalignant prostate epithelial cells induces cell cycle arrest and apoptosis. Br. J. Cancer 91, 1842–1850. Scaltriti, M., et al. (2004c). Intracellular clusterin induces G2-M phase arrest and cell death in PC-3 prostate cancer cells1. Cancer Res. 64, 6174–6182. Scaltriti, M., et al. (2006). Molecular classification of green tea catechin-sensitive and green tea catechin-resistant prostate cancer in the TRAMP mice model by quantitative real-time PCR gene profiling. Carcinogenesis 27, 1047–1053. Shannan, B., et al. (2006). Challenge and promise: roles for clusterin in pathogenesis, progression and therapy of cancer. Cell Death Differ. 13, 12–19. Sowery, R. D., et al. (2008). Clusterin knockdown using the antisense oligonucleotide OGX011 re-sensitizes docetaxel-refractory prostate cancer PC-3 cells to chemotherapy. BJU Int. 102, 389–397. Sporn, M. B., et al. (1976). Prevention of chemical carcinogenesis by vitamin A and its synthetic analogs (retinoids). Fed. Proc. 35, 1332–1338. Suuronen, T., et al. (2007). Epigenetic regulation of clusterin/apolipoprotein J expression in retinal pigment epithelial cells. Biochem. Biophys. Res. Commun. 357, 397–401. Takase, O., et al. (2008). Inhibition of NF-kappaB-dependent Bcl-xL expression by clusterin promotes albumin-induced tubular cell apoptosis. Kidney Int. 73, 567–577. Tannock, I. F., et al. (2004). Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N. Engl. J. Med. 351, 1502–1512. Tchernitsa, O. I., et al. (2004). Transcriptional basis of KRAS oncogene-mediated cellular transformation in ovarian epithelial cells. Oncogene 23, 4536–4555. Thomas-Tikhonenko, A., et al. (2004). Myc-transformed epithelial cells down-regulate clusterin, which inhibits their growth in vitro and carcinogenesis in vivo. Cancer Res. 64, 3126–3136. Thompson, I. M., and Ankerst, D. P. (2007). Prostate-specific antigen in the early detection of prostate cancer. CMAJ 176, 1853–1858. Thompson, I. M., et al. (2003). The influence of finasteride on the development of prostate cancer. N. Engl. J. Med. 349, 215–224.
The Clusterin Paradox in Prostate Cancer
19
Thompson, I. M., et al. (2005). Screening for prostate cancer: Opportunities and challenges. Surg. Oncol. Clin. N. Am. 14, 747–760. Tomlins, S. A., et al. (2007). Integrative molecular concept modeling of prostate cancer progression. Nat. Genet. 39, 41–51. Trougakos, I. P., and Gonos, E. S. (2002). Clusterin/apolipoprotein J in human aging and cancer. Int. J. Biochem. Cell Biol. 34, 1430–1448. Trougakos, I. P., and Gonos, E. S. (2006). Regulation of clusterin/apolipoprotein J, a functional homologue to the small heat shock proteins, by oxidative stress in ageing and age-related diseases. Free Radic. Res. 40, 1324–1334. William, W. N., Jr., et al. (2009). Molecular targets for cancer chemoprevention. Nat. Rev. Drug Discov. 8, 213–225. Xie, M. J., et al. (2002). Expression of clusterin in human pancreatic cancer. Pancreas 25, 234–238. Xie, D., et al. (2005). Up-regulated expression of cytoplasmic clusterin in human ovarian carcinoma. Cancer 103, 277–283. Zellweger, T., et al. (2002). Enhanced radiation sensitivity in prostate cancer by inhibition of the cell survival protein cluster. Clin. Cancer Res. 8, 3276–3284. Zhang, L. Y., et al. (2003). Loss of clusterin both in serum and tissue correlates with the tumorigenesis of esophageal squamous cell carcinoma via proteomics approaches. World J. Gastroenterol. 9, 650–654.
The Role of Clusterin (CLU) in Malignant Transformation and Drug Resistance in Breast Carcinomas Maximino Redondo, Teresa Tellez, and Marı´a Jose Roldan Department of Biochemistry, Hospital Costa del Sol, CIBER Epidemiologı´a y Salud Pu´blica, Universidad de Ma´laga, Marbella, Spain
I. II. III. IV. V. VI.
Introduction Localization of CLU in Breast Carcinomas Relationship Between Cytoplasmic CLU and Apoptosis in Breast Carcinomas Role of CLU in Tumorigenesis and Progression of Breast Carcinomas Prognostic Significance of CLU Expression in Breast Carcinomas CLU and Resistance to Treatment in Breast Carcinomas A. CLU and Resistance to Chemotherapy B. CLU and Resistance to Antiestrogen C. CLU and Dexamethasone VII. Conclusions References Breast cancer is the main cause of cancer-related death among women in Western countries. Current research is focused on identifying antiapoptotic proteins which could be a possible target for novel chemotherapeutic drugs. Secretory clusterin (sCLU) is an extracellular chaperone that has been functionally implicated in DNA repair, cell-cycle regulation, apoptotic cell death and tumorigenesis. The implication of sCLU in carcinogenesis and the progression of breast carcinomas make it an interesting gene, worthy of investigation. It has been reported to present powerful antiapoptotic activity and to perform a prosurvival function with most therapeutic treatments for breast cancer. This review summarizes our current understanding of the role of CLU in tumorigenesis, progression, and response to treatment in breast carcinomas. # 2009 Elsevier Inc.
ABBREVIATIONS AS-ODN, Antisense-oligodeoxynucleotides; CLU, Clusterin; nCLU, Nuclear clusterin; sCLU, Secreted clusterin; ER, Endoplasmic reticulum; ERþ, Estrogen receptor positive; HCC, Hepatocellular carcinoma; HDI, Histone deacetylase inhibitors; HSPs, Heat-shock proteins; IR, Ionizing radiation; MHC, Major histocompatibility complex; PSA, Prostatic specific antigen;
Advances in CANCER RESEARCH Copyright 2009, Elsevier Inc. All rights reserved.
21
0065-230X/09 $35.00 DOI: 10.1016/S0065-230X(09)05002-7
Maximino Redondo et al.
22
siRNA, Small-interfering RNA; SCID, Severe combined immune deficiency; TUNEL, Terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end-labeling
I. INTRODUCTION Breast cancer is the most common cancer among women, excluding nonmelanoma skin cancers, and is the commonest cause of cancer death in women worldwide. One in three women who are diagnosed with cancer will be diagnosed with breast cancer. Breast cancer rates have risen by about 30% in the past 25 years in Western countries, due in part to increased screening, which detects the cancer in earlier stages. In the United States, though, breast cancer rates decreased by 10% between 2000 and 2004, due in part to a reduction in the use of hormone replacement therapy. Although breast cancer rates are rising in many Western countries, deaths from the disease have decreased in some countries as a result of improved screening and treatment (Ravdin et al., 2007). The lifetime probability of developing breast cancer in developed countries is about 4.8%, according to the American Cancer Society (the probability is about 13% for any type of cancer). In developing countries, the lifetime probability of developing breast cancer is about 1.8% (Jemal et al., 2005). In approximately one-third of women with nodal-negative breast cancer, the disease recurs, and about two-thirds of women with nodal-positive breast cancer experience a tumor relapse during the10 years following local-regional therapy (Early Breast Cancer Trialists’ Collaborative Group, 2000). These data highlight the need for more sensitive and specific prognostic markers, allowing better identification of patients who bear a high risk of tumor relapse and may profit from more closely meshed follow-up and from a more aggressive therapy. Clusterin (CLU) is a ubiquitously distributed glycoprotein in mammals that was first described in rat Sertoli cells in 1982 (Kissinger et al., 1982) and in ram rete testis fluid (Blaschuk et al., 1983; Fritz et al., 1983). Subsequently, it was found to be the most potently induced gene in the rat ventral prostate after androgen ablation (Bettuzzi et al., 1989). CLU has been shown to present important roles in various physiopathological processes and found to be altered in various human carcinomas (Trougakos and Gonos, 2002). In a similar way to small heat-shock proteins (HSPs), secreted CLU (sCLU) has been shown to interact with a variety of unrelated proteins that had been subjected to misfolding under raised or reduced temperatures, thus
Clusterin and Breast Cancer
23
establishing the chaperone activity of this molecule (Humphreys et al., 1999) (see chapter “Cell protective functions of secretory CLU (sCLU)” of Vol. 104). Various observations indicate an association of CLU expression with contradictory functions, including cell survival, tumor progression, treatment resistance in vivo, and apoptosis (Gleave and Miyake, 2005; Trougakos et al., 2004). These apparently ambiguous functions have been attributed to the existence of two different but related CLU protein forms, a glycosylated and an apparently nonglycosylated form (see chapters “Clusterin (CLU): From one gene and two transcripts to many proteins” and “Nuclear CLU (nCLU) and the fate of the cell” of Vol. 104). In particular, a shorter form (nuclear CLU (nCLU)) with an apparent size ranging from 45 to 50 kDa and targeting the nucleus has also been identified. Enhanced expression of nCLU is generally associated with cell death (Leskov et al., 2003; Pucci et al., 2004). The proapoptotic activity of nCLU is also dependent on calcium ions. Depletion of intracellular calcium causes extensive death in prostate carcinoma cells (Caccamo et al., 2005) (see chapter “Regulation of CLU expression by calcium” of Vol. 104). The mechanisms of production of nCLU are still not clear. Tentative explanations proposed are (i) alternative translation initiation (Moretti et al., 2007; Scaltriti et al., 2004); (ii) alternative splicing (Leskov et al., 2003). In any case, nCLU protein does not undergo alpha/beta cleavage, nor does this protein appear to be extensively glycosylated (Reddy et al., 1996; Scaltriti et al., 2004). Confocal microscopy experiments have revealed an apparently inactive nuclear CLU form in the cytoplasm of nonirradiated cells (Yang et al., 2000) that translocate to nuclear regions after ionizing radiation, colocalizing with the nuclear Ku70/86 heterodimer involved in apoptosis induction and DNA repair (Yang et al., 1999, 2000). The binding of nCLU and Ku70 can lead to the formation of Ku70–Ku80 dimers that can affect DNA nonhomologous repair and eventually lead to genomic instability and cell death.
II. LOCALIZATION OF CLU IN BREAST CARCINOMAS In breast carcinomas, nuclear staining of CLU is found infrequently. In a recent study, only 3 out of 114 invasive tumors presented nuclear staining by immunohistochemistry and in situ hybridization (Redondo et al., 2000). In another study by Kru¨ger et al. (2007), CLU immunoreactivity was also found to be restricted to the cytoplasm of breast tumor cells. These observations concordantly demonstrate that only the cytoplasmic isoform of CLU is
24
Maximino Redondo et al.
expressed in breast carcinoma cells, which may potentially render a survival advantage to them. Similar results have been found in ovarian carcinomas (Xie et al., 2005a), in renal cell carcinomas (Kurahashi et al., 2005), and in colon carcinoma and adenoma (Andersen et al., 2007; Xie et al., 2005b). Interestingly, Pucci et al. (2004) observed the expression of nCLU in normal epithelial cells, and a shift in subcellular localization to the cytoplasm in tumors. Other studies have failed to observe nuclear expression in normal colon tissues (Andersen et al., 2007; Xie et al., 2005b). The reason for this discrepancy is unknown, although it may be hypothesized that nCLU-positive cells will be very rapidly cleared off the tissue by apoptotic death and thus their detection would be easier under conditions in which nCLU is acutely induced, but more difficult in steady-state conditions. Anyhow, the relative lack of production of the nCLU in breast cancer cells suggests that full-length CLU cDNA and/or protein precursor might be differentially processed in normal versus transformed cells (Moretti et al., 2007). Neoplastic transformation might require the prosurvival sCLU action prevailing over the proapoptotic properties of nCLU. The distribution of CLU isoform expression found in tumors may support the hypothesis that protection from apoptosis is due either to the disappearance of the nuclear form in tumors or to the inhibition of the translocation in the nucleus. In fact, enhanced tumor cell survival is correlated with the loss of the proapoptotic form and also with the overexpression of the secreted form which might be involved in the heightened invasive ability and motility and is probably relevant in enhancing the metastatic potential.
III. RELATIONSHIP BETWEEN CYTOPLASMIC CLU AND APOPTOSIS IN BREAST CARCINOMAS Studies performed in tissue tumor samples have shown that CLU gene expression in the cytoplasm of breast cancer cells is not directly associated with programmed cell death in breast tumors. In fact, a lower apoptotic index in CLU-positive tumors than in negative ones is found in breast carcinomas (Redondo et al., 2000). In three reports (Chen et al., 2003, Pucci et al., 2004, Xie et al., 2005a), a correlation between CLU and apoptosis was also found. The TUNEL assay, in which cells undergoing apoptosis were stained, showed a negative correlation between CLU expression and the apoptotic index in these tumors. In experimental studies, Sensibar et al. (1995) reported a significant increase in cell death after transfection of CLU antisense oligonucleotides into LnCaP cells. In addition, it was observed that, by using small-interfering RNA, CLU knockdown induced a significant reduction of human tumor cell growth and greater
Clusterin and Breast Cancer
25
rates of spontaneous endogenous apoptosis (Trougakos et al., 2004). On the other hand, other papers support the evidence that CLU knockout alone is not sufficient to reduce the proliferation of tumour cells, or to induce spontaneous apoptosis (Gleave et al., 2001; Miyake et al., 2000a, 2001a, 2005a; Yamanaka et al., 2005). In laryngeal carcinomas, a statistical significant association between the expression of cytoplasmic CLU and bcl-2 proteins has been found (Redondo et al., 2006). In the report by Trougakos et al. (2004), CLU knockdown by siRNA was associated with downregulation of bcl-2 in two sarcoma cell lines. The exact mechanism of this association has yet to be elucidated (Trougakos et al., 2004). In breast carcinoma, to date, no report of this correlation has been published. However, the precise biological function of CLU in mechanisms of cell death/survival remains to be defined.
IV. ROLE OF CLU IN TUMORIGENESIS AND PROGRESSION OF BREAST CARCINOMAS In breast carcinomas, CLU expression has been related to tumorigenesis and progression (Redondo et al., 2000). In addition to breast carcinomas (Kru¨ger et al., 2007; Redondo et al., 2000; Zhang et al., 2006), overexpression of sCLU has been found in the majority of tumors investigated, including prostate cancer (July et al., 2002; Miyake et al., 2004; Zellweger et al., 2003), ovarian carcinoma (Xie et al., 2005a), lung carcinoma (July et al., 2004), hepatocellular carcinoma (Aigelsreiter et al., 2009; Kang et al., 2004), renal cell carcinomas (Miyake et al., 2002; Parczyk et al., 1994), melanoma (Busam et al., 2006), bladder carcinoma (Miyake et al., 2001a, 2002), cervical carcinoma (Watari et al., 2008), and colon cancers (Chen et al., 2003; Kevans et al., 2009; Pucci et al., 2004). A few reports do, however, suggest decreased sCLU levels in some cancers (Bettuzzi et al., 2002; Chen et al., 2004; Santilli et al., 2003; Scaltriti et al., 2004; Thomas-Tikhonenko et al., 2004; Xie et al., 2002; Zhang et al., 2003). In prostate carcinomas, it was observed an accumulation of sCLU in specific, well-defined areas of the stromal compartment of the tumors, while prostate cancer cells were devoid of CLU expression (Scaltriti et al., 2004). A very recent report (Chayka et al., 2009) presented evidence of the potential role of CLU as a tumor suppressor gene in neuroblastomas. The dual nature of CLU as a tumor suppressor and as a promoter may not be contradictory. The outcome may depend largely on the protein form available and prevailing in the cell, as well as the time and context of CLU expression during cell lifetime and the microenvironment (Trougakos et al., 2009a).
26
Maximino Redondo et al.
Redondo et al. (2000) examined the expression of this molecule in a large cohort of breast tumors, and found that CLU is expressed in the malignant epithelium from early to late stages of carcinogenesis in the breast. When measured in serum, CLU was significantly higher in breast carcinoma patients than in their controls (Doustjalali et al., 2004). Furthermore, CLU has been associated with spontaneous breast cancer in mice (Sun et al., 2007). Upregulation of this gene in breast cancer is closely associated with the different steps of tumor progression from normal tissue to premalignant and malignant breast lesions, with a greater expression in lymph node metastasis (Redondo et al., 2000). Thus, only 19% of benign lesions presented positive staining for CLU. In contrast, the frequency of CLU-positive samples increased in atypical hyperplasias (47%), intraductal carcinomas (49%), and invasive carcinomas (53%) (Fig. 1). Furthermore, CLU expression is associated with the progression from primary to metastatic carcinomas in lymph nodes. Redondo et al. (2000) reported that 80% of metastatic nodes had positive CLU expression, and 67% of primary carcinomas without CLU expression became positive in lymph node metastases, while 88% of the CLU-positive primary carcinomas were also immunoreactive in metastases. Very similar results have been obtained in ovarian and colorectal carcinomas (Xie et al., 2005a,b). Anyhow, the work by Chen et al. (2004) shows that the morphologically normal cells close to the colon tumor cells are also CLU-positive in their cytoplasm. Interestingly, the overexpression of CLU in hepatocellular carcinomas increases cell migration and the formation of metastatic tumor nodules in the liver (Lau et al., 2006). The latter study also reported that the frequency of CLU overexpression increased significantly in metastatic HCCs compared with that in primary tumors. These findings show that the upregulation of this gene is associated with the different step of tumor progression. However, different results have been obtained in prostate carcinomas concerning CLU expression in tumor cells (July et al., 2002; Steinberg et al., 1997). Moreover, although most reports have indicated that CLU is upregulated in neoplastic colorectal tissues (Andersen et al., 2007; Kevans et al., 2009; Pucci et al., 2004; Xie et al., 2005b) others have indicated downregulation (Chen et al., 2004). A possible explanation for this discrepancy may be the existence of different CLU forms, and the possibility that they may undergo specific changes during the different phases of neoplastic transformation. Consistently with this, it was recently suggested that CLU is epigenetically downregulated in prostate cancer and more generally in neoplasia disease, through promoter hypermethylation (Lund et al., 2006; Rauhala et al., 2008). As mentioned above, some authors have suggested that CLU secreted by stromal cells could be involved in tissue remodeling processes during stromal compartment involution due to cancer progression (Scaltriti et al., 2004). With regard to osteosarcoma cell lines, Trougakos et al. (2005) proposed that although
Clusterin and Breast Cancer
27
A
B
C
Fig. 1 Immunohistochemical analysis of CLU expression in proliferative lesions, DCIS, and invasive carcinomas. (A) Atypical ductal hyperplasia showing positive staining. (B) Strong cytoplasmic CLU staining is seen in a ductal carcinoma in situ. (C) Cytoplasmic expression of CLU in a poorly differentiated breast carcinoma. All sections were counterstained with hematoxylin.
28
Maximino Redondo et al.
both extracellular and intracellular CLU at low/moderate levels are cytoprotective, CLU may become highly cytostatic and/or cytotoxic if it accumulates intracellularly in high amounts, either by direct synthesis or by uptake from the extracellular milieu. CLU probably protects carcinoma cells and contributes to the highly metastatic phenotype of those cells. Its presence at cell surfaces and its ubiquitous presence in virtually all types of biological fluids may protect cell membranes exposed to deleterious components. This protective effect may favor the metastatic phenotype of tumor cells, which may migrate through tissues and fluids without excessive damage (Jordan Stark et al., 1992). Thus, CLU induction is a reactive response to environmental changes. Accordingly, it could be thought of as an extracellular version of a heat-shock protein (Humphreys et al., 1999) (See chapters “The chaperone action of CLU and its putative role in quality control of extracellular protein folding” and “Cell protective functions of secretory CLU (sCLU)” of Vol. 104). Overexpression of CLU (Fig. 1) may thus represent an acquired phenotypic feature which facilitates local invasion and dissemination of breast tumor cells; in fact, a positive correlation between CLU expression and tumor size has been found (Redondo et al., 2000). In colorectal carcinomas, a significant positive correlation between the overexpression of CLU and advanced clinical stage has also been observed (Xie et al., 2005b). Therefore, CLU could be an important factor in determining the aggressive nature of a given breast tumor. In this sense, CLU has also been associated with negativity for estrogen and progesterone receptors and with high histological grade (Redondo et al., 2000). In another study of breast carcinomas (Kru¨ger et al., 2007), CLU expression was also found to correlate with some prognostic factors, like grading and Ki67 labeling index. However, the percentage of tumors with positive CLU expression was only found in 26% of cases. In prostate cancer, CLU levels have also been correlated with pathological grade on both biopsy and radical prostatectomy specimens (Steinberg et al., 1997). In laryngeal tumors its expression was associated with local invasiveness (Redondo et al., 2006, 2009). Redondo et al. (2000, 2006) compared the expression of CLU mRNA by in situ hybridization with its protein expression by immunohistochemistry in the breast and also in laryngeal carcinomas. Another study has reported differences between the two methods in astrocytes and neurons (Pasinetti et al., 1994). In breast carcinomas, only 2 cases of 30 randomly chosen ones were positive by in situ hybridization and negative by immunohistochemistry (Redondo et al., 2000). Similar results were found in laryngeal carcinomas, where only one tumor presented RNA expression in cytoplasm without CLU protein expression (Redondo et al., 2006). Different CLU forms or differences in antibody affinity according to cell type or state may explain these results.
Clusterin and Breast Cancer
29
Most authors are in agreement with the recent suggestion that tumor cell survival is connected with the overexpression of sCLU and the loss of nCLU (Pucci et al., 2004). This theory has been supported by data which suggest that cells must suppress sCLU to stimulate cell death. One proposed mechanism for this death process is through the activation of the p53 tumor suppressor gene (Criswell et al., 2003) (see chapter “Regulation of CLU gene expression by oncogenes, epigenetic regulation and the role of CLU in tumorigenesis” of this volume). P53 can suppress basal as well as IR-induced sCLU expression in both MCF-7 (breast cancer) and HCT116 (colon cancer) cells by repressing CLU promoter activity and transcription (Criswell et al., 2003). CLU expression has also been reported to be directly regulated by B-MYB, a transcription factor involved in the regulation of cell survival, proliferation, and differentiation (Cervellera et al., 2000). Another transcription factor that has been reported to be regulated by CLU is NF-B. The modulation of NF-B activity is believed of importance for cell survival, motility, proliferation, and transformation. The loss of CLU expression in cells that depend on NF-B activity for proliferation or chemoresistance could lead to tumor progression (Santilli et al., 2003). Moreover, increased apoptosis in sCLU-depleted cells is correlated with altered ratios of proapoptotic to antiapoptotic Bcl-2 protein family members (Trougakos et al., 2009b). In the cytoplasm, sCLU depletion disrupts the Ku70–Bax complex and triggers Bax activation and relocation to the mitochondria. Therefore, sCLU binds and stabilizes the Ku70–Bax protein complex that serves as a cytosol retention factor for Bax. Raised sCLU levels may enhance tumorigenesis by interfering with Bax proapoptotic activities and contribute to one of the major characteristics of cancer cells, their resistance to apoptosis (Trougakos et al., 2009b). CLU has also been shown to inhibit c-mycinduced apoptosis (Zhang et al., 2005). Very interesting is the relationship of CLU with the expression of class I major histocompatibility complex antigens (MHC class I) in breast carcinomas. Tumor cells may gain the ability to evade immune responses by downregulation or the loss of expression of MHC class I molecules. In an attempt to identify the relation of MHC antigenic loss to factors influencing Fasmediated apoptosis by T lymphocytes, Redondo et al. (2003) studied the possible relationship with cytoplasmic CLU expression. Comparative analysis revealed a statistically significant association between MHC class I antigen expression and CLU expression in in situ and invasive carcinomas, which may influence the lysis of tumoral cells, since CLU has been shown to inhibit Fas-mediated apoptosis in some systems (Miyake et al., 2001b). On the other hand, in metastatic tumor cells the situation is reversed, as CLU is upregulated while MHC class I antigens are downregulated (Redondo et al., 2003). The presence of CLU expression and the lack of HLA class I antigens in metastatic nodes may reflect an important survival advantage of
30
Maximino Redondo et al.
metastatic cells. It could be hypothesized that the simultaneous presence of several mechanisms involved in tumor immune evasion must be the result of a progressive selection of characteristics that are advantageous for tumor survival in a competent host.
V. PROGNOSTIC SIGNIFICANCE OF CLU EXPRESSION IN BREAST CARCINOMAS Resistance to adjuvant therapies and disease recurrence could reliably be predicted by assessing biochemical factors strictly related to tumor cell biology and tumor aggressiveness. Although it has been reported that CLU protein is related to carcinogenesis and the progression of breast and other carcinomas (Chen et al., 2003; Redondo et al., 2000; Xie et al., 2005a), it is necessary to evaluate its influence on the survival of cancer patients. Very few studies have reported on the relationship between CLU expression and survival in human tumors. It is tempting to speculate that tumors presenting the aforementioned molecule would be prone to a poorer clinical course than tumors with no such expression, and indeed, most studies that have analyzed survival rates for CLU-positive and CLU-negative cancers have reported CLU positivity to have a poor prognostic correlation (Kang et al., 2004; Kevans et al., 2009; Kru¨ger et al., 2006; Kurahashi et al., 2005; Miyake et al., 2002, 2005b; Watari et al., 2008). Anyhow, the fact that in some studies CLU was an independent predictor factor while in others CLU was not an independent predictor of biochemical recurrence or poor prognosis may further complicate the picture. In breast carcinoma, cytoplasmic CLU also correlates with a poor prognosis in survival analysis (Kru¨ger et al., 2007; Zhang et al., 2006). Zhang et al. (2006) reported that the survival time of patients with CLU expression was significantly shorter than that of patients with no detectable CLU expression. According to Kru¨ger et al. (2007), CLU immunoreactivity shows an independent prognostic relevance concerning the prediction of relapse-free survival. However, in a series of breast carcinomas (Redondo et al., 2000), CLU was found to be related to carcinogenesis and progression but it was not an independent prognostic marker in this series. One report on colon carcinoma has presented evidence that CLU is associated with decreased survival (Kevans et al., 2009). In this work, survival was significantly associated with intensity and percentage of epithelial cytoplasmic staining, and intensity and percentage of stromal cytoplasmic staining. In prostate carcinomas, CLU expression in stromal cells, but not in cancer cells, has also been used as a possible predictor for biochemical recurrence following radical prostatectomy (Pins et al., 2004). In invasive cervical cancer and nonpapillary renal cell carcinoma,
Clusterin and Breast Cancer
31
overexpression of CLU is an independent prognostic factor (Miyake et al., 2002; Watari et al., 2008). Contrary to the results reported for most groups, a favorable course has been reported in pancreatic adenocarcinoma (Xie et al., 2002) and nonsmall cell carcinomas (Albert et al., 2007). This may be related to a number of clinical and technical factors. There is great variability in the posttranslational modifications of CLU in different species and in different tissue types (Appel et al., 1996; Jenne and Tschopp, 1992). This contributes to the expression of distinct glyco-forms. More research examining expression levels of the different forms of CLU in these cancers is needed. Therefore, there is clinical evidence implicating cytoplasmic CLU expression with poor outcome in several human tumors. In these cases, determining CLU may help to identify patients at higher risk of tumor recurrence, who would be most likely to benefit from additional adjuvant therapy after surgery. In fact, CLU is a key molecule in inducing resistance to adjuvant therapy in breast carcinomas (Biroccio et al., 2005; Redondo et al., 2007; So et al., 2005).
VI. CLU AND RESISTANCE TO TREATMENT IN BREAST CARCINOMAS A. CLU and Resistance to Chemotherapy Metastatic breast cancer is usually not a curable condition. However, systemic (bodywide) treatment can prolong life, delay the progression of the cancer, relieve cancer-related symptoms, and improve quality of life. Options for systemic treatment in women with metastatic breast cancer include endocrine therapy, conventional chemotherapy, and a special class of biologic agents that target a specific protein, called HER2, that is found in some breast cancer cells. There are no well-defined factors that predict whether a breast cancer will respond to chemotherapy. Chemotherapy may be recommended when a tumor lacks both hormone receptors and HER2. Among women with metastatic breast cancer who have not previously received chemotherapy for metastatic disease, between 50% and 75% will respond to an initial course of chemotherapy. Therefore, chemotherapy resistance is a major problem in the management of patients with breast cancer and the underlying basis for most cancer deaths. Clinical resistance of solid tumors such as breast cancer is likely to be multifactorial and heterogeneous. One of the primary goals in breast cancer research is to develop new ways of inhibiting cancer cell
32
Maximino Redondo et al.
growth, in part by improving the effectiveness of existing cancer treatment regimens. It has been hypothesized that sCLU gene silencing using siRNA or other techniques may ultimately be developed into attractive antitumor therapies (Trougakos et al., 2004). It has been shown that sCLU knockdown in human cancer cells, using siRNA-mediated CLU gene silencing, induces a significant reduction in cellular growth and higher rates of spontaneous endogenous apoptosis (Trougakos et al., 2004), although (as discussed before: Gleave et al., 2001; Miyake et al., 2000a, 2001a, 2005a; Yamanaka et al., 2005) several evidences shows that CLU silencing alone do not affect cell growth or apoptosis rate. Recent findings support the concept that silencing sCLU expression can enhance the cytotoxicity of various chemotherapeutic agents (Gleave and Jansen, 2003; Trougakos and Gonos, 2002) as well as ionizing radiation (Criswell et al., 2003) (see chapters “CLU and chemoresistance” and “Regulation of CLU gene expression by oncogenes, epigenetic regulation and the role of CLU in tumorigenesis” of this volume). It is now accepted that the primary function of the 67–80 kDa sCLU protein form is cytoprotective and it is well known that resistance to cancer treatment is mediated, at least in part, by enhanced expression of cell survival proteins that facilitate tumor progression (Mallory et al., 2005; Miyake et al., 2000b). sCLU provides cytoprotection for IR-exposed MCF-7 breast cancer cells, which is not observed after siRNA, to silence the CLU gene (Criswell et al., 2003). The novel therapeutic strategy of silencing sCLU expression to overcome resistance to cancer therapy is of interest for the treatment of cancers that overexpress sCLU, as is the case in kidney (Zellweger et al., 2001a), bladder (Muramaki et al., 2009), prostate (July et al., 2002), breast (Redondo et al., 2007; So et al., 2005), and lung tumors (July et al., 2004) and in ovarian carcinomas (Park et al., 2008). To explore the potential of the CLU inhibition approach in breast cancer therapy, the cytotoxic interaction between antisense CLU oligonucleotide or anticlusterin antibody and the drugs commonly used in breast cancer treatment such as doxorubicin and paclitaxel have been analyzed in vitro using the breast carcinoma cell lines MCF-7 and MDA-MB-231 (Redondo et al., 2007). The effect of cytotoxic treatment on the level of CLU expression in breast tumor cell lines has also been evaluated, as CLU is known to be highly upregulated in various tissues undergoing apoptotic cell death (Kyprianou et al., 1991; Sensibar et al., 1991). As expected, CLU expression in MCF-7 cells was found to increase considerably after cytotoxic treatment, suggesting that CLU upregulation is likely to be an adaptative response that mediates chemoresistance. The capacity of anti-CLU treatment to sensitize breast carcinoma cells to chemotherapy was also investigated, and a search was made for drug combinations that produce additive cytotoxicity. Antisense CLU oligonucleotides or anti-CLU antibodies efficiently inhibited
Clusterin and Breast Cancer
33
CLU expression in MCF-7 and MDA-MB-231 cell lines, and this activity was associated with a decrease in cell viability. A sequence control oligonucleotide and nonimmune IgG fraction did not increase the effect of chemotherapy, which suggests that the sensitization of cells to apoptosis was due to the specific downregulation of CLU. These findings confirm the cytoprotective function of CLU in breast carcinoma cells, and suggest there is a role for anti-CLU therapy in the treatment of breast carcinomas which mainly express CLU protein. Other reports have confirmed the chemosensitivity to paclitaxel and doxorubicin in breast cancer cell lines (Mallory et al., 2005; So et al., 2005). In addition, in other tumors, anti-CLU therapy with antisense oligonucleotides (Chung et al., 2004; Gleave and Miyake, 2005; Miyake et al., 2001a; Zellweger et al., 2001b) or siRNA (Trougakos et al., 2004) has been evaluated as a potential therapeutic agent. Furthermore, cervical cancer cells expressing high levels of CLU are reported to become more sensitive to paclitaxel after CLU silencing through siRNA and antisense strategies (Park et al., 2008). In advanced breast cancer, AS-ODN targeting CLU enhances the growthinhibitory effect of Trastuzumab, an HER-2-targeted monoclonal antibody used in the clinical management of advanced breast cancer patients (Biroccio et al., 2005). Only the combination of AS-ODN and Trastuzumab leads to an activation of apoptosis, which was not observed with either agent alone. It is a noteworthy finding that raises even more expectations concerning the usefulness of the AS-ODN technology. Because the suppression of CLU expression renders human cancer cells sensitive to chemotherapeutic drug-mediated apoptosis, it is currently an antisense target in clinical trials (Chi et al., 2005, 2008; Chia et al., 2009). Preclinical trials have also shown that inhibition of CLU levels using antisense oligonucleotides increases apoptosis after conventional chemotherapeutic treatments (Gleave and Miyake, 2005; Miyake et al., 2005a). However, the task to reduce sCLU levels without affecting nCLU levels seems not easy. At the moment, production of a siRNA targeting a specific CLU form has not been demonstrated (see chapter “Clusterin (CLU): From one gene and two transcripts to many proteins” of Vol. 104). Theoretically, specific siRNA knockdown of sCLU levels would enhance cytotoxic agent efficacy, whereas siRNA against nCLU would confers cytoprotection against a variety of chemotherapeutic agents. OGX-011 (Oncogenex Technologies Inc., Vancouver, BC, Canada) is a second generation phosphorothioate antisense oligonucleotide that is complementary to the CLU mRNA translation initiation site and strongly inhibits CLU expression in both in vitro and in vivo laboratory models (Zellweger et al., 2001b). Thus, in preclinical efficacy studies, the antisense oligonucleotide OGX-011 has been shown to significantly enhance the therapeutic effect of hormone, chemo- and radiation therapy in a variety
34
Maximino Redondo et al.
of tumor models (Gleave and Jansen, 2003). In phase I clinical trials, OGX-011 is reported to be a potent suppressor of CLU expression in prostate cancer tissues, in combination with androgen deprivation therapy (Chi et al., 2005). This trial had a unique design in those patients with localized prostate cancer. OGX-011 were administered prior to radical prostatectomy, and drug tissue level and serum CLU expression were determined for each patient. Concentrations of OGX-011 associated with preclinical effect were measured in tumor tissue and 90% suppression of CLU was achieved. OGX-011 presents favorable pharmacokinetic properties and clinically demonstrable biological activity when combined with standard doses of docetaxel in patients with prostate, breast, and lung cancers (Chi et al., 2008). Thus, in a phase II trial, the combination of OGX-11 and docexatel was well tolerated and clinical activity was observed in patients with metastatic breast cancer, but there were insufficient responses to meet the criteria for proceeding to the second stage of accrual (Chia et al., 2009).
B. CLU and Resistance to Antiestrogen Antiestrogen represents the first line of therapy in the treatment of estrogen receptor-positive (ERþ) breast cancer patients, and its introduction has significantly contributed to the reduction of mortality observed in recent decades. Unfortunately, up to 40% of patients develop resistance associated with progression and frequently die as a result of metastatic breast cancer. The molecular events leading to pharmacological resistance are not completely understood. Although different mechanisms have been proposed to explain the phenomena of resistance, its relatively frequent occurrence represents a challenge for clinicians and is still incompletely understood. Different proteins, which are upregulated after cytotoxic treatment, have been considered as potential targets to overcome the problem of resistance in breast cancer; these include members of the bcl-2 family, survivin, HSPs, Akt, and CLU. A link between CLU overexpression and resistance to androgenic withdrawal (Gleave and Miyake, 2005), radiotherapy and chemotherapy has been reported in cases of prostate cancer (Zellweger et al., 2003), but also in breast cancer cells treated with trastuzumab or paclitaxel (Biroccio et al., 2005; Redondo et al., 2007; So et al., 2005). Interestingly, CLU overexpression has been associated with hallmarks of aggressiveness such as negative estrogen and progesterone receptor status in breast cancer (Redondo et al., 2000). In fact, the estrogen-independent cell line MDA-MB231 presents high levels of CLU expression (Redondo et al., 2007). In one report (Toffanin et al., 2008), basal CLU levels are higher in antiestrogenresistant cell line T47 than in the ERþ sensitive cell line MCF-7, but CLU is upregulated following antiestrogen treatment independently of the
Clusterin and Breast Cancer
35
sensitivity of the cell line. In the fairly sensitive cell line MCF-7, upregulation of CLU could be interpreted as an adaptive survival mechanism adopted by the cell to counteract growth inhibition effects. Such an upregulation has already been observed in several studies (Chen et al., 1996; Redondo et al., 2007; Warri et al., 1993) and might represent a general defense mechanism of cancer cells toward cytostatic drugs. This hypothesis is supported by other studies demonstrating that CLU is upregulated during conditions of cell stress, similar to what has been observed in breast cancer cells following trastuzumab treatment (Biroccio et al., 2005) and in prostate cancer cells after androgen ablation (July et al., 2002). Moreover, it has been demonstrated that overexpression of CLU by stable transfection protects human cancer cells from apoptosis induced by different triggers, such as radiotherapy (Zellweger et al., 2002) and tumor necrosis factor alpha (Sensibar et al., 1995). This adaptive overexpression could increase cell survival, thereby promoting progression and the acquisition of genomic instability. Furthermore, in studies by Toffanin et al. (2008) and Cappelletti et al. (2008), the downregulation of cytoplasmic CLU restores sensitivity to toremifene in the antiestrogen-resistant cell line T47D. Such results do not demonstrate a direct involvement of CLU in antiestrogen-resistant mechanisms, but strongly support the concept that CLU may modulate response to antiestrogens and therefore that it represents an interesting pharmaceutical target. CLU levels increase after treatment with antiestrogens irrespective of the constitutive sensitivity profile of the cell line (Toffanin et al., 2008). In another work, CLU treatment has been shown to increase the efficacy of tamoxifen in the MCF-7 cell line (Redondo et al., 2007). Combinations of CLU treatment with tamoxifen performed better than the respective singleagent treatment alone. The observation that by downregulating CLU, an estrogen-receptor-positive cell line (T47D) resistant to selective receptor modulators could be made sensitive to toremifene (Toffanin et al., 2008) represents an interesting contribution that should be prospectively confirmed by determining CLU expression in addition to estrogen receptors before starting treatment with antiestrogens. Such results, if confirmed, could provide a rationale for modulating sensitivity to antiestrogens by combining traditional therapy with strategies targeting CLU in vivo.
C. CLU and Dexamethasone Glucocorticoids (such as dexamethasone) are routinely used in the clinical application of chemotherapy to prevent adverse effects. In breast cancer cell lines, Wu et al. (2004) reported the inhibitory action of glucocorticoids on chemotherapy-induced apoptosis, which also raises a clinically relevant
36
Maximino Redondo et al.
question as to whether pretreatment with glucocorticoids might interfere with the therapeutic efficacy of chemotherapy. Glucocorticoids play a major role in attenuating the inflammatory response. These steroid hormones are able to induce apoptosis in cells of the hematopoietic system such as the monocytes, macrophages, and T lymphocytes that are involved in the inflammation reaction. In contrast, it has recently been discovered that in glandular cells such as the mammary gland epithelia, hepatocytes and ovarian follicular cells, and in fibroblasts, glucocorticoids protect against the apoptotic signals evoked by cytokines, cAMP, tumor suppressors, and death genes. It is well known that the antiapoptotic effect of glucocorticoids is exerted by the modulation of survival genes such as Bcl-2, Bcl-x(L), and NF-B, in a cell type-specific manner (Amsterdam and Sasson, 2002). In breast cancer, CLU is induced after treatment with dexamethasone, and therefore, it may be one of these genes responsible for the antiapoptotic effect of glucocorticoids (Redondo et al., 2007). A similar response by CLU after treatment with glucocorticoids was obtained on hemangioma in vitro (Hasan et al., 2003). These results, together with the facts that dexamethasone inhibits chemotherapy-induced cytotoxicity and that antisense oligonucleotides to CLU increase chemotherapy-induced cytotoxicity despite the previous administration of dexamethasone (Redondo et al., 2007) clearly show that dexamethasone modulates the expression of CLU. This finding is in agreement with the known anti-inflammatory properties of dexamethasone in upregulating the expression of other complement inhibitors (Imai et al., 2004).
VII. CONCLUSIONS One of the primary goals in breast cancer research is to identify prognostic markers of cancer recurrence and to develop new ways of improving the effectiveness of existing cancer treatment regimens. Cytoplasmic CLU has a role in breast tumorigenesis and progression, although its clinical utility as a survival prognostic factor remains to be clarified. The presence of many proteins forms with diverse and opposite effects constitutes a further obstacle in understanding the biological function of this protein. The CLU protein influences apoptotic activity and therefore any alterations it undergoes may affect resistance to chemo- and radiotherapy. Several studies have firmly established a role for sCLU as a cell survival factor with enhanced presence following tamoxifen therapy and chemotherapy, and which inhibits tumor cell death in breast carcinomas. The inhibition of CLU, using antisense oligonucleotides and antibodies, enhances the cytotoxic effects of chemotherapy agents, including paclitaxel and doxorubicin. However, in a phase II trial, the combination of CLU antisense oligonucleotides and docexatel was
Clusterin and Breast Cancer
37
well tolerated and clinical activity was seen in patients with metastatic breast cancer, but there were an insufficient number of responses to meet the criteria for proceeding to the second stage of accrual. In addition, glucocorticoids may influence breast cancer behavior via the upregulation of CLU, which might play a major role in the effects of dexamethasone, protecting breast cancer cells from the effects of chemotherapy. Blocking CLU reverses the drug’s unwanted effects of dexamethasone regarding cancer cell survival. More research is required to increase our understanding of the extent to which and the mechanisms by which CLU is involved in cancer development, providing the basis for earlier and more accurate cancer diagnosis, prognosis, and therapeutic intervention.
ACKNOWLEDGMENTS We thank Dr. Rafael Funez for providing Fig. 1 and Avelina Bautista for her positive and useful suggestions. This work was partially supported by Fondo de Investigaciones Sanitarias (FIS 06/1062, Spain).
REFERENCES Aigelsreiter, A., Janig, E., Sostaric, J., Pichler, M., Unterthor, D., Halasz, J., Lackner, C., Zatloukal, K., and Denk, H. (2009). Clusterin expression in cholestasis, hepatocellular carcinoma and liver fibrosis. Histopathology 54, 561–570. Albert, J. M., Gonzalez, A., Massion, P. P., Chen, H., Olson, S. J., Shyr, Y., Diaz, R., Lambright, E. S., Sandler, A., Carbone, D. P., Putnam, J.B, Jr., Johnson, D. H., et al. (2007). Cytoplasmic clusterin expression is associated with longer survival in patients with resected non small cell lung cancer. Cancer Epidemiol. Biomarkers Prev. 16, 1845–1851. Amsterdam, A., and Sasson, R. (2002). The anti-inflammatory action of glucocorticoids is mediated by cell type specific regulation of apoptosis. Mol. Cell. Endocrinol. 189, 1–9. Andersen, C. L., Schepeler, T., Thorsen, K., Birkenkamp-Demtro¨der, K., Mansilla, F., Aaltonen, L. A., Laurberg, S., and Ørntoft, T. F. (2007). Clusterin expression in normal mucosa and colorectal cancer. Mol. Cell. Proteomics 6, 1039–1048. Appel, D., Pilarsky, C., Graichen, R., and Koch-Brandt, C. (1996). Sorting of gp80 (GPIII, clusterin), a marker protein for constitutive apical secretion in Madin–Darby canine kidney (MDCK) cells, into the regulated pathway in the pheochromocytoma cell line PC12. Eur. J. Cell Biol. 70, 142–149. Bettuzzi, S., Hiipakka, R. A., Gilna, P., and Liao, S. T. (1989). Identification of an androgenrepressed mRNA in rat ventral prostate as coding for sulphated glycoprotein 2 by cDNA cloning and sequence analysis. Biochem. J. 257, 293–296. Bettuzzi, S., Scorcioni, F., Astancolle, S., Davalli, P., Scaltriti, M., and Corti, A. (2002). Clusterin (SGP-2) transient overexpression decreases proliferation rate of SV40-immortalized human prostate epithelial cells by slowing down cell cycle progression. Oncogene 21, 4328–4334.
38
Maximino Redondo et al.
Biroccio, A., D’Angelo, C., Cansen, B., Gleave, M. E., and Zupi, G. (2005). Antisense clusterin oligodeoxynucleotides increase the response of HER-2 gene amplified breast cancer cells to Trastuzumab. Cell. Physiol. 204, 463–469. Blaschuk, O., Burdzy, K., and Fritz, I. B. (1983). Purification and characterization of a cellaggregating factor (clusterin), the major glycoprotein in ram rete testis fluid. J. Biol. Chem. 258, 7714–7720. Busam, K. J., Kucukgol, D., Eastlake-Wade, S., Frosina, D., Delgado, R., and Jungbluth, A. A. (2006). Clusterin expression in primary and metastatic melanoma. J. Cutan. Pathol. 33, 619–623. Caccamo, A. E., Scaltriti, M., Caporali, A., D’Arca, D., Corti, A., Corvetta, D., Sala, A., and Bettuzzi, S. (2005). Ca2þ depletion induces nuclear clusterin, a novel effector of apoptosis in immortalized human prostate cells. Cell Death Differ. 12, 101–104. Cappelletti, V., Gariboldi, M., De Cecco, L., Toffanin, S., Reid, J. F., Lusa, L., Bajetta, E., Celio, L., Greco, M., Fabbri, A., Pierotti, M. A., and Daidone, M. G. (2008). Patterns and changes in gene expression following neo-adjuvant anti-estrogen treatment in estrogen receptor-positive breast cancer. Endocr. Relat. Cancer 15, 439–449. Cervellera, M., Raschella, G., Santilli, G., Tanno, B., Ventura, A., Mancini, C., Sevignani, C., Calabretta, B., and Sala, A. (2000). Direct transactivation of the anti-apoptotic gene apolipoprotein J (clusterin) by B-MYB. J. Biol. Chem. 275, 21055–21060. Chayka, O., Corvetta, D., Dews, M., Caccamo, A. E., Piotrowska, I., Santilli, G., Gibson, S., Sebire, N. J., Himoudi, N., Hogarty, M. D., Anderson, J., Bettuzzi, S., et al. (2009). Clusterin, a haploinsufficient tumor suppressor gene in neuroblastomas. J. Natl. Cancer Inst. 101, 663–677. Chen, H., Tritton, T. R., Kenny, N., Absher, M., and Chiu, J. F. (1996). Tamoxifen induces TGFbeta 1 activity and apoptosis of human MCF-7 breast cancer cells in vitro. J. Cell. Biochem. 61, 9–17. Chen, X., Halberg, R. B., Ehrhardt, W. M., Torrealba, J., and Dove, W. F. (2003). Clusterin as a biomarker in murine and human intestinal neoplasia. Proc. Natl. Acad. Sci. USA 100, 9530–9535. Chen, T., Turner, J., McCarthy, S., Scaltriti, M., Bettuzzi, S., and Yeatman, T. J. (2004). Clusterin-mediated apoptosis is regulated by adenomatous polyposis coli and is p21 dependent but p53 independent. Cancer Res. 64, 7412–7419. Chi, K. N., Eisenhauer, E., Fazli, L., Jones, E. C., Goldenberg, S. L., Powers, J., Tu, D., and Gleave, M. E. (2005). A phase I pharmacokinetic and pharmacodynamic study of OGX-011, a 2’-methoxyethyl antisense oligonucleotide to clusterin, in patients with localized prostate cancer. J. Natl. Cancer Inst. 97, 1287–1296. Chi, K. N., Siu, L. L., Hirte, H., Hotte, S. J., Knox, J., Kollmansberger, C., Gleave, M., Guns, E., Powers, J., Walsh, W., Tu, D., and Eisenhauer, E. (2008). A phase I study of OGX-011, a 2’methoxyethyl phosphorothioate antisense to clusterin, in combination with docetaxel in patients with advanced cancer. Clin. Cancer Res. 14, 833–839. Chia, S., Dent, S., Ellard, S., Ellis, P. M., Vandenberg, T., Gelmon, K., Powers, J., Walsh, W., Seymour, L., and Eisenhauer, E. A. (2009). Phase II trial of OGX-011 in combination with docetaxel in metastatic breast cancer. Clin. Cancer Res. 15, 708–713. Chung, J., Kwak, C., Jin, R. J., Lee, C. H., Lee, K. H., and Lee, S. E. (2004). Enhanced chemosensitivity of bladder cancer cells to cisplatin by suppression of clusterin in vitro. Cancer Lett. 203, 155–161. Criswell, T., Klokov, D., Beman, M., Lavik, J. P., and Boothman, D. A. (2003). Repression of IR-inducible clusterin expression by the p53 tumor suppressor protein. Cancer Biol. Ther. 2, 372–380. Doustjalali, S. R., Yusof, R., Yip, C. H., Looi, L. M., Pillay, B., and Hashim, O. H. (2004). Aberrant expression of acute-phase reactant proteins in sera and breast lesions of patients with malignant and benign breast tumors. Electrophoresis 25, 2392–2401.
Clusterin and Breast Cancer
39
Early Breast Cancer Trialists’ Collaborative Group(2000). Favourable and unfavourable effects on long-term survival of radiotherapy for early breast cancer: An overview of the randomised trials. Lancet 355, 1757–1770. Fritz, I. B., Burdzy, C., Setchell, B., and Blaschuck, O. (1983). Ram rete testes fluid contains a protein (clusterin) which influence cell–cell interactions in vitro. Biol. Reprod. 28, 1173–1188. Gleave, M., and Jansen, B. (2003). Clusterin and IGFBPs as antisense targets in prostate cancer. Ann. N. Y. Acad. Sci. 1002, 95–104. Gleave, M., and Miyake, H. (2005). Use of antisense oligonucleotides targeting the cytoprotective gene, clusterin, to enhance androgen- and chemo-sensitivity in prostate cancer. World J. Urol. 23, 38–46. Gleave, M. E., Miyake, H., Zellweger, T., Chi, K., July, L., Nelson, C., and Rennie, P. (2001). Use of antisense oligonucleotides targeting the antiapoptotic gene, clusterin/testosteronerepressed prostate message 2, to enhance androgen sensitivity and chemosensitivity in prostate cancer. Urology 58, 39–49. Hasan, Q., Tan, S. T., Xu, B., and Davis, P. F. (2003). Effects of five commonly used glucocorticoids on haemangioma in vitro. Clin. Exp. Pharmacol. Physiol. 30, 140–144. Humphreys, D. T., Carver, J. A., Easterbroock Smith, S. B., and Wilson, M. R. (1999). Clusterin has chaperone like activity similar to that of small heat shock protein. J. Biol. Chem. 274, 6875–6881. Imai, M., Hwang, H. Y., Norris, J. S., and Tomlinson, S. (2004). The effect of dexamethasone on human mucin 1 expression and antibody-dependent complement sensitivity in a prostate cancer cell line in vitro and in vivo. Immunology 111, 291–297. Jemal, A., Murray, T., Ward, E., Samuels, A., Tiwari, R. C., Ghafoor, A., Feuer, E. J., and Thun, M. J. (2005). Cancer statistics, 2005. CA Cancer J. Clin. 55, 10–30. Jenne, D. E., and Tschopp, J. (1992). Clusterin: The intriguing guises of a widely expressed glycoprotein. Trends Biochem. Sci. 17, 154–159. Jordan Stark, T. C., Witte, D. P., Aronow, B., and Harmony, J. A. K. (1992). Apolipoprotein J: A membrane policeman? Curr. Opin. Lipidol. 3, 75–85. July, L. V., Akbari, M., Zellweger, T., Jones, E. C., Goldenberg, S. L., and Gleave, M. E. (2002). Clusterin expression is significantly enhanced in prostate cancer cells following androgen withdrawal therapy. Prostate 50, 179–188. July, L. V., Beraldi, E., So, A., Fazli, L., Evans, K., English, J. C., and Gleave, M. E. (2004). Nucleotide-based therapies targeting clusterin chemosensitize human lung adenocarcinoma cells both in vitro and in vivo. Mol. Cancer Ther. 3, 223–232. Kang, Y. K., Hong, S. W., Lee, H., and Kim, W. H. (2004). Overexpression of clusterin in human hepatocellular carcinoma. Hum. Pathol. 35, 1340–1346. Kevans, D., Foley, J., Tenniswood, M., Sheahan, K., Hyland, J., O’Donoghue, D., Mulcahy, H., and O’Sullivan, J. (2009). High clusterin expression correlates with a poor outcome in stage II colorectal cancers. Cancer Epidemiol. Biomarkers Prev. 18, 393–399. Kissinger, C., Skinner, M. K., and Griswold, M. D. (1982). Analysis of Sertoli cell-secreted proteins by two-dimensional gel electrophoresis. Biol. Reprod. 27, 233–240. Kru¨ger, S., Mahnken, A., Kausch, I., and Feller, A. C. (2006). Value of clusterin immunoreactivity as a predictive factor in muscle-invasive urothelial bladder carcinoma. Urology 67, 105–109. Kru¨ger, S., Ola, V., Fischer, D., Feller, A. C., and Friedrich, M. (2007). Prognostic significance of clusterin immunoreactivity in breast cancer. Neoplasma 54, 46–50. Kurahashi, T., Muramaki, M., Yamanaka, K., Hara, I., and Miyake, H. (2005). Expression of the secreted form of clusterin protein in renal cell carcinoma as a predictor of disease extension. BJU Int. 96, 895–899.
40
Maximino Redondo et al.
Kyprianou, N., English, H. F., Davidson, N. E., and Isaacs, J. T. (1991). Programmed cell death during regression of the MCF 7 human breast cancer following estrogen ablation. Cancer Res. 51, 162–166. Lau, S. H., Sham, J. S., Xie, D., Tzang, C. H., Tang, D., Ma, N., Hu, L., Wang, Y., Wen, J. M., Xiao, G., Zhang, W. M., Lau, G. K., et al. (2006). Clusterin plays an important role in hepatocellular carcinoma metastasis. Oncogene 25, 1242–1250. Leskov, K. S., Klokov, D. Y., Li, J., Kinsella, T. J., and Boothman, D. A. (2003). Synthesis and functional analyses of nuclear clusterin, a cell death protein. J. Biol. Chem. 278, 11590–11600. Lund, P., Weisshaupt, K., Mikeska, T., Jammas, D., Chen, X., Kuban, R. J., Ungethu¨m, U., Krapfenbauer, U., Herzel, H. P., Scha¨fer, R., Walter, J., and Sers, C. (2006). Oncogenic HRAS suppresses clusterin expression through promoter hypermethylation. Oncogene 25, 4890–4903. Mallory, J. C., Crudden, G., Oliva, A., Saunders, C., Stromberg, A., and Craven, R. J. (2005). A novel group of genes regulates susceptibility to antineoplastic drugs in highly tumorigenic breast cancer cells. Mol. Pharmacol. 68, 1747–1756. Miyake, H., Chi, K. N., and Gleave, M. E. (2000a). Antisense TRPM-2 oligodeoxynucleotides chemosensitize human androgen-independent PC-3 prostate cancer cells both in vitro and in vivo. Clin. Cancer Res. 6, 1655–1663. Miyake, H., Nelson, C., Rennie, P. S., and Gleave, M. E. (2000b). Acquisition of chemoresistant phenotype by overexpression of the antiapoptotic gene testosterone-repressed prostate message-2 in prostate cancer xenograft models. Cancer Res. 60, 2547–2554. Miyake, H., Hara, I., Kamidono, S., and Gleave, M. E. (2001a). Synergistic chemosensitization and inhibition of tumor growth and metastasis by the antisense oligodeoxynucleotide targeting clusterin gene in a human bladder cancer model. Clin. Cancer Res. 7, 4245–4252. Miyake, H., Hara, S., Zellweger, T., Kamidono, S., Gleave, M. E., and Hara, I. (2001b). Acquisition of resistance to Fas-mediated apoptosis by overexpression of clusterin in human renal-cell carcinoma cells. Mol. Urol. 5, 105–111. Miyake, H., Hara, S., Arakawa, S., Kamidono, S., and Hara, I. (2002). Over expression of clusterin is an independent prognostic factor for nonpapillary renal cell carcinoma. J. Urol. 167, 703–706. Miyake, H., Hara, I., Gleave, M. E., and Eto, H. (2004). Protection of androgen-dependent human prostate cancer cells from oxidative stress-induced DNA damage by overexpression of clusterin and its modulation by androgen. Prostate 61, 318–323. Miyake, H., Yamanaka, K., Muramaki, M., Hara, I., and Gleave, M. E. (2005a). Therapeutic efficacy of adenoviral-mediated p53 gene transfer is synergistically enhanced by combined use of antisense oligodeoxynucleotide targeting clusterin gene in a human bladder cancer model. Neoplasia 7, 171–179. Miyake, H., Yamanaka, K., Muramaki, M., Kurahashi, T., Gleave, M., and Hara, I. (2005b). Enhanced expression of the secreted form of clusterin following neoadjuvant hormonal therapy as a prognostic predictor in patients undergoing radical prostatectomy for prostate cancer. Oncol. Rep. 14, 1371–1375. Moretti, R. M., Marelli, M. M., Mai, S., Cariboni, A., Scaltriti, M., Bettuzzi, S., and Limonta, P. (2007). Clusterin isoforms differentially affect growth and motility of prostate cells: Possible implications in prostate tumorigenesis. Cancer Res. 67, 10325–10333. Muramaki, M., So, A., Hayashi, N., Sowery, R., Miyake, H., Fujisawa, M., and Gleave, M. E. (2009). Chemosensitization of gemcitabine-resistant human bladder cancer cell line both in vitro and in vivo using antisense oligonucleotide targeting the anti-apoptotic gene, clusterin. BJU Int. 103, 384–390. Parczyk, K., Pilarsky, C., Rachel, U., and Koch-Brandt, C. (1994). Gp80 (Clusterin; TRPM2) mRNA level is enhanced in human renal clear cell carcinomas. J. Cancer Res. Clin. Oncol. 120, 186–188.
Clusterin and Breast Cancer
41
Park, D. C., Yeo, S. G., Wilson, M. R., Yerbury, J. J., Kwong, J., Welch, W. R., Choi, Y. K., Birrer, M. J., Mok, S. C., and Wong, K. K. (2008). Clusterin interacts with Paclitaxel and confers Paclitaxel resistance in ovarian cancer. Neoplasia 10, 964–972. Pasinetti, G. M., Johnson, S. A., Oda, T., Rozovsky, I., and Finch, C. E. (1994). Clusterin (SGP-2): A multifunctional glycoprotein with regional expression in astrocytes and neurons of the adult rat brain. J. Comp. Neurol 339, 387–400. Pins, M. R., Fiadjoe, J. E., Korley, F., Wong, M., Rademaker, A. W., Jovanovic, B., Yoo, T. K., Kozlowski, J. M., Raji, A., Yang, X. J., and Lee, C. (2004). Clusterin as a possible predictor for biochemical recurrence of prostate cancer following radical prostatectomy with intermediate Gleason scores: A preliminary report. Prostate Cancer Prostatic Dis. 7, 243–248. Pucci, S., Bonanno, E., Pichiorri, F., Angeloni, C., and Spagnoli, L. G. (2004). Modulation of different clusterin isoforms in human colon tumorigenesis. Oncogene 23, 2298–2304. Rauhala, H. E., Porkka, K. P., Sarama¨ki, O. R., Tammela, T. L., and Visakorpi, T. (2008). Clusterin is epigenetically regulated in prostate cancer. Int. J. Cancer 123, 1601–1609. Ravdin, P. M., Cronin, K. A., Howlader, N., Berg, C. D., Chlebowski, R. T., Feuer, E. J., Edwards, B. K., and Berry, D. A. (2007). The decrease in breast-cancer incidence in 2003 in the United States. N. Engl. J. Med. 356, 1670–1674. Reddy, K. B., Jin, G., Karode, M. C., Harmony, J. A., and Howe, P. H. (1996). Transforming growth factor beta (TGF beta)-induced nuclear localization of apolipoprotein J/clusterin in epithelial cells. Biochemistry 35, 6157–6163. Redondo, M., Villar, E., Torres-Mun˜oz, J., Tellez, T., Morell, M., and Petito, C. K. (2000). Overexpression of clusterin in human breast carcinoma. Am. J. Pathol. 157, 393–399. Redondo, M., Garcı´a, J., Villar, E., Rodrigo, I., Perea-Milla, E., Serrano, A., and Morell, M. (2003). Major histocompatibility complex status in breast carcinogenesis and relationship to apoptosis. Hum. Pathol. 34, 1283–1289. Redondo, M., Esteban, F., Gonzalez-Moles, M. A., Delgado-Rodriguez, M., Nevado, M., Torres-Mun˜oz, J. E., Tellez, T., Villar, E., Morell, M., and Petito, C. K. (2006). Expression of the antiapoptotic proteins clusterin and bcl-2 in laryngeal squamous cell carcinomas. Tumour Biol. 27, 195–200. Redondo, M., Te´llez, T., Roldan, M. J., Serrano, A., Garcı´a-Aranda, M., Gleave, M. E., Hortas, M. L., and Morell, M. (2007). Anticlusterin treatment of breast cancer cells increases the sensitivities of chemotherapy and tamoxifen and counteracts the inhibitory action of dexamethasone on chemotherapy-induced cytotoxicity. Breast Cancer Res. 9, R86. Redondo, M., Funez, R., and Esteban, F. (2009). Apoptosis in the development and treatment of laryngel cancer: Role of p53, bcl-2 and clusterin. In “Apoptosis in Carcinogenesis and Chemotherapy,” (G. Chen and P. B. S. Lai, Eds.), Vol. 10, pp. 237–249. Springer, Netherlands. Santilli, G., Aronow, B. J., and Sala, A. (2003). Essential requirement of apolipoprotein J (clusterin) signaling for IkappaB expression and regulation of NF-kappaB activity. J. Biol. Chem. 278, 38214–38219. Scaltriti, M., Bettuzzi, S., Sharrard, R. M., Caporali, A., Caccamo, A. E., and Maitland, N. J. (2004). Clusterin overexpression in both malignant and nonmalignant prostate epithelial cells induces cell cycle arrest and apoptosis. Br. J. Cancer 91, 1842–1850. Sensibar, J. A., Griswold, M. D., Sylvester, S. R., Buttyan, R., Bardin, C. W., Cheng, C. Y., Dudek, S., and Lee, C. (1991). Prostatic ductal system in rats: regional variation in localization of an androgen-repressed gene product, sulfated glycoprotein-2. Endocrinology 128, 2091–2102. Sensibar, J. A., Sutkowski, D. M., Raffo, A., Buttyan, R., Griswold, M. D., Sylvester, S. R., Kozlowskim, J. M., and Lee, C. (1995). Prevention of cell death induced by tumor necrosis factor alpha in LNCaP cells by overexpression of sulfated glycoprotein-2 (clusterin). Cancer Res. 55, 2431–2437.
42
Maximino Redondo et al.
So, A., Sinnemann, S., Huntsmann, D., Fazli, L., and Gleave, M. (2005). Knockdown of the cytoprotective chaperone, clusterin, chemosensitizes human breast cancer cells both in vitro and in vivo. Mol. Cancer Ther. 4, 1837–1849. Steinberg, J., Oyasu, R., Lang, S., Sintich, S., Rademaker, A., Lee, C., Kozlowski, J. M., and Sensibar, J. A. (1997). Intracellular levels of SGP-2 (clusterin) correlate with tumor grade in prostate cancer. Clin. Cancer Res. 3, 1707–1711. Sun, B., Zhang, S., Zhang, D., Liu, Y., Li, Y., Rong, Z., Zhu, Y., and Jia, X. (2007). Clusterin is associated with spontaneous breast cancer in TA2 mice. FEBS Lett. 581, 3277–3282. Thomas-Tikhonenko, A., Viard-Leveugle, I., Dews, M., Wehrli, P., Sevignani, C., Yu, D., Ricci, S., El-Deiry, W., Aronow, B., Kaya, G., Saurat, J. H., and French, L. E. (2004). Myctransformed epithelial cells down-regulate clusterin, which inhibits their growth in vitro and carcinogenesis in vivo. Cancer Res. 64, 3126–3136. Toffanin, S., Daidone, M. G., Miodini, P., De Cecco, L., Gandellini, P., and Cappelletti, V. (2008). Clusterin: A potential target for improving response to antiestrogens. Int. J. Oncol. 33, 791–798. Trougakos, I. P., and Gonos, E. S. (2002). Clusterin/Apolipoprotein J in human aging and cancer. Int. J. Biochem. Cell Biol. 34, 1430–1448. Trougakos, I. P., So, A., Jansen, B., Gleave, M. E., and Gonos, E. S. (2004). Silencing expression of the clusterin/apolipoprotein J gene in human cancer cells using small interfering RNA induces spontaneous apoptosis, reduced growth ability, and cell sensitization to genotoxic and oxidative stress. Cancer Res. 64, 1834–1842. Trougakos, I. P., Lourda, M., Agiostratidou, G., Kletsas, D., and Gonos, E. S. (2005). Differential effects of clusterin/apolipoprotein J on cellular growth and survival. Free Radic. Biol. Med. 38, 436–449. Trougakos, I. P., Djeu, J. Y., Gonos, E. S., and Boothman, D. A. (2009a). Advances and challenges in basic and translational research on clusterin. Cancer Res. 69, 403–406. Trougakos, I. P., Lourda, M., Antonelou, M. H., Kletsas, D., Gorgoulis, V. G., Papassideri, I. S., Zou, Y., Margaritis, L. H., Boothman, D. A., and Gonos, E. S. (2009b). Intracellular clusterin inhibits mitochondrial apoptosis by suppressing p53-activating stress signals and stabilizing the cytosolic Ku70–Bax protein complex. Clin. Cancer Res. 15, 48–59. Warri, A. M., Huovinen, R. L., Laine, A. M., Martikainen, P. M., and Harkonen, P. L. (1993). Apoptosis in toremifene-induced growth inhibition of human breast cancer cells in vivo and in vitro. J. Natl. Cancer Inst. 85, 1412–1418. Watari, H., Ohta, Y., Hassan, M. K., Xiong, Y., Tanaka, S., and Sakuragi, N. (2008). Clusterin expression predicts survival of invasive cervical cancer patients treated with radical hysterectomy and systematic lymphadenectomy. Gynecol. Oncol. 108, 527–532. Wu, W., Chaudhuri, S., Brickley, D. R., Pang, D., Karrison, T., and Conzen, S. D. (2004). Microarray analysis reveals glucocorticoid-regulated survival genes that are associated with inhibition of apoptosis in breast epithelial cells. Cancer Res. 64, 1757–1764. Xie, M. J., Motoo, Y., Su, S. B., Mouri, H., Ohtsubo, K., Matsubara, F., and Sawabu, N. (2002). Expression of clusterin in human pancreatic cancer. Pancreas 25, 234–238. Xie, D., Lau, S. H., Sham, J. S., Wu, Q. L., Fang, Y., Liang, L. Z., Che, L. H., Zeng, Y. X., and Guan, X. Y. (2005a). Up-regulated expression of cytoplasmic clusterin in human ovarian carcinoma. Cancer 103, 277–283. Xie, D., Sham, J. S., Zeng, W. F., Che, L. H., Zhang, M., Wu, H. X., Lin, H. L., Wen, J. M., Lau, S. H., Hu, L., and Guan, X. Y. (2005b). Oncogenic role of clusterin overexpression in multistage colorectal tumorigenesis and progression. World J. Gastroenterol. 11, 3285–3289. Yamanaka, K., Gleave, M. E., Hara, I., Muramaki, M., and Miyake, H. (2005). Synergistic antitumor effect of combined use of adenoviral-mediated p53 gene transfer and antisense oligodeoxynucleotide targeting clusterin gene in an androgen-independent human prostate cancer model. Mol. Cancer Ther. 4, 187–195.
Clusterin and Breast Cancer
43
Yang, C. R., Yeh, S., Leskov, K., Odegaard, E., Hsu, H. L., Chang, C., Kinsella, T. J., Chen, D. J., and Boothman, D. A. (1999). Isolation of Ku70-binding proteins (KUBs). Nucleic Acids Res. 27, 2165–2174. Yang, C. R., Leskov, K., Hosley-Eberlein, K., Criswell, T., Pink, J. J., Kinsella, T. J., and Boothman, D. A. (2000). Nuclear clusterin/XIP8, an X-ray-induced Ku70-binding protein that signals cell death. Proc. Natl. Acad. Sci. USA 97, 5907–5912. Zellweger, T., Miyake, H., July, L. V., Akbari, M., Kiyama, S., and Gleave, M. E. (2001a). Chemosensitization of human renal cell cancer using antisense oligonucleotides targeting the antiapoptotic gene clusterin. Neoplasia 3, 360–367. Zellweger, T., Miyake, H., Cooper, S., Chi, K., Conklin, B. S., Monia, B. P., and Gleave, M. E. (2001b). Antitumor activity of antisense clusterin oligonucleotides is improved in vitro and in vivo by incorporation of 2’-O-(2-methoxy)ethyl chemistry. J. Pharmacol. Exp. Ther. 298, 934–940. Zellweger, T., Chi, K., Miyake, H., Adomat, H., Kiyama, S., Skov, K., and Gleave, M. E. (2002). Enhanced radiation sensitivity in prostate cancer by inhibition of the cell survival protein clusterin. Clin. Cancer Res. 8, 3276–3284. Zellweger, T., Kiyama, S., Chi, K., Miyake, H., Adomat, H., Skov, K., and Gleave, M. E. (2003). Overexpression of the cytoprotective protein clusterin decreases radiosensitivity in the human LNCaP prostate tumour model. BJU Int. 92, 463–469. Zhang, L. Y., Ying, W. T., Mao, Y. S., He, H. Z., Liu, Y., Wang, H. X., Liu, F., Wang, K., Zhang, D. C., Wang, Y., Wu, M., Qian, X. H., et al. (2003). Loss of clusterin both in serum and tissue correlates with the tumorigenesis of esophageal squamous cell carcinoma via proteomics approaches. World J. Gastroenterol. 9, 650–654. Zhang, H., Kim, J. K., Edwards, C. A., Xu, Z., Taichman, R., and Wang, C. Y. (2005). Clusterin inhibits apoptosis by interacting with activated Bax. Nat. Cell Biol. 7, 909–915. Zhang, S., Zhang, D., Zhu, Y., Guo, H., Zhao, X., and Sun, B. (2006). Clusterin expression and univariate analysis of overall survival in human breast cancer. Technol. Cancer Res. Treat. 5, 573–578.
CLU and Colon Cancer. The Dual Face of CLU: From Normal to Malignant Phenotype P. Mazzarelli,* Sabina Pucci,* and L. G. Spagnoli*,{ *Department of Biopathology, Institute of Anatomic Pathology, University of Rome “Tor Vergata”, 00133 Rome, Italy { IRCCS San Raffaele Pisana, Rome, Italy
I. Introduction: Genes and Proteins in Colorectal Cancer II. Genetic Instability and Control of DNA Damage: DNA Double-Strand Breaks Repair III. Clonal Expansion: Apoptosis Inhibition A. Clusterin (CLU), a Multifunctional Protein Influenced by the Cellular Context IV. CLU in Colorectal Cancer Progression: sCLU and Apoptosis Escape A. CLU–Ku–Bax Localization in Colon Cancer V. CLU as a New Biomarker for Colon Cancer Screening VI. Conclusions and Future Perspectives References The transition from normal to malignant phenotype implies the activation of some pathways that underlie the aberrant clone expansion. In some way, the conventional function of proteins involved in DNA repair, cell death/growth induction, vascularization, and metabolism is inhibited or shifted toward other pathways by soluble mediators that orchestrate such change depending on the microenvironment conditions. The adenoma–carcinoma sequence of the colon represents one of the most well studied and characterized models of human tumor progression. In this section, we focus our attention on defined pathways that underlie the initiation, promotion, and progression of colon cancer, conferring aggressiveness to the neoplastic cells. Clusterin (CLU) is a pleiotropic protein with a broad range of functions. It has recently drawn much attention because of its association with cancer promotion and metastasis. It is involved in prosurvival and apoptosis processes that are carried out by two different forms. sCLU is cytoprotective and its prosurvival function is the basis of the current Phase I/II clinical trials. In colorectal cancer an increase of sCLU expression occurs, whereas the nuclear proapoptotic form is downregulated. Several controversial data have been published on colon cancer discussing its role as tumor suppressor or prosurvival factor in colon cancer. Here, we report the dynamic interaction of the different forms of CLU with their partners DNA-repair protein Ku70 and proapoptotic factor Bax during colon cancer progression, which seems to be a crucial point for the neoplastic cell fate. We also highlight that the appearance and the progressive increase of the sCLU in colorectal tumors correlate to a significant increase of CLU in serum and stool of patients. On the basis of results obtained by CLU immuno-dosage in blood and stool of colon cancer patients, we report that sCLU could represent a diagnostic molecular marker for colon cancer screening. # 2009 Elsevier Inc.
Advances in CANCER RESEARCH Copyright 2009, Elsevier Inc. All rights reserved.
45
0065-230X/09 $35.00 DOI: 10.1016/S0065-230X(09)05003-9
46
P. Mazzarelli et al.
I. INTRODUCTION: GENES AND PROTEINS IN COLORECTAL CANCER Colorectal cancer is a significant cause of morbidity and mortality in Western populations. This cancer develops as a result of the pathologic transformation of normal colonic epithelium to an adenomatous polyp and ultimately an invasive cancer. The multistep progression requires years and possibly decades and is accompanied by a number of well-characterized genetic alterations. Chronic inflammation, as in inflammatory bowel disease, may predispose patients to malignancy. Mutations in two classes of genes, tumor-suppressor genes and protooncogenes, impart a proliferative advantage to cells and contribute to development of the malignant phenotype (Gryfe et al., 1997). Inactivating mutations of both copies (alleles) of the adenomatous polyposis coli (APC) gene, a tumor-suppressor gene on chromosome 5q, mark one of the earliest events in colorectal carcinogenesis. Germline mutation of the APC gene and subsequent somatic mutation of the second APC allele cause the inherited familial adenomatous polyposis syndrome (FAP). This syndrome is characterized by the presence of hundreds to thousands of colonic adenomatous polyps. If these polyps are left untreated, colorectal cancer develops. Mutation leading to dysregulation of the K-ras protooncogene is also an early event in colon cancer formation. Conversely, loss of heterozygosity on the long arm of chromosome 18 (18q) occurs later in the sequence of development from adenoma to carcinoma, and this mutation may predict poor prognosis. Loss of the 18q region is thought to contribute to inactivation of the DCC tumor-suppressor gene. More recent evidence suggests that other tumor-suppressor genes, DPC4 and MADR2 of the transforming growth factor beta (TGF-) pathway, also may be inactivated by allelic loss on chromosome 18q. In addition, mutation of the tumor-suppressor gene p53 on chromosome 17p appears to be a late phenomenon in colorectal carcinogenesis. This mutation may allow the growing tumor with multiple genetic alterations to evade cell-cycle arrest and apoptosis. Neoplastic progression is probably accompanied by additional genetic events, which are indicated by allelic loss on chromosomes 1q, 4p, 6p, 8p, 9q, and 22q in 25–50% of colorectal cancers. Moreover a third class of genes, DNA-repair genes, has been implicated in tumorigenesis of colorectal cancer. Study findings suggest that DNA mismatch repair deficiency, due to germline mutation of the hMSH2, hMLH1, hPMS1, or hPMS2 genes, contributes to development of hereditary nonpolyposis colorectal cancer (HNPCC). The majority of tumors in patients with this disease, and 10–15% of sporadic colon cancers display microsatellite instability (MSI), also know as the replication error positive (RERþ) phenotype. These tumors
CLU Dual Face in Colon Cancer
47
are characterized by genetic instability at microsatellite loci. Although colorectal cancer cells are characterized by specific microsatellite alterations, the same four different signaling pathways, WNT/Wingless pathway, K-ras pathway, TGF- pathway and p53 pathway, could be implicated in tumor progression. These alterations contribute to the adenoma–carcinoma transition. Moreover changes in DNA methylation pattern, in sense of hypermethylation, have been shown to inactivate genes associated with DNA-damage responses and DNA repair, MLH1, MLH3 MSH6, and SFN (Loukola et al., 2000; Taylor et al., 2006), contributing to colon cancer development. The epigenetic hypermethylation instability is strictly linked to genetic instability.
II. GENETIC INSTABILITY AND CONTROL OF DNA DAMAGE: DNA DOUBLE-STRAND BREAKS REPAIR Genetic instability causes genetic heterogeneity, that is a peculiar feature of tumors and fundamental in cancer progression. The majority of tumors, with no exception for colorectal cancers, show no obvious familiar inheritance suggesting that multiple low penetrance genes segregating in the human population confer cancer susceptibility and resistance to environmental carcinogens. These low penetrance genes play a key role in DNAdamage repair, in apoptosis induction, in immune response efficiency, and may act combinatorially in a dosage-dependent manner, to confer predisposition of cancer insurgence. In fact, environmental insult or mutations that alter checkpoint genes involved in DNA-damage repair and survival pathways, could select cells that proliferate more quickly than those stopped to repair damage. Moreover, increased DNA synthesis is associated with extensive genetic damage. High levels of DNA synthesis together with chromosomal and MSI in tumors strongly suggest that alteration in DNA-repair machinery and apoptosis may contribute to uncontrolled and error-prone DNA synthesis. As reported above, the efficiency of DNA repair is crucial to maintain the genome homeostasis, preventing malignant transformation and tumor insurgence (Difilippantonio et al., 2000). Double-strand breaks (DSBs) are the most hazardous lesions occurring in the genome of eukaryotic organisms. These lesions could take place during DNA replication, meiosis, and immune system development. Not only colorectal cancer but also breast, endometrial, and gastric carcinomas display increased risk of development in subjects with germline mutations at the DNA–DSBs repair system (BRCA1, BRCA2, ATM, etc.).
48
P. Mazzarelli et al.
The DSBs repair requires the homologous recombination (HR) and nonhomologous end joining (NHEJ). The NHEJ DSBs repair involves the activity of Ku70/80 protein heterodimer, sensor of the damage (Gottlieb and Jackson, 1993). In fact, the first character of the NHEJ is the DNAdependent protein kinase (DNA-PK), a serine–threonine kinase consisting of a 470 kDa catalytic subunit (DNA-PKcs) and the regulatory protein, called Ku, which is composed of 70 and 86 kDa subunits. The heterodimer Ku, first described as a nuclear autoantigen, is a regulatory factor of DNA replication and transcription. The Ku heterodimer binds the ends of various types of DNA discontinuity, and is involved in the repair of DNA breaks caused by an incorrect DNA replication, V(D)J recombination, isotype switching, physiological oxidations, ionizing irradiation, and some chemotherapeutic drug effects (Blunt et al., 1995; Jackson and Jeggo, 1995). The interaction of Ku with ends of DNA has been extensively studied. Ku binds with high affinity to free ends of double-stranded DNA as well as to nicked DNA hairpins and dumbbell structures in vitro and in vivo in nuclear extracts. The principal role of Ku proteins is to take care of the homeostasis of the genome being involved in telomere maintenance, regulation of apoptosis induction, specific gene transcription, DNA replication, and cell-cycle regulation. The function of this caretaker gene is to suppress chromosomal aberrations, translocation, and aneuploidy. Ku was originally reported to be a nuclear protein, consistent with its functions as a subunit of DNA-PK. However, several studies have revealed the cytoplasmic or cell surface localization of Ku proteins in various cell types (Prabhakar et al., 1990). The subcellular localization of Ku70 and Ku86 changes during the cell-cycle progression (Koike et al., 1999), and nuclear translocation of Ku70 precedes that of Ku86 in late telophase/early G1 phase. Furthermore, changes in subcellular localization of Ku could be controlled by various external growth-regulating stimuli (Fewell and Kuff, 1996). Recently, it has been demonstrated a Ku DNA-binding activity in the cytoplasmic compartment of highly invasive bladder and breast tumors and metastatic nodes (Pucci et al., 2001), whereas the nuclear activity related to the DNA-repair system, was impaired. Experimental data further reported an inactivation of Ku DNA-binding activity, essential for genomic stability, in colon cancer progression models, in breast and in bladder carcinomas. A dysfunction of this protective activity let the aberrant cell clone growing. In highly infiltrative and metastatic tumors of the colon, breast and bladder, the impaired DNA-repair activity is due to the loss of Ku86 (Pucci et al., 2001) and to the Ku70 shifting from the nucleus to the cytoplasm. The shift from the nucleus to the cytoplasm of the Ku70/80 proteins in tumor cells could represents a mechanism to inhibit cell death through the cooperative interaction with sCLU, giving rise to a new chemoresistant clone with a more aggressive phenotype.
CLU Dual Face in Colon Cancer
49
III. CLONAL EXPANSION: APOPTOSIS INHIBITION A. Clusterin (CLU), a Multifunctional Protein Influenced by the Cellular Context The cooperative interactions among proteins involved in DNA repair, apoptosis induction, and the influence of the microenvironment on their activity play a central role to understand the mechanisms that underlie the clonal expansion. Partners and regulatory proteins of Ku activity are evidenced in the last few years. In this view, CLU and the balance between its different forms has been shown to be one of the main player involved in colon cancer progression, being the regulator of Ku70/80 DNA double-strand breaks repair and Bax-dependent apoptosis induction. CLU expression is markedly upregulated both in vitro and in vivo in response to various cell stress conditions. These include heat shock, UV radiation, oxidative stress, and pathologic states, such as neurodegenerative disorders, multiple sclerosis, atherosclerosis, myocardial infarction, and cancer. The presence of different CLU protein isoforms (nCLU and sCLU) and their functions within the cell was a much debated question. Nuclear clusterin (nCLU) (XIP8), was firstly described as an X-rayinduced Ku70-binding protein (KUBs) that signals cell death (Leskov et al., 2003; Yang, et al., 2000). Its role in apoptosis induction has been further described (Leskov et al., 2003; Pucci et al., 2009a,b). In normal cells, after an irreversible cell damage, nCLU cooperates with Ku70 to induce apoptotic death, activating the translocation of Bax to mitochondria. Confocal microscopy experiments revealed an apparently inactive nCLU form in the cytoplasm of nonirradiated cells (Yang et al., 2000) that translocates to the nucleus after ionizing radiation, colocalizing with nuclear Ku70/86 heterodimer involved in DNA repair and apoptosis induction (Yang et al., 1999, 2000). Data on the preferential induction of the proapoptotic clusterin form after ionizing radiation (Leskov et al., 2003), suggest that the transcription of one of the two mRNA forms is closely linked to the cellular state and could be influenced by intracellular and extracellular milieu (such as cytokines, growth factors, and stress-inducing agents) (O’Sullivan et al., 2003; Pucci et al., 2004a,b, 2009a,b; Reddy et al., 1996; Yang et al., 1999). Ku70 DNA end-joining protein has been shown to suppress apoptosis by sequestering Bax from mitochondria. The regulation of its sequestering interaction with Bax would be regulated by Ku70 acetylation state. It has been found that the acetylation of lysine at the C-terminus of the protein is sufficient to completely block the ability of Ku70 to suppress Bax-mediated apoptosis (Cohen et al., 2004). The regulation of the proapoptotic factor Bax is relevant for the development and progression of cancer (Evan and
P. Mazzarelli et al.
50 B
A
Nucleus p70 p80
Cytoplasm BAX
BAX
p70 p80
p70 sCLU
p70 nClu
Fig. 1 Ku70/80–CLU–Bax: Physiological interactions. (A) Bax is localized inactive in the cytoplasm in normal, undamaged cell interacting with the Ku70 protein C-terminus. This status determines its inability to give rise to apoptotic event. sCLU stabilizes the Ku70–Bax interaction in the cytoplasm acting as cytoprotectant. (B) After DNA damage inducing DNA double-strand breaks repair (UV treatment, ionizing radiation, etc.) Ku70 allows the translocation of Bax to the mitochondria.
Vousden, 2001). Following its activation, Bax homodimerizes translocating into the mitochondrial membrane and leading to the release of deathpromoting factors such as cytochrome c, in the cytoplasmic compartment. Bax is localized physiologically inactive in the cytoplasm in normal, undamaged cells interacting with the Ku70 protein C-terminus (Fig. 1). This status determines its inability to homodimerize and give rise to apoptotic key events. Overexpression of Ku70 in vitro blocks the Bax-induced apoptosis under some variety of stimuli in epithelial cells. After a UV treatment inducing DNA damage, the DNA double-strand-breaks repair sensor Ku70 allows the translocation of Bax to the mitochondria and its homodimerization after its sterical modification. This important function of Ku as regulator of Bax-mediated release of several death-promoting factors is in agreement with its role as caretaker in the nucleus. On the other hand, CLU seems to play an important role in cell survival pathways and in cell death escape, stabilizing the Ku70–Bax interaction in the cytoplasm that in pathological condition could lead to the survival of the aberrant cell clone. Overall, the dynamic interaction among CLU, Ku70, and Bax seems to have an important role in both tumor insurgence and its progression (Pucci et al., 2009a,b) (Fig. 2).
IV. CLU IN COLORECTAL CANCER PROGRESSION: sCLU AND APOPTOSIS ESCAPE Cell survival and cell death represent key processes in cancer development and progression. These processes could be both regulated by the balanced expression of the different CLU forms involved in antagonistic action that
CLU Dual Face in Colon Cancer
51 Neoplastic cell
BAX Ku70
sCLU
sCLU
Cytoplasm
No apoptosis, no repair
Fig. 2 Ku70–Bax–CLU pathological interaction. Apoptosis escaping. The shift of clusterin forms production, the loss of ku80, and the cytoplasmic relocalization of ku70 are related to cell death inhibition and cancer progression.
turns the cell fate. Hence a large number of studies have focused their interest on CLU in tumors and tumor progression models and its controversial role in cancer progression was ruled out focusing on the CLU different forms functions and their action in normal and in neoplastic cell processes. Evidence of the upregulation of CLU expression in intestinal tumors was reported by Chen et al. (2003). The authors investigated the relationship between CLU expression, APC function, cell proliferation, and apoptosis. CLU gene was identified as upregulated in murine and human colon cancer. Wild-type and B6-Min mice were investigated, the last carrying the multiple intestinal neoplasia (Min) mutation in the adenomatous polyposis coli (APC) gene. This line provides an experimental model of human familial intestinal cancer progression. Loss of tumor suppressor APC function initiates tumorigenesis in the intestine. The APC protein is involved in the degradation of -catenin within the cytoplasm, thus the loss of WT APC antigen leads to enhanced levels of cytoplasmic -catenin protein. A strong positive association was found between elevation of CLU expression and loss of APC function in tumor cells. The authors found CLU expression much stronger in murine tumors than in normal tissues. Tumor cells are normally poorly differentiated during uncontrolled proliferation. Lack of differentiation factors in most tumor cells with elevated CLU expression suggested that CLU could be a sensitive and stable histological indicator for murine and human intestinal tumors representing a useful diagnostic marker for colon cancer disease. Elevated CLU expression was maintained in both murine and human invasive adenocarcinomas indicating that this protein plays a role in the maintenance and/or progression of tumors. High levels of
52
P. Mazzarelli et al.
CLU were also detected in normal human colon crypts adjacent to the adenomas and adenocarcinomas, whereas they failed to reveal CLU in normal crypts far from the tumors and in tumor-free colonic tissues. Recent reports suggest the apparent dichotomy of function may be related to two different isoforms, one secreted and cytoplasmic, the other nuclear. To clarify the functional role of CLU in regulating apoptosis, Bettuzzi and his collaborators examined its expression in human colon cancer tissues and in human colon cancer cell lines. They additionally explored its expression and activity using models of APC- and chemotherapy-induced apoptosis (Chen et al., 2004). They found a decrease of CLU RNA and protein levels in colon cancer tissues largely devoid of wild-type APC when compared with matched normal tissue controls, suggesting a means for invasive cancers to avoid apoptosis. Conversely, induction of apoptosis by expression of wild-type APC or by treatment with chemotherapy led to increased clusterin RNA and protein levels localizing to apoptotic nuclei. They observed that transient transfection of CLU to colon cancer cell lines directly enhanced basal and chemotherapyinduced apoptosis. CLU-induced apoptosis was inhibited by antisense CLU and was found to be highly dependent on p21 but not p53 expression, yet a deficit in p21 can be subverted by CLU transfection. Collectively, these data support the hypothesis that nCLU function is proapoptotic when induced by APC or chemotherapy in the context of p21 expression. Absent of p21, CLU in not induced, and apoptosis is significantly inhibited. These data support a potential therapeutic role for CLU in enhancing chemotherapy-induced apoptosis and in promoting apoptosis in cells deficient in p21. Other findings were reported by Thomas-Tikhonenko et al. (2004). He demonstrated that Myc-transformed epithelial cells model downregulated CLU and that CLU could inhibit cell growth in vitro and prevent carcinogenesis in vivo. Indeed, in this experimental model, CLU transient overexpression decreased cell accumulation in Myc-transduced colonocytes suggesting a potential role of CLU as tumor suppressor. The debated role of CLU in colon cancer lately was attributed to the differential expression of CLU forms displaying antagonistic functions (nCLU and sCLU) conciliating the “tumor suppressor” and the “tumor promoting” role of this protein in cancer. Several experimental data have shown a strong correlation between a differential shift of the two CLU isoforms and tumoral progression. Our report (Pucci et al., 2004a,b) provided the first link among the unbalanced overexpression of sCLU form, the disappearance of nCLU form, and colorectal cancer progression. In fact, immunohistochemical analysis, performed on 30 bioptic and surgical samples of colorectal tumors, showed a nuclear localization of CLU in normal colonic mucosa, and a complete loss of nCLU in the advanced stages of colon cancer (Dukes C, D). In addition, the progression toward the advanced stages of cancers led to an overexpression
CLU Dual Face in Colon Cancer
53
of the highly glycosylated cytoplasmic form. In particular, colonic adenomas presented positive staining both in the nuclei and in the cytoplasm and its expression was significantly increased, as compared with normal mucosa. sCLU expression strongly increased in noninvasive carcinomas (Dukes stage A, B). The immunohistochemical observation of highly aggressive and metastatic tumors (Dukes C, D) showed that CLU could also be released in the extracellular space. Western blot analysis displayed the presence of different CLU isoforms using an anti-CLU -chain antibody. psCLU precursor form was present both in normal and tumoral tissues. The nCLU form was evident in normal mucosa, whereas it was completely lost in the tumoral tissues. The 40 kDa CLU, corresponding to the secreted form (sCLU), was present in normal tissues and it was overexpressed in the cancer samples. Moreover, the apoptotic index was inversely related to the increase of sCLU expression and to the tumor stage. In addition, in vitro experiments confirmed that in colon cancer cell CLU was extracellularly released and that the form released in the extracellular space corresponded to the sCLU. In vitro experiments were performed to determine whether the translocation of the CLU from the cytoplasm to the nucleus could be modulated by a cytostatic and proapoptotic treatment, restoring the physiological balance of the two CLU isoforms. In vitro studies confirmed a shift of the different isoforms after cytostatic treatment in colon cancer cells, related to the apoptotic induction. The cytostatic treatment with somatostatin in colon carcinoma cells (Caco2) induced a strong increase of nCLU in the nucleus. In addition, ex vivo isolated cells from normal mucosa and colorectal cancer tissues of the same patients confirmed the restore of nCLU isoform following antiproliferative treatment, concurrent to apoptosis induction. Overall, the overexpression of the sCLU in the cytoplasm of highly infiltrating tumors (and metastatic nodes), was due to a shift of CLU forms expression in cancer cells driven by exogenous growth regulatory factors.
A. CLU–Ku–Bax Localization in Colon Cancer In view of the emerging role of CLU, Ku70, and Bax interactions in tumor development and progression, the expression, localization, and physical interaction of Bax, Ku70, Ku86 were also investigated in human colorectal cancers (n ¼ 50) (Pucci et al., 2009a,b) (Fig. 3). A tumor-specific modulation of these protein factors was found in human colon cancer. Bax showed only faint cytoplasmic staining in normal mucosae (70% of controls), whereas it was overexpressed in the cytoplasm of quite all carcinomas (P ¼ 0.04). Ku70 staining was strongly positive in the nuclei of normal mucosa aside the neoplasia. In node-negative carcinomas, Ku70 expression
P. Mazzarelli et al.
54 A
B
C
Ku70
Ku80
CLU
Bax
Fig. 3 Tumor-specific modulation of ku70/80, CLU and Bax proteins in human colon cancer. Bax showed faint cytoplasmic staining in normal mucosae (A) and it was overexpressed in the cytoplasm of all carcinomas (B, C). Ku70 staining was strongly positive in the nuclei of normal mucosa (A). In node-negative carcinomas (B) Ku70 expression slightly decreased and it localized mainly in the nucleus. In node-positive carcinomas (C) Ku70 staining was distributed between nucleus and cytoplasm. The expression of Ku86 was positive in the nuclei of control tissues (A). Nuclear Ku86 expression was strongly decreased in node-negative tumors (B). No staining for Ku86 was found in the nucleus or in the cytoplasm of node-positive carcinomas (C). CLU isoforms expression was reported in Pucci et al. (2004a,b).
slightly decreased and it localized in the nucleus, while 11 out of 28 cases displayed a cytoplasmic staining as well. In node-positive carcinomas, Ku70 staining was not altered in total amount, compared with node-negative tumors, but it was distributed between nucleus and cytoplasm. In all cases, Ku70 was positive in the cytoplasmic compartment. The expression of Ku86 was positive in the nuclei of control tissues. Nuclear Ku86 expression was strongly decreased in A–B stage tumors. No staining for Ku86 was found in the nucleus or in the cytoplasm of node-positive carcinomas (C–D stages). Interestingly, Ku86 expression was lost in metastatic nodes. CLU isoforms expression confirmed previous data (Pucci et al., 2004a,b). Double immunofluorescence analysis showed that strong nuclear Ku70 staining in normal mucosa and faint Bax staining in the cytoplasm. Advanced stage carcinomas (C–D stage) showed increased levels of Ku70 and Bax and CLU proteins. Triple immunostaining and confocal analysis demonstrated the Ku70–CLU–Bax colocalization in the cytoplasm. This data suggests that in highly aggressive tumours the interaction of Ku70 and CLU with Bax permanently inhibits Bax activation and its subsequent
CLU Dual Face in Colon Cancer
55
heterodimerization and translocation into the mitochondria. This condition in advanced tumor stage leads to apoptosis escape. In vitro experiments, reported also in chapter “CLU and tumor microenvironment” of this volume, demonstrated that in colon cancer progression this physical interaction among Ku70–CLU and Bax are not irreversible and it is strongly influenced by the tumor microenvironment, suggesting that apoptosis escape could be related to exogenous factors, such as IL-6 and VEGF and TGF- present in the extracellular milieu of the tumoral mass (Pucci et al., 2009a,b). As previously mentioned, a physiological growth regulatory factor such as Somatostatin induces apoptosis after 24 h of treatment in colon cancer cell line Caco-2, determining the release of Bax from sCLU and Ku70 (Pucci et al., 2004a,b). In addition, Somatostatin treatment induced also the shift of CLU forms production inducing the upregulation of the proapoptosis nCLU. An antithetic effect was obtained treating Caco-2 with IL-6 or VEGF165a, microenvironmental factors involved in tumor progression and metastasis. In fact a strong upregulation of sCLU production and an increase in Ku–CLU–Bax binding were observed, confirming that these interactions that regulate the Bax-dependent cell death could be driven by exogenous and endogenous factors that could be determine the cell fate. From these findings, it seems that the differential shift of CLU isoform production, the loss of Ku80, and the cytoplasmic relocalization of Ku70 and sCLU overexpression are related to cell death inhibition and colorectal cancer progression. Others studies focused on CLU different forms production and their function in colon cancer. The study of Chen et al. (2004) highlighted the function of nCLU in colorectal cancer tissues and colon cancer cell lines. nCLU RNA and protein levels were decreased in colon cancer tissue, compared with normal mucosa as means of apoptosis escaping. The author analyzes APC status associated with CLU. Most colon cancer lack functional APC protein and the data suggest diminished CLU expression in these samples. The expression of WT APC or chemotherapy treatment associated to increased levels of CLU and apoptosis. Apoptosis induced by CLU was p21 dependent. In addition, it was shown that the depletion of sCLU did not affect significantly the growth rate. Data of Chen are consistent with results reported by Pucci et al. Chen T. analyzed in particular the nuclear form at protein level (60 kDa by Western blot analysis). Also the primers used to detect mRNA levels matched for the splicing isoform of nCLU variant. Xie et al. (2005) also confirmed the overexpression of cytoplasmic staining of CLU, on human tissue microarrays which contained 85 advanced colorectal cancer (Dukes B, C, and D). A significant positive correlation between overexpression of CLU and clinical stage was observed (P < 0.01). Nevertheless the same authors failed to detect the nuclear staining neither in normal nor in neoplastic colonocytes. They also showed an inverse relation between the
56
P. Mazzarelli et al.
cytoplasmic Clu overexpression and the apoptotic index (TUNEL assay). In fact, the frequency of high apoptotic index was significantly higher in tumors with a normal expression of CLU, than that in cases which overexpress CLU (P < 0.01). In addition, the cell proliferation in colorectal cancer (evaluated with ki-67 expression) positively correlated with CLU expression. In light of the above, sCLU overexpressed in highly aggressive tumors and metastatic nodes, being correlated to cell matrix formation, cell membrane remodeling, and cell–cell adhesion, could represent a potential predictive marker for colon carcinoma aggressiveness.
V. CLU AS A NEW BIOMARKER FOR COLON CANCER SCREENING At present, colon cancer is second only to lung cancer in men and to breast carcinoma in women, for incidence and mortality in western countries. The higher incidence per age is observed between the sixth and seventieth decade, while 60% of the patients survive up to 5 years. The most important reason for the low percentage of recoveries is due to the fact that when the primary tumor is removed, a high number of patients have already developed micrometastases, principally at liver. Therefore, methods for early screening are requested. Genetic counseling, predictive molecular testing, and when indicated, endoscopic surveillance at appropriate intervals should be offered to individuals from families at high risk for colorectal cancer (HNPCC or FAP). At present, the early diagnosis protocols (secondary prevention) consist of rectal exploration, determination of fecal occult blood, and rectosigmoidoscopy periodically performed on individuals of 45 years of age and older and nonsymptomatic. Periodic pan-colonoscopy is the only procedure for early diagnosis of neoplasia on individuals with positive familiar history for colorectal cancer (CRC), on patients with already a neoplasia or affected by syndrome with a high risk of neoplasia insurgence, that are part of the socalled “at risk population.” Randomized controlled trials (RCTs) have shown that annual or biennial screening in asymptomatic people over the age of 50 years using fecal occult blood test (FOBTs), can reduce CRC mortality by 15–33%. Nevertheless FOBT, utilized for early colon carcinoma diagnosis in clinical practice, yields frequent false-negative and false-positive results that lower screening effectiveness and raise program costs. On the basis of the above, new molecular pathogenetic markers, that would overcome the restrictions of the invasive methods used at present such as colonoscopy, are needed to improve the efficacy, sensitivity, and specificity of the
CLU Dual Face in Colon Cancer
57
early diagnosis test. Moreover, molecular markers would help to stratify more selectively the cohort of patients who really need colonoscopy. The use of CLU as a diagnostic marker in some pathological conditions such as type II diabetes and several coronary pathologies has already been described (Trougakos et al., 2002). There were just few previous attempts to determine CLU by ELISA in tumoral pathologies, specifically in the blood of prostate carcinoma patients (Morrissey et al., 2001). Moreover, CLU level in blood and urine has been demonstrated to be a potential marker for bladder and for kidney tumors, being directly related to the dimension of the neoplasia (Stejskal and Fiala, 2006). In a recent paper, we highlighted that the appearance and the progressive increase of the CLU cytoplasmic isoform in tumors correlated to a release of CLU in the extracellular space. In this paper, we demonstrated that sCLU upregulated in the neoplastic colonocytes was also secreted in the intestinal lumen (Pucci et al., 2009a,b). In an ex vivo experiment, isolated cells of healthy and neoplastic colonic mucosa were collected and after 72 h sCLU-level culture supernatant was determined. A significant increase of CLU level (2.9 times) was found in the culture supernatant of tumoral cells, compared to normal colonocytes of the same patient. The increased release of sCLU in tumoral cell supernatant confirmed that the overexpression previously observed in situ was strongly correlated to an increase of CLU release. Furthermore, in order to investigate if CLU release from colon cancer cells could effectively affect the total amount of the circulating protein, human colon cancer cells, Caco-2, were underskin injected in nude mice. Before inoculating Caco-2 cells, blood was collected from each mouse in order to evaluate the endogenous basal level of CLU before tumor cells injection. Mice were sacrificed at the day 15th, 20th, and 25th after tumor injection, in order to evaluate CLU level in relation of tumor size. Blood was collected, tumor was removed, and tumor size was evaluated. The level of CLU was significantly increased in blood of tumor-injected mice as compared to uninjected mice; moreover, an increased level of CLU was correlated to the dimension of the tumors suggesting its potential value as new biomarker for colorectal cancer screening. sCLU level was evaluated in the serum and stool samples of CRC patients and age-matched controls. The Dot blot analysis on human sera from colorectal cancer patients (CRC, n ¼ 35) and no cancerous subjects (controls, n ¼ 25) displayed statistically significant differences in CLU levels. In fact, CLU concentration was 82.8 26.9 g/ml in CRC cancers and 57.8 19.3 g/ml in controls (CRC vs. controls: P ¼ 0.0002). In order to avoid the interference of the increased level of CLU in blood due to other nontumoral or tumoral diseases (cancer of breast, prostate, testicle, ovary, SNC, hemo-lymphopoietic system), the level of CLU was
58
P. Mazzarelli et al.
determined in stool of the colorectal cancer patients. Dot blot analysis of fecal extracts from cancer patients (n ¼ 28) as compared to controls (n ¼ 25), provided significant differences with mean values of 47.5 19.6 and 26.8 12.8 g/g, respectively (CRC vs. controls: P < 0.000). A significant correlation between CLU values in stool and colorectal cancer stages was found (P ¼ 0.05). These results demonstrated that sCLU efficiently discriminates between colorectal cancer disease and nonneoplastic controls. In fact, the receiver operating characteristic (ROC) curves provided several cut off points to show the trade-off between sensitivity and specificity, at different cut off values. For Dot blot assay in blood, the optimal cutoff corresponded to 55.6% sensitivity and 100% specificity, whereas the stool test reached 66.7% sensitivity and 84% specificity at the selected cut off value, as reported above. In addition, a recent report confirmed that increased levels of sCLU correlated with poor survival in a population of 251 CRC patients, stage II. Recently, Kevans et al. (2009) studied and reported the same by tissue microarray and immunohistochemistry. The adverse outcome of stage II colorectal cancer correlated with epithelial and stromal sCLU immunostaining in tumor tissues. Taken together, these data suggest a potential role of sCLU as a biomarker for colon cancer screening and relapse of the disease.
VI. CONCLUSIONS AND FUTURE PERSPECTIVES Despite the original hypothesis that CLU is a marker for programmed cell death, several experiments and clinical studies have demonstrated conflicting findings on the role of CLU in tumors. Experimental results obtained in SCID mice injected with CLU transfected human renal carcinoma cells indicate that CLU overexpression may contribute both to enhance cancer cell survival, preventing apoptosis, and to increase the metastatic potential. Moreover, in vitro studies showed that CLU overexpression stimulates cell motility and invasive ability in human renal cell line. Recent findings on the opposite function of CLU different forms contributed to clarify the conflicting data on its function inside and outside the cell. Collectively these data suggest that sCLU upregulation plays a protective role against apoptosis induced by various kinds of stimuli and thereby may confer an aggressive phenotype during cancer progression. The observation on CLU expression throughout the different steps of colon carcinoma progression demonstrated the presence of the nuclear form in the nuclei of the normal mucosa. As the nuclear form has been demonstrated to be involved
CLU Dual Face in Colon Cancer
59
in cell-cycle regulation and apoptosis induction this result suggests that in a normal cell proliferative state of the colonic mucosa this protein could be probably involved in cell-cycle regulation and apoptosis induction involving the regulation of Bax activation. In colon cancer, the upregulated sCLU isoform is extracellularly released both in blood and in stool and a sensitive method was assessed to detect it, highlighting its value as new biomarker for a noninvasive colon cancer screening. There is a consensus that CRC screening is effective to prevent this disease in many cases. Due to CRC screening, the incidence of this tumor has dropped in recent years. There is less consensus regarding optimal screening strategies, as sensitivity, specificity, and patient acceptance limit current options. To overcome these barriers a range of approaches, including proteomics-based testing, stool genetic testing, radiological imaging, and enhanced endoscopies have been the focus of intense research. Presently, colonoscopy with a sensitivity of 97% and a specificity of 98% for colon cancer and a 90% sensitivity for adenomas of at least 1 cm diameter is considered the gold standard for colon cancer diagnosis and offers the potential to both diagnose and remove premalignant lesions, but it is associated with patient discomfort, complications, variable sensitivity given through the experience of the endoscopies and high costs. A useful diagnostic assay must be sensitive, must detect cancer at the onset and it must have a high specificity to minimize false positives that necessitate expensive and invasive examination. Stool testing, unlike other conventional screening approaches, is noninvasive and requires no cathartic preparation. New stool tests for CRC diagnosis have been recently developed displaying a higher sensitivity as compared to FOBT, whereas specificity is still to be defined. In particular, specificities of about 95% have been reported for tests based on detection of genetic mutations occurring in the tumoral tissues but not in the early stage and these are not present in all cases. On the other hand, markers such as calprotectin, may represent both a marker of cancer disease and of bowel inflammation, leading to nearly 30% false positive results. Recently a high-specific serum testing for colon cancer-specific antigen 2 and 4 (CCSA-2 and -4) has been proposed, but the limitation of this test is that not all colon cancers may express the NMP CCSA-2 and -4 (20–30%) and therefore a multiple marker testing is needed. Data obtained by stool analysis by Pucci et al. clearly point out that the increase of CLU in cancer patients is significant not only compared to healthy subjects but also compared to patients affected by systemic or bowel inflammatory pathologies and benign lesions of the colon. Moreover, data obtained by Dot blot in stool of cancer patients showed a positive correlation between sCLU values and stage of disease. Furthermore, data on animal model point out that the increase of sCLU level correlates with tumor size, suggesting a role of sCLU as a new marker
60
P. Mazzarelli et al.
of onset, prognosis, and relapse of colon cancer. Hence, these results suggest the potential applicative role of CLU detection to improve the effectiveness and efficiency appeal for large-scale clinical cancer screening. Moreover, studies on the molecular mechanisms that regulate the activation of CLU promoter and CLU isoforms shifting could provide new molecular targets for specific antineoplastic therapies.
REFERENCES Blunt, T., Finnie, N. J., Taccioli, G. E., Smith, G. C., Demengeot, J., and Gottlieb, T. M. (1995). Defective DNA-dependent protein kinase activity is linked to V(D)J recombination and DNA repair defects associated with the murine scid mutation. Cell 80, 813–823. Chen, X., Halberg, R. B., Ehrhardt, W. M., Torrealba, J., and Dove, W. F. (2003). Clusterin as a biomarker in murine and human intestinal neoplasia. PNAS 100(16), 9530–9535. Chen, T., Turner, J., McCarthy, S., Scaltriti, M., Bettuzzi, S., and Yeatman, T. J. (2004). Clusterinmediated apoptosis is regulated by adenomatous polyposis coli and is p21 dependent but p53 independent. Cancer Res. 64(20), 7412–7419. Cohen, H. Y., Lavu, S., Bitterman, K. J., Hekking, B., Imahiyerobo, T. A., Miller, C., Frye, R., Ploegh, H., Kessler, B. M., and Sinclair, D. A. (2004). Acetylation of the C terminus of Ku70 by CBP and PCAF controls Bax-mediated apoptosis. Mol. Cell 13, 627–638. Difilippantonio, M. J., Zhu, J., Chen, H. T., Meffre, E., Nussenzweig, M. C., Max, E. E., Ried, T., and Nussenzweig, A. (2000). DNA repair protein Ku80 suppresses chromosomal aberrations and malignant transformation. Nature 404, 510–514. Evan, G. I., and Vousden, K. H. (2001). Proliferation cell cycle and apoptosis in cancer. Nature 441, 342–348. Fewell, J. W., and Kuff, E. L. (1996). Intracellular redistribution of Ku immunoreactivity in response to cell–cell contact and growth modulating components in the medium. J. Cell Sci. 109, 1937–1946. Gottlieb, T. M., and Jackson, S. P. (1993). The DNA dependent protein kinase: Requirement for DNA ends and association with Ku antigen. Cell 72, 131–142. Gryfe, R., Swallow, C., Bapat, B., Redston, M., and Gallinger, S. (1997). Couture molecular biology of colorectal cancer. 5. J. Curr. Probl. Cancer 21, 233–300. Jackson, S. P., and Jeggo, P. A. (1995). DNA double-strand break repair and V(D)J recombination: Involvement of DNA-PK. Trends Biochem. Sci. 20, 412–415. Kevans, D., Foley, J., Tenniswood, M., Sheahan, K., Hyland, J., O’Donoghue, D., Mulcahy, H., and O’Sullivan, J. (2009). High clusterin expression correlates with a poor outcome in stage II colorectal cancers. Cancer Epidemiol. Biomarkers Prev. 18, 393–399. Koike, M., Awaji, T., Kataoka, M., Tsujimoto, G., Kartasova, T., Koike, A., and Shiomi, T. (1999). Differential subcellular localization of DNA-dependent protein kinase components Ku and DNA-PKcs during mitosis. J. Cell Sci. 112(Pt. 22), 4031–4039. Leskov, K. S., Klokov, D. Y., Li, J., Kinsella, T. J., and Boothman, D. A. (2003). Synthesis and functional analyses of nuclear clusterin, a cell death protein. J. Biol. Chem. 278, 11590–11600. Loukola, A., Vilkki, S., Singh, J., Launonen, V., and Aaltonen, L. A. (2000). Germline and somatic mutation analysis of MLH3 in MSI-positive colorectal cancer. Am. J. Pathol. 157(2), 347–352. Morrissey, C., Lakins, J., Moquin, A., Hussain, M., and Tenniswood, M. (2001). An antigen capture assay for the measurement of serum clusterin concentrations. J. Biochem. Biophys. Methods 48, 13–21.
CLU Dual Face in Colon Cancer
61
O’Sullivan, J., Whyte, L., Drake, J., and Tenniswood, M. (2003). Alterations in the posttranslational modification and intracellular trafficking of clusterin in MCF-7 cells during apoptosis. Cell Death Differ. 10, 914–927. Prabhakar, B. S., Allaway, G. P., Srinivasappa, J., and Notkins, A. L. (1990). Cell surface expression of the 70-kDa component of Ku, a DNA-binding nuclear autoantigen. J. Clin. Invest. 86, 1301–1305. Pucci, S., Mazzarelli, P., Rabitti, C., Giai, M., Gallucci, M., Flammia, G., Alcini, A., Altomare, V., and Fazio, V. M. (2001). Tumor specific modulation of Ku70/80 DNA binding activity in breast and bladder human tumor biopsies. Oncogene 20, 739–747. Pucci, S., Bonanno, E., Pichiorri, F., Angeloni, C., and Spagnoli, L. G. (2004a). Modulation of different clusterin isoforms in human colon tumorigenesis. Oncogene 23(13), 2298–2304. Pucci, S., Bonanno, E., Pichiorri, F., Mazzarelli, P., and Spagnoli, L. G. (2004b). The expression and the nuclear activity of the caretaker gene ku86 are modulated by somatostatin. Eur. J. Histochem. 48(2), 103–110. Pucci, S., Bonanno, E., Sesti, F., Mazzarelli, P., Mauriello, A., Ricci, F., Biondi Zoccai, G., Rulli, F., and Galata`, G. (2009a). Spagnoli LG Clusterin in stool: A new biomarker for colon cancer screening? Am. J. Gastroenterol. 104, 1–9. Pucci, S., Mazzarelli, P., Sesti, F., Boothman, A. D., and Spagnoli, L. G. (2009b). Interleukin-6 affects cell death escaping mechanisms acting on Bax–Ku70–Clusterin interactions in human colon cancer progression. Cell Cycle 8(3), 473–481. Reddy, K. B., Karode, M. C., Harmony, A. K., and Howe, P. H. (1996). Transforming growth factor beta (TGF beta)-induced nuclear localization of apolipoprotein J/clusterin in epithelial cells. Biochemistry 35, 309–314. Stejskal, D., and Fiala, R. R. (2006). Evaluation of serum and urine clusterin as a potential tumor marker for urinary bladder cancer. Neoplasma 53(4), 343–346. Taylor, N. P., Zighelboim, I., Huettner, P. C., Powell, M. A., Gibb, R. K., Rader, J. S., Mutch, D. G., Edmonston, T. B., and Goodfellow, P. J. (2006). DNA mismatch repair and TP53 defects are early events in uterine carcinosarcoma tumorigenesis. Mod. Pathol. 19(10), 1333–1338. Thomas-Tikhonenko, A., Viard-Leveugle, I., Dews, M., Wehrli, P., Sevignani, C., Yu, D., Ricci, S., el-Deiry, W., Aronow, B., Kaya, G., Saurat, J. H., and French, L. E. (2004). Myc-transformed epithelial cells down-regulate clusterin, which inhibits their growth in vitro and carcinogenesis in vivo. Cancer Res. 64, 3126–3136. Trougakos, I. P., Poulakou, M., Stathatos, M., Chalikia, A., Melidonis, A., and Gonos, E. S. (2002). Serum levels of the senescence biomarker clusterin/apolipoprotein J increase significantly in diabetes type II and during development of coronary heart disease or at myocardial infarction. Exp. Gerontol. 37, 1175–1187. Xie, D., Sham, J., Zeng, W. F., Che, L. H., Zhang, M., Wu, H. X., Lin, H. L., Wen, J. M., Lau, H., Hu, L., and Guan, X. Y. (2005). Oncogenic role of clusterin overexpression in multistage colorectal tumorigenesis and progression. World J. Gastroenterol. 11(21), 3285–3289. Yang, C. R., Yeh, S., Leskov, K., Odegaard, E., Hsu, H. L., and Chang, C. (1999). Isolation of Ku70-binding proteins (KUBs). Nucleic Acids Res. 27, 2165–2174. Yang, C. R., Leskov, K., Hosley-Eberlein, K., Criswell, T., Pink, J. J., Kinsella, T. J., and Boothman, D. A. (2000). Nuclear clusterin/XIP8, an X-ray-induced Ku70-binding protein that signals cell death. Proc. Natl. Acad. Sci. USA 97, 5907–5912.
Clusterin (CLU) and Lung Cancer F. Panico,* F. Rizzi,{ L. M. Fabbri,* Saverio Bettuzzi,{ and F. Luppi* *Department of Oncology, Hematology and Respiratory Diseases, Section of Respiratory Diseases, University of Modena and Reggio Emilia, 41100 Modena, Italy { Dipartimento di Medicina Sperimentale, Sezione di Biochimica, Biochimica Clinica e Biochimica dell’Esercizio Fisico, Universita` di Parma, Via Volturno 39-43100 Parma and Istituto Nazionale Biostrutture e Biosistemi (I.N.B.B.), Rome, Italy
I. II. III. IV. V. VI.
Introduction Carcinogenesis Progression/Metastasis Treatment CLU Expression and Prognosis in Patients with Lung Cancer Conclusions References Lung cancer is the leading cause of cancer-related mortality. It is categorized into two histological groups that have distinct clinical behaviors, the nonsmall cell lung cancers (NSCLC) and the small cell lung cancer (SCLC). When identified at an early stage, NSCLC is treated by surgical resection. However, patients who undergo surgical resection still have a relative low survival rate, primarily for tumor recurrence. Unfortunately, advances in cytotoxic therapy have reached a plateau and new approaches to treatment are needed together with new and better parameters for more accurate prediction of the outcome and more precise indication of the efficacy of the treatment. Several in vitro studies have examined the role of Clusterin (CLU) in carcinogenesis, lung cancer progression, and response to chemo- and radiotherapy. Studies performed in lung cancer cell lines and animal models showed that CLU is upregulated after exposure to chemo- and radiotherapy. A potential role proposed for the protein is cytoprotective. In vitro, CLU silencing by antisense oligonucleotides (ASO) and small-interfering RNAs (siRNA) directed against CLU mRNA in CLU-rich lung cancer cell lines sensitized cells to chemotherapy and radiotherapy and decreased their metastatic potential. In vivo, a recent work analyzed the prognostic role of CLU in NSCLC, showing that CLU-positive patients with lung cancer had a better overall survival and disease-free survival than those with CLU-negative tumors. These data are contradictory to the promising in vitro results. From the results of these studies we may hypothesize that in early-stage lung cancers CLU represents a positive biomarker correlating with better overall survival. In advanced patients, already treated with chemo- and radiotherapy, the induction of CLU may confer resistance to the treatments. However, many studies are needed to better understand the role of CLU in early-stage and advanced lung cancers with the aim to discriminate patients and specific local conditions that could benefit for a CLU knocking down treatment. # 2009 Elsevier Inc.
Advances in CANCER RESEARCH Copyright 2009, Elsevier Inc. All rights reserved.
63
0065-230X/09 $35.00 DOI: 10.1016/S0065-230X(09)05004-0
F. Panico et al.
64
I. INTRODUCTION Unfortunately, lung cancer still remains the predominant cause of cancerrelated death in both women and men in the western world, with 166,000 deaths in 2008. The sum total of lung cancer deaths exceeds those of prostate, breast, and colon cancer combined (Jemal et al., 2008). From the histological point of view, it is categorized into the following two groups with distinct clinical behaviors: (i) noncsmall cell lung cancer (NSCLC); (ii) small cell lung cancer (SCLC). NSCLC accounts for 80–85% of all lung cancer cases, and includes adenocarcinoma (35–45%), squamous cell carcinoma (25–35%), and large cell carcinoma (<10%). The treatment of choice for early-stage NSCLC is surgical resection, which confers a 5-year survival rate in 80% of patients (Smythe, 2003). Unfortunately, only 20– 30% of patients are candidates for curative surgery due to the spread of disease at presentation (Carney and Hansen, 2000). The disease is aggressive, because early stages I and II NSCLC patients who undergo surgical resection still have a relatively “low” survival rate (60–70%) at 5 years (Jemal et al., 2006; Paik et al., 2004). Prostate, breast, and colorectal carcinoma have all demonstrated significant improvements in 5-year survival over time and are currently 99%, 88%, and 64%, respectively, although prostate cancer is known to be a slow progression disease. In contrast, the 5-year survival rate for lung cancer has remained relatively stable at 15% (Borczuk et al., 2009). There are several potential explanations for the disparity between lung cancer survival and that of other common tumors, including late detection and histological heterogeneity. Currently, over 75% of new lung cancer diagnoses are made in patients presenting with distant or regional metastatic disease (Jemal et al., 2006). SCLC accounts for 15–20% of all lung cancers, and is characterized by rapid growth and early metastasis. Thus, surgical resection is rarely possible. Chemotherapy and/or radiotherapy remain the only treatment options for SCLC and advanced NSCLC, but intensive therapeutic regimes offer only a short-term survival benefit (Ohe, 2004). Unfortunately, advances in cytotoxic therapy have reached a plateau and new approaches to treat these diseases are needed. Histologically, prostate, breast, and colorectal carcinomas are uniformly adenocarcinoma and treatment is primarily determined by clinical stage, at times modified by results of molecular assays (Paik et al., 2004; Romond et al., 2005). In contrast, only 30% of lung carcinoma is adenocarcinoma. Yet for the most part, until recently, NSCLC (squamous, large cell, and adenocarcinoma) were treated similarly, regardless of the biological heterogeneity associated with histology. It is likely that poor historical lung cancer
CLU and Lung Cancer
65
response rates may in part be attributable to a relatively homogenous approach to a heterogeneous disease. Ongoing efforts are underway to identify clinically relevant biological properties of tumors that will facilitate individualized lung cancer treatment. This research takes advantage of technical advances that allow rapid high-throughput assays to interrogate the genome (mRNA, microRNA, copy number, mutation analyses), proteome, and epigenome. Clusterin (CLU) has been implicated in a variety of physiologic and pathologic processes, including cancer (Shannan et al., 2006). This gene coding for multiple protein products is involved in numerous physiological processes important for carcinogenesis and tumor growth, including apoptotic cell death, cell-cycle regulation, DNA repair, cell adhesion, tissue remodeling, lipid transportation, membrane recycling, and immune system regulation. Early studies seemed to establish that enhanced CLU gene expression was a marker of apoptotic cell loss (Lakins et al., 1998). Importantly, the level of expression of CLU has been studied in various human malignancies, often with conflicting results. Among these, bladder (Miyake et al., 2001), kidney (Hara et al., 2001), prostate (Bettuzzi et al., 2002; July et al., 2002; Miyake et al., 2003, 2004; Scaltriti et al., 2004a; Zellweger et al., 2003), colon (Pucci et al., 2004), breast (Leskov et al., 2003; Redondo et al., 2000), and lung tumors (July et al., 2004). Oncomine (www. oncomine.org) is a powerful web application that integrates and unifies high-throughput cancer profiling data obtained with the Affimetrix platform so that target expression across a large volume of cancer types, subtypes, and experiments can be assessed online in seconds. In three independent DNA microarray studies performed in patients, CLU was found downregulated when adenocarcinoma, squamous cell lung carcinoma, and SCLC were compared to normal mached control cells. In this review, we will focus on the potential role of CLU in lung cancer carcinogenesis, progression, prognosis, and treatment.
II. CARCINOGENESIS A key event in the development of lung cancer is escaping from the normal control of cell proliferation causing uncontrolled growth of transformed cells. Fifteen percent of lifetime smokers develop lung cancer, but 10% of lung cancers occur in never-smokers (Spitz et al., 2003). In nonsmokers, exposure to second hand smoke or to other lung carcinogens such as radon, asbestos, arsenic, or air pollution may be contributory. In both smokers and nonsmokers, genetic polymorphisms in genes associated with carcinogen metabolism, DNA damage/repair, and cell-cycle control may influence lung cancer
66
F. Panico et al.
susceptibility and modify the injury response associated with exposure to the dozens of carcinogens contained in tobacco smoke (Hecht, 2002; Schwartz et al., 2007). While the heritable component of lung cancer risk due to these genetic polymorphisms is most notable in early-onset cases occurring with less total smoke exposure, due to the limited number of genes studied, the overall attributable risk remains small even in this group of patients (Schwartz et al., 2007; Vineis et al., 2007). Because of tobacco smoke exposure, the bronchial epithelium of smokers is subject to field cancerization with alterations of airway cell DNA. Such alterations bring about oncogene activation, tumor suppressor gene silencing, and widespread loss of heterozygosity, all of which drive distinct gene expression signatures. While it is unclear which molecular alterations are required for cancer onset and progression, in some instances carcinogenesis leads to malignant transformation of lung cancer progenitor cells. This malignant transformation may also require contributions from other systemic cell types, such as bone marrow-derived stem cells (Avital et al., 2007; Houghton et al., 2004). CLU is believed to be implicated in multiple cellular processes. Recent observations indicated an association of CLU expression with diverse and sometimes contradictory functions, such as cell survival, tumor progression, acquisition of drug resistance, or apoptosis (Gleave and Miyake, 2005; Pucci et al., 2004; Trougakos and Gonos, 2002). Although CLU is an apoptosis-associated gene, a precise relationship between this gene activity and programmed cell death has not been elucidated clearly until the role of nuclear CLU (nCLU) has been described (Leskov et al., 2001, 2003; Scaltriti et al., 2004b; Yang et al., 2000). Several forms of CLU proteins have been identified in mammalian cells (Shannan et al., 2006; Trougakos and Gonos, 2002). At least two major forms affect the fate of the cell: the secretory glycoprotein (sCLU; see chapters “The chaperone action of CLU and its putative role in quality control of extracellular protein folding” and “Cell protective functions of secretory CLU (sCLU)” of Vol. 104), usually excreted in body fluids, and the nuclear (nCLU; see chapter “Nuclear CLU (nCLU) and the fate of the cell” of Vol. 104) form, usually localized in the nucleus of cells doomed to die. The mechanisms producing nCLU are still unclear (Rizzi and Bettuzzi, 2008) (please see also chapter “Clusterin (CLU): From one gene and two transcripts to many proteins” of Vol. 104 to this regard). It has been hypothesized that the two forms may derive from: (i) different mRNAs produced by alternative splicing (Leskov et al., 2003); (ii) from the same transcript using two alternative translation start sites (Moretti et al., 2007; Scaltriti et al., 2004b); and (iii) by posttranslational modification of a protein precursor (Caccamo et al., 2005). Several studies showed that nCLU acts as a prodeath signal, inhibiting cell growth and survival (Criswell et al., 2003;
CLU and Lung Cancer
67
Lakins et al., 1998; Leskov et al., 2003; Trougakos and Gonos, 2002; Yang et al., 1999). However, other studies have shown that sCLU expression may exert cytoprotective actions (Criswell et al., 2003; Miyake et al., 2001, 2004; Trougakos and Gonos, 2002; Zellweger et al., 2002, 2003). In a recent article, Shannan et al. suggested that tumor cell survival is connected with overexpression of sCLU and loss of nCLU (Shannan et al., 2006). This hypothesis is apparently supported by Cai and colleagues, which explored genes sensitivity to the anticancer drugs Navelbine and Docetaxel both in SCLC and NSCLC cell strains (Cai et al., 2005). Using microarray technology, they observed that CLU was overexpressed in NSCLC cell strains that were resistant to the previously reported chemotherapeutic agents, suggesting a prosurvival role for CLU at least in NSCLC cells. These results suggest a role of sCLU in promoting cell survival in lung cancer. Again, these findings— exploring the role of CLU after chemotherapeutic treatment—could be contradictory to the clinical data above reported, but it must be considered that the latter data have been obtained in basal conditions.
III. PROGRESSION/METASTASIS Accumulating evidences clearly indicate that perturbation of the integrity of integrated signaling networks, which positively or negatively regulate various cellular processes to maintain homeostasis of the lung, leads to the carcinogenesis and progression of lung cancer. In the end, the accumulated genetic and epigenetic alterations are thought to similarly confer various capabilities on lung cancer cells, including escaping from growth inhibitory signals and telomeres shortening, resistance to apoptosis, sustained stimuli for proliferation and angiogenesis and invasive and metastatic features (Osada and Takahashi, 2002). Increasing evidence points to a crucial role of epithelial–mesenchymal transition (EMT) in tumor progression, which would endow carcinoma cells with invasive and metastatic properties. EMT is a process by which epithelial cells modulate their phenotype and acquire mesenchymal-like properties via disruption of intercellular adhesion and enhancement of cell motility. EMT is normally seen in tissue morphogenesis during embryonic development. In addition, EMT is involved in some fibrosing conditions subsequent to tissue injury. Because it is known that processes required for embryonic development, if inappropriately activated in adult life, can lead to pathological conditions, the hypothesis that a reactivation of some aspects of the embryonic EMT program underlies the mechanism of tumor invasion has been raised (Guarino et al., 2007; Yang and Weinberg, 2008). In a recent study, Chou and colleagues showed the involvement of CLU in modulating invasiveness of cancer cells, discovering that CLU levels
68
F. Panico et al.
positively correlated with the degree of invasiveness in human lung adenocarcinoma cell lines in vitro. In fact, CLU promoted cell migration and invasion in human lung adenocarcinoma (Chou et al., 2009). In this work, the authors observed that CLU-rich cells displayed a spindle-shape morphology, while those with low CLU levels were cuboidal in shape, suggesting the ability of cells with spindle-shape morphology to migrate. CLU silencing by small-interfering (si) RNA in a highly invasive, CLU-rich lung adenocarcinoma cell line induced a mesenchymal-to-epithelial transition, as observed by the following morphological features: acquisition of a cuboidal morphology, increased E-cadherin, and reduction in fibronectin expression. These cells showed also a reduced migration and decreased ability to metastatize. Therefore, the authors concluded that CLU expression may mediate EMT and invasiveness of human lung adenocarcinoma cells through the activation of some intracellular pathways, particularly ERK–Slug pathway (Chou et al., 2009). These results are in line with those of French and colleagues, reporting that CLU gene is widely expressed during murine embryogenesis, nearly exclusively in the developing epithelia and often in association with distinct stages of their differentiation. Particularly in the lung, CLU gene was selectively and transiently expressed by differentiating bronchial epithelial cells. In fact, while CLU mRNA was highly expressed in the lung of 14.5 and 16.5 daysold embryos, it was drastically reduced in 18.5 days-old embryos in which the process of branching morphogenesis was complete (French et al., 1993). Similarly, an increase of cytoplasmic CLU was reported in the progression of hepatocellular carcinoma (Lau et al., 2006) and colorectal carcinoma (Xie et al., 2005). In particular, CLU correlated with an increase in tumor cell migration and a decrease in the apoptotic index of the tumor. Taken together, these results suggest a role for CLU in promoting neoplastic cell migration, invasiveness, and metastatization, but these data obtained in cell systems have to be taken with circumvention in consideration of the fact that (i) contradictory reports are also available in the literature (Moretti et al., 2007); (ii) data obtained in vivo in tissue specimens taken from patients in real clinical settings may suggest an alternative scenario.
IV. TREATMENT Treatment of lung cancer depends on the cell type (NSCLC vs. small cell), tumor stage, and patient’s overall condition. Patients with stage I, II, or III NSCLC are generally treated with curative intent, using surgery, chemotherapy, radiation therapy (RT), or a combined modality approach. However, patients deemed suitable for curative treatment will still maintain a high rate of relapse (Jemal et al., 2008). Patients with advanced disease are often treated
CLU and Lung Cancer
69
with systemic chemotherapy and, more recently, with biological therapy. On the basis of the hypothesis previously reported in this review concerning a role for CLU in promoting neoplastic cell survival and invasiveness, a few studies have attempted the silencing of CLU gene both in vitro and in vivo. Okano et al. observed that increased serum levels of CLU were detected in patients with NSCLC compared with control patients (Okano et al., 2006). Furthermore, Chi and colleagues showed that an antisense oligonucleotide (ASO), complementary to CLU mRNA (called OGX-011) decreased serum CLU levels at a biologically active dose, suggesting a potential curative treatment by this molecule. Therefore, OGX-011 was used in combination with the chemotherapeutic agent Docetaxel in a clinical trial. Specific toxicity was described as mild or moderate for the most part, with additional toxicity being qualitatively and quantitatively in keeping with what would be expected with Docetaxel alone. However, gastrointestinal side effects such as diarrhea and mucositis have been found increased at higher doses of OGX-011 (Chi et al., 2008). Gastrointestinal toxicity has been observed previously with other antisense therapeutics and could be a potential class effect. However, this increased toxicity may also be indicative of targeted/side effect of OGX-011, because it has been postulated that CLU plays a role in protection of the mucosal barrier (Gassler et al., 2001). A phase II study with OGX-011 in combination with Gemcitabine/Platinum has been recently closed in patients with stage III/IV NSCLC (http://ClinicalTrial.gov) and the recruitment was stopped. The final report is expected for the end of 2009. July and colleagues have evaluated CLU levels in a subset of lung cancer tissues. They reported that CLU was expressed in 82% of the lung tumors originating from bronchial epithelial lining (adenocarcinoma, squamous cell carcinoma, and large cell carcinoma) without significant association between CLU level and tumor type (July et al., 2004). CLU staining was confined to the cytoplasm. Also a lung adenocarcinoma cell line, A549, showed high levels of cytoplasmic CLU expression. The authors reported that CLU silencing in A549 by a specific ASO and a small-interfering RNA (siRNA) sensitized lung cancer cells to chemotherapy both in vitro and in vivo. In fact, in a mouse model in which tumor was induced by A549 inoculation, synergistic effects of combined use of CLU ASO plus Paclitaxel was also observed. Systemic administration of CLU ASO plus Paclitaxel (or Gemcitibine) suppressed A549 tumor growth compared to treatment with mismatch control oligonucleotides plus Paclitaxel. Detection of increased apoptosis after combined ASO and chemotherapy by DNA fragmentation analysis in A549 cells suggests that decrease in tumor progression rates after combined CLU ASO plus Paclitaxel resulted from enhanced chemotherapy-induced apoptosis rather than decreased cell proliferation (July et al., 2004). However, the results reported by the authors shown that
F. Panico et al.
70
well-differentiated adenocarcinomas had a greater number of CLU-positive cells than those of poorly differentiated adenocarcinomas. In contrast, in squamous cell carcinoma, the number of CLU-positive cells was much higher in poorly differentiated tumors than in well-differentiated. These results suggest that CLU may have a different pattern of expression in different histological subtypes of lung cancers, suggesting that an analysis of CLU expression in a well-defined subset of lung cancers may help to better understand the role of CLU in lung tumorigenesis and to discriminate patients that could benefit from a treatment with ASO against CLU (Table 1). Furthermore, Cao and colleagues reported that CLU silencing by OGX011 in a lung cancer cell line (H460 cell line) sensitized cells to radiation by increasing apoptosis and decreasing cell viability. In vivo, the combination of OGX-011 with radiotherapy resulted in a significant growth delay and vascular regression. The authors have explained their results by suggesting that CLU ASO had a preferential effect on the secreted form of CLU, allowing nCLU to continue to function as an apoptotic agent (Cao et al., 2005). This result is somehow puzzling, because the target sequence of OGX-011 is exon 2 of CLU gene, which is common to all CLU isoform transcripts known so far (please see chapter “Clusterin (CLU): From one gene and two transcripts to many proteins” of Vol. 104). Therefore, there is no rationale explaining how it is possible to specifically ablate sCLU while inducing nCLU expression. As a matter of fact, the hypothesis that nCLU is coded by a specific mRNA originated by alternative splicing of isoform 1 transcript lacking exon 2 has never been confirmed in colon and prostate cancer cells (please see chapter “Clusterin (CLU): From one gene and two transcripts to many proteins” of Vol. 104). Table 1 CLU Immunostaining in Lung Cancer Tissue Microarray (July et al., 2004) No. of cases
Adenocarcinoma
Squamous cell carcinoma
Large cell carcinoma Small cell carcinoma
Well differentiated Moderately differentiated Poorly differentiated Well differentiated Moderately differentiated Poorly differentiated N/A N/A
Clusterin
Mean
Examined/ No. of total cases 18/44 13/44
Positive cells 40.9% 29.5%
Intensity 1.2 1.3
13/44 12/79
29.5% 13.5%
1.3 1.0
31/79
41.5%
1.2
36/79 25/25 1/1
29.5% 100% 0
1.2 1.24 0.00
CLU and Lung Cancer
71
V. CLU EXPRESSION AND PROGNOSIS IN PATIENTS WITH LUNG CANCER The data reported above are of interest, but in the majority of the cases they are related to cell systems or cancer cell lines inoculated in a suitable mouse host. Data from tissue specimens obtained from patients in a real clinical setting should be more informative. Over 75% of new lung cancer diagnoses are in patients who present with distant or regional metastatic disease. Early diagnosis remains an elusive goal with current research focused on identifying lung cancer-specific molecular alterations in accessible sites such as the airway, breath (Powell, 2008), and blood (Maheswaran et al., 2008). Because of tobacco smoke exposure, the bronchial airway epithelium of smokers is subject to field cancerization with alterations of airway cell DNA. Such alterations bring about oncogene activation, tumor suppressor gene silencing, and widespread loss of heterozygosity that is associated with preneoplastic lesions and cancer. The distinction of benign smoking-associated field changes from high-risk molecular alterations is a challenge for early diagnosis strategies and has been addressed in recent studies (Jonsson et al., 2008). When identified at an early stage, NSCLC is primarily treated by surgical resection, which is potentially curative. However, 30–60% of patients with stage IB to IIIA NSCLC die within 5 years after surgery, primarily from tumor recurrence (Raso and Wistuba, 2007). Therefore, although TNM staging is the established prognostic feature, better parameters are urgently needed for more accurate prediction of the outcome and more precise indication of the efficacy of the treatment(s) (Niklinski et al., 2001). Given the high recurrence rate of the tumor, many authors have looked for clinical and pathological markers that can guide therapy as reliable prognostic indicators, particularly helping to identify patients likely to benefit from adjunct therapies. In this view, several studies have examined the prognostic value of CLU expression in various malignancies with conflicting results. Only one study investigated the prognostic role of CLU in a series of resected NSCLC, suggesting a positive prognostic effect for CLU expression. In fact, the authors found that CLU immunostaining predicted overall survival and disease-free survival independent of the clinical stage and of the tumor histological type (Fig. 1). Furthermore, the authors found also a significant association between CLU staining and histological type, with positive CLU staining observed in 57% adenocarcinoma, 21% of squamous cell carcinoma, and 35% of other NSCLC tumors (Albert et al., 2007). These data are in substantial agreement with data retrieved from Oncomine (www.oncomine. org). In three independent DNA microarray studies performed in patients, CLU was found downregulated when adenocarcinoma, squamous cell lung
F. Panico et al.
72 A
Overall survival
B
1
Recurrence-free survival 1
Recurrence-free distribution function
Negative Positive
Survival distribution function
0.8
0.6
0.4
0.2
0.8
0.6
0.4
0.2
P = 0.0108
Negative Positive
P = 0.0231
0
0 0
1
2
3
4
5
Years after diagnosis
6
7
0
1
2
3
4
5
6
7
Years after diagnosis
Fig. 1 Kaplan–Meier survival curve for NSCLC patients. (A) Overall survival and (B) recurrence-free survival for positive versus negative CLU staining. Log-rank test comparing the two curves for overall survival and recurrence-free survival revealed statistically significant differences (P ¼ 0.018 and P ¼ 0.0231, respectively) (Albert et al., 2007).
carcinoma, and SCLC were compared to normal matched control cells. Furthermore, the Garber lung study shows that CLU expression is higher in lung adenocarcinoma cells when compared to large cell lung cancer, SCLC, and squamous cell lung carcinoma (www.oncomine.org). Therefore, the promising data obtained working in lung cancer cell lines do not fit well with the real clinical situation, in which CLU expression seems to be a favorable prognostic marker.
VI. CONCLUSIONS There are several studies reported in the literature about the potential role of CLU in lung cancer, but results are often conflicting. Studies conducted in vitro using lung cancer cell lines—both obtained from NSCLC and SCLC—showed high levels of CLU expression in lung cancer cells. In these cell lines, CLU seems to have a cytoprotective role. Particularly, cytoplasmic CLU should be involved in cell resistance against chemo- (July et al., 2004) and radiotherapy (Cao et al., 2005). Recent studies
73
CLU and Lung Cancer
showed that CLU expression is inducted after chemo- and radiotherapy, contributing to decrease chemo- and radiotherapy efficacy (Cai et al., 2005). Furthermore, high levels of cytoplasmic CLU in pulmonary neoplastic cells positively correlate with the degree of invasiveness and metastatic ability of these tumors (Chou et al., 2009). The cytoprotective role of CLU in the lung has also been observed by Heller and colleagues, which induced an inflammatory response in isolated lungs obtained from rabbit. In this system, the inoculation of 2.5 g/ml of sCLU reduced the cellular damage caused by oxidative stress (Heller et al., 2003). Similarly, we showed previously in vitro that cigarette smoke induced CLU synthesis and release in the extracellular milieu. In the same work, we found that sCLU may have a protective effect against cigarette smoke-induced oxidative stress in lung fibroblasts (Carnevali et al., 2006). This result is in agreement with a general consensus about the cytoprotective role of sCLU in the extracellular compartment, probably by acting as extracellular chaperone (please see chapters “The chaperone action of CLU and its putative role in quality control of extracellular protein folding” and “Cell protective functions of secretory CLU (sCLU)” of Vol. 104). Clinical studies may depict a different story. A recent study analyzed the prognostic role of CLU in patients with early NSCLC who underwent pulmonary resection, showing that high CLU levels in lung tissue correlated with a better prognosis after 3 years from surgical intervention. These results appear to be conflicting compared to in vitro studies and with those obtained from animal models (Table 2). Various explanations may justify this discrepancy: (i) cancer cell lines may not be the best bench to challenge an hypothesis on CLU action, because they usually display many DNA mutations often related to resistance and positive selection during cell culturing which may potently alter cell homeostasis; (ii) the type of antibodies used in these studies
Table 2 Schematic Results for CLU Expression from In Vitro and In Vivo Studies in Lung Tumorigenesis Carcinogenesis In vitro studies Cai et al. (2005) Chou et al. (2009) July et al. (2004) Cao et al. (2005) In vivo studies Okano et al. (2006) Albert et al. (2007) "Clusterin overexpression. #Clusterin downregulation.
Progression
" " " " #
#
Treatment
" " "
74
F. Panico et al.
performed in human lung biopsies do not distinguish different CLU protein forms. Data available on Oncomine database show that CLU is downregulated in tumors as compared to normal lung (www.oncomine.org). We have further investigated this issue. Our more recent study shows that CLU is downregulated in lung cancer, with a further reduction of its expression during the progression of the malignancy (manuscript in preparation). In addition, we have confirmed the positive prognostic value of CLU in patients with resected NSCLC.
ACKNOWLEDGMENTS F. P., L. M. F., and F. L. are supported by unrestricted grants from Associazione per lo Studio dei Tumori e delle Malattie Polmonari (ASTMP, Padova, Italy), Consorzio Ferrara Ricerche (CFR, Ferrara, Italy) and Programma di Ricerca Regione-Universita` 2007-2009; F. R. and S. B. are supported by FIL 2008 and FIL 2009, University of Parma, Italy, AIRC (UK) Grant No. 06-711 and Istituto Nazionale Biostrutture e Biosistemi (INBB), Roma, Italy.
REFERENCES Albert, J. M., et al. (2007). Cytoplasmic clusterin expression is associated with longer survival in patients with resected non small cell lung cancer. Cancer Epidemiol. Biomarkers Prev. 16, 1845–1851. Avital, I., et al. (2007). Donor-derived human bone marrow cells contribute to solid organ cancers developing after bone marrow transplantation. Stem Cells 25, 2903–2909. Bettuzzi, S., et al. (2002). Clusterin (SGP-2) transient overexpression decreases proliferation rate of SV40-immortalized human prostate epithelial cells by slowing down cell cycle progression. Oncogene 21, 4328–4334. Borczuk, A. C., et al. (2009). Genomics of lung cancer. Proc. Am. Thorac. Soc. 6, 152–158. Caccamo, A. E., et al. (2005). Ca2þ depletion induces nuclear clusterin, a novel effector of apoptosis in immortalized human prostate cells. Cell Death Differ. 12, 101–104. Cai, L., et al. (2005). Effects of navelbine and docetaxel on gene expression in lung cancer cell strains. Acta Pharmacol. Sin. 26, 1017–1024. Cao, C., et al. (2005). Clusterin as a therapeutic target for radiation sensitization in a lung cancer model. Int. J. Radiat. Oncol. Biol. Phys. 63, 1228–1236. Carnevali, S., et al. (2006). Clusterin decreases oxidative stress in lung fibroblasts exposed to cigarette smoke. Am. J. Respir. Crit. Care Med. 174, 393–399. Carney, D. N., and Hansen, H. H. (2000). Non-small-cell lung cancer—stalemate or progress? N. Engl. J. Med. 343, 1261–1262. Chi, K. N., et al. (2008). A phase I study of OGX-011, a 2’-methoxyethyl phosphorothioate antisense to clusterin, in combination with docetaxel in patients with advanced cancer. Clin. Cancer Res. 14, 833–839. Chou, T. Y., et al. (2009). Clusterin silencing in human lung adenocarcinoma cells induces a mesenchymal-to-epithelial transition through modulating the ERK/Slug pathway. Cell Signal. 21, 704–711.
CLU and Lung Cancer
75
Criswell, T., et al. (2003). Repression of IR-inducible clusterin expression by the p53 tumor suppressor protein. Cancer Biol. Ther. 2, 372–380. French, L. E., et al. (1993). Murine clusterin: Molecular cloning and mRNA localization of a gene associated with epithelial differentiation processes during embryogenesis. J. Cell Biol. 122, 1119–1130. Gassler, N., et al. (2001). Expression of clusterin in Crohn’s disease of the terminal ileum. Histol. Histopathol. 16, 755–762. Gleave, M., and Miyake, H. (2005). Use of antisense oligonucleotides targeting the cytoprotective gene, clusterin, to enhance androgen- and chemo-sensitivity in prostate cancer. World J. Urol. 23, 38–46. Guarino, M., et al. (2007). The role of epithelial–mesenchymal transition in cancer pathology. Pathology 39, 305–318. Hara, I., et al. (2001). Introduction of clusterin gene into human renal cell carcinoma cells enhances their resistance to cytotoxic chemotherapy through inhibition of apoptosis both in vitro and in vivo. Jpn J. Cancer Res. 92, 1220–1224. Hecht, S. S. (2002). Cigarette smoking and lung cancer: Chemical mechanisms and approaches to prevention. Lancet Oncol. 3, 461–469. Heller, A. R., et al. (2003). Clusterin protects the lung from leukocyte-induced injury. Shock 20, 166–170. Houghton, J., et al. (2004). Gastric cancer originating from bone marrow-derived cells. Science 306, 1568–1571. Jemal, A., et al. (2006). Cancer statistics, 2006. CA Cancer J. Clin. 56, 106–130. Jemal, A., et al. (2008). Cancer statistics, 2008. CA Cancer J. Clin. 58, 71–96. Jonsson, S., et al. (2008). Chromosomal aneusomy in bronchial high-grade lesions is associated with invasive lung cancer. Am. J. Respir. Crit. Care Med. 177, 342–347. July, L. V., et al. (2002). Clusterin expression is significantly enhanced in prostate cancer cells following androgen withdrawal therapy. Prostate 50, 179–188. July, L. V., et al. (2004). Nucleotide-based therapies targeting clusterin chemosensitize human lung adenocarcinoma cells both in vitro and in vivo. Mol. Cancer Ther. 3, 223–232. Lakins, J., et al. (1998). Clusterin biogenesis is altered during apoptosis in the regressing rat ventral prostate. J. Biol. Chem. 273, 27887–27895. Lau, S. H., et al. (2006). Clusterin plays an important role in hepatocellular carcinoma metastasis. Oncogene 25, 1242–1250. Leskov, K. S., et al. (2001). When X-ray-inducible proteins meet DNA double strand break repair. Semin. Radiat. Oncol. 11, 352–372. Leskov, K. S., et al. (2003). Synthesis and functional analyses of nuclear clusterin, a cell death protein. J. Biol. Chem. 278, 11590–11600. Maheswaran, S., et al. (2008). Detection of mutations in EGFR in circulating lung-cancer cells. N. Engl. J. Med. 359, 366–377. Miyake, H., et al. (2001). Synergistic chemsensitization and inhibition of tumor growth and metastasis by the antisense oligodeoxynucleotide targeting clusterin gene in a human bladder cancer model. Clin. Cancer Res. 7, 4245–4252. Miyake, H., et al. (2003). Resistance to cytotoxic chemotherapy-induced apoptosis in human prostate cancer cells is associated with intracellular clusterin expression. Oncol. Rep. 10, 469–473. Miyake, H., et al. (2004). Protection of androgen-dependent human prostate cancer cells from oxidative stress-induced DNA damage by overexpression of clusterin and its modulation by androgen. Prostate 61, 318–323. Moretti, R. M., et al. (2007). Clusterin isoforms differentially affect growth and motility of prostate cells: Possible implications in prostate tumorigenesis. Cancer Res. 67, 10325–10333.
76
F. Panico et al.
Niklinski, J., et al. (2001). Prognostic molecular markers in non-small cell lung cancer. Lung Cancer 34(Suppl. 2), S53–S58. Ohe, Y. (2004). Chemoradiotherapy for lung cancer: Current status and perspectives. Int. J. Clin. Oncol. 9, 435–443. Okano, T., et al. (2006). Plasma proteomics of lung cancer by a linkage of multi-dimensional liquid chromatography and two-dimensional difference gel electrophoresis. Proteomics 6, 3938–3948. Osada, H., and Takahashi, T. (2002). Genetic alterations of multiple tumor suppressors and oncogenes in the carcinogenesis and progression of lung cancer. Oncogene 21, 7421–7434. Paik, S., et al. (2004). A multigene assay to predict recurrence of tamoxifen-treated, nodenegative breast cancer. N. Engl. J. Med. 351, 2817–2826. Powell, C. A. (2008). Waiting to exhale. Am. J. Respir. Crit. Care Med. 177, 246–247. Pucci, S., et al. (2004). Modulation of different clusterin isoforms in human colon tumorigenesis. Oncogene 23, 2298–2304. Raso, M. G., and Wistuba, I. I. (2007). Molecular pathogenesis of early-stage non-small cell lung cancer and a proposal for tissue banking to facilitate identification of new biomarkers. J. Thorac. Oncol. 2, S128–S135. Redondo, M., et al. (2000). Overexpression of clusterin in human breast carcinoma. Am. J. Pathol. 157, 393–399. Rizzi, F., and Bettuzzi, S. (2008). Targeting Clusterin in prostate cancer. J. Physiol. Pharmacol. 59(Suppl. 9), 265–274. Romond, E. H., et al. (2005). Trastuzumab plus adjuvant chemotherapy for operable HER2positive breast cancer. N. Engl. J. Med. 353, 1673–1684. Scaltriti, M., et al. (2004a). Clusterin (SGP-2, ApoJ) expression is downregulated in low- and high-grade human prostate cancer. Int. J. Cancer 108, 23–30. Scaltriti, M., et al. (2004b). Intracellular clusterin induces G2-M phase arrest and cell death in PC-3 prostate cancer cells1. Cancer Res. 64, 6174–6182. Schwartz, A. G., et al. (2007). The molecular epidemiology of lung cancer. Carcinogenesis 28, 507–518. Shannan, B., et al. (2006). Challenge and promise: Roles for clusterin in pathogenesis, progression and therapy of cancer. Cell Death Differ. 13, 12–19. Smythe, W. R. (2003). Treatment of stage I non-small cell lung carcinoma. Chest 123, 181S–187S. Spitz, M. R., et al. (2003). Genetic susceptibility to lung cancer: The role of DNA damage and repair. Cancer Epidemiol. Biomarkers Prev. 12, 689–698. Trougakos, I. P., and Gonos, E. S. (2002). Clusterin/apolipoprotein J in human aging and cancer. Int. J. Biochem. Cell. Biol. 34, 1430–1448. Vineis, P., et al. (2007). Evidence of gene gene interactions in lung carcinogenesis in a large pooled analysis. Carcinogenesis 28, 1902–1905. Xie, D., et al. (2005). Oncogenic role of clusterin overexpression in multistage colorectal tumorigenesis and progression. World J. Gastroenterol. 11, 3285–3289. Yang, J., and Weinberg, R. A. (2008). Epithelial–mesenchymal transition: At the crossroads of development and tumor metastasis. Dev. Cell 14, 818–829. Yang, C. R., et al. (1999). Isolation of Ku70-binding proteins (KUBs). Nucleic Acids Res. 27, 2165–2174. Yang, C. R., et al. (2000). Nuclear clusterin/XIP8, an X-ray-induced Ku70-binding protein that signals cell death. Proc. Natl. Acad. Sci. USA 97, 5907–5912. Zellweger, T., et al. (2002). Enhanced radiation sensitivity in prostate cancer by inhibition of the cell survival protein clusterin. Clin. Cancer Res. 8, 3276–3284. Zellweger, T., et al. (2003). Overexpression of the cytoprotective protein clusterin decreases radiosensitivity in the human LNCaP prostate tumour model. BJU Int. 92, 463–469.
Clusterin and Chemoresistance Julie Y. Djeu and Sheng Wei Department of Immunology, H. Lee Moffitt Cancer Center, Tampa, Florida 33612, USA
I. II. III. IV. V. VI.
Introduction Association of sCLU with Chemoresistance Association of sCLU with Multidrug Resistance Association of sCLU with Resistance to Irradiation and Oxidative Stress Association of sCLU with Progressive Tumors Association of sCLU with Resistance to Targeted Therapy (TNF, FAS, TRAIL, HDAC Inhibitors, Herceptin) VII. Induction by Genotoxic and Oxidative Stresses VIII. Mechanism of Induction of and Cytoprotection by CLU IX. Strategies to Blockade CLU for Chemosensitization in Cancer Cells References Resistance to anticancer agents is one of the primary impediments to effective cancer therapy. Chemoresistance occurs not only to clinically established therapeutic agents but also to novel targeted therapeutics. Both intrinsic and acquired mechanisms have been implicated in drug resistance but it remains controversial which mechanisms are responsible that lead to failure of therapy in cancer patients. Recent focus has turned to clusterin (CLU) as a key contributor to chemoresistance to anticancer agents. Its role has been documented in prostate cancer for paclitaxel/docetaxel resistance as well as in renal, breast, and lung tumor cells. Moreover, it is abnormally upregulated in numerous advanced stage and metastatic cancers spanning prostate, renal, bladder, breast, head and neck, colon, cervical, pancreatic, lung carcinomas, melanoma, and lymphoma. It is noteworthy that only the cytoplasmic/secretory clusterin form (sCLU), and not the nuclear form, is expressed in aggressive late stage tumors, which is in line with its antiapoptotic function. Most significantly, sCLU expression is documented to lead to broad-based resistance to other unrelated chemotherapeutic agents such as doxorubicin, cisplatin, etoposide, and camphothecin. Resistance to targeted death-inducing molecules, tumor necrosis factor, Fas and TRAIL, or histone deacetylase inhibitors can also be mediated by sCLU. Expression of sCLU may be an adaptive response to genotoxic and oxidative stresses but this adaptive response could pose a threat in malignant cells being treated with cytotoxic agents by enhancing their survival potential. The actual mechanisms for sCLU induction are unclear but STAT1 is required for its constitutive upregulation in docetaxel-resistant tumor cells. Known as a protein chaperone, sCLU appears to stabilize Ku70/Bax complexes, sequestering Bax from its ability to induce mitochondrial release of cytochrome c that triggers cell apoptosis. Thus, sCLU has a key role in preventing apoptosis induced by cytotoxic agents and has the potential to be targeted for cancer therapy. # 2009 Elsevier Inc.
Advances in CANCER RESEARCH Copyright 2009, Elsevier Inc. All rights reserved.
77
0065-230X/09 $35.00 DOI: 10.1016/S0065-230X(09)05005-2
78
Julie Y. Djeu and Sheng Wei
I. INTRODUCTION Cancer is a daunting disease to cure especially when it is diagnosed at an advanced stage that has a high risk of progression to metastasis. Chemoresistance to both standard anticancer agents and novel targeted therapeutics is a key barrier and remains one of the most pressing issues as the disseminated tumor becomes refractory to the drug, eventually failing all clinically proven drugs available for the tumor type (Borst et al., 2007). Understanding the mechanisms of resistance may therefore lead to improved cancer therapeutics. Intrinsic pathways already existent within the tumor cell may participate in resisting cell death by cytotoxic drugs, yet other new pathways triggered during drug treatment can also play a role in preventing cell death. Despite intense effort to unravel the intrinsic and extrinsic pathways that mediate chemoresistance, it is still unclear which specific process is dominant in tumor cell survival. Clusterin (CLU), in its cytoplasmic secretory form (sCLU), has the unique property in mediating chemoresistance to numerous unrelated anticancer agents and its presence has been observed in a variety of solid tumors and lymphoma (Trougakos et al., 2009a). In this chapter, evidence will be provided that elucidates the role of sCLU in chemoresistance and the potential of targeting sCLU to overcome drug resistance in the clinic will be discussed.
II. ASSOCIATION OF sCLU WITH CHEMORESISTANCE For almost a decade since its discovery, CLU was primarily considered a marker of cell death because of its appearance initially described in castration-induced programmed cell death in the normal rat prostate (Bettuzzi et al., 1989; Leger et al., 1987) and later in other organ systems undergoing massive apoptosis. For example, CLU expression is induced in renal tubule tissue damaged by ligation, embryonic fetal cells in regressing interdigital tissue in the forming forelimbs, and murine bladder tumors undergoing cytotoxic death during cyclophosphamide treatment (Buttyan et al., 1989). However, this view was overturned when a key set of experiments analyzing cell death induced by tumor necrosis factor (TNF) in LNCAP human prostate tumor cells provided the first report that CLU may actually be cytoprotective (Sensibar et al., 1995). The evidence came from the following observations. Upon TNF treatment, CLU did rise but declined prior to observation of cell death. More significantly, transfection with antisense
Clusterin and Chemoresistance
79
CLU to deplete it in LNCAP tumor cells resulted in increased apoptosis and the reverse was seen with CLU overexpression, which endowed the tumor cells with the ability to survive better and resist the cytotoxic effect of TNF (Sensibar et al., 1995). Thus, CLU plays a critical role in protection against TNF-induced cell death. This seminal finding was reproduced in another human prostate tumor cell line, PC3, which constitutively expressed more sCLU than LNCAP and defined the linkage between sCLU and resistance to TNF-induced apoptosis (Sintich et al., 1999). Clearly, purified sCLU added to LNCAP resulted in its ability to resist TNF cytotoxicity while pretreatment of PC3 tumor cells with anti-CLU antibodies sensitized them to TNFmediated death. Thus, extracellular sCLU is responsible for the protective effects against TNF. In addition to TNF resistance, androgen independence can also be attributed to sCLU. Prostate cancer is initially treated with surgery or irradiation for localized disease but long-term disease control can only be achieved with hormonal therapy that suppress androgen receptor signaling (Loblaw et al., 2007; Taplin, 2007). Prostate tumor cells are exquisitely dependent on androgen and its receptor for growth signaling, but therapies directed against this pathway inevitably fail as resistance occurs (Tammela, 2004). In order to understand this phase of progression in prostate cancer, LNCAP human prostate tumor cells, which are androgen-dependent, were employed to investigate the participation of sCLU. LNCAP, which does not constitutively express sCLU, is normally highly sensitive to apoptosis upon androgen withdrawal from the culture medium. However, when stably transfected with the CLU gene, these tumor cells were found to gain the ability to survive and resist androgen ablation (Miyake et al., 2000c). In another androgen-dependent tumor model, Shionogi murine tumors in male mice usually undergo complete regression upon castration but recurrence is common after a month with accompanying androgen-independence. These recurrent tumor cells were found to express sCLU in both its cytoplasmic 60 kDa form and secretory heterodimeric 40 kDa forms which are those established to be critical for cell survival (Miyake et al., 2000c). In vivo administration of antisense CLU oligonucleotides into Shionogi tumor-bearing mice was demonstrated to significantly accelerate tumor regression and substantially delayed the development of androgen-independent tumors. These findings indicate that sCLU is instrumental in acting as an antiapoptotic agent and facilitates survival and growth of tumors that no longer require androgen for their maintenance. Using these two tumor cell lines, sCLU was also implicated in the development of chemoresistance to paclitaxel (Miyake et al., 2000b). Once androgen independence gains a foothold in prostate tumor cells, the drug
80
Julie Y. Djeu and Sheng Wei
of choice is proven to be the taxanes, including paclitaxel (taxol) and docetaxel (Petrylak et al., 2004; Tannock et al., 2004). This is because androgen-independence is associated with the induction of bcl-2, an antiapoptotic protein (McDonnell et al., 1992). Although the taxanes primarily work through microtubule disruption, they also are highly effective in disrupting bcl-2 phosphorylation, required for its antiapoptotic function (Haldar et al., 1997; Scatena et al., 1998). This property led to the clinical use of taxanes, particularly, docetaxel, in the treatment of advanced refractory prostate cancer (Petrylak et al., 2004; Tannock et al., 2004). Nevertheless, it soon became obvious that resistance to paclitaxel or docetaxel can often occur, leading to treatment failure and the spread of metastasis, particularly to bone (Galletti et al., 2007). LNCAP is highly sensitive to paclitaxel in vitro, but upon transfection with sCLU, it was found to withstand such treatment and resisted apoptotic cell death. In vivo in nude mice, parental human LNCAP tumors readily regressed upon castration and administration of paclitaxel, but sCLU-overexpressing LNCAP survived such treatment. Data complementing these observations were also obtained in the sCLU-positive Shionogi tumors. Administration of antisense CLU alone did not cause tumor regression in mice bearing syngeneic Shionogi tumors, but this treatment together with paclitaxel was highly effective. Thus, the conclusion can be reached that sCLU overexpression helps to create a chemoresistant phenotype and sCLU ablation via specific antisense oligonucleotides may be required to chemosensitize resistant tumors to paclitaxel in hormone refractory prostate tumors. To prove this concept, androgen-independent PC3 prostate tumors were tested in vitro and in vivo in nude mice for susceptibility to paclitaxel. Apparently, PC3 tumor cells naturally express sCLU and it was confirmed that blockade of CLU via specific siRNA was first needed before these tumors could respond to paclitaxel to show shrinkage (Miyake et al., 2000a). This finding with PC3 tumor cells was also reproduced by others (Trougakos et al., 2004). The same property of CLU in chemoresistance to paclitaxel was observed in other tumor types, including renal, breast, and lung carcinoma. Pretreatment of Caki-2 human renal carcinoma cells with antisense-CLU greatly enhanced chemosensitivity to paclitaxel in vitro and in vivo in nude mice (Zellweger et al., 2001). Using another model of breast cancer where taxanes are the established choice for management of metastatic disease, antisense-CLU effectively chemosensitized MCF7 and MD-MB231 breast tumor cells to paclitaxel-induced apoptosis (So et al., 2005). Such results were also obtained with human A549 lung carcinoma cells responding to paclitaxel (July et al., 2004). Thus, the potential of targeting sCLU in sensitizing tumor cells to chemotherapy has become an attractive new modality for cancer treatment.
Clusterin and Chemoresistance
81
III. ASSOCIATION OF sCLU WITH MULTIDRUG RESISTANCE To further investigate the extent of chemoresistance conferred by sCLU, investigators began to explore other commonly used chemotherapeutic agents for cancer treatment. It was quickly found that cisplatin sensitivity could be modulated by sCLU. Cisplatin and its derivatives have a broad range of activity in malignant disease, covering testicular, ovarian, small cell and nonsmall cell lung, cervical, head and neck, colorectal and bladder cancers (Martin et al., 2008). They work by binding to DNA and forming DNA adducts leading to intrastand or interstand cross-links, thus disrupting normal DNA structure and impairing proper DNA replication resulting in cell death (Rabik and Dolan, 2007). The development of cisplatin resistance is an unavoidable threat and several important modes of resistance have been uncovered, based on DNA repair enzymes (Martin et al., 2008). In addition to these DNA repair mechanisms, however, resistance by other processes can also be a potent deterrent to cytotoxic death by cisplatin. In examining bladder cancer, it was found that antisense-CLU could enhance chemosensitivity to cisplatin in KoTCC-1 human bladder tumor cells in vitro (Miyake et al., 2001a). As in PC3 prostate tumor cells, antisense CLU alone did not affect KoTCC-1 survival or proliferation in vitro. However, antisense CLU plus cisplatin treatment clearly suppressed tumor cell growth via induction of apoptosis. This method of CLU blockade was also effective in vivo, as systemic administration of CLU-specific antisense oligonucleotides greatly enhanced cisplatin sensitivity of KoTCC-1 in nude mice as compared to control oligonucleotides, leading to retardation in tumor growth. A similar strategy used in Caki-l renal carcinoma cells confirmed that antisense-CLU can provide chemosensitization against cisplatin (Lee et al., 2002). In another detailed analysis of SKOV3 ovarian tumor cells, it was definitely demonstrated that transfection with the nuclear form of CLU induced apoptosis while transfection with the sCLU form in the same ovarian tumor cells promoted survival against cisplatin, thus leaving no doubt as to the dual forms and functions of CLU (Wei et al., 2009). It became clear then that CLU could mediate multidrug resistance to a broad range of unrelated chemotherapeutic agents. For example, CLU overexpression in Mel-Juso melanoma cells was associated with an increase in drug resistance not only to paclitaxel but also to cisplatin and 5 fluorouracil in vitro (Hoeller et al., 2005). In addition, 5182 melanoma cells that constitutively express sCLU, grew progressively in nude mice but this growth could be stemmed by in vivo treatment with antisense-CLU which allowed for chemosensitivity to dacarbazine-induced apoptosis and improved tumor
82
Julie Y. Djeu and Sheng Wei
responses (Hoeller et al., 2005). Similar use of antisense CLU also raised sensitivity to gemcitabine in human bladder koTCC-1 tumor cells in vitro and in vivo (Miyake et al., 2004a). Moreover, human fibrosarcoma cells transfected with sCLU were reportedly resistant to etoposide as well as campthothecin (Zhang et al., 2005). In another study, in order to analyze chemoresistance, doxorubicin-resistant human osteosarcoma cells were developed by culture in increasing levels of the drug (Lourda et al., 2007). These resistant tumor cells were found to show significantly less cell death normally induced by paclitaxel, cisplatin, or camphotecin. In yet another study, it was shown that DU145 and PC3 human prostate tumor cells already cultivated in docetaxel to develop resistance, were also resistant to TRAIL-induced cell death (Sallman et al., 2007). This acquisition to TRAIL resistance was due to expression of sCLU, as shown by restoration of TRAIL sensitivity upon knockdown of CLU gene expression by specific siRNA. It can thus be concluded that multidrug resistance to a wide array of therapeutic agents used for management of cancer can be achieved by upregulation of sCLU in tumor cells.
IV. ASSOCIATION OF sCLU WITH RESISTANCE TO IRRADIATION AND OXIDATIVE STRESS It has become apparent that CLU may also be protective against radiation therapy or oxidative stress. In androgen-dependent LNCAP tumor cells, the overexpression of CLU renders them significantly less sensitive to irradiation in vitro, as compared to nontransfected parental LNCAP tumor cells (Zellweger et al., 2002). On the other hand, antisense CLU-specific oligonucleotides can reduce the expression of CLU in androgen-independent PC3 tumor cells and sensitize them to radiation-induced cell death. Radiation treatment of antisense-CLU transfected PC3 tumor cells induced a higher rate of apoptosis than the same treatment in mismatch control-transfected PC3. Thus, CLU can act as a cell survival protein that mediates radioresistance by prevention of apoptosis. In addition, CLU may also participate in resistance to oxidative stress. Oxidative stress is a major factor associated with the progression of prostate cancer via accumulation of DNA damage. It was shown that transfection of CLU into LNCAP tumor cells can suppress hydrogen peroxide (H2O2)-induced apoptosis (Miyake et al., 2004b). This protection against oxidative stress was also mirrored in human osteosarcoma cells (Trougakos et al., 2004). U-2OS osteosarcoma cells constitutively express high levels of sCLU and show resistance to both H2O2 and doxorubicin. Upon antisense-CLU transfection, these cells now gain sensitivity to both oxidative and genotoxic stress. Conversely, osteosarcoma cells that
Clusterin and Chemoresistance
83
have low CLU expression, when cultured in doxorubicin to develop drugresistant cell lines, acquire resistance to H2O2 and both types of resistance in the same cells was mediated by the upregulation of sCLU (Lourda et al., 2007). It is pertinent that human diploid fibroblasts, transfected with sCLU, can resist H2O2- and ethanol-mediated stress-induced premature senescence (Dumont et al., 2002).
V. ASSOCIATION OF sCLU WITH PROGRESSIVE TUMORS From the above findings, it stands to reason that, in cancer patients, tumors expressing sCLU are likely to display more aggressive behavior and respond less well to chemotherapy or radiation therapy. To address whether CLU is spontaneously upregulated in cancer, a wide distribution of human tumor biopsies were analyzed by numerous laboratories. One of the earliest observations of CLU overexpression was made in human gliomas where analysis of differential gene display between benign and malignant tissues identified a markedly increased level of CLU mRNA expression in astrocytomas and glioblastomas (Danik et al., 1991). This was confirmed by in situ hybridization to detect CLU mRNA in the tissues. As a control, endothelial cells in the vasculature were analyzed and were found to be CLU negative. In renal carcinoma, tumor tissues were reported to contain threefold more CLU-specific mRNA than the adjacent normal tissue (Parczyk et al., 1994). Following these observations, a number of other tumor types were analyzed. With the availability of antibodies against CLU, subsequent reports also focused on CLU protein expression and its localization within the cell. In immunohistochemical analysis of CLU protein in 40 human prostate tumor specimens, it became clear that CLU is steadily increased as the grade of tumor rises, reflected by the Gleason Score, and it is restricted to the cytoplasm with little presence in the nucleus (Steinberg et al., 1997). Normal prostate tissue had no CLU staining while benign hyperplastic tissue from the prostate showed a weak sCLU staining. Its staining intensified with higher forms of malignancy. This provided the first indication that protection from apoptosis by CLU may account in part for biologically aggressive behavior. Others have since confirmed that sCLU is present in prostate tumor tissues from radical prostatectomy of cancer patients and its level is even higher in androgen-independent tumors in a sample size of 128 (July et al., 2002). Another study with a sample size of 172 prostate tumor archival material also showed that sCLU correlates with Gleason Score (Miyake et al., 2006).
84
Julie Y. Djeu and Sheng Wei
Similar observations were repeated in a comprehensive study of a large set of human breast cancer specimens, including 34 benign, 8 atypical hyperplastic, 18 in situ carcinoma, 54 invasive carcinoma, and 8 metastatic breast archival specimens (Redondo et al., 2000). Analysis of 40 nonneoplastic glandular epithelia was included as a control and none of them expressed CLU. On the other hand, CLU was upregulated primarily in the cytoplasm and was closely associated with a corresponding increase in tumor progression, developing from normal tissue toward premalignant and advancing onto the malignant phenotype. In fact, highest CLU expression was in the lymph node metastasis, thus linking sCLU as a phenotypic determinant of the aggressive nature of breast cancer. This early work was confirmed by another report in a breast tumor tissue microarray that represents 379 samples (So et al., 2005). Upregulation of sCLU or its mRNA is now reported in tumor specimens representing ovarian (Hough et al., 2001; Xie et al., 2005), renal (Zellweger et al., 2001), colon (Pucci et al., 2004), lung (July et al., 2004), melanoma (Hoeller et al., 2005), pancreas (Mourra et al., 2007), and cervical cancer (Watari et al., 2008). In the colon, it is clearly demonstrated the transition of CLU detection from the nucleus to the cytoplasm reflects a growing aggression of the malignancy (Pucci et al., 2004). It has thus become a universal observation that sCLU can be detected in most solid tumors and its level, particularly in the cytoplasm, corresponds with progressing stages of the disease. It is interesting that sCLU is absent in most leukemias and lymphomas, and is only detected in anaplastic large cell lymphoma (Wellmann et al., 2000). Analysis of 198 well-characterized lymphomas, including T cell, B cell, Hodgkin lymphomas, as well as 31 established leukemia/lymphoma cell lines demonstrated that only one category of lymphoma, anaplastic large cell lymphoma, overexpressed the CLU gene and immunostaining localized it to the cytoplasm.
VI. ASSOCIATION OF sCLU WITH RESISTANCE TO TARGETED THERAPY (TNF, FAS, TRAIL, HDAC INHIBITORS, HERCEPTIN) In addition to drug resistance to clinically relevant chemotherapeutic agents and ionizing radiation, sCLU also has the power to protect against reagents that specifically target molecules involved in cell death. A set of TNF-related proteins released or expressed by both immune and nonimmune cells, such as TNF, FAS, and TRAIL, has the unique ability to trigger cell death by binding its specific receptors on target cells and inducing a common signal pathway that leads to caspase activation and apoptosis
Clusterin and Chemoresistance
85
(Papenfuss et al., 2008). This property has led to the pursuit of these molecules as anticancer agents. However, sCLU has the potency to block the function of these death receptors. In fact, the antiapoptotic function of sCLU was actually first uncovered by the analysis of TNF-induced cell death in LNCAP prostate tumor cells, as discussed in the beginning of this chapter (Sensibar et al., 1995). This original observation suggested that CLU might also protect against other death-inducing molecules of the same family. Thus, another study took up Fas as the targeting agent and reported that overexpression of CLU in a human renal carcinoma cell line, ACHN, indeed, prevented apoptosis that is normally achieved with Fas ligation through anti-Fas antibody (Miyake et al., 2001b). In addition, TRAIL resistance in human DU145 and PC3 tumor cells was traced to sCLU expression (Sallman et al., 2007). Another surface receptor whose targeting has produced remarkable clinical responses is the Her2 growth factor receptor. This receptor is displayed at an abnormally high level in various cancers including breast and ovary and Herceptin/Trastuzumab which is a specific humanized monoclonal antibody targeting the Her2 receptor is highly effective in treatment of cancers expressing this receptor (Nahta et al., 2006). However, resistance also develops against it. One study has indicated that use of antisense CLU prior to treatment with Herceptin can enhance the sensitivity to this drug, suggesting that CLU may also play a negative role in this targeted therapy (Biroccio et al., 2005). In addition, histone deacetylase (HDAC) inhibitors are the new generation chemotherapeutic agents that modify DNA-related gene transcription, and are proving to have efficacy against a variety of cancers (Kelly and Marks, 2005). However, CLU can interfere with its efficacy. Human breast tumor cells, MDA-MB231 and MDA-MB468, which were highly resistant to HDAC inhibitors, upon transfection with siRNA-CLU, became sensitive to cell death induced by TSA, a potent HDAC inhibitor (Liu et al., 2009). Overall, it is now evident that sCLU is a powerful mediator of cell survival that can block the effects of almost all known therapeutic agents. Conversely, its suppression can sensitize tumor cells against these reagents. Thus, it is important to elucidate the mechanism of its induction for expression.
VII. INDUCTION BY GENOTOXIC AND OXIDATIVE STRESSES One of the earliest studies on TNF-induced cell death provided the first demonstration that TNF induces it, not to induce apoptosis as was widely believed at that time, but to prevent cell death as a cytoprotective reaction
86
Julie Y. Djeu and Sheng Wei
against a toxic insult (Sensibar et al., 1995). This seminal report led to numerous reassessments of tumor cells responding to various cytotoxic agents, and it was confirmed that all toxic agents triggered the expression of CLU. Because CLU was first discovered as a survival protein in prostate tumor cells, a push was made to investigate if androgen ablation, irradiation, and paclitaxel treatment which are common strategies for treatment of this cancer, upregulated CLU. All of these modalities were found to readily induce sCLU expression in human prostate tumor cells. In terms of hormonal ablation, androgen ablation linkage with CLU induction in prostate tumor cell lines (Cochrane et al., 2007) were corroborated by analysis of human prostate tumor specimens taken from patients after hormonal therapy in comparison to those without treatment (Gleave et al., 2001). More significantly, needle biopsies of prostate tumors obtained prior to neoadjuvant hormonal therapy were compared to the radical prostectomy specimens from the same patients after varying lengths of treatment (July et al., 2002). It was clearly shown that the levels of sCLU in the treated samples were markedly higher than those before treatment, thus suggesting that sCLU expression is an adaptive response to provide cytoprotection against the anticancer regimen. In addition to androgen withdrawal in prostate cancer, estrogen withdrawal or paclitaxel treatment in breast tumor cells was also found to induce sCLU (So et al., 2005). Such a response was also elicited by docetaxel treatment of prostate tumor cells (Patterson et al., 2006). Irradiation had the same effect as reported in both prostate and breast tumor cells (Criswell et al., 2003; Zellweger et al., 2002). Other chemotherapeutic agents, such as cisplatin (Lee et al., 2002; Miyake et al., 2001a), doxorubicin (Lourda et al., 2007; Trougakos et al., 2004) as well as specific targeting agents such as Herceptin (Biroccio et al., 2005) and HDAC inhibitors (Liu et al., 2009; Ranney et al., 2007) were also identified to be capable of inducing sCLU. Lastly, sCLU is also a responsive gene to oxidative stress, under H2O2 or ethanol treatment (Trougakos et al., 2004). What is becoming apparent, then, is that sCLU is an adaptive response to not only genotoxic and cytotoxic stress but also oxidative stress. It is thus a unique survival protein that is called upon within a cell to withstand any damaging insult.
VIII. MECHANISM OF INDUCTION OF AND CYTOPROTECTION BY CLU Given that the true biological nature of CLU is beginning to be better understood and the assignment of nCLU for proapoptotic function and sCLU for prosurvival function has finally gained consensus among the vested investigators (Trougakos et al., 2009a), the molecular mechanisms for its
Clusterin and Chemoresistance
87
production and function are still relatively unknown. In terms of its induction, its gene has been cloned and analysis of its promoter region has revealed several transcription regulators reported to be involved in gene transcription. Transcription factors that have been shown to interact with the CLU promoter and regulate its function include Egr-1 (Criswell et al., 2005), AP-1 (Jin and Howe, 1997, 1999), Heat Shock Factor 1/2 (Loison et al., 2006), Cdx1 (Suh et al., 2001), B-MYB (Cervellera et al., 2000), and c-MYC (Thomas-Tikhonenko et al., 2004). In terms of STAT-related transcription factors, STAT1 but not STAT3 is identified to be involved in CLU gene expression (Patterson et al., 2006). In addition, p53 and Nkx3.1 tumor suppressor genes appear to negatively regulate CLU gene expression (Criswell et al., 2003; Song et al., 2009) as does the b-catenin/Wnt pathway (Schepeler et al., 2007). How does sCLU mediate its cytoprotective function? It is apparently linked to its ability to bind Ku70/Bax complexes. A seminal report demonstrated that sCLU specifically binds activated Bax sequestering it from translocation to the mitochondria to induce cytochrome c release and apoptosis (Zhang et al., 2005). Others have corroborated this finding and demonstrated that sCLU binds and stabilizes the Ku70/Bax complex in the cytoplasm, retaining it as a complex and preventing its release (Pucci et al., 2009; Trougakos et al., 2009b). Thus, it is of significance that its mechanism of action is linked to the Bcl-2 family of proteins that are potent in controlling the fate of a cell. In this case, sCLU clearly associated with a specific member, Bax. It is also of interest that the mechanism of action of nCLU is to bind the Ku70/Bax complex (Sawada et al., 2003; Yang et al., 2000). Although Ku70 is established to bind Bax in the cytoplasm to prevent its activation (Sawada et al., 2003), it is also found in the nucleus where Ku70 was originally discovered as a critical component of the DNA repair machinery (Wang et al., 1998). Thus, the proapoptotic function of nCLU might be to sequester, in this case, Ku70, which is critical for repair in DNA double-strand breaks in the nucleus. It is intriguing that sCLU works in the cytoplasm by stabilizing Bax/Ku70 to sequester Bax from inducing apoptosis, while nCLU works in the nucleus by stabilizing Bax/Ku70 to sequester Ku70 from its DNA repair function. Another pathway by which sCLU might act is via NF-B. Being a protein stabilizer, sCLU apparently can also stabilize iKb, thus preventing its degradation which is needed to release the p50/p65 NF-B heterodimer for entry into the nucleus to act as a transcription factor (Santilli et al., 2003; Takase et al., 2008). Another means by which sCLU can affect cell survival is via its receptor. Interestingly, sCLU has been shown to bind its cell surface receptor, megalin, in a rat prostate cell line and induce AKT activation which then can phosphorylate Bad, causing a decrease in cytochrome c release, thus favoring cell survival (Ammar and Closset, 2008).
88
Julie Y. Djeu and Sheng Wei
IX. STRATEGIES TO BLOCKADE CLU FOR CHEMOSENSITIZATION IN CANCER CELLS It is clear that resistance to anticancer drugs is a major obstacle in the cure of cancer patients. Multidrug resistance often develops against clinically useful chemotherapeutics and it is also becoming evident that resistance against newer targeted therapeutics can occur. A wide spectrum of intrinsic and extrinsic mechanisms has been proposed for the development of multidrug resistance but it is difficult and time-consuming to attack each mechanism to prevent drug resistance. The development of strategies to circumvent drug resistance poses a frustrating challenge. The emerging realization that sCLU is common to many advanced cancer types and that it can mount resistance to a large array of chemotherapeutic compounds with unrelated mechanisms has brought the impetus to target sCLU to treat multidrug resistance. Antisense technology has facilitated such targeting in the clinic. A second generation phosphorothioate antisense oligonucleotide complementary to the CLU mRNA translation initiation site has been developed which showed high efficiency in vitro in blocking CLU expression and in chemosensitization to several drugs, including paclitaxel (Miyake et al., 2000a),cisplatin (Miyake et al., 2001a) and gemcitabine (Miyake et al., 2004a). This antisense-CLU reagent, OGX-011, has already been entered into Phase I clinical trials together with docetaxel administration in patients with advanced stages of cancer, including prostate, ovarian, renal, lung, bladder, and breast cancers (Chi et al., 2005, 2008). Serum sCLU levels were effectively reduced without serious toxicity. Of the 32 patients who were evaluable, 2 with hormone-refractory prostate cancer had a partial response, 11 had stable disease up to 6 months, and 1 breast cancer patient had a complete response (Chi et al., 2008). Of 14 hormone-refractory prostate cancer patients treated, 3 showed a decline in PSA. Based on these initial successes, a Phase II trial has been instituted. Another independent study of OGX-011 Phase II trial on 15 patients with measurable metastatic breast cancer has been published and again confirmed that the toxicity was no greater than docetaxel alone (Chia et al., 2009). However, although some clinical responses were seen, they were not beyond those expected with docetaxel alone. It is too soon to predict the efficacy of such antisense therapy in cancer patients but the promise of disrupting sCLU to achieve better chemosensitivity to established drugs is still an attractive goal to pursue. Instead of direct targeting of sCLU, focusing on compounds that can interrupt sCLU gene expression might also be a useful strategy. This area, however, remains unexplored and must await better knowledge of CLU
Clusterin and Chemoresistance
89
mechanism of induction and expression, and the availability of compounds that can target the identified processes. In summary, sCLU has emerged as a potent adaptive response to cell stress, either induced by cytotoxic agents, ionizing radiation, or targeted therapy. It is part of the normal cell response to stress as a cytoprotective reaction but in malignant cells, this property works to the advantage of the tumor. Because it confers survival advantage to cancer cells and is readily induced by therapeutic agents, sCLU targeting as a means to chemosensitization toward clinically established drugs could be a potent strategy to overcome drug resistance.
REFERENCES Ammar, H., and Closset, J. L. (2008). Clusterin activates survival through the phosphatidylinositol 3-kinase/Akt pathway. J. Biol. Chem. 283, 12851–12861. Bettuzzi, S., Hiipakka, R. A., Gilna, P., and Liao, S. T. (1989). Identification of an androgenrepressed mRNA in rat ventral prostate as coding for sulphated glycoprotein 2 by cDNA cloning and sequence analysis. Biochem. J. 257, 293–296. Biroccio, A., et al. (2005). Antisense clusterin oligodeoxynucleotides increase the response of HER-2 gene amplified breast cancer cells to Trastuzumab. J. Cell. Physiol. 204, 463–469. Borst, P., et al. (2007). What makes tumors multidrug resistant? Cell Cycle 6, 2782–2787. Buttyan, R., et al. (1989). Induction of the TRPM-2 gene in cells undergoing programmed death. Mol. Cell. Biol. 9, 3473–3481. Cervellera, M., et al. (2000). Direct transactivation of the anti-apoptotic gene apolipoprotein J (clusterin) by B-MYB. J. Biol. Chem. 275, 21055–21060. Chi, K. N., et al. (2005). A phase I pharmacokinetic and pharmacodynamic study of OGX-011, a 20 -methoxyethyl antisense oligonucleotide to clusterin, in patients with localized prostate cancer. J. Natl. Cancer Inst. 97, 1287–1296. Chi, K. N., et al. (2008). Multicenter phase II study of combined neoadjuvant docetaxel and hormone therapy before radical prostatectomy for patients with high risk localized prostate cancer. J. Urol. 180, 565–570, discussion 570. Chia, S., et al. (2009). Phase II trial of OGX-011 in combination with docetaxel in metastatic breast cancer. Clin. Cancer Res. 15, 708–713. Cochrane, D. R., et al. (2007). Differential regulation of clusterin and its isoforms by androgens in prostate cells. J. Biol. Chem. 282, 2278–2287. Criswell, T., et al. (2003). Repression of IR-inducible clusterin expression by the p53 tumor suppressor protein. Cancer Biol. Ther. 2, 372–380. Criswell, T., et al. (2005). Delayed activation of insulin-like growth factor-1 receptor/Src/ MAPK/Egr-1 signaling regulates clusterin expression, a pro-survival factor. J. Biol. Chem. 280, 14212–14221. Danik, M., et al. (1991). Human gliomas and epileptic foci express high levels of a mRNA related to rat testicular sulfated glycoprotein 2, a purported marker of cell death. Proc. Natl. Acad. Sci. USA 88, 8577–8581. Dumont, P., et al. (2002). Overexpression of apolipoprotein J in human fibroblasts protects against cytotoxicity and premature senescence induced by ethanol and tert-butylhydroperoxide. Cell Stress Chaperones 7, 23–35.
90
Julie Y. Djeu and Sheng Wei
Galletti, E., et al. (2007). Paclitaxel and docetaxel resistance: Molecular mechanisms and development of new generation taxanes. ChemMedChem 2, 920–942. Gleave, M. E., et al. (2001). Use of antisense oligonucleotides targeting the antiapoptotic gene, clusterin/testosterone-repressed prostate message 2, to enhance androgen sensitivity and chemosensitivity in prostate cancer. Urology 58, 39–49. Haldar, S., et al. (1997). Bcl2 is the guardian of microtubule integrity. Cancer Res. 57, 229–233. Hoeller, C., et al. (2005). Clusterin regulates drug-resistance in melanoma cells. J. Invest. Dermatol. 124, 1300–1307. Hough, C. D., et al. (2001). Coordinately up-regulated genes in ovarian cancer. Cancer Res. 61, 3869–3876. Jin, G., and Howe, P. H. (1997). Regulation of clusterin gene expression by transforming growth factor beta. J. Biol. Chem. 272, 26620–26626. Jin, G., and Howe, P. H. (1999). Transforming growth factor beta regulates clusterin gene expression via modulation of transcription factor c-Fos. Eur. J. Biochem. 263, 534–542. July, L. V., et al. (2002). Clusterin expression is significantly enhanced in prostate cancer cells following androgen withdrawal therapy. Prostate 50, 179–188. July, L. V., et al. (2004). Nucleotide-based therapies targeting clusterin chemosensitize human lung adenocarcinoma cells both in vitro and in vivo. Mol. Cancer Ther. 3, 223–232. Kelly, W. K., and Marks, P. A. (2005). Drug insight: Histone deacetylase inhibitors—development of the new targeted anticancer agent suberoylanilide hydroxamic acid. Nat. Clin. Pract. Oncol. 2, 150–157. Lee, C. H., et al. (2002). Suppression of clusterin expression enhanced cisplatin-induced cytotoxicity on renal cell carcinoma cells. Urology 60, 516–520. Leger, J. G., et al. (1987). Characterization and cloning of androgen-repressed mRNAs from rat ventral prostate. Biochem. Biophys. Res. Commun. 147, 196–203. Liu, T., et al. (2009). Over-expression of clusterin is a resistance factor to the anti-cancer effect of histone deacetylase inhibitors. Eur. J. Cancer 45, 1846–1854. Loblaw, D. A., et al. (2007). Initial hormonal management of androgen-sensitive metastatic, recurrent, or progressive prostate cancer: 2006 update of an American Society of Clinical Oncology practice guideline. J. Clin. Oncol. 25, 1596–1605. Loison, F., et al. (2006). Up-regulation of the clusterin gene after proteotoxic stress: Implication of HSF1-HSF2 heterocomplexes. Biochem. J. 395, 223–231. Lourda, M., et al. (2007). Development of resistance to chemotherapeutic drugs in human osteosarcoma cell lines largely depends on up-regulation of Clusterin/Apolipoprotein J. Int. J. Cancer 120, 611–622. Martin, L. P., et al. (2008). Platinum resistance: The role of DNA repair pathways. Clin. Cancer Res. 14, 1291–1295. McDonnell, T. J., et al. (1992). Expression of the protooncogene bcl-2 in the prostate and its association with emergence of androgen-independent prostate cancer. Cancer Res. 52, 6940–6944. Miyake, H., et al. (2000a). Antisense TRPM-2 oligodeoxynucleotides chemosensitize human androgen-independent PC-3 prostate cancer cells both in vitro and in vivo. Clin. Cancer Res. 6, 1655–1663. Miyake, H., et al. (2000b). Acquisition of chemoresistant phenotype by overexpression of the antiapoptotic gene testosterone-repressed prostate message-2 in prostate cancer xenograft models. Cancer Res. 60, 2547–2554. Miyake, H., et al. (2000c). Testosterone-repressed prostate message-2 is an antiapoptotic gene involved in progression to androgen independence in prostate cancer. Cancer Res. 60, 170–176.
Clusterin and Chemoresistance
91
Miyake, H., et al. (2001a). Synergistic chemsensitization and inhibition of tumor growth and metastasis by the antisense oligodeoxynucleotide targeting clusterin gene in a human bladder cancer model. Clin. Cancer Res. 7, 4245–4252. Miyake, H., et al. (2001b). Acquisition of resistance to Fas-mediated apoptosis by overexpression of clusterin in human renal-cell carcinoma cells. Mol. Urol. 5, 105–111. Miyake, H., et al. (2004a). Synergistic antitumor activity by combined treatment with gemcitabine and antisense oligodeoxynucleotide targeting clusterin gene in an intravesical administration model against human bladder cancer kotcc-1 cells. J. Urol. 171, 2477–2481. Miyake, H., et al. (2004b). Protection of androgen-dependent human prostate cancer cells from oxidative stress-induced DNA damage by overexpression of clusterin and its modulation by androgen. Prostate 61, 318–323. Miyake, H., et al. (2006). Expression of clusterin in prostate cancer correlates with Gleason score but not with prognosis in patients undergoing radical prostatectomy without neoadjuvant hormonal therapy. Urology 68, 609–614. Mourra, N., et al. (2007). Clusterin is highly expressed in pancreatic endocrine tumours but not in solid pseudopapillary tumours. Histopathology 50, 331–337. Nahta, R., et al. (2006). Mechanisms of disease: Understanding resistance to HER2-targeted therapy in human breast cancer. Nat. Clin. Pract. Oncol. 3, 269–280. Papenfuss, K., et al. (2008). Death receptors as targets for anti-cancer therapy. J. Cell. Mol. Med. 12, 2566–2585. Parczyk, K., et al. (1994). Gp80 (clusterin; TRPM-2) mRNA level is enhanced in human renal clear cell carcinomas. J. Cancer Res. Clin. Oncol. 120, 186–188. Patterson, S. G., et al. (2006). Novel role of Stat1 in the development of docetaxel resistance in prostate tumor cells. Oncogene 25, 6113–6122. Petrylak, D. P., et al. (2004). Docetaxel and estramustine compared with mitoxantrone and prednisone for advanced refractory prostate cancer. N. Engl. J. Med. 351, 1513–1520. Pucci, S., et al. (2004). Modulation of different clusterin isoforms in human colon tumorigenesis. Oncogene 23, 2298–2304. Pucci, S., et al. (2009). Interleukin-6 affects cell death escaping mechanisms acting on Bax– Ku70–Clusterin interactions in human colon cancer progression. Cell Cycle 8, 473–481. Rabik, C. A., and Dolan, M. E. (2007). Molecular mechanisms of resistance and toxicity associated with platinating agents. Cancer Treat. Rev. 33, 9–23. Ranney, M. K., et al. (2007). Multiple pathways regulating the anti-apoptotic protein clusterin in breast cancer. Biochim. Biophys. Acta 1772, 1103–1111. Redondo, M., et al. (2000). Overexpression of clusterin in human breast carcinoma. Am. J. Pathol. 157, 393–399. Sallman, D. A., et al. (2007). Clusterin mediates TRAIL resistance in prostate tumor cells. Mol. Cancer Ther. 6, 2938–2947. Santilli, G., et al. (2003). Essential requirement of apolipoprotein J (clusterin) signaling for IkappaB expression and regulation of NF-kappaB activity. J. Biol. Chem. 278, 38214–38219. Sawada, M., et al. (2003). Ku70 suppresses the apoptotic translocation of Bax to mitochondria. Nat. Cell Biol. 5, 320–329. Scatena, C. D., et al. (1998). Mitotic phosphorylation of Bcl-2 during normal cell cycle progression and Taxol-induced growth arrest. J. Biol. Chem. 273, 30777–30784. Schepeler, T., et al. (2007). Clusterin expression can be modulated by changes in TCF1-mediated Wnt signaling. J. Mol. Signal. 2, 6. Sensibar, J. A., et al. (1995). Prevention of cell death induced by tumor necrosis factor alpha in LNCaP cells by overexpression of sulfated glycoprotein-2 (clusterin). Cancer Res. 55, 2431–2437.
92
Julie Y. Djeu and Sheng Wei
Sintich, S. M., et al. (1999). Cytotoxic sensitivity to tumor necrosis factor-alpha in PC3 and LNCaP prostatic cancer cells is regulated by extracellular levels of SGP-2 (clusterin). Prostate 39, 87–93. So, A., et al. (2005). Knockdown of the cytoprotective chaperone, clusterin, chemosensitizes human breast cancer cells both in vitro and in vivo. Mol. Cancer Ther. 4, 1837–1849. Song, H., et al. (2009). Loss of Nkx3.1 leads to the activation of discrete downstream target genes during prostate tumorigenesis. Oncogene 28, 3307–3319. Steinberg, J., et al. (1997). Intracellular levels of SGP-2 (Clusterin) correlate with tumor grade in prostate cancer. Clin. Cancer Res. 3, 1707–1711. Suh, E., et al. (2001). Clusterin gene transcription is activated by caudal-related homeobox genes in intestinal epithelium. Am. J. Physiol. Gastrointest. Liver Physiol. 280, G149–G156. Takase, O., et al. (2008). Inhibition of NF-kappaB-dependent Bcl-xL expression by clusterin promotes albumin-induced tubular cell apoptosis. Kidney Int. 73, 567–577. Tammela, T. (2004). Endocrine treatment of prostate cancer. J. Steroid Biochem. Mol. Biol. 92, 287–295. Tannock, I. F., et al. (2004). Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N. Engl. J. Med. 351, 1502–1512. Taplin, M. E. (2007). Drug insight: Role of the androgen receptor in the development and progression of prostate cancer. Nat. Clin. Pract. Oncol. 4, 236–244. Thomas-Tikhonenko, A., et al. (2004). Myc-transformed epithelial cells down-regulate clusterin, which inhibits their growth in vitro and carcinogenesis in vivo. Cancer Res. 64, 3126–3136. Trougakos, I. P., et al. (2004). Silencing expression of the clusterin/apolipoprotein j gene in human cancer cells using small interfering RNA induces spontaneous apoptosis, reduced growth ability, and cell sensitization to genotoxic and oxidative stress. Cancer Res. 64, 1834–1842. Trougakos, I. P., et al. (2009a). Advances and challenges in basic and translational research on clusterin. Cancer Res. 69, 403–406. Trougakos, I. P., et al. (2009b). Intracellular clusterin inhibits mitochondrial apoptosis by suppressing p53-activating stress signals and stabilizing the cytosolic Ku70–Bax protein complex. Clin. Cancer Res. 15, 48–59. Wang, J., et al. (1998). A model for Ku heterodimer assembly and interaction with DNA. Implications for the function of Ku antigen. J. Biol. Chem. 273, 31068–31074. Watari, H., et al. (2008). Clusterin expression predicts survival of invasive cervical cancer patients treated with radical hysterectomy and systematic lymphadenectomy. Gynecol. Oncol. 108, 527–532. Wei, L., et al. (2009). Roles of clusterin in progression, chemoresistance and metastasis of human ovarian cancer. Int. J. Cancer 125, 791–806. Wellmann, A., et al. (2000). Detection of differentially expressed genes in lymphomas using cDNA arrays: identification of clusterin as a new diagnostic marker for anaplastic large-cell lymphomas. Blood 96, 398–404. Xie, D., et al. (2005). Up-regulated expression of cytoplasmic clusterin in human ovarian carcinoma. Cancer 103, 277–283. Yang, C. R., et al. (2000). Nuclear clusterin/XIP8, an x-ray-induced Ku70-binding protein that signals cell death. Proc. Natl. Acad. Sci. USA 97, 5907–5912. Zellweger, T., et al. (2001). Chemosensitization of human renal cell cancer using antisense oligonucleotides targeting the antiapoptotic gene clusterin. Neoplasia 3, 360–367. Zellweger, T., et al. (2002). Enhanced radiation sensitivity in prostate cancer by inhibition of the cell survival protein clusterin. Clin. Cancer Res. 8, 3276–3284. Zhang, H., et al. (2005). Clusterin inhibits apoptosis by interacting with activated Bax. Nat. Cell Biol. 7, 909–915.
CLU “In and Out”: Looking for a Link Sabina Pucci,* P. Mazzarelli,* C. Nucci,{ F. Ricci,{ and L. G. Spagnoli*,z {
*Department of Biopathology, Institute of Anatomic Pathology, University of Rome “Tor Vergata”, 00133 Rome, Italy Department of Biopathology, Section of Ophthalmology, University of Rome “Tor Vergata”, 00133 Rome, Italy z IRCCS San Raffaele Pisana, Rome, Italy
I. Introduction II. Normal and Cancer Microecosystem III. Microenvironment Effects on CLU Expression (the “IN” Effect) A. TGF-, the Primum Movens B. Hypoxia Inducible Factor: Altered of IL-6 and VEGF-A165 Expression in the Microenvironment C. IL-6 D. VEGF-A E. IL-6, VEGF-A165, and Cell Death Escape in Colon Cancer Cell: Acting on sCLU Induction IV. sCLU Effects on Microenvironment (the “OUT” Effect): Up- and Downstream Signals V. Conclusions and Future Perspectives References Cancer cells need to interact synergistically with their surrounding microenvironment to form a neoplasm and to progress further to colonize distant organs. The microenvironment can exert profound epigenetic effects on cells through cell-derived interactions between cells, or through cell-derived factors deposited into the microenvironment. Tumor progression implies immune-escaping and triggers several processes that synergistically induce a cooperation among transformed and stromal cells, that compete for space and resources such as oxygen and nutrients. Therefore, the extra cellular milieu and tissue microenviroment heterotypic interactions cooperate to promote tumor growth, angiogenesis, and cancer cell motility, through elevated secretion of pleiotropic cytokines and soluble factors. Clusterin (CLU), widely viewed as an enigmatic protein represents one of the numerous cellular factors sharing the intracellular information with the microenvironment and it has also a systemic diffusion, tightly joining the “In and the Out” of the cell with a still debated variety of antagonistic functions. The multiplicity of names for CLU is an indication of the complexity of the problem and could reflect, on one hand its multifunctionality, or alternatively could mask a commonality of function. The posited role for CLU, further supported as a cytoprotective prosurvival chaperone-like molecule, seems compelling, in contrast its tumor suppressor function, as a guide of the guardians of the genome (DNA-repair proteins Ku70/80, Bax cell death inducer), could really reflect the balanced expression of its different forms, most certainly depending on the Advances in CANCER RESEARCH Copyright 2009, Elsevier Inc. All rights reserved.
93
0065-230X/09 $35.00 DOI: 10.1016/S0065-230X(09)05006-4
Sabina Pucci et al.
94
intra- and extracellular microenvironment cross talk. The complicated balance of cytokines network and the regulation of CLU forms production in cancer and stromal cells undoubtedly represent a potential link among adaptative responses, genomic stability, and bystander effect after oxidative stresses and damage. This review focuses on the tumor–microenvironment interactions strictly involved in controlling local cancer growth, invasion, and distant metastases that play a decisive role in the regulation of CLU different forms expression and release. In addition, we focus on the pleiotropic action of the extracellular form of this protein, sCLU, that may play a crucial role in redirecting stromal changes, altering intercellular communications binding cell surface receptors and contributing to influence the secretion of chemokines in paracrine and autocrine fashion. Further elucidation of CLU functions inside and ouside (“in and out”) of cancer cell are warranted for a deeper understanding of the interplay between tumor and stroma, suggesting new therapeutic cotargeting strategies. # 2009 Elsevier Inc.
I. INTRODUCTION Cooperation through the sharing of diffusible factors of tumor microenvironment and the redirection of some specific guardian pathways raises new questions about tumorigenesis and has implication on designing new therapeutic approaches. Tissue microenvironment strongly influences tumorigenesis and neovascularization, redirecting some pathways versus a persisting prosurvival state. The tumor microenvironment consisting of tumoral, immune, stromal, and inflammatory cells, all of which produce cytokines, growth factors, and adhesion molecules, may promote tumor progression and metastases. Among the tumor-associated cells are endothelial cells and pericytes that together form the neovasculature, which supplies tumor cells with nutrients and oxygen. In addition, mesenchymal cells of the tumoral stroma are already apparent in tumors at its earliest stages suggesting a coevolution between stromal and cancer cells that leads to clonal expansions. This cancer stromal coevolution leads to support additional microenvironmental changes to foster tumor growth. Given the striking similarities between the phenotypes of aggressive tumors and embryonic stem cells particularly with respect to specific signaling pathways underlying their intriguing plasticity is not surprising that the gradients of growth factors leads long distant final target. The cytokines produced by cancer cells function to create optimal growth conditions within the tumor microenvironment, while cytokines secreted by stromal cells may influence behavior of malignant cells (Lewis et al., 2006). While the normal development of tissue and organs in the embryo is coordinated by a complex equilibrium between positively and negatively acting cues and signals given by the microenvironment, the dysregulated expression of potent embryonic morphogens in some aggressive cancer cell type in absence of normal negative regulators has demonstrated a strong induction of uncontrolled proliferation, increased survival, dedifferentiation, and plasticity that promotes malignant
CLU “In and Out”: Looking for a Link
95
transformation and contributes to metastasis. These effects could be transient, as seen in signaling pathways regulating cellular proliferation, or associated to more stable events, such as cell fate determination or differentiation. In tumor microenvironment as in a microecosystem, all cell population intimately interacts with one another and plays an important role in inflammatory and proangiogenic processes, promoting tumor cell proliferation. Interestingly, there is a tightened association between chronic inflammation and tumor insurgence or progression. Chronic inflammation is mediated via a persistent and continuous stimulus and the resulting prolonged exposure to inflammatory cytokines has the potential to promote tumor growth through the induction of angiogenesis, DNA damage, events that favor tumor invasion. Inflammatory and stromal cells communicate via cytokines and chemokines. Cytokines regulate growth, trafficking, signaling, and differentiation of both stromal and tumor cells. It has been observed (Soufla et al., 2006) that the different steps of cancer progression are signed by a gradient of differential cytokine cocktails that, in turn, epigenetically activate different pathways, either in cancer and in stromal cells, influencing primary genes transcription. The epigenetic influence of the microenvironment on tumoral cells is curiously observed, not only on the induction of trans-signaling cascade pathways affecting gene acetylation and transcription, but also on the fine regulation of tumor-suppressor microRNAs and also on the alternative RNA splicing pattern, leading to the expression of specific protein isoforms. Alternative splicing is a major mechanism for modulating the expression of cellular and viral genes and enables a single gene to increase its coding capacity. An explicative example in cancer is the changes in the microenviroment mediators induced by hypoxia, a hallmark of cancer progression. In fact, hypoxia induces preferential expression of one isoform of vascular endothelial growth factor (VEGF-A165) (Panutsopulos et al., 2003). The different products of VEGF-A, (120, 183, 165a, 165b) are produced by alternative splicing between exon 5 and 8 of the VEGF-A gene, are differently conditioned by microenvironment, as a result of cytokines and growth factors influence. The differential splicing in cancer and stromal cells is also strongly connected with the preferential clusterin (CLU) forms production, and also in this case it could be influenced by external cytokines, growth factors, and environmental condition (pH, ROS, Ca2þ,etc.), that guide the production of the predominant form needed in a particular district depending on the specific tumoral context. It has been demonstrated that the different CLU forms production is influenced by different soluble factors activating different signaling cascade (Criswell et al., 2005; Patterson et al., 2006; Pucci et al., 2004a,b, 2008, 2009) which in turn may be involved in orchestrating the differential step
96
Sabina Pucci et al.
of carcinogenesis, inducing a prosurvival response, membrane remodeling, cell–cell adhesion, and cancer cell motility. Both forms of CLU are involved in cancerogenesis. The prevailing data suggest that in the first step the cell could attempt to block the transformation, therefore the prodeath form nCLU functions as a tumor suppressor, while its repression during cancer progression could favor the prosurvival sCLU isoform, that could be extracellularly released in the microenvironment. In normal cells, the balanced production could be affected also by ionizing radiation (IR) in a dose-dependent manner (Klokov et al., 2004). In fact, high doses induce the proapoptotic CLU isoform, whereas low doses (>0.02 Gy) induce the prosurvival isoform involved in the adaptative response. This particular function of CLU in tumors, as demonstrated by D.A. Boothman and his group, is strongly involved in the acquisition of radio- and chemoresistance to various chemotherapeutic agents, including docetaxel, cisplatin, doxorubicin, and camptothecin used for treatment of breast, colon, renal, bladder, lung, and prostate cancer. The prosurvival function of sCLU is on the basis of current phase I/II clinical trials in prostate, lung, and breast cancer. In tumor progression, the induction of sCLU requires its transcriptional activation and de novo synthesis, which is strongly influenced by growth factors and cytokines, such as TGF-1, TGF-2, IL-6, IGF-1, and VEGFA165 prevailing in the tumoral context (Criswell et al., 2005; Pucci et al., 2009). In the present review, we shall concentrate on the molecular cross talk between cancer cells and stroma that influences the differential production of CLU forms (in the cell) and the potential action that sCLU could exerts out, in the microenvironment, to favor tumor growth and tumor cell migration. Taken together, these observations underscore the influence of the host microenvironmental changes on the survival and the behavior of malignant tumors, adhering to the older Paget’s concept that the soil could influence the success and the behavior of the seed.
II. NORMAL AND CANCER MICROECOSYSTEM Development and homeostasis of tissue is dynamically and finely modulated by a complex network of cellular communication that influences cell behavior and fate, throughout the embryogenesis and beyond. During gastrulation, epithelial cells in the ectoderm change into mesenchymal cells, which invade the primitive steak and insert themselves between ectoderm and endoderm. Some of these mesenchymal cells will participate to epithelial stroma establishment, maintaining a lifelong relationship with citotype. The organ and tissue formations are coordinated by epithelial–mesenchymal
CLU “In and Out”: Looking for a Link
97
reciprocal interactions and the epithelial response to systemic hormones and growth factors are often mediated by the mesenchyme. Many data demonstrated in vitro the cross talk between these cell types. Among tumor-associated mesenchymal cells, are endothelial and pericytes that together form the neovasculature, which is fundamental to supply nutrients, oxygen, evacuating wastes, and carbon dioxide. Moreover, fibroblasts and myofibroblasts overshadowed. This communication could be physically supported by the heterotypic cell–cell interactions for a short-range signal that do not need a prompt diffusion in the surrounding area, or by secreted molecules such as cytokines, chemokines, growth factors, proteinases and their inhibitors, and lipid products. There is growing evidence that tumors are promoted and sustained by active inflammatory signals from the surrounding microenvironment as an alteration of the normal tissues homeostasis. In 1850, Rudolf Virchow was the first to describe the tumor-promoting effect of chronic irritation or inflammation. Prominent examples include the association between infection with Helicobacter pylori and gastric cancer, discovered by the Nobel prize laureates Barry Marshall and Robin Warren; papilloma virus and cofactors like Chlamydia or herpes simplex 2 infection and cervical cancer; and the predisposition of patients with Chron’s disease to colorectal cancer, prostate chronic inflammation, and prostate cancer. The persistence of these pathobiological factors concur to determine an active chronic inflammation acting in a paracrine manner to induce angiogenesis, as well as activation of surrounding stromal cell types, fibroblasts, smooth muscle cells, and adipocytes, leading to the secretion of growth factors and proteases. Activated fibroblast in the stroma promotes tumor progression by producing stroma-modulating growth factors. In particular, members of VEGF family, platelet-derived growth factors (PDGF), epidermal growth factors receptors (EGFR) ligands, interleukins such as IL-6, IL-1, IL-8, and transforming growth factors- (TGF-). These factors disrupt the normal tissues and act in a paracrine manner to induce stromal reaction, angiogenesis, and inflammatory response. The altered expression of growth factors is associated, as a result of their autocrine effect on tumor cells, with the production of proteolytic enzymes and matrix metallo-proteinases (MMPs) which start the remodeling of the promigratory cell matrix components. TGF-, VEGF, PDGF, and fibroblast growth factor signaling pathways are involved in the process of neoangiogenesis, whereas insulin-like growth factor-1 (IGF-1), epidermal growth factor, CXC12, and IL-6 play active roles in the cancer progression and in the formation of distal metastasis of many epithelial cancers. It seems that the different phases of cancer progression are accompanied by gradients of different components of the microenvironment that prelude and induce the subsequent step and cooperate to confer different
98
Sabina Pucci et al.
aggressiveness to the tumor. CLU is one of the targets of this alteration of the balanced production of cytokines and growth factors, that signs the transition between normal and malignant phenotype. In 2003, De Wever and Mareel (2003) proposed two closely interactive pathways in their model of the cross talk between cancer cells and stromal tissue, namely the efferent and afferent pathways. In the efferent pathway, cancer cells trigger a reactive response in the stroma by releasing soluble factors such as TGF- and PDGF. These factors can directly or indirectly trans-differentiate fibroblasts into myofibroblasts or induce epithelial– mesenchymal transition (EMT) in the surrounding cancer-associated stroma, resulting in cells that exhibit increased expression of vimentin, unchanged levels of smooth muscle -actin, and decreased expression of calponin, which together constitute the characteristic myofibroblast phenotype. Cancer cells are also capable of inducing neoplastic transformation in the stromal cells of the host organ. This efferent pathway seems necessary and may serve as an early event in cancer progression. In the afferent pathway, cancer cells respond to modified stromal cells in the surrounding microenvironment. Reactive stroma exerts multiple effects on the behavior of cancer cells. Reactive stromal cells release soluble factors, secrete solid matrix components, repress cell apoptosis, increase motility and invasion, and guide progression and distal spread.
III. MICROENVIRONMENT EFFECTS ON CLU EXPRESSION (THE “IN” EFFECT) A. TGF-b, the Primum Movens In normal unstressed tissues, the basal release of TGF- by local source may suffice for the maintenance of the homeostasis. However, under conditions of tissues injury, TGF- is abundantly released by blood platelet cells and various stromal components. TGF superfamily includes: TGF-, bone morphogenic protein (BMPs) and activins. TGF- plays multifunctional roles in regulating cell cycle, apoptosis, differentiation, and extracellular matrix (ECM) remodeling. The different isoforms of TGF- seems to play different antagonistic role during tumor growth and progression, being differentially expressed during tumorigenesis. The controversial role of TGF- could be attributed to the presence of different isoforms and receptors members and the prevailing expression of one on the others could determine the inhibition or development of tumor formation (Tian and Shiemann, 2009). The inhibition of the signal transduction of TGF- receptor II has been shown to induce the progression to malignancy in epithelial
CLU “In and Out”: Looking for a Link
99
cells (Browmick et al., 2001). Cancer cells that lose the tumor suppressive arm of the TGF- pathway accrue tumorigenic effects that directly enhance tumor growth and invasion (Massague`, 2008). Conversely, the overexpression of TGF- in skin papillomas of a transgenic mouse model is associated with progression to metastatic tumors, mediated by both autocrine and paracrine signaling. Experimental animal models have demonstrated that cancer invasion is stimulated by wound healing stroma. This observation implies that growth factors involved in wound healing such as TGF- and PDGF play an important role in tumor growth and invasion as well. TGF- and PDGF are two factors secreted by a wide range of cancer cells and mediate the first interaction among tumor cells and stromal fibroblast. Reactive stroma not only play an important role during cancer initiation and progression, but also in determining whether TGF- suppresses or promotes tumor formation. TGF- exerts its antitumor activities by regulating the behavior of epithelial cells and adjacent fibroblast, which synthesize and secrete a variety of cytokines, growth factors, and ECM proteins that mediate homeostasis and suppress cancer development. Thus, the inactivation of paracrine TGF- signaling between adjacent epithelial and stromal compartment promotes cellular transformation, as well induces the growth, survival, and motility of developing neoplasm. In turn, cancer cells trigger a reactive response in the stroma inducing the release of factors, such as TGF- and PDGF that directly or indirectly could trans-differentiate fibroblast into myofibroblast or induce EMT in the surrounding cancer stroma, with an increased production of vimentin, unchanged expression of -actin decreased expression of calponin (De Wever and Mareel, 2003). A controversial correlation between high inflammatory infiltrate and aggressiveness of the tumor has been observed by pathologists in the middle of nineteenth century. The dual and ambiguous role of TGF- during cancer progression is further demonstrated by the TGF- activation of nuclear Foxp3 in the T cell infiltrate, inducing CD25þFoxP3þ T cells the determinant players of tumor immune escaping during cancer progression. These data effort the involvement of TGF- in immune dormancy in cancer favoring cancer progression. Moreover, TGF- affects Foxp3 activation and its nuclear translocation also in cancer cells, where its intriguing role has still to be clarified. TGF-2 treatment but not TGF-1 induced the upregulation of the forkhead factor Foxp3 and determined its translocation from the cytoplasm to the nucleus of the neoplastic cell. The Foxp3 expression and nuclear translocation induced by TGF-2 and IL-10 has been associated with worse overall survival in breast cancer (Merlo et al., 2009). Interesting data were obtained after TGF-2 treatment to a hepatocarcinoma cell line, Hep-G2. In TGF-2-treated Hep-G2 cells, we observed an
100
Sabina Pucci et al.
increased expression of one specific isoform of Foxp3. Moreover, TGF- 2 treatment induced a nuclear translocation of this factor (Pucci et al., unpublished data). TGF-2 treatment induces the expression and the nuclear translocation of Foxp3 also in pancreatic ductal adenocarcinoma cells and tumors. The TGF-2 effect could be mimicked by ectopic expression of a constitutively active TGF- type I receptor/AK5 mutant (Hinz et al., 2007). Furthermore, the expression of Foxp3 induced by TGF-2 in cancer cells could affect, by an unknown paracrine action, also the immune response dormancy of the tumor T cell infiltrate. Coculture of Foxp3-expressing tumor cells with naı¨ve T cells, completely inhibited T cell proliferation, but not their activation. This effect indicates that pancreatic carcinoma cells share growth suppressive effects with T-reg and suggest a new mechanism of immune evasion induced by TGF-2 action underlying the dual role of this growth factor in cancer initiation and progression. In this complex cross talk, factors that directly and indirectly affect CLU production will be examined. It seems that TGF- is closely involved in the regulation of CLU different forms expression, depending on the stage and state of neoplastic disease. The TGF- input influences the expression of different CLU forms in cancer cells. Hence, TGF- and its dichotomous nature during tumorigenesis could also be related to the differential induction of the antagonistic CLU different forms. TGF-1 signaling pathways are activated by ligand binding to the cell surface transforming growth factor receptor type II (TGFRII), the activation leads to a phosphorylation of the TGF receptor type I. The signaling cascade involved the migration to the nucleus via SMAD (SMAD2 and SMAD3) superfamily proteins inducing the transcription of target genes. However, variant branches of Smad-independent pathways coexist with the canonical paths in response to TGF-. Smad-independent modes of TGF signaling involve the activation of the transcription factor AP-1 and EGR-1. AP-1 has been proposed to have a role in the CLU upregulation after TGF- treatment in normal cells (Jin and Howe, 1997; Reddy et al., 1996). The induction of nuclear localization of CLU in normal cells after TGF- treatment indicates the TGF- role in maintaining the normal tissue homeostasis and its functional link in the tumor suppressor action of CLU. A striking accordance exists between CLU and TGF- expression during mouse embryogenesis, cardiac valve morphogenesis, and in various pathophysiological conditions, such as atherosclerosis and Alzheimer’s disease. On the other hand, in advanced stage cancers the truncated form of CLU, produced by differential splicing event, is extracellularly released, probably through the stimulus of different cocktail of cytokines indirectly induced in cancer progression. A reciprocal control exists between CLU and TGF- signaling. It is known that CLU interacts with TGF- type II receptor (TGF-RII). Moreover,
CLU “In and Out”: Looking for a Link
101
experimental data demonstrated that CLU regulates TGF- signaling pathway by modulating the stability of Smad2/3 proteins. CLU siRNA repressed TGF--induced transcriptional activity and decreased the amount of Smad2/3 proteins in hepatocarcinoma Hep3B cells (Lee et al., 2008). It has been demonstrated that CLU increased Smad2/3 phosphorylation and also stabilized Smad2/3 proteins, probably modulating their turnover via proteosomal degradation (Lee et al., 2008). In yeast two-hybrid analysis and cell lysates, CLU is able to bind the receptors TGF-RI and TGF-RII. However, the sCLU binding occurred in the intracellular portion of the receptor, suggesting that the cytoplasmic, rather than the secreted form may be involved in TGF-1 signaling. It has been demonstrated that the treatment with TGF-1 induced an increase of both sCLU and nCLU proteins in TGFRII-proficient VACO-400RII human colon cancer cells, whereas no induction of CLU was found in VACO-400 cells bearing a mutation of TGF-RI (Boothman, Koklov et al., unpublished data). Moreover, TGF-1-induced CLU production resulted in growth arrest but not apoptosis. This effect of TGF-1 is in accordance to its main function during tissue formation and after tissue damage or in the first phases of tumor growth in order to maintain tissue homeostasis inhibiting proliferation and, in case of damaging agents, activating cytoprotective pathways as defense mechanisms. In contrast, the presence of cytokines and growth factors such as IL-6 and VEGF-A165, that secondly becomes abundant in the tumor microenvironment act influencing sCLU production favoring the prosurvival pathways of the neoplastic cell. The increased expression of IL-6 and VEGFA165 is strongly influenced by oxidative stress and hypoxia.
B. Hypoxia Inducible Factor: Altered of IL-6 and VEGFA165 Expression in the Microenvironment The discovery of the hypoxia-inducible (HIF-1) transcription factor (Semenza, 2003) and the finding that it directly controls the transcription of almost every enzyme of glycolysis provided a potential molecular mechanism for the effects of hypoxia on the glucose metabolism of tumors. The accumulating data suggest that the altered metabolism of tumor cells is genetically controlled by the very mutations that give rise to cancer. In recent years, several studies have demonstrated that oncogenic and tumor suppressor mutations found in a wide variety of human cancers can directly activate HIF-1 independently of hypoxia. HIF-1 is a heterodimer transcription factor that consists of two proteins HIF-1 and HIF-1. The HIF complex binds to HREs (hypoxia responsive elements) in promoters of target genes, stimulating transcription of key angiogenic factors such as VEGF and
102
Sabina Pucci et al.
angiopoietin-2 which are overexpressed in tumors (and essential for tumor angiogenesis). In addition, nearly all the enzymes of glycolysis are increased by HIF-1. As a consequence, the glycolytic shift would be caused by a specific transcriptional program. Signaling pathways activated by growth factors and deregulated in cancer lead to an increase in HIF-1 protein levels. Studies on VHL tumor suppressor (which is mutated in familial and sporadic clear cell renal carcinomas) show that VHL encodes for a subunit of ubiquitin ligase protein involved in the degradation of HIF-1. In particular, HIF-1 is not hydroxylated and can escape polyubiquitylation (Ub–Ub, mediated by the von Hippel–Lindau protein (pVHL) ubiquitin–ligase complex) and degradation even under normoxic conditions. HIF-1 then translocates to the nucleus where, together with HIF-1, forms the active HIF complex that induces the expression of genes that support tumor growth and spreading and might decrease apoptosis (Semenza, 2006). Presently HIF activation in tumoral tissues is considered a prognostic factor associated with radiation resistance and poor prognosis representing a hallmark of cancer progression. HIF-1 protein synthesis is regulated by activation of the phosphatidylinositol-3-kinase (PI3K) and ERK mitogen-activated protein kinase (MAPK) pathways. These pathways can be activated by signaling via receptor tyrosine kinases, nonreceptor tyrosine kinases, or G-protein-coupled receptors. In particular, PI3K is directly activated by both Ras and growth factor receptor tyrosine kinase (EGFR-Her2/Neu) mutationally activated in the majority of human cancers. The downstream effector of PI3K is AKT protein kinase. Overall, HIF-1 is overexpressed in human cancers as a result of intratumoral hypoxia as well as genetic alterations, such as gainof-function mutations in oncogenes (amplification of Her2/Neu) and lossof-function mutations in tumor-suppressor genes (VHL and PTEN). HIF-1 overexpression in cancer is associated with treatment failure and increased mortality. In preclinical studies, inhibition of HIF-1 activity has marked effects on tumor growth. Efforts are underway to identify inhibitors of HIF-1 and to test their efficacy as anticancer therapeutics. The level of IL-6 in tumor microenvironment and in cancer cells could be strongly influenced by the HIF. Of particular interest is the cooperation of IL-6 and HIF-1 and their action on tumor cells behavior and cell death escape. It has been observed that in critical conditions (hypoxia, oxidative stress), the activation of STAT3 influences the preferential expression of VEGF-A165a, leading to the inhibition of programmed cell death inducing Bcl-2. Moreover, it has been shown that an increased formation of IL-6–sIL-6R complexes that interact with gp130 on the cell membrane (trans-signaling) leads to the enhanced expression and nuclear translocation of STAT3, which can cause induction of antiapoptotic genes, such as Bax antagonist, Bcl-xL (Mitsuyama et al., 2007). In this view,
103
CLU “In and Out”: Looking for a Link
IL-6, induced by HIF, seems to contribute to a key mechanism of tumor development and progression through the inhibition of cell pathways leading to apoptosis. Moreover, IL-6 could directly or indirectly influence and promote tumor growth and vascularization synergizing with the HIF-1 in the induction of VEGF-A165 expression. These complex network influences also CLU production in neoplastic cells. We will discuss firstly the effect of exogenous IL-6 on CLU different forms expression in colon cancer cells and the IL-6 induced physical interactions among sCLU, Ku70, and Bax. Finally, we report results on the cooperative interaction between IL-6 and VEGFA165 in sCLU form-mediated cell death escape (Fig. 1).
C. IL-6 IL-6 is involved in several processes, explaining its long list of synonyms (B cell stimulatory factor-2, B cell differentiation factor, T cell-replacing factor, interferon-2, 26-kDa protein, hybridoma growth factor, interleukin
Inflammatory cells (myo)fibroblasts
Normal stroma
ECM
Angiogenesis Lymphangiogenesis
Tumor stroma
Cancer cell
Normal cell
IL-6 VEGFA165a
Apoptosis escaping
Aggressiveness acquisition
sCLU
DNA repair impaired
Fig. 1 Tumor microenvironmental factors. IL-6–VEGF-A165a cooperation in cancer progression.
104
Sabina Pucci et al.
hybridoma plasmacytoma factor 1, plasmacytoma growth factor, hepatocyte-stimulating factor, macrophage granulocyte-inducing factor 2, cytotoxic T cell differentiation factor, and thrombopoietin). IL-6 is now well recognized for its role in the acute-phase inflammatory response which is characterized by production of a variety of hepatic proteins termed acute phase proteins (e.g., C-reactive protein, serum amyloid A, fibrinogen, complement, alpha1-antitrypsin) (Scheller et al., 2006). In addition to its role in the acute-phase response, IL-6 is important for the development of specific immunologic responses. Moreover, it appears to play an important role in bone metabolism through induction of osteoclastogenesis and osteoclast activity. IL-6 induces differentiation of activated, but not resting, B cells culminating in production of immunoglobulin. Along with B cell differentiation, IL-6 stimulates proliferation of thymic and peripheral T cells and in cooperation with IL-1, induces T cell differentiation to cytolytic T cells and activates natural killer cells. These observations emphasize the importance of IL-6 in both nonspecific and specific immune responses. In addition to the activities described above, IL-6 functions in a wide variety of other systems including the reproductive system by participating in spermatogenesis, skin proliferation, megakaryocytopoiesis, macrophage differentiation, and neural cell differentiation and proliferation. Furthermore, IL-6 levels are directly correlated with aging in a variety of species, thus it may play an important role in the aging process and in the aging-related disorders including, Alzheimer’s disease, arteriosclerosis, and thyroiditis. Intriguingly, dietary restriction, the only experimental intervention that reproducibly prolongs maximum lifespan in mammals can restore to the young phenotype a variety of physiologic parameters, including IL-6 secretion and serum levels. IL-6 may be an important mediator of several infectious and autoimmune diseases. These include human immunodeficiency virus, rheumatoid arthritis, Castleman’s disease, and the paraneoplastic symptoms associated with cardiac myxoma. Because of its multidimensional and complex actions, dysregulation of IL-6 results in a myriad of disorders (Keller et al., 1996) including a variety of neoplastic processes. The involvement of this cytokine in the regulation of neoplastic cell growth has been recently characterized. Levels of both IL-6 and its receptor increase during tumorigenesis in many tissues. Activation of the JAK/STAT (Janus kinase/signal transducer and activator of transcription), MAPK and PI3K/AKT signaling pathways has been reported in various cancer cell lines in response to IL-6 (Culig et al., 2005). It may affect cancer progression by its actions on cell adhesion and motility, thrombopoiesis, tumor-specific antigen expression, and cancer cell proliferation. Depending on the cell type and the presence or absence of IL-6R, IL-6 can either inhibit or stimulate cancer cell proliferation. A great variety of tumor types are stimulated by IL-6, including melanoma, renal cell carcinoma,
CLU “In and Out”: Looking for a Link
105
prostate carcinoma, Kaposi’s sarcoma, ovarian carcinoma, lymphoma and leukaemia, and multiple myeloma. In many of these tumors, IL-6R has been detected and a direct proliferative signal has been proposed. Yet, when tumor cells are devoid of IL-6R, a tumor inhibiting effect of IL-6 has been demonstrated, presumably because of its immune enhancing properties. Studies using an anti-IL-6 antibody have reported induction of apoptosis, inhibition of tumor proliferation, and elimination of the progression to androgen-independent status in a prostate cancer xenograft model (Wallner et al., 2006).
D. VEGF-A The human VEGF-A gene is organized into eight exons, separated by seven introns and is localized in chromosome 6p21.3. Alternative splicing of the human VEGF-A gene give rises to at least six different transcripts, encoding isoforms of 121, 145, 165, 183, 189, and 206 amino acid residues. Multiple protein forms are encoded through alternative exon splicing. All transcripts contain the exon 5, codify for the signal sequence and core VEGF binding or VEGF/PDGF homology domain and exon 8, with diversity generated through the alternative splicing of exons 6 and 7. Exon 6 encodes a heparin-binding domain, while exons 7 and 8 encode a domain that mediates binding to neuropilin-1 (NP1) and heparin exons. Several additional minor splice variants also have been described including VEGF-145, VEGF-162, and VEGF-165b, a variant reported to have an antagonistic effect on VEGF-165a-induced mitogenesis. VEGF-165b recently identified, display different activities in respect of its isoforms 165a, it is not mitogenic and it does not increase proliferation, but its functions are still not well characterized (Pucci et al., 2008). It is known that IL-6 is involved in regulation of VEGF expression as well as neuroendocrine differentiation in prostate tissue (Culig et al., 2005). It is expressed mainly by stromal cells in the prostate, although both stroma and epithelium express the IL-6 receptor (IL-6R). Key upstream signals for VEGF regulation via HIF-1 may be differentially regulated in cancer and may synergize with other mechanisms of VEGF upregulation. Overall, the presence of IL-6 induced VEGF by the activation of NF-b and AP-1. The AP-1 activation resulted in the increased expression of both, VEGF and IL-6 regulated in paracrine and in autocrine fashion. Concomitantly, it was found that increased level of IL-6 and VEGF are closely linked to increased nuclear protein levels of HIF-1 and enhanced nuclear transcription factor DNA binding activity to a hypoxia responsible element located in the VEGF promoter. Especially two isoform are involved in tumor progression: VEGF-A165 and VEGF-120. The characterization of the VEGF-A165 and VEGF-120 variants is a relevant improvement in
106
Sabina Pucci et al.
discovering the regulatory pathway of this abundant growth factor especially for the design of new targets anticancer chemotherapy. Recently blocking molecules are successfully used in anticancer therapy. Anti-VEGF therapy with bevacizumab can increase overall survival and/or progression-free survival in patients with colorectal, breast, lung cancer, or glioblastoma multiforme when combined with cytotoxic agents. The balance and the interaction of IL-6 and VEGF-165 and in particular its isoform 165a has been observed to cooperate in neoangiogenesis and tumor cell survival. The molecular mechanisms involved influences CLU forms balance inducing a strong induction of sCLU form.
E. IL-6, VEGF-A165, and Cell Death Escape in Colon Cancer Cell: Acting on sCLU Induction There is growing evidence that IL-6 may play a crucial role in the uncontrolled intestinal chronic inflammatory process, leading to colon cancer initiation. IL-6 regulates neoplastic cell growth in autocrine and paracrine fashion, although data on the possible relationship between IL-6 production and tumor progression are conflicting. In colon cancer progression, we observed that the production of IL-6 released as by the tumor itself as by tumor-associated macrophages, and VEGF-A165, could influence tumor cell proliferation, favor apoptotic escaping and cell migration. The presence of IL-6 influences cell survival acting on the sCLU–Ku–Bax physical interactions. As we previously reported in the chapter 3 of this volume, Bax is localized in a physiologically inactive form in the cytoplasm of normal undamaged cells, where it heterodimerizes with the C-terminus of Ku70, a protein that participates in the repair of DNA double-strand breaks (DSBs), caused by V (D)J recombination, isotype switching, physiological oxidations, IR, and chemotherapeutic agents that target DNA (Gottlieb and Jackson, 1993). The ability of Ku70 to sequester Bax is a main determinant in preventing this proapoptotic protein from homodimerizing, thereby abrogating key apoptotic initiation events. The regulation of the ability of Ku70 to sequester Bax in the cytoplasm seems to be regulated by the lysine acetylation state within its C-terminus region (Cohen et al., 2004). Changes in subcellular localization of Ku can, apparently, be controlled by various external growthregulating stimuli, suggesting biological functions for the nuclear Ku70/86 heterodimer driven by microenvironment-soluble mediators (Pucci et al., 2001; Pucci et al., 2004a,b). Thus, changes in the microenvironment play a central role in redirecting pathways involved in DNA repair and cell death, affecting tumorigenesis. In normal cell Ku86 activation and translocation
CLU “In and Out”: Looking for a Link
107
into the nucleus could be regulated or stimulated by the induction of nuclear Clusterin (nCLU)–Ku70 interactions. nCLU binds the Ku70 subunit after sublethal damage induction allowing Bax to homodimerize. On the other hand, cytoprotective sCLU seems to play an important role in cell survival pathways stabilizing the Ku70–Bax interaction in the cytoplasm. In this scenario, we observed that soluble mediators, in particular IL-6, could actively affect colon cancer progression targeting the prosurvival pathways of neoplastic cell. We found HIF-1 activated in the nucleus of human colon cancer biopsies correlated to an increased level of IL-6. Regarding the expression of VEGF-A165 in advanced colon cancer we found an upregulation of VEGF-A165 a expression with an evident disequilibrium between the production of the two, a and b, isoforms of VEGF-A165 as compared to normal tissues. Notably, the VEGF-A165 isoform b, known to display an antagonistic action in respect to the isoform a, was completely lost. Moreover, a tumor-specific modulation of Bax, Ku, and CLU expression and subcellular localization in human colon cancer tissues was reported in chapter 3 of this volume. In vitro experiments confirmed that the expression level and the unconventional compartment localization of these proteins could be driven by exogenous factors. IL-6 and VEGF-A165 treatment of a colon cancer cell line, Caco-2, modulated the expression of genes involved in tumor invasion and apoptosis, observed by microarrays. In particular, IL-6 downmodulated Bax expression at mRNA level. Concomitantly, IL-6 exposure influenced Bax also at protein level acting on the Bax–Ku70–sCLU physical interactions in the cytoplasm, by affecting the Ku70 acetylation and phosphorylation state, thus leading to the inhibition of Bax proapoptotic activity. In addition, we found that IL-6 treatment induced a significant downregulation of Ku86 and a strong increase of sCLU. The downregulation or the loss of Ku86 as observed in advanced stage colon carcinomas give raise to an impaired DNA repair process with accumulation of genetic alterations leading to higher aggressiveness acquisition. Concomitantly, the accumulation of sCLU in the cytoplasm cells bound to Ku70 and Bax in IL-6 treated cells, as demonstrated by coimmunoprecipitation experiments (Pucci et al., 2009), inhibits the activation of the proapoptotic process. In contrast, the Caco-2 cells treatment with somatostatin, the physiological growth regulatory hormone, was able to restore apoptosis, demonstrating that Ku70–Bax–CLU interactions could be dynamically modulated by microenvironmental factors. Strikingly, we observed that the cooperation between IL-6 and VEGFA165 influenced the expression of tumor suppressing miRNAs affecting the epigenetic HDAC-1 activity and the EMT, turning the neoplastic cell from epithelial to mesenchymal, strongly correlated to the aggressiveness
Sabina Pucci et al.
108
acquisition of many types of cancers (Pucci, unpublished data; Sullivan et al., 2009). The effect of these factors on EMT in cancer cell is strongly supported by sCLU overproduced and released from neoplastic cells versus the stromal component that cooperate to this transformation, as reported in the next paragraph describing the downstream effects of sCLU on the microenvironment. These still obscure molecular interactions in cancer cell that oppose prosurvival sCLU and proapoptotic factors underlie the relevant role of microenvironmental factors, in the complicated cross talk among molecules that could effectively turn the cell fate (Fig. 2).
IV. sCLU EFFECTS ON MICROENVIRONMENT (THE “OUT” EFFECT): UP- AND DOWNSTREAM SIGNALS CLU represents one of the numerous soluble factors which share the inner information of cell with the microenvironment. Among the others, one characteristic of CLU is its ability to interact with a wide array of components both in the serum and on the cell surface, such as complement regulatory proteins, lipid molecules, immunoglobulin, -amyloid peptide (Shinoura et al., 1994). It also binds to the cell surface of Staphylococcus Normal mucosa
IL-6
Normal mucosa
Colon cancer
VEGFA165a/b
HIF 1a
VEGFA165b
VEGFA165a
Fig. 2 HIF-1, IL-6, and VEGF-A165 in colon cancer.
IL-6
CLU “In and Out”: Looking for a Link
109
aureus (Partridge et al., 1996) and it seems that such interaction may be an important bacterial virulence determinant for S. aureus. On the other hand, CLU binds to membrane-type MMP subgroup MT6-MMP/MMP-25, expressed in neutrophils and in brain tumors. MT6-MMP interacts with CLU forming a complex which regulates the neutrophil function, preventing the destruction of the host normal tissues (Matsuda et al., 2003). Moreover, in CLU-deficient mice affected by autoimmune myocarditis, a more extensive tissue damage caused by immune reaction was observed in the inflammatory site, compared to wild-type mice (McLaughlin et al., 2000). It was reported that exogenous CLU promotes cell growth. However, how CLU promotes cell growth remains largely unknown. Exogenous CLU stimulates Ras-dependent Raf-1/mitogen-activated protein kinase MEK/ERK activation. Shim et al. investigate the signaling pathway and related molecules underlying astrocyte proliferation by CLU. CLU-induced astrocyte proliferation and ERK1/2 phosphorylation were abrogated by an inhibitor of EGFR, or EGFR small interfering RNA. These results suggest that CLU requires EGFR activation to deliver its mitogenic signal through the Ras/Raf-1/MEK/ ERK signaling cascade in astrocytes (Shim et al., 2009). However, Reddy et al. (1996) showed that Clu/ApoJ did not interact directly with EGFR, whereas it associated with both TGF- type I and type II receptors. Extracellular CLU was found to be cytoprotective when cells were challenged with apoptotic stimuli. CLU cDNA transfected into prostate cancer cells increased the resistance to apoptosis induced by tumor necrosis factor- treatment (Sensibar et al., 1995). Moreover, the increased expression of CLU in prostate cancer correlated with tumor grade. By CLU antisense therapy, Miyake et al. (2000) showed that extracellular CLU is an antiapoptotic gene involved in progression of prostate cancer. sCLU exerts its cell protective function through binding to a cellular receptor. As reported by Koch-Brandt megalin was the first identified CLU receptor (Bartl et al., 2001) and CLU–megalin interaction-induced signaling was the cause of cell protection in prostate cells treated with TNF- apoptotic stimuli. CLU binding to its receptor activated the PI3 kinase/Akt pathway and produced multiple protein phosphorylation. However, the inhibition of this pathway did not block the protective effect indicating that additional pathways may be involved in the protection by CLU-induced signaling (Ammar and Closset, 2008). PI3K and its major downstream kinase, Akt, play key roles in many aspects of tumorigenesis, such as cellular proliferation, survival, and migration. Constitutive activation of the PI3K–Akt pathway is closely associated with cancer cell resistance to chemotherapeutic agents. Deactivation of this pathway has been shown to increase the efficacy of many anticancer drugs, targeting a wide range of cellular components. It seems that cancer cellsecreted IGF-1 and extracellular CLU constitute the regulatory system of
110
Sabina Pucci et al.
PI3K–Akt pathway. The interplay between CLU and IGF-1 produces an effect opposite to that resulting from the CLU/megalin interaction. In fact, Jo et al. (2008) showed that sCLU associated with IGF-1 and inhibits its binding to the IGF-1 receptor and hence negatively regulates the PI3K–Akt pathway. This inhibitory function of CLU appears to prefer IGF-1, as it fails to exert any effects on epidermal growth factor signaling. Therefore, CLU represents a positive or negative regulator of PI3/Akt pathway depending on its interacting partner (CLU receptor Megalin and IGF-1, respectively). Intriguingly, the uptake of glucose is a process mediated by the PI3K–Akt pathway and its availability, often limited in the tumor microenvironment, can be a deciding factor for cell proliferation. Under this condition, those cells with a higher “resistance” would gain growth advantages and thus would be positively selected. We previously reported that CLU level is tightly associated with various cellular stress responses. Therefore, its secretion may reflect the cellular adaptive responses to endure adverse environmental conditions (i.e., by suppressing its own growth and that of surrounding cells). In a mouse model of prostate cancer, it has been shown that epithelial cancer cells initiate and promote the clonal expansion of stromal fibroblasts that lack the p53 tumor suppressor gene (Hill et al., 2005), indicating that the cancer cell-derived factors initially impose selective pressures (i.e., antiproliferation) on neighboring cells. The presence of CLU in the tumor microenvironment can contribute to such selective pressures. It is of note that PI3 kinase/Akt pathway regulates the HIF-1 protein synthesis and furthermore the activation of MAPK and PI3K/AKT signaling has been reported in various cancer cell lines in response to IL-6 (Culig et al., 2005). It implies that not only IL-6 could promote cancer progression affecting Bax, Ku, and CLU interactions (Pucci et al., 2009), but also CLU could modulate HIF-1 protein levels and enhance IL-6 effects in a positive feedback loop. Hence, cancer cell-derived CLU may provide a molecular framework to further dissect the complex relationships between cancer cells and their environment. During EMT, epithelial cells downregulate their intercellular adhesion, lose the apical–basal polarity, and undergo morphological changes from a monolayer of cuboidal-shaped cells to dispersed, spindle-shaped fibroblastlike cells. The expression of differentiation markers switches from cell–cell junction proteins such as E-cadherin to mesenchymal markers including fibronectin and vimentin. Furthermore, the stationary cells convert to migratory cells capable of invasion through ECM. The mechanism underlying EMT involves coordination of multiple signaling pathways, including TGF- (Heldin et al., 2009), receptor tyrosine kinase/Ras signaling pathways, other autocrine factors (e.g., EGF, HGF, and IGF), Wnt, Notch, Hedgehog, and NF-B signaling pathways (Bates and Mercurio, 2005). These pathways exert their EMT-inducing effect through modulating
CLU “In and Out”: Looking for a Link
111
transcriptional regulators, to repress epithelial genes such as those encoding proteins of intercellular junctions, or activate genes pivotal for cells to acquire migratory and invasive properties. It was demonstrated that CLU modulates the EMT in human lung adenocarcinoma cell lines (Chou et al., 2009). CLUrich cells displayed a spindle-shape morphology while those with low CLU levels were cuboidal in shape. Moreover, CLU silencing by siRNA in highly invasive CLU-rich lung adenocarcinoma induced a mesenchymal-to-epithelial transition (MET) evidenced by the spindle-to-cuboidal morphological change, increased E-cadherin expression, and decreased fibronectin expression.
V. CONCLUSIONS AND FUTURE PERSPECTIVES In this review, we emphasize the importance of cross talk between stroma and epithelia in carcinogenesis. Reactive stroma is induced after epithelia changes and their coevolution determines the release of the microenvironmental factors, strong determinant of tumor behavior. CLU behaves as an important player of the cell fate within the cancer cell and as relevant soluble factor of the microenvironment still conditioning the fate and the proliferation rate of surrounding cancer cell in a paracrine fashion. Its production and the release are finely regulated by the microenvironment. Further study is required to fully understand the complex interaction between cancer cell and the tumor microenvironment that leads to sCLU production and prosurvival pathways activation. Target therapy to the stromal compartment as well to epithelia is expected to be clinically promising, and further elucidations on the molecular mechanisms underlying tumor–stroma interactions may yield novel therapeutic targets for anticancer therapy.
REFERENCES Ammar, H., and Closset, J. L. (2008). Clusterin activates survival through the phosphatidylinositol 3-kinase/Akt pathway. J. Biol. Chem. 283(19), 12851–12861. Bartl, M. M., Luckenbach, T., Bergner, O., Ullrich, O., and Koch-Brandt, C. (2001). Multiple receptors mediate apoJ-dependent clearance of cellular debris into nonprofessional phagocytes. Exp. Cell Res. 271(1), 130–141. Bates, R. C., and Mercurio, A. M. (2005). The epithelial–mesenchymal transition (EMT) and colorectal cancer progression. Cancer Biol. Ther. 4(4), 365–370. Browmick, N. A., et al. (2001). Tumor derived TGF- signalling in fibroblast modulated the oncogenic potential of adjacent cells. Science 303, 848–851. Chou, T. Y., Chen, W. C., Lee, A. C., Hung, S. M., Shih, N. Y., and Chen, M. Y. (2009). Clusterin silencing in human lung adenocarcinoma cells induces a mesenchymal-to-epithelial transition through modulating the ERK/Slug pathway. Cell Signal. 21(5), 704–711. Cohen, H. Y., Lavu, S., Bitterman, K. J., Hekking, B., Imahiyerobo, T. A., Miller, C., Frye, R., Ploegh, H., Kessler, B. M., and Sinclair, D. A. (2004). Acetylation of the C terminus of Ku70 by CBP and PCAF controls Bax-mediated apoptosis. Mol. Cell 13, 627–638.
112
Sabina Pucci et al.
Criswell, T., Beman, M., Araki, S., Leskov, K., Cataldo, E., Mayo, L. D., and Boothman, D. A. (2005). Delayed activation of insulin-like growth factor-1 receptor/Src/MAPK/Egr-1 signaling regulates clusterin expression, a pro-survival factor. J. Biol. Chem. 280(14), 14212–14221. Culig, Z., Steiner, H., Bartsch, G., and Hobisch, A. (2005). Interleukin-6 regulation of prostate cancer cell growth. J. Cell. Biochem. 95, 497–505. De Wever, O., and Mareel, M. (2003). Role of tissue stroma in cancer cell invasion. J. Pathol. 200(4), 429–447. Gottlieb, T. M., and Jackson, S. P. (1993). The DNA dependent protein kinase: Requirement for DNA ends and association with Ku antigen. Cell 72, 131–142. Heldin, C. H., Landstro¨m, M., and Moustakas, A. (2009). Mechanism of TGF-beta signaling to growth arrest, apoptosis, and epithelial-mesenchymal transition. Curr. Opin. Cell Biol. 21(2), 166–176. Hill, R., Song, Y., Cardiff, R. D., and Van Dyke, T. (2005). Selective evolution of stromal mesenchyme with p53 loss in response to epithelial tumorigenesis. Cell 123, 1001–1011. Hinz, S., Pagerols-Raluy, L., Oberg, H. H., Ammerpohl, O., Gru¨ssel, S., Sipos, B., Gru¨tzmann, R., Pilarsky, C., Ungefroren, H., Saeger, H. D., Klo¨ppel, G., Kabelitz, D., et al. (2007). Foxp3 expression in pancreatic carcinoma cells as a novel mechanism of immune evasion in cancer. Cancer Res. 67, 8344–8350. Jin, G., and Howe, P. H. (1997). Regulation of clusterin gene expression by transforming growth factor beta. J. Biol. Chem. 272, 26620–26626. Jo, H., Jia, Y., Subramanian, K. K., Hattori, H., and Luo, H. R. (2008). Cancer cell-derived clusterin modulates the phosphatidylinositol 3’-kinase-Akt pathway through attenuation of insulin-like growth factor 1 during serum deprivation. Mol. Cell. Biol. 28(13), 4285–4299. Keller, E. T., Wanagat, J., and Ershler, W. B. (1996). Molecular and cellular biology of Interleukin-6 and its receptor. Front. Biosci. 1, 340–357. Klokov, D., Criswell, T., Leskov, K. S., Araki, S., Mayo, L., and Boothman, D. A. (2004). IRinducible clusterin gene expression: a protein with potential roles in ionizing radiation-induced adaptive responses, genomic instability, and bystander effects. Mutat. Res. 568(1), 97–110. Lee, K. B., Jeon, J. H., Choi, I., Kwon, O. Y., Yu, K., and You, K. H. (2008). Clusterin, a novel modulator of TGF-beta signaling, is involved in Smad2/3 stability. Biochem. Biophys. Res. Commun. 366(4), 905–909. Lewis, A. M., Varghese, S., Xu, H., and Alexander, H. R. (2006). Interleukin-1 and cancer progression: The emerging role of interleukin-1 receptor antagonist as a novel therapeutic agent in cancer treatment. J. Transl. Med. 4, 48. Massague`, J. (2008). TGF in cancer. Cell 134, 215–230. Matsuda, A., Itoh, Y., Koshikawa, N., Akizawa, T., Yana, I., and Seiki, M. (2003). Clusterin, an abundant serum factor, is a possible negative regulator of MT6-MMP/MMP-25 produced by neutrophils. J. Biol. Chem. 278(38), 36350–36357. McLaughlin, L., Zhu, G., Mistry, M., Ley-Ebert, C., Stuart, W. D., Florio, C. J., Groen, P. A., Witt, S. A., Kimball, T. R., Witte, D. P., Harmony, J. A., and Aronow, B. J. (2000). Apolipoprotein J/clusterin limits the severity of murine autoimmune myocarditis. J. Clin. Invest. 106(9), 1105–1113. Merlo, A., Casalini, P., Carcangiu, M. L., Malventano, C., Triulzi, T., Me`nard, S., Tagliabue, E., and Balsari, A. (2009). FOXP3 expression and overall survival in breast cancer. J. Clin. Oncol. 27, 1746–1752. Mitsuyama, K., Matsumoto, S., Masuda, J., Yamasakii, H., Kuwaki, K., Takedatsu, H., and Sata, M. (2007). Therapeutic strategies for targeting the IL-6/STAT3 cytokine signaling pathway in inflammatory bowel disease. Anticancer Res. 27, 3749–3756. Miyake, H., Nelson, C., Rennie, P. S., and Gleave, M. E. (2000). Testosterone-repressed prostate message-2 is an antiapoptotic gene involved in progression to androgen independence in prostate cancer. Cancer Res. 60(1), 170–176.
CLU “In and Out”: Looking for a Link
113
Panutsopulos, D., Zafiropoulos, A., Krambovitis, E., Kochiadakis, G. E., Igoumenidis, N. E., and Spandidos, D. A. (2003). Peripheral monocytes from diabetic patients with coronary artery disease display increased bFGF and VEGF mRNA expression. J. Transl. Med. 1, 6. Partridge, S. R., Baker, M. S., Walker, M. J., and Wilson, M. R. (1996). Clusterin, a putative complement regulator, binds to the cell surface of Staphylococcus aureus clinical isolates. Infect. Immun. 64(10), 4324–4329. Patterson, S. G., Wei, S., Chen, X., Sallman, D. A., Gilvary, D. L., Zhong, B., Pow-Sang, J., Yeatman, T., and Djeu, J. Y. (2006). Novel role of Stat1 in the development of docetaxel resistance in prostate tumor cells. Oncogene 25(45), 6113–6122. Pucci, S., Mazzarelli, P., Rabitti, C., Giai, M., Gallucci, M., Flammia, G., Alcini, A., Altomare, V., and Fazio, V. M. (2001). Tumor specific modulation of Ku70/80 DNA binding activity in breast and bladder human tumor biopsies. Oncogene 20, 739–747. Pucci, S., Bonanno, E., Pichiorri, F., Angeloni, C., and Spagnoli, L. G. (2004a). Modulation of different clusterin isoforms in human colon tumorigenesis. Oncogene 23(13), 2298–2304. Pucci, S., Bonanno, E., Pichiorri, F., Mazzarelli, P., and Spagnoli, L. G. (2004b). The expression and the nuclear activity of the caretaker gene ku86 are modulated by somatostatin. Eur. J. Histochem. 48(2), 103–110. Pucci, S., Mazzarelli, P., Missiroli, F., Regine, F., and Ricci, F. (2008). Neuroprotection: VEGF, IL-6, and clusterin: the dark side of the moon. Prog. Brain Res. 173, 555–573. Pucci, S., Mazzarelli, P., Sesti, F., Boothman, A. D., and Spagnoli, L. G. (2009). Interleukin-6 affects cell death escaping mechanisms acting on Bax–Ku70–Clusterin interactions in human colon cancer progression. Cell Cycle 8(3), 473–481. Reddy, K. B., Jin, G., Karode, M. C., Harmony, J. A., and Howe, P. H. (1996). Transforming growth factor beta (TGF beta)-induced nuclear localization of apolipoprotein J/clusterin in epithelial cells. Biochemistry 35(19), 6157–6163. Scheller, J., Ohnesorge, N., and Rose-John, S. (2006). Interleukin-6 trans-signalling in chronic inflammation and cancer. Scand. J. Immunol. 63(5), 321–329. Semenza, G. L. (2003). Targeting HIF-1 for cancer therapy. Nat. Rev. Cancer 3, 721–732. Semenza, G. L. (2006). VHL and p53: Tumor suppressors team up to prevent cancer. Mol. Cell 22(4), 437–439. Sensibar, J. A., Sutkowski, D. M., Raffo, A., Buttyan, R., Griswold, M. D., Sylvester, S. R., Kozlowski, J. M., and Lee, C. (1995). Prevention of cell death induced by tumor necrosis factor a in LNCaP cells by overexpression of sulfated glycoprotein-2 (clusterin). Cancer Res. 55, 2431–2437. Shim, Y. J., Shin, Y. J., Jeong, S. Y., Kang, S. W., Kim, B. M., Park, I. S., and Min, B. H. (2009). Epidermal growth factor receptor is involved in clusterin-induced astrocyte proliferation. Neuroreport 20(4), 435–439. Shinoura, N., Heffelfinger, S. C., Miller, M., Shamraj, O. I., Miura, N. H., Larson, J. J., DeTribolet, N., Warnick, R. E., Tew, J. J., and Menon, A. G. (1994). RNA expression of complement regulatory proteins in human brain tumors. Cancer Lett. 86(2), 143–149. Soufla, G., Porichis, F., Sourvinos, G., Vassilaros, S., Spandidos, D. A., De Wever, O., and Mareel, M. (2006). Transcriptional deregulation of VEGF, FGF2, TGF-beta1, 2, 3 and cognate receptors in breast tumorigenesis. Role of tissue stroma in cancer cell invasion. Cancer Lett. 235(1), 100–113. Sullivan, N. J., Sasser, A. K., Axel, A. E., Vesuna, F., Raman, V., Ramirez, N., Oberyszyn, T. M., and Hall, B. M. (2009). Interleukin-6 induces an epithelial–mesenchymal transition phenotype in human breast cancer cells. Oncogene 28(33), 2940–2947. Tian, M., and Schiemann, W. P. (2009). The TGF-beta paradox in human cancer: An update. Future Oncol 5(2), 259–271. Wallner, L., Dai, J., Escara-Wilke, J., Zhang, J., Yao, Z., Lu, Y., Trikha, M., Nemeth, J., Zaki, M. H., and Keller, E. T. (2006). Inhibition of interleukin-6 with CNTO328, an antiinterleukin-6 monoclonal antibody, inhibits conversion of androgen-dependent prostate cancer to an androgen-independent phenotype in orchiectomized mice. Cancer Res. 66, 3087–3095.
Regulation of CLU Gene Expression by Oncogenes and Epigenetic Factors: Implications for Tumorigenesis Arturo Sala,* Saverio Bettuzzi,{ Sabina Pucci,z Olesya Chayka,* Michael Dews,} and Andrei Thomas-Tikhonenko} *Molecular Haematology and Cancer Biology Unit, Institute of Child Health, University College London, United Kingdom Dipartimento di Medicina Sperimentale, Sezione di Biochimica, Biochimica Clinica e Biochimica dell’Esercizio Fisico, Universita` di Parma, Via Volturno 39-43100 Parma and Istituto Nazionale Biostrutture e Biosistemi (I.N.B.B.), Rome, Italy z Department of Biopathology, Institute of Anatomic Pathology, University of Rome "Tor Vergata," Rome, Italy } Department of Pathology and Laboratory Medicine, University of Pennsylvania and Children’s Hospital of Philadelphia, Philadelphia, Pennsylvania 19104, USA
{
I. Introduction II. Regulation of CLU Expression by Transforming Oncogenes: Early Evidence III. Regulation of CLU Expression by Protooncogenic Transcription Factors A. MYC B. MYB C. NF- B D. Egr1 E. Stat1 F. GLI and TCF IV. Oncogenic Signaling and Epigenetic Regulation of CLU Expression V. Concluding Remarks References In no other field has the function of clusterin (CLU) been more controversial than in cancer genetics. After more than 20 years of research, there is still uncertainty with regard to the role of CLU in human cancers. Some investigators believe CLU to be an oncogene, others—an inhibitor of tumorigenesis. However, owing to the recent efforts of several laboratories, the role of CLU in important cellular processes like proliferation, apoptosis, differentiation, and transformation is beginning to emerge. The “enigmatic” CLU is becoming less so. In this chapter, we will review the work of research teams interested in understanding how CLU is regulated by oncogenic signaling. We will discuss how and under what circumstances oncogenes and epigenetic factors modify CLU expression, with important consequences for mammalian tumorigenesis. # 2009 Elsevier Inc.
Advances in CANCER RESEARCH Copyright 2009, Elsevier Inc. All rights reserved.
115
0065-230X/09 $35.00 DOI: 10.1016/S0065-230X(09)05007-6
Arturo Sala et al.
116
I. INTRODUCTION The role of clusterin (CLU) in cancer has been the matter of debate for many years. There are many reports, mainly based on studies with cancer cell lines, indicating that CLU is involved in promotion of tumorigenesis and in conferring resistance to chemotherapeutic drugs (Chi et al., 2008; Chung et al., 2004; Miyake et al., 2003; Sallman et al., 2007). However, more recent studies using mouse models of neuroblastoma and prostate cancer have established that an important function of CLU is to restrict tumor development (Bettuzzi et al., in press; Chayka et al., 2009). While some of the contrasting results observed so far could be explained by the use of different types of cell lines, reagents or procedures, we suggest here that CLU, lying at the crossroad of life and death, is at the same time an oncogene and a tumour suppressor gene. This concept will be developed and clarified in the course of this review. The categorization of genes in strict functional classes clearly does not reflect the complexity of biological systems. The boundaries dividing gene functions are becoming blurred and to classify genes as oncogenes or tumor suppressors is, in the light of the more recent literature, an anachronism. Classical tumor suppressor genes like pRb, PML, and p21WAF-Cip are now found to promote human cancer in specific contexts (Cote et al., 1998; Ito et al., 2008; Morris et al., 2008; Viale et al., 2009). Conversely, protooncogenes like E2F1 and MYB have been shown to restrict tumor growth or promote the maintenance of normal cell division (Morris et al., 2008; Pierce et al., 1999; Tarasov et al., 2008). Additionally, it is useful to keep in mind that association (“post hoc”) does not imply causation (“propter hoc”). Just because a gene is overexpressed or underexpressed in certain tumors, one cannot conclude that it is driving or inhibiting neoplastic growth, respectively. Instead, its deregulation might be a defense mechanism that the host employs to maintain tissue homeostasis. Research on CLU could well be a case study in cancer gene complexity. CLU is the prototypical multifunctional gene: it was found to regulate apoptosis, cell–cell interactions, protein stability, cell signaling, proliferation and, finally, transformation. In spite of the multiple functions that have been ascribed to CLU, its genetic inactivation in mice is well tolerated and animals develop and live normally (McLaughlin et al., 2000). Since CLU expression in mammalian cells is highly modulated by certain pathological processes or exposure to physical and chemical agents, it is tempting to speculate that CLU is mainly required to respond to exogenous or endogenous stress signals. Indeed, CLU knockout mice are more susceptible than
Regulation of CLU Gene Expression by Oncogenes and Epigenetic Factors
117
wild-type mice to experimentally induced autoimmune diseases, and fibroblasts derived from CLU knockout mice are more sensitive to thermal injury (McLaughlin et al., 2000; Santilli et al., 2005). In cancer, expression of CLU has been shown to be either up- or downmodulated, although the data available on the Oncomine Web site, which represents a very large and growing collection of cDNA microarray experiments, shows that in the most cancer types CLU is downregulated (Fig. 1). It is still unclear whether the opposing observations published in the literature are caused by technical reasons—that is, use of different antibodies, cell lines, patients, etc.—or they reflect the fact that CLU can be a tumor suppressor and promoter, at the same time, depending on the specific biology of the disease and its phase of progression. As we try to avoid the dualistic, Cartesian classification of cancer genes into oncogenes and tumor suppressors, we submit that a gene can inhibit tumor growth under untreated conditions (i.e., be “pro host”) but at the same time act “antipatient” by rendering tumors resistant to chemo-, radio-, or biological therapy. Conversely, there are well-recognized examples when the oncogene initiates or even promotes tumor growth (or acts “antihost”) but at the same time confers chemosensitivity (or acts “propatient”). While these distinctions have been largely absent from the CLU literature, for other cancer genes these complexities are well appreciated. The c-Myc protooncogene is a case in point. At the cellular level many putative Myc target genes pertain to cell proliferation. Among them are ornithine decarboxylase, cyclins A and D2, cdc25A, cdk4, Id2, and telomerase. In addition, entry into the cell cycle is facilitated by repression of several genes, such as assorted cdk inhibitors, gadd45, etc. (reviewed recently in Meyer and Penn, 2008). Consistent with these observations, activation of Myc forces quiescent fibroblasts to reenter the cell cycle (Eilers et al., 1989), and rodent fibroblasts with targeted disruption of Myc are severely deficient in cell proliferation (Mateyak et al., 1997). Furthermore, at least in mice decreased expression of Myc results in hypoplasia (Trumpp et al., 2001). Conversely, when overexpressed in many transgenic settings, c-Myc initiates tumor growth with very high penetrance (Morgenbesser and DePinho, 1994). Furthermore, c-Myc is overexpressed in a variety of spontaneous human cancers, making it a classical oncogene in the view of both mouse and human geneticists. This “antihost” properties of Myc are balanced to some extent by the potentially “propatient” propensity of Myc to induce apoptosis (reviewed in Evan and Vousden, 2001), both intrinsic and extrinsic. To accomplish the former, Myc activates p53 via the ARF pathway (Zindy et al., 1998). In
Arturo Sala et al.
118 +
P -value threshold: 1E-4
Adrenal Bladder Brain Head-neck Liver Others Peripheral nervous system Spinal cord Testis Tonsil Colon Renal Breast CellLine Leukemia Lung Lymphoma Multicancer Prostate Sarcoma Central nervous systen Ovarian Parathyroid Gastric Melanoma Blood Bone marrow Cardiac Muscle Neuroblastoma Oral Placenta Seminoma Skin Tongue Vulva
Normal vs. normal 1 1 2 4 1 1 2
−
Cancer vs. normal
Outlier rank threshold: 50 Cancer vs. cancer
Histologic Molecular subtype alteration
Tumor stage
Misc
1
1 4 1 2 1
2
Tumor grade
+
− Outlier
1 2
2
1 1
1
2 4 3 1
1 1
2
1
1 1 2 2 3 1 7
1 1 1 1 2 3 7 1 7
1 2 4 2 1 6 5 3 6
2 1 2
2 1
2 4
1 5
2 1 3
9
6 1
1
2 1 6 1
1 3 3
1
1 1
1 1
2 1 1 1
1
1 1
1 2 1 2 2 1 2 1 1 1 1
Fig. 1 CLU expression in primary tumors. Expression of CLU mRNA in Affymetrix experiments as represented on the Oncomine Web site (www.Oncomine.org). Shades of blue color indicate underexpression, whereas shades of red indicate overexpression. The intensity of the color is proportional to the statistical significance of the difference. Numbers indicate how many independent experiments show a significant difference in each tissue. Note that in the “Cancer vs. Normal” column (green rectangle) all the transformed tissues, with the exception of brain, show significant downregulation of CLU with respect to the corresponding normal samples.
addition, Myc appears to potentiate the extrinsic pathway triggered by ligation of a death receptor. Indeed, Myc has been shown to participate in apoptosis induced by Fas/CD95 ligand (Hueber et al., 1997), TNF- (Klefstrom et al., 1994), and TRAIL (Ricci et al., 2004).
Regulation of CLU Gene Expression by Oncogenes and Epigenetic Factors
119
To the extent that cytotoxic drugs inflict DNA damage and activate the intrinsic apoptotic pathway, one might predict that deregulation of Myc would be associated with chemosensitivity. Indeed, as exemplified by studies on human colon carcinoma, low-level Myc amplification (combined with wild-type p53 expression) increases susceptibility to 5-fluorouracil in vivo (Arango et al., 2001). Similarly, human Burkitt’s lymphomas with Myc overexpression appear to be intrinsically sensitive to TRAIL in the clinical setting (reviewed in Finnberg and El-Deiry, 2008). The CLU situation has many interesting parallels with the Myc story. Not only is CLU downregulated by c-Myc but it has been reported to mediate TRAIL resistance in prostate cancer cells (Sallman et al., 2007). Thus, it is tempting to extend this parallel and propose that CLU is to Myc as yin is to yang. According to this framework, CLU inherently inhibits cell proliferation and neoplastic growth (i.e., acts “prohost”) but confers resistance to therapy (i.e., acts “antipatient”)—hence its overexpression under certain conditions. Taking this view helps make sense of some famously contradictory data. For example, the group lead by Martin Gleave has published a number of studies which suggest that expression of CLU is enhanced in human prostate cancer and antisense oligonucleotides targeting CLU expression inhibit prostate tumorigenesis in vivo and in vitro (Chi et al., 2008; Miyake et al., 2005). These results are contrasted by the work of the Bettuzzi group, which showed reduced expression of CLU during mouse and human prostate cancer progression (Caporali et al., 2004; Scaltriti et al., 2004b). The analysis of several gene expression studies available in the Oncomine database shows that there is a significant downregulation of the CLU mRNA in almost all types of cancer, as compared to matched normal tissue controls, corroborating the hypothesis that CLU expression is generally silenced in primary (i.e., frequently untreated) human cancers (Fig. 1). As a further indication that CLU expression might be inactivated in mammalian tumorigenesis, CLU KO mice are more prone than wild-type counterparts to oncogene-induced tumorigenesis (Chayka et al., 2009; Thomas-Tikhonenko et al., 2004). This introduces the theme of this assay, namely, how CLU is regulated by oncogenic signaling and what role CLU plays in inhibiting mammalian cell transformation, as opposed to potentially promoting chemoresistance.
II. REGULATION OF CLU EXPRESSION BY TRANSFORMING ONCOGENES: EARLY EVIDENCE Early reports demonstrated that expression of CLU is modulated during cell transformation. CLU expression was reported to be increased in gliomas compared to normal brain tissue (Danik et al., 1991). In 1993, the group of
120
Arturo Sala et al.
Michael Sporn observed that expression of CLU was increased after malignant transformation of the rat prostate caused by chemical carcinogenesis (Kadomatsu et al., 1993). It should be noted that only the mRNA expression was detected in these early studies, leaving open the question of whether CLU protein was also upregulated. Further evidence that CLU might be important in human tumorigenesis originates from the observation that CLU expression is often modulated during apoptosis. A theory was elaborated suggesting that CLU is secreted during injury as a survival response in the face of apoptosis (Koch-Brandt and Morgans, 1996). Consequently, some research groups started to investigate whether oncogenic transcription factors could induce CLU, facilitating cell survival, transformation, and/or resistance to chemotherapeutic drug killing. The first evidence that CLU expression is modulated by oncogenic activity was published in 1989 when it was first reported that a thermally inducible gene, called T64, was activated in avian cells by retroviral oncogenes with protein kinase activity such as v-src, v-fps, and v-mil (Michel et al., 1989). Sequencing of T64 revealed that it was the avian orthologue of rat CLU. Subsequent investigations revealed that induction by the oncogenic kinases was dependent on the AP-1 binding site present in close proximity to the CLU transcription start site. Similarly, Herault et al. (1992) found that the gene most strongly overexpressed upon Rous sarcoma virus infection in quail neuroretina cells was CLU. Mutation of the TGACTCA motif in the CLU promoter abolished CAT activity of the reporter suggesting that the AP-1 binding site was required for induction by Src. The role of AP-1 (a complex containing the Jun and Fos oncoproteins) in regulating CLU expression was confirmed later on in other contexts. For example, it was shown that TGF- positively modulates CLU expression via activation of an AP-1 site in the mammalian CLU promoter (Jin and Howe, 1999). In this work, the authors proposed that the mechanism of activation is the removal of the trans-repression effect of c-Fos by TGF- . In another study, exposure of HaCaT keratinocyte cells to vanadium was shown to induce apoptosis, c-Fos expression, and a switch from secreted to nuclear CLU (Markopoulou et al., 2009). Ectopic expression of c-fos also induced apoptosis and nuclear CLU expression in HaCaT cells. The authors inferred that c-Fos controls the ratio of cytoplasmic versus nuclear fraction of CLU. However, it has not been resolved whether c-fos directly regulates the levels of the different CLU protein isoforms, or apoptosis resulting from c-Fos overexpression is actually causing the isoform switch. Claudia Koch-Brandt’s group was one of the first to study the role of two classical protooncogenes, namely c-MYC and Ha-RAS, in regulation of CLU expression. It was reported that overexpression of Ha-RAS, but not of c-MYC, in the rat embryo fibroblast cell line Rat-1 caused repression of
Regulation of CLU Gene Expression by Oncogenes and Epigenetic Factors
121
CLU expression at the mRNA level (Klock et al., 1998). There had been no attempt to understand the mechanism by which Ha-RAS was causing the inhibitory effect, but this was clarified in subsequent studies that will be discussed later. This early evidence linking the activity of protooncogenes to CLU expression and the emerging role of CLU as a modulator of apoptosis prompted many other groups to study the relationship between oncogenic transcription factors and CLU.
III. REGULATION OF CLU EXPRESSION BY PROTOONCOGENIC TRANSCRIPTION FACTORS Transcription factors are the essential molecular tools, with which the cell is able to respond to changing environmental conditions, stress, differentiating stimuli, or proliferative cues. Transcription factors can be tissue-specific or ubiquitously expressed, and oncogenic versions can be found in both categories. In the following sections, we will discuss in detail which oncogenic transcription factors have been found to regulate CLU and the biological consequences of its deregulation.
A. MYC MYC is a small family of transcription factors composed of the prototype member, c-MYC, the neuronal-specific MYCN and the less-studied L-MYC. C-MYC is a major player in human tumorigenesis and its function in human cancer has been discussed in detail in many reviews (see references above and also (Lutz et al., 2002; Vita and Henriksson, 2006; Yaylim-Eraltan et al., 2008). Although it was initially thought that c-MYC could not regulate the expression of CLU (Klock et al., 1998) the group lead by Andrei ThomasTikhonenko reported that ectopic levels of c-MYC could strongly repress the expression of CLU in murine colonocytes or human keratinocytes. One of the most interesting observations in this paper is that forced overexpression of CLU could inhibit, at least in part, c-MYC-dependent tumorigenesis. Indeed, CLU could attenuate proliferation of colonocytes transformed by c-MYC, and mice with a disrupted CLU gene were more prone to develop papillomas as a consequence of exposure to carcinogens (ThomasTikhonenko et al., 2004). The concept that CLU could behave as an inhibitor of cell proliferation was not without precedent. Bettuzzi et al. (2002) showed that forced overexpression of CLU induced cell-cycle arrest of human prostate cells in vitro.
122
Arturo Sala et al.
Neuronal MYC (MYCN) is also a negative regulator of CLU. It has been recently shown (Chayka et al., 2009) that CLU is downregulated in the pediatric cancer neuroblastoma. Neuroblastoma is characterized by the amplification of MYCN, which is necessary and sufficient to induce transformation of embryonal sympathetic cells into malignant neuroblasts. In tumors with amplified MYCN, CLU is strongly downregulated and MYCN appears to induce CLU downregulation at least in part through transcriptional induction of the six-microRNA cluster miR-17-92 (composed of miR-17, -18, -19a/b, -20, and -92) (Dews et al., 2006; O’Donnell et al., 2005). These and other microRNAs are short noncoding RNAs that can specifically decrease protein output by decreasing translation and/or by mRNA destabilization (Mendell, 2008). In Chayka et al. it was demonstrated that the MYCN-induced miR-17-92 cluster downregulates CLU expression in neuroblastoma cells. The MYCN–CLU axis is functionally important, since mice with a disrupted CLU gene are more prone to the formation of neuroblastomas induced by transgenic expression of MYCN, thus suggesting that CLU is a repressor of MYCN tumorigenesis (Chayka et al., 2009). A still unpublished study is yielding evidence suggesting that MYCN can also directly repress transcription of CLU through an E-box in the CLU 50 -flanking region which is conserved in different mammalian species. More careful examination of the connection between miR-17-92 cluster members and CLU has revealed several surprises. While the Miranda algorithm (John et al., 2004) predicts binding sites for several members of the miR-17-92 cluster within the 30 -UTR of human CLU, these predictions could not be confirmed experimentally using the luciferase sensor assay or gain-of-function microRNA mimic screens (Dews et al., submitted for publication). This suggests that CLU may not be a direct molecular target for miR17-92 and that instead this cluster targets an upstream activator of CLU expression. As mentioned above, in some cell lines CLU can be induced by the TGF- signaling pathway (Jin and Howe, 1997, 1999). This idea had been also promulgated by David Boothman and his colleagues (Bey et al., 2006). Thus, it was tempting to propose that perhaps downregulation of CLU by miR-17-92 is in fact lack of activation by TGF- . Indeed, very recent work from the Thomas-Tikhonenko laboratory demonstrated that Myc-overexpressing cells contain defects in several key components of the TGF- signaling pathway, including TGF- receptor II and activating Smads (Bierie and Moses, 2006; Massague, 2008). As predicted previously (Volinia et al., 2006), miR-17-5p and miR-20 reduce levels of the type II TGF- receptor (TGFBR2) and miR-18 was found to target Smad4 and in some cell lines—Smad2. Overall, weakened TGF- signaling in Myc and/or miR-17-92 overexpressing cells resulted in very poor induction of CLU by TGF- .
Regulation of CLU Gene Expression by Oncogenes and Epigenetic Factors
123
B. MYB MYB, similarly to MYC, is a family of transcription factors which includes the tissue-specific c-MYB and A-MYB and the ubiquitous B-MYB, a positive regulator of cell proliferation and survival (Lipsick et al., 2001; Oh and Reddy, 1999; Sala and Watson, 1999). Interestingly, B-MYB is overexpressed or amplified in various types of human cancer suggesting that it too is a protooncogene (Nakajima et al., 2008; Raschella et al., 1999; Sala and Watson, 1999). In the Sala laboratory, it has been shown that B-MYB binds to and positively regulates the CLU promoter through a MYB-consensus sequence. It has also been shown that CLU mediates, at least in part, the antiapoptotic effects of B-MYB. B-MYBinduced CLU can confer resistance to doxorubicin killing of human LAN5 neuroblastoma cells. Furthermore, thermal injury is more pronounced in fibroblasts transfected with a construct expressing dominant-negative B-MYB, which also blunts thermal induction of CLU (Cervellera et al., 2000; Santilli et al., 2005). These results are in agreement with evidence correlating decreased expression of secreted CLU and B-MYB with apoptosis induced by all-trans-retinoic acid in smooth muscle cells (Orlandi et al., 2005).
C. NF-k B NF-B is a multifunctional transcription factor that has central importance in immunity and cancer. NF-B is activated in response to external stimuli—such as engagement of the TNF- receptor by its ligand, and by the IKK kinases alpha, beta and gamma (the latter known as NEMO) which phosphorylate the inhibitors of B (IBs), liberating the transcriptionally active NF-B molecule (Gilmore, 2006; Perkins, 2007). The first evidence that NF-B regulates CLU expression was provided by Kenneth Marcu and coworkers. In their study, the authors carried out a systematic analysis to isolate all NF-B target genes in mouse embryo fibroblasts. They used a molecular inhibitor of NF-B in the presence or absence of TNF-, a classical NF-B inducer. Among the plethora of genes activated by NF-B, CLU was one of the most highly regulated (Li et al., 2002). Interestingly, knockout of either one of the three IKKs resulted in lack of activation of CLU, suggesting that its activation is dependent on the whole NF-B signalsome. These results were later confirmed by another group that showed that CLU can be induced in glial and astrocyte cells by the bacterial lipopolysaccharide LPS (Saura et al., 2003). LPS is a known activator of NF-B, and
124
Arturo Sala et al.
the use of aspirin or MG132 as indirect means to inhibit NF-B resulted in the inhibition of CLU expression. Intriguingly, it was later shown that CLU regulates NF-B activity in a negative manner by stabilizing IBs (Devauchelle et al., 2006; Santilli et al., 2003; Savkovic et al., 2007; Takase et al., 2008a,b). This leads to the hypothesis that CLU participates in a negative loop in which transcriptional activation of CLU is evoked to dampen NF-B activity. This would be especially important when there is a need to control the secretion of potentially harmful cytokines regulated by NF-B. This hypothesis is corroborated by the study in which it has been shown that abnormally low CLU levels cause excessive NF-B activation and pathological cytokine secretion in rheumatoid arthritis (Devauchelle et al., 2006).
D. Egr1 The group lead by David Boothman was the first to show that secreted CLU is induced by ionizing irradiation (Yang et al., 2000). The same group later showed that irradiation leads to the activation of a signaling pathway that emanates from two growth factors receptors: EGFR and IGFR. It was demonstrated that IGFR, but not EGFR, mediates the induction of secreted CLU in response to irradiation (Criswell et al., 2005). Notably, the Src/Map kinase cascade that is triggered downstream of IGFR ultimately signals to the transcription factor Egr1, which, in turn, binds to the CLU promoter and induces upregulation of CLU mRNA. In this context, secreted CLU is induced as a protective response to damaging stress since knockdown of CLU by RNAi accelerates cell death.
E. Stat1 Stats are a group of transcription factors implicated in transducing survival or apoptotic signaling downstream of a class of receptor-associated molecules called JAKs. In an Affymetrix screen to search for genes involved in conferring resistance to the chemotherapeutic drug docetaxel, Djeu and coworkers identified CLU and Stat-1 as docetaxel-inducible genes that inhibit drug-induced apoptosis. Interestingly, Stat-1 seems to lie upstream of CLU since its depletion by siRNA induces a 50% reduction of CLU expression in prostate cancer cells (Patterson et al., 2006). It is not clear whether Stat-1 can directly regulate CLU gene expression, but the presence of putative Stat-binding sites in the CLU promoter suggests that this could be the case.
Regulation of CLU Gene Expression by Oncogenes and Epigenetic Factors
125
F. GLI and TCF Recent studies have placed CLU downstream of the two protooncogenic transcription factors activated by the signaling molecules Hedgehog and Wnt, GLI-2 and TCF-1, respectively. Hedgehog and Wnt play important roles in normal development and cancer (Jiang and Hui, 2008; Nusse, 1992; Polakis, 2000). Signaling emanating from these developmental factors is relayed to nuclear transcription factors belonging to the GLI and TCF families. Abnormal activation of GLI is often detected in medulloblastomas with disruption of the Sonic Hedgehog antagonist Patched, which results in constitutive activation of Hedgehog signaling (Villavicencio et al., 2000). TCF family members are usually activated in epithelial tumors in which Wnt signaling is increased by stabilization of -catenin, an essential partner in transcriptional regulation (Ilyas, 2005; Rask et al., 2003). The group led by Marin Gleave has found that knockdown of GLI-2 in prostate cancer cells results in suppression of proliferation and increased apoptosis and concurrent inactivation or activation of several genes. Interestingly, CLU protein expression increased after treatment with GLI-2 antisense oligonucleotides, although CLU mRNA expression did not change (Narita et al., 2008). In another study, it was found that TCF-1 mediates activation of a short mRNA isoform of CLU (Schepeler et al., 2007). In both studies, no attempts were made to understand the functional role of CLU in these signaling pathways and whether CLU is a positive or negative modulator of Wnt and Hedgehog pathways remains to be determined.
IV. ONCOGENIC SIGNALING AND EPIGENETIC REGULATION OF CLU EXPRESSION As mentioned in previous paragraphs, c-Myc, N-Myc, and Ras cause silencing of CLU expression, probably facilitating tumorigenesis. The mechanisms of repression by Myc family members appear to be complex and are still a matter of active investigation. Recent studies suggest that RAS-mediated silencing is epigenetic. Analysis of gene expression in rat fibroblasts transformed with activated Ha-RAS revealed that several genes, including CLU, are silenced. Interestingly, RAS first induces deacetylation of the CLU promoter followed by methylation of a CpG island located in proximity of the transcription start site via MEK/ERK signaling (Lund et al., 2006). Curiously, as mentioned in the previous section, the Boothman group has shown that an IGFR-dependent MEK–ERK–EGR signaling pathway mediates activation of CLU by ionizing radiation (Criswell et al., 2005). The apparent contradiction could be explained if one hypothesized that the
126
Arturo Sala et al.
MEK–ERK pathway could feed into different downstream effectors in a stimulus-dependent manner (i.e., irradiation-induced MEK–ERK activates gene transcription via transcription factors whereas ras-induced MEK–ERK inactivates gene expression by inducing histone deacetylases). Other research groups have observed epigenetic silencing of CLU in transformed cells and cancer. For example, Nuutinen et al. have shown that CLU is silenced by gene methylation and deaceylation in human neuroblastoma or neuronal cell lines (Nuutinen et al., 2005). In murine and human prostate cancer cell lines CLU expression is silenced by gene methylation and/or histone deacetylation (Rauhala et al., 2008). In line with these results, the Bettuzzi group had previously shown that CLU expression is downregulated during progression of human and murine prostate cancer and that CLU promotes slowdown of prostate cell proliferation (Bettuzzi et al., 2000, 2002; Caporali et al., 2004; Scaltriti et al., 2004b). Moreover, CLU is one of the genes most highly induced by histone deacetylase inhibitors and inhibitors of DNA methylation in tumor endothelial cells. Most notably, suppression of CLU expression by shRNA drives increased proliferation, migration, and sprouting of tumor endothelial cells (Hellebrekers et al., 2007). Overall, these results invoke a scenario in which oncogenic stimuli provoke chromatin rearrangements that result in suppression of genes, like CLU, that are implicated in restraining tumor proliferation and angiogenesis. Indeed, most recent work from Thomas-Tikhonenko laboratory provides direct evidence that CLU overexpression severely limit neovascularization of murine and human colon carcinomas (Dews et al., submitted for publication) potentially affecting tumor metabolism (see chapter “The shifting balance between CLU forms during tumor progression,” of vol. 104).
V. CONCLUDING REMARKS These are exciting times for researchers studying CLU and cancer. CLU is emerging as an important player in human cancer, although its role is more complex than anticipated. Regarded initially as a mere extracellular chaperone or a scavenging protein, CLU has been proven to be an important mediator of cell signaling as well. Its documented ability to interfere with NF-B, PI3 kinase, or MAP kinase signaling could perhaps explain its role as a tumor modifier. Cancer cells often hijack cellular signaling to their advantage and become “addicted” to a specific molecular pathway. By interfering in a negative or a positive manner with such pathways, CLU could either promote or restrict neoplastic disease. In the light of recent evidence gathered using mouse models of human cancer where CLU has been genetically
Regulation of CLU Gene Expression by Oncogenes and Epigenetic Factors
127
ablated, we hypothesize that CLU is mainly required to restrict the early stages of mammalian tumorigenesis and metastatic spread while assisting established tumors in becoming chemo- and radioresistant. While the mechanism by which CLU acts as a tumor suppressor gene is not entirely clear, there is some evidence to suggest that suppression of the NF-B signaling could be involved. It is tempting to speculate that very aggressive clones of cancer cells arising after chemotherapeutic drug treatments or natural selection could reactivate the expression of CLU. This hypothesis was recently corroborated in experimental models in which initial upregulation of CLU was found to induce clonogenic toxicity, thus killing the majority of prostate cancer cells, while the rare surviving clones were expressing CLU solely in the cytoplasm. This could lead to the development of antiapoptotic properties and the ability to survive the mitotic catastrophe, if only at the cost of acquiring an altered phenotype with impaired mitosis, endoreduplication, and genetic instability (Scaltriti et al., 2004a,c). Therefore, high expression of secreted or cytoplasmic CLU could be advantageous because it confers increased resistance to killing by anticancer drugs or enhances tumor cell survival in specific niches. The opposite roles played by CLU in early versus late stages of tumorigenesis could also explain why epigenetic inactivation of CLU, but not gene rearrangements or mutations, is commonly detected in mammalian cancers.
ACKNOWLEDGMENTS Research in the authors’ laboratories has been supported by: Sporting Aiding Medical Research for Kids (SPARKS) and Neuroblastoma Society (A.S.); FIL 2008 and FIL 2009, University of Parma, Italy, AICR (UK) Grant No. 06-711, and Istituto Nazionale Biostrutture e Biosistemi (INBB), Roma, Italy (S.B.); US National Institutes of Health grant CA 122334 (ATT).
REFERENCES Arango, D., Corner, G. A., Wadler, S., Catalano, P. J., and Augenlicht, L. H. (2001). c-myc/p53 interaction determines sensitivity of human colon carcinoma cells to 5-fluorouracil in vitro and in vivo. Cancer Res. 61, 4910–4915. Bettuzzi, S., Davalli, P., Astancolle, S., Carani, C., Madeo, B., Tampieri, A., Corti, A., Saverio, B., Pierpaola, D., Serenella, A., Cesare, C., Bruno, M., et al. (2000). Tumor progression is accompanied by significant changes in the levels of expression of polyamine metabolism regulatory genes and clusterin (sulfated glycoprotein 2) in human prostate cancer specimens. Cancer Res. 60, 28–34.
128
Arturo Sala et al.
Bettuzzi, S., Scorcioni, F., Astancolle, S., Davalli, P., Scaltriti, M., and Corti, A. (2002). Clusterin (SGP-2) transient overexpression decreases proliferation rate of SV40-immortalized human prostate epithelial cells by slowing down cell cycle progression. Oncogene 21, 4328–4334. Bey, E. A., Wuerzberger-Davis, S. M., Pink, J. J., Yang, C. R., Araki, S., Reinicke, K. E., Bentle, M. S., Dong, Y., Cataldo, E., Criswell, T. L., Wagner, M. W., Li, L., et al. (2006). Mornings with art, lessons learned: Feedback regulation, restriction threshold biology, and redundancy govern molecular stress responses. J. Cell. Physiol. 209, 604–610. Bierie, B., and Moses, H. L. (2006). Tumour microenvironment: TGF : The molecular Jekyll and Hyde of cancer. Nat. Rev. Cancer 6, 506–520. Caporali, A., Davalli, P., Astancolle, S., D’Arca, D., Brausi, M., Bettuzzi, S., and Corti, A. (2004). The chemopreventive action of catechins in the TRAMP mouse model of prostate carcinogenesis is accompanied by clusterin over-expression. Carcinogenesis 25, 2217–2224. Cervellera, M., Raschella, G., Santilli, G., Tanno, B., Ventura, A., Mancini, C., Sevignani, C., Calabretta, B., and Sala, A. (2000). Direct transactivation of the anti-apoptotic gene apolipoprotein J (clusterin) by B-MYB. J. Biol. Chem. 275, 21055–21060. Chayka, O., Corvetta, D., Dews, M., Caccamo, A. E., Piotrowska, I., Santilli, G., Gibson, S., Sebire, N. J., Himoudi, N., Hogarty, M. D., Anderson, J., Bettuzzi, S., et al. (2009). Clusterin, a haploinsufficient tumour suppressor gene in neuroblastomas. J. Natl. Cancer Inst. 101, 663–677. Chi, K. N., Zoubeidi, A., and Gleave, M. E. (2008). Custirsen (OGX-011): A second-generation antisense inhibitor of clusterin for the treatment of cancer. Expert Opin. Investig. Drugs 17, 1955–1962. Chung, J., Kwak, C., Jin, R. J., Lee, C. H., Lee, K. H., and Lee, S. E. (2004). Enhanced chemosensitivity of bladder cancer cells to cisplatin by suppression of clusterin in vitro. Cancer Lett. 203, 155–161. Cote, R., Dunn, M., Chatterjee, S., Stein, J., Shi, S., Tran, Q., Hu, S., Xu, H., Groshen, S., Taylor, C., Skinner, D., and Benedict, W. (1998). Elevated and absent pRb expression is associated with bladder cancer progression and has cooperative effects with p53. Cancer Res. 58, 1090–1094. Criswell, T., Beman, M., Araki, S., Leskov, K., Cataldo, E., Mayo, L. D., and Boothman, D. A. (2005). Delayed activation of insulin-like growth factor-1 receptor/Src/MAPK/Egr-1 signaling regulates clusterin expression, a pro-survival factor. J. Biol. Chem. 280, 14212–14221. Danik, M., Chabot, J. G., Mercier, C., Benabid, A. L., Chauvin, C., Quirion, R., and Suh, M. (1991). Human gliomas and epileptic foci express high levels of a mRNA related to rat testicular sulfated glycoprotein 2, a purported marker of cell death. Proc. Natl. Acad. Sci. USA 88, 8577–8581. Devauchelle, V., Essabbani, A., De Pinieux, G., Germain, S., Tourneur, L., Mistou, S., Margottin-Goguet, F., Anract, P., Migaud, H., Le Nen, D., Lequerre, T., Saraux, A., et al. (2006). Characterization and functional consequences of underexpression of clusterin in rheumatoid arthritis. J. Immunol. 177, 6471–6479. Dews, M., Homayouni, A., Yu, D., Murphy, D., Sevignani, C., Wentzel, E., Furth, E. E., Lee, W. M., Enders, G. H., Mendell, J. T., and Thomas-Tikhonenko, A. (2006). Augmentation of tumor angiogenesis by a Myc-activated microRNA cluster. Nat. Genet. 38, 1060–1065. Eilers, M., Picard, D., Yamamoto, K. R., and Bishop, J. M. (1989). Chimaeras of Myc oncoprotein and steroid receptors cause hormone-dependent transformation of cells. Nature 340, 66–68. Evan, G. I., and Vousden, K. H. (2001). Proliferation, cell cycle and apoptosis in cancer. Nature 411, 342–348. Finnberg, N., and El-Deiry, W. S. (2008). TRAIL death receptors as tumor suppressors and drug targets. Cell Cycle 7, 1525–1528.
Regulation of CLU Gene Expression by Oncogenes and Epigenetic Factors
129
Gilmore, T. D. (2006). Introduction to NF-kappaB: Players, pathways, perspectives. Oncogene 25, 6680–6684. Hellebrekers, D. M., Melotte, V., Vire, E., Langenkamp, E., Molema, G., Fuks, F., Herman, J. G., Van, C. W., Griffioen, A. W., and Van, E. M. (2007). Identification of epigenetically silenced genes in tumor endothelial cells. Cancer Res. 67, 4138–4148. Herault, Y., Chatelain, G., Brun, G., and Michel, D. (1992). V-src-induced-transcription of the avian clusterin gene. Nucleic Acids Res. 20, 6377–6383. Hueber, A. O., Zornig, M., Lyon, D., Suda, T., Nagata, S., and Evan, G. I. (1997). Requirement for the CD95 receptor-ligand pathway in c-Myc-induced apoptosis. Science 278, 1305–1309. Ilyas, M. (2005). Wnt signalling and the mechanistic basis of tumour development. J. Pathol. 205, 130–144. Ito, K., Bernardi, R., Morotti, A., Matsuoka, S., Saglio, G., Ikeda, Y., Rosenblatt, J., Avigan, D. E., Teruya-Feldstein, J., and Pandolfi, P. P. (2008). PML targeting eradicates quiescent leukaemia-initiating cells. Nature 453, 1072–1078. Jiang, J., and Hui, C. C. (2008). Hedgehog signaling in development and cancer. Dev. Cell 15, 801–812. Jin, G., and Howe, P. H. (1997). Regulation of clusterin gene expression by transforming growth factor . J. Biol. Chem. 272, 26620–26626. Jin, G., and Howe, P. H. (1999). Transforming growth factor regulates clusterin gene expression via modulation of transcription factor c-Fos. Eur. J. Biochem. 263, 534–542. John, B., Enright, A. J., Aravin, A., Tuschl, T., Sander, C., and Marks, D. S. (2004). Human microRNA targets. PLoS Biol. 2, e363. Kadomatsu, K., Anzano, M. A., Slayter, M. V., Winokur, T. S., Smith, J. M., and Sporn, M. B. (1993). Expression of sulfated glycoprotein 2 is associated with carcinogenesis induced by N-nitroso-N-methylurea in rat prostate and seminal vesicle. Cancer Res. 53, 1480–1483. Klefstrom, J., Vastrik, I., Saksela, E., Valle, J., Eilers, M., and Alitalo, K. (1994). c-Myc induces cellular susceptibility to the cytotoxic action of TNF-alpha. EMBO J. 13, 5442–5450. Klock, G., Storch, S., Rickert, J., Gutacker, C., and Koch-Brandt, C. (1998). Differential regulation of the clusterin gene by Ha-ras and c-myc oncogenes and during apoptosis. J. Cell. Physiol. 177, 593–605. Koch-Brandt, C., and Morgans, C. (1996). Clusterin: A role in cell survival in the face of apoptosis? Prog. Mol. Subcell. Biol. 16, 130–149. Li, X., Massa, P. E., Hanidu, A., Peet, G. W., Aro, P., Savitt, A., Mische, S., Li, J., and Marcu, K. B. (2002). IKK, IKK , and NEMO/IKK are each required for the NF-Bmediated inflammatory response program. J. Biol. Chem. 277, 45129–45140. Lipsick, J. S., Manak, J., Mitiku, N., Chen, C. K., Fogarty, P., and Guthrie, E. (2001). Functional evolution of the Myb oncogene family. Blood Cells Mol. Dis. 27, 456–458. Lund, P., Weisshaupt, K., Mikeska, T., Jammas, D., Chen, X., Kuban, R. J., Ungethum, U., Krapfenbauer, U., Herzel, H. P., Schafer, R., Walter, J., and Sers, C. (2006). Oncogenic HRAS suppresses clusterin expression through promoter hypermethylation. Oncogene 25, 4890–4903. Lutz, W., Leon, J., and Eilers, M. (2002). Contributions of Myc to tumorigenesis. Biochim. Biophys. Acta 1602, 61–71. Markopoulou, S., Kontargiris, E., Batsi, C., Tzavaras, T., Trougakos, I., Boothman, D. A., Gonos, E. S., and Kolettas, E. (2009). Vanadium-induced apoptosis of HaCaT cells is mediated by c-fos and involves nuclear accumulation of clusterin. FEBS J. 276, 3784–3799. Massague, J. (2008). TGF in cancer. Cell 134, 215–230. Mateyak, M. K., Obaya, A. J., Adachi, S., and Sedivy, J. M. (1997). Phenotypes of c-Mycdeficient rat fibroblasts isolated by targeted homologous recombination. Cell Growth Differ. 8, 1039–1048.
130
Arturo Sala et al.
McLaughlin, L., Mistry, M., Zhu, G., Ley-Ebert, C., Stuart, W. D., Florio, C. J., Groen, P. A., Witt, S. A., Kimball, T. R., Witte, D. P., Harmony, J. A., and Aronow, B. J. (2000). Apolipoprotein J/Clusterin limits the severity of murine autoimmune myocarditis. J. Clin. Invest. 106, 1105–1113. Mendell, J. T. (2008). miRiad roles for the miR-17–92 cluster in development and disease. Cell 133, 217–222. Meyer, N., and Penn, L. Z. (2008). Reflecting on 25 years with MYC. Nat. Rev. Cancer 8, 976–990. Michel, D., Gillet, G., Volovitch, M., Pessac, B., Calothy, G., and Brun, G. (1989). Expression of a novel gene encoding a 51.5 kD precursor protein is induced by different retroviral oncogenes in quail neuroretinal cells. Oncogene Res. 4, 127–136. Miyake, H., Hara, I., Kamidono, S., Gleave, M. E., and Eto, H. (2003). Resistance to cytotoxic chemotherapy-induced apoptosis in human prostate cancer cells is associated with intracellular clusterin expression. Oncol. Rep. 10, 469–473. Miyake, H., Hara, I., and Gleave, M. E. (2005). Antisense oligodeoxynucleotide therapy targeting clusterin gene for prostate cancer: Vancouver experience from discovery to clinic. Int. J. Urol. 12, 785–794. Morgenbesser, S. D., and DePinho, R. A. (1994). Use of transgenic mice to study myc family gene function in normal mammalian development and in cancer. Semin. Cancer Biol. 5, 21–36. Morris, E. J., Ji, J. Y., Yang, F., Di Stefano, L., Herr, A., Moon, N. S., Kwon, E. J., Haigis, K. M., Naar, A. M., and Dyson, N. J. (2008). E2F1 represses beta-catenin transcription and is antagonized by both pRB and CDK8. Nature 455, 552–556. Nakajima, T., Yasui, K., Zen, K., Inagaki, Y., Fujii, H., Minami, M., Tanaka, S., Taniwaki, M., Itoh, Y., Arii, S., Inazawa, J., and Okanoue, T. (2008). Activation of B-Myb by E2F1 in hepatocellular carcinoma. Hepatol. Res. 38, 886–895. Narita, S., So, A., Ettinger, S., Hayashi, N., Muramaki, M., Fazli, L., Kim, Y., and Gleave, M. E. (2008). GLI2 knockdown using an antisense oligonucleotide induces apoptosis and chemosensitizes cells to paclitaxel in androgen-independent prostate cancer. Clin. Cancer Res. 14, 5769–5777. Nusse, R. (1992). The Wnt gene family in tumorigenesis and in normal development. J. Steroid Biochem. Mol. Biol. 43, 9–12. Nuutinen, T., Suuronen, T., Kyrylenko, S., Huuskonen, J., and Salminen, A. (2005). Induction of clusterin/apoJ expression by histone deacetylase inhibitors in neural cells. Neurochem. Int. 47, 528–538. O’Donnell, K. A., Wentzel, E. A., Zeller, K. I., Dang, C. V., and Mendell, J. T. (2005). c-Mycregulated microRNAs modulate E2F1 expression. Nature 435, 839–843. Oh, I. H., and Reddy, E. P. (1999). The myb gene family in cell growth, differentiation and apoptosis. Oncogene 18, 3017–3033. Orlandi, A., Pucci, S., Ciucci, A., Pichiorri, F., Ferlosio, A., and Spagnoli, L. G. (2005). Modulation of clusterin isoforms is associated with all-trans retinoic acid-induced proliferative arrest and apoptosis of intimal smooth muscle cells. Arterioscler. Thromb. Vasc. Biol. 25, 348–353. Patterson, S. G., Wei, S., Chen, X., Sallman, D. A., Gilvary, D. L., Zhong, B., Pow-Sang, J., Yeatman, T., and Djeu, J. Y. (2006). Novel role of Stat1 in the development of docetaxel resistance in prostate tumor cells. Oncogene 25, 6113–6122. Perkins, N. D. (2007). Integrating cell-signalling pathways with NF-kappaB and IKK function. Nat. Rev. Mol. Cell. Biol. 8, 49–62. Pierce, A. M., Schneider-Broussard, R., Gimenez-Conti, I. B., Russell, J. L., Conti, C. J., and Johnson, D. G. (1999). E2F1 has both oncogenic and tumor-suppressive properties in a transgenic model. Mol. Cell. Biol. 19, 6408–6414. Polakis, P. (2000). Wnt signaling and cancer. Genes Dev. 14, 1837–1851.
Regulation of CLU Gene Expression by Oncogenes and Epigenetic Factors
131
Raschella, G., Cesi, V., Amendola, R., Negroni, A., Tanno, B., Altavista, P., Tonini, G. P., De Bernardi, B., and Calabretta, B. (1999). Expression of B-myb in neuroblastoma tumors is a poor prognostic factor independent from MYCN amplification. Cancer Res. 59, 3365–3368. Rask, K., Nilsson, A., Brannstrom, M., Carlsson, P., Hellberg, P., Janson, P. O., Hedin, L., and Sundfeldt, K. (2003). Wnt-signalling pathway in ovarian epithelial tumours: Increased expression of beta-catenin and GSK3beta. Br. J. Cancer 89, 1298–1304. Rauhala, H. E., Porkka, K. P., Saramaki, O. R., Tammela, T. L., and Visakorpi, T. (2008). Clusterin is epigenetically regulated in prostate cancer. Int. J. Cancer 123, 1601–1609. Ricci, S., Jin, Z., Dews, M., Yu, D., Thomas-Tikhonenko, A., Dicker, D. T., and El-Deiry, W. S. (2004). Direct repression of FLIP expression by c-myc is a major determinant of TRAIL sensitivity. Mol. Cell. Biol. 24, 8541–8555. Sala, A., and Watson, R. (1999). B-Myb protein in cellular proliferation, transcription control, and cancer: Latest developments. J. Cell. Physiol. 179, 245–250. Sallman, D. A., Chen, X., Zhong, B., Gilvary, D. L., Zhou, J., Wei, S., and Djeu, J. Y. (2007). Clusterin mediates TRAIL resistance in prostate tumor cells. Mol. Cancer Ther. 6, 2938–2947. Santilli, G., Aronow, B. J., and Sala, A. (2003). Essential requirement of apolipoprotein J (clusterin) signaling for IkappaB expression and regulation of NF-kappaB activity. J. Biol. Chem. 278, 38214–38219. Santilli, G., Schwab, R., Watson, R., Ebert, C., Aronow, B. J., and Sala, A. (2005). Temperaturedependent modification and activation of B-MYB: Implications for cell survival. J. Biol. Chem. 280, 15628–15634. Saura, J., Petegnief, V., Wu, X., Liang, Y., and Paul, S. M. (2003). Microglial apolipoprotein E and astroglial apolipoprotein J expression in vitro: Opposite effects of lipopolysaccharide. J. Neurochem. 85, 1455–1467. Savkovic, V., Gantzer, H., Reiser, U., Selig, L., Gaiser, S., Sack, U., Kloppel, G., Mossner, J., Keim, V., Horn, F., and Bodeker, H. (2007). Clusterin is protective in pancreatitis through antiapoptotic and anti-inflammatory properties. Biochem. Biophys. Res. Commun. 356, 431–437. Scaltriti, M., Bettuzzi, S., Sharrard, R. M., Caporali, A., Caccamo, A. E., and Maitland, N. J. (2004a). Clusterin overexpression in both malignant and nonmalignant prostate epithelial cells induces cell cycle arrest and apoptosis. Br. J. Cancer 91, 1842–1850. Scaltriti, M., Brausi, M., Amorosi, A., Caporali, A., D’Arca, D., Astancolle, S., Corti, A., and Bettuzzi, S. (2004b). Clusterin (SGP-2, ApoJ) expression is downregulated in low- and highgrade human prostate cancer. Int. J. Cancer 108, 23–30. Scaltriti, M., Santamaria, A., Paciucci, R., and Bettuzzi, S. (2004c). Intracellular clusterin induces G(2)-M phase arrest and cell death in PC-3 prostate cancer cells. Cancer Res. 64, 6174–6182. Schepeler, T., Mansilla, F., Christensen, L. L., Orntoft, T. F., and Andersen, C. L. (2007). Clusterin expression can be modulated by changes in TCF1-mediated Wnt signaling. J. Mol. Signal. 2, 6. Takase, O., Marumo, T., Hishikawa, K., Fujita, T., Quigg, R. J., and Hayashi, M. (2008a). NFkappaB-dependent genes induced by proteinuria and identified using DNA microarrays. Clin. Exp. Nephrol. 12, 181–188. Takase, O., Minto, A. W., Puri, T. S., Cunningham, P. N., Jacob, A., Hayashi, M., and Quigg, R. J. (2008b). Inhibition of NF-kappaB-dependent Bcl-xL expression by clusterin promotes albumin-induced tubular cell apoptosis. Kidney Int. 73, 567–577. Tarasov, K. V., Tarasova, Y. S., Tam, W. L., Riordon, D. R., Elliott, S. T., Kania, G., Li, J., Yamanaka, S., Crider, D. G., Testa, G., Li, R. A., Lim, B., et al. (2008). B-MYB is essential for normal cell cycle progression and chromosomal stability of embryonic stem cells. PloS ONE 3, e2478.
132
Arturo Sala et al.
Thomas-Tikhonenko, A., Viard-Leveugle, I., Dews, M., Wehrli, P., Sevignani, C., Yu, D., Ricci, S., El-Deiry, W. S., Aronow, B., Kaya, G., Saurat, J.-H., and French, L. E. (2004). Myc-transformed epithelial cells down-regulate clusterin which inhibits their growth in vitro and carcinogenesis in vivo. Cancer Res. 64, 3126–3136. Trumpp, A., Refaeli, Y., Oskarsson, T., Gasser, S., Murphy, M., Martin, G. R., and Bishop, J. M. (2001). c-Myc regulates mammalian body size by controlling cell number but not cell size. Nature 414, 768–773. Viale, A., De Franco, F., Orleth, A., Cambiaghi, V., Giuliani, V., Bossi, D., Ronchini, C., Ronzoni, S., Muradore, I., Monestiroli, S., Gobbi, A., Alcalay, M., et al. (2009). Cell-cycle restriction limits DNA damage and maintains self-renewal of leukaemia stem cells. Nature 457, 51–56. Villavicencio, E. H., Walterhouse, D. O., and Iannaccone, P. M. (2000). The sonic hedgehogpatched-gli pathway in human development and disease. Am. J. Hum. Genet. 67, 1047–1054. Vita, M., and Henriksson, M. (2006). The Myc oncoprotein as a therapeutic target for human cancer. Semin. Cancer Biol. 16, 318–330. Volinia, S., Calin, G. A., Liu, C. G., Ambs, S., Cimmino, A., Petrocca, F., Visone, R., Iorio, M., Roldo, C., Ferracin, M., Prueitt, R. L., Yanaihara, N., et al. (2006). A microRNA expression signature of human solid tumors defines cancer gene targets. Proc. Natl. Acad. Sci. USA 103, 2257–2261. Yang, C. R., Leskov, K., Hosley-Eberlein, K., Criswell, T., Pink, J. J., Kinsella, T. J., and Boothman, D. A. (2000). Nuclear clusterin/XIP8, an X-ray-induced Ku70-binding protein that signals cell death. Proc. Natl. Acad. Sci. USA 97, 5907–5912. Yaylim-Eraltan, I., Bozkurt, N., Ergen, A., Zeybek, U., Ozturk, O., Arikan, S., Erbil, Y., Uslu, I., Camlica, H., and Isbir, T. (2008). L-myc gene polymorphism and risk of thyroid cancer. Exp. Oncol. 30, 117–120. Zindy, F., Eischen, C. M., Randle, D. H., Kamijo, T., Cleveland, J. L., Sherr, C. J., and Roussel, M. F. (1998). Myc signaling via the ARF tumor suppressor regulates p53-dependent apoptosis and immortalization. Genes Dev. 12, 2424–2433.
Conclusions and Perspectives Saverio Bettuzzi Dipartimento di Medicina Sperimentale, Sezione di Biochimica, Biochimica Clinica e Biochimica dell’Esercizio Fisico, Universita` di Parma, Via Volturno 39-43100 Parma and Istituto Nazionale Biostrutture e Biosistemi (I.N.B.B.), Rome, Italy
I. II. III. IV.
Introduction Take Home Messages from Chapters of Vols. 104 and 105 General Consensus Open Issues A. Regulation of CLU Gene Expression, Production of CLU Protein Forms, and Cell Trafficking B. How Important is the Tissue Context in CLU Action During Cell Transformation? C. How Critical is the Role of CLU Alone? D. Which are the other Pieces of the Puzzle? V. Possible Clinical Implementations VI. Future Research References Since its first discovery, CLU has been fascinating many researchers around the world, probably because CLU has always been a surprise. In many studies, CLU showed up unexpected as a secondary but significant observation in the hands of scientists which probably had no specific intention to study it. Too often the first descriptions of the structure, action, and biological meaning of CLU had to be changed in the light of novel and rather surprising findings, making what was thought to be well established, actually obsolete. CLU is a biological object still in movement. To understand more, we will have to cope correctly with this challenge. As an Ariadne’s thread, following CLU we will challenge our static ideas pointing to alternative views which will possibly lead us to a different thinking and a much better understanding of the complexity of cell transformation. In this final chapter, we will provide “take home messages” from the chapters of vols. 104 and 105 concerning the most up to date knowledge about CLU. Then we will provide some agreements and consensus in this field, wherever possible. Highlighting possible faults or misunderstandings, we will define the open issues which will very likely drive the research on CLU in the near future. Finally, we will address how this novel knowledge about CLU may possibly result in important implementations, from basic science all the way to the clinical setting, in the fight against cancer. # 2009 Elsevier Inc.
I. INTRODUCTION Since its first discovery, CLU has been fascinating many researchers around the world. The reason for that probably relays on the fact that CLU has always been a surprise. In too many of the studies reported in the Advances in CANCER RESEARCH Copyright 2009, Elsevier Inc. All rights reserved.
133
0065-230X/09 $35.00 DOI: 10.1016/S0065-230X(09)05008-8
134
Saverio Bettuzzi
literature CLU showed up unexpected, as a sort of a secondary but significant observation made from scientists which probably had no specific intention to study it. And too often the first descriptions of the structure, action, and biological meaning of CLU had to be changed in the light of novel and rather surprising findings. What was thought to be well established, actually it was not. CLU is an object still in movement. What we are trying to depict here is not a static picture of CLU, but a movie in which CLU is acting under our eyes. But, besides a theoretical interest, good for the academy, what is good for the people in practical? CLU had been constantly challenging who was trying to understand, so why should we keep studying this odd gene? The reward can be actually very important, because CLU is a sort of an Ariadne’s thread, challenging our static ideas, pointing to alternative views, obliging us to a different thinking: in other words, constantly leading us to a much better understanding of the complexity of such an important process as cell transformation. To understand more, we will have to cope correctly with this challenge by finding agreements and consensus wherever possible, highlighting possible faults or misunderstandings, defining the open issues which will drive the new research on CLU. What we have been able to discover up to date will possibly allow soon important implementations in the clinical setting. This is what we will try to address in this final chapter.
II. TAKE HOME MESSAGES FROM CHAPTERS OF VOLS. 104 AND 105 Introduction: CLU has been found differentially expressed and nearly ubiquitous in tissues and body fluids. A critical review of the story of its discovery shows that CLU is a single copy gene in mammalian genomes, but the proteomic profile of CLU was found rather complex since the first studies. Therefore, CLU has to be considered, rather than a single protein, more as a family of protein products of almost unknown structure with different intra- and extracellular localization. Clusterin (CLU): From one gene and two transcripts to many proteins: The molecular mechanism of production of these protein forms remains unclear. Alternative transcriptional initiation, possibly driven by two distinct promoters, may produce at least two distinct CLU mRNA isoforms, differing in their unique first exon, named Isoform 1 and Isoform 2. A third transcript, named Isoform 11036, has been recently found as one of the most probable mRNA variants. Three in-frame ATGs have been located in these transcripts, which may be functionally activated under specific conditions in
Conclusions
135
an alternative fashion. There are no direct experimental evidences describing the structure of possibly different/alternative CLU protein forms originated by these transcripts. The shifting balance between CLU forms during tumor progression: It has been found that cell transformation is not only strictly linked to changes in the expression of tumor suppressors and oncogenes, but important metabolic changes also occur. These are instrumental for initial survival of cancer cells, mostly when lack of oxygen is not compensated by neoangiogenesis. In this view, induction of anaerobic glycolysis is fundamental. Later, also the spreading of disease requires late metabolic changes which will be responsible for progression and clinical emerging of cancer. In this view, several metabolic genes have been found implicated in this multistep process, among which are fatty acid synthase (FASN) and carnitine palmitoyl transferase I (CPT I). An intriguing link between these metabolic shifts and a change in the balance between nuclear and secreted forms of CLU (nCLU/ sCLU) is suggested. The shifting balance between CLU forms during tumor progression would affects the fate of the cell and be instrumental to the metabolic shift occurring in the different steps of tumor progression. For these reasons, determination of CLU expression in early phases of cell transformation has been found to be a useful biomarker for molecular diagnosis and prognosis of prostate cancer in both animal models and clinical settings. Regulation of CLU expression by calcium: The production of at least two CLU forms (secreted, sCLU and nuclear, nCLU) is differentially regulated by changes in Ca2þ fluxes inside the cell. The range of intracellular Ca2þ concentrations affecting the expression of CLU gene is very narrow. The availability of Ca2þ also affects important posttranscriptional and posttranslational events which are fundamental for the synthesis, processing, and intracellular localization of CLU. Ca2þ may importantly affect the final destiny of the cell by inducing nCLU translocation in the nucleus (apoptosis) or sCLU excretion (survival). Nuclear CLU (nCLU) and the fate of the cell: In the vast majority of the experimental systems studied, specific stimuli may inhibit the secretion and the maturation of CLU, leading at least one protein product to target the nucleus (nCLU). We know little about the structure of nCLU. Cell transfection experiments showed that absence of the leader peptide is sufficient to produce biological relevant amounts of nCLU. The biological effect caused by nuclear targeting of CLU is induction of cell death in both benign and cancer cells, at least at initial stage of transformation. Cells resistant to apoptosis seem to inhibit CLU entering the nucleus. Nuclear targeting of CLU seems to be independent from the three NLS signals present in its sequence, therefore the mechanism of shuttling in and out the nucleus still needs to be elucidated.
136
Saverio Bettuzzi
The chaperone action of CLU and its putative role in quality control of extracellular protein folding: Fully processed and secreted CLU (sCLU) has been found to be an extracellular chaperone with a putative role in quality control of extracellular protein folding. Secreted CLU may protect extracellular spaces from protein deposition which is potentially pathological. Many serious human diseases are characterized by the deposition of extracellular protein aggregates. This knowledge may lead to the development of novel strategies to combat these diseases. Cell protective functions of secretory CLU (sCLU): The secreted form of CLU (sCLU) seems to be cytoprotective in many experimental systems. The activity of sCLU includes prevention of lipid oxidation and cell damage in blood vessels, removal of dead cell remnants in regressing tissues, and clearance of harmful extracellular molecules, such as amyloid beta, by endocytosis or transcytosis. The activity of sCLU may depend by its propensity to bind to a wide spectrum of hydrophobic molecules. In addition, sCLU may bind specific cell surface receptors also modulating signal transduction. CLU, chronic inflammation, and autoimmunity: Recent data show that sCLU can directly and indirectly affect inflammation and immunity in several ways: (i) regulating complement activation; (ii) as a negative regulator of NF-B; (iii) promoting pro- or antiapoptotic activity according to the specific CLU protein form produced (i.e., sCLU/nCLU); (iv) by modulating cell differentiation; (v) by regulating major proinflammatory cytokines such as TNF- and IL-6. For instance, sCLU is an inhibitor of complement activation, protecting cells or tissues from membrane attack complement complex. In addition, secretion of sCLU is stimulated by endotoxin (LPS), tumor necrosis factor (TNF), and interleukin (IL)-1. All these properties of sCLU have a strong impact in autoimmune diseases. In particular, intracellular CLU affects NF-B signaling and apoptosis by inducing IB stabilization. Thus, reduced production of intracellular forms of CLU, often observed in pathological conditions, could be a key mechanism responsible for the greater activation status of NF-B found in several inflammatory conditions, such as those also occurring in the synovium. CLU regulation by IL-6 could have a strong relevance in autoimmune conditions. In this exciting field, the next challenges would be to unravel the mechanisms inducing downregulation of CLU expression in RA synovium and discovering the precise role of CLU and its protein forms in the differentiation of cells such as Th-17. In addition, we need to know how important are the posttranslational modification of CLU for autoantibodies recognition. The definition of the role of CLU in the regulation of apoptosis, drug resistance, and proinflammatory response of targeted cells will be fundamental also in understanding autoimmune reaction. CLU as a novel biosensor of reactive oxygen species: Sequence comparison studies of CLU gene and promoter showed a very high degree of
Conclusions
137
conservation among mammalian species. Considering the wide tissue distribution of CLU, its implication in normal processes like development and differentiation and the absence of functional CLU gene polymorphisms in humans clearly indicate that the protein has evolved in vertebrates to accomplish one or more functions of fundamental biological importance. Based on recent observations, CLU seems to be a sensitive biosensor of generic environmental insults and particularly oxidative stress. Oxidative stress is the driving force of most age-related diseases including cancer, and CLU has been functionally implicated in all these pathological conditions. The main mission of CLU would be to cease the deleterious effects of oxidative stress. If this is true, increased CLU gene expression during ageing should reflect the general oxidative status of the subject rather than its chronological age. In favor of this hypothesis is the finding that CLU is upregulated in lymphocyte samples from old donors as compared to their young counterparts. Moreover, in healthy centenarians CLU gene expression level is significantly lower than those found in old donors. On this basis, sCLU could be a valuable prognostic biomarker of increased body stress and disease progression. Also, sCLU, or other CLU protein forms, may represent promising targets for the development of future therapeutic strategies against pathologies related to oxidative stress such as cancer, inflammation, neurodegenerative diseases, or cataract. Besides its role as extracellular chaperone, to better understand this exciting field it will be fundamental to assess its specific role in mitochondria activity, its implications in the regulation of cell signaling pathways, its oxidation status during ageing and finally the mechanism of its responsiveness to oxidative stress, which is probably regulating the production of different CLU forms. CLU and prostate cancer: The role of CLU in prostate tumorigenesis is probably the most highly controversial. Many research teams have produced data on this issue. Possible biases explaining controversial reports may be due to different experimental models (i.e., cell lines and culturing conditions) or tools (i.e., antibodies against CLU) or also to different ways to interpret the same data. Therefore, contradictions and alternative hypothesis still exist. Understanding is further complicated by the existence of different protein form of CLU, that is sCLU and nCLU, with different biological meaning. Nevertheless, it should be now fully acknowledged the fundamental information that CLU is downregulated in vivo in prostate cancer and in the vast majority of other cancers as well (please see the publicly available Oncomine database at http://www.oncomine.org). At the moment, upregulation of CLU has been confirmed only in vitro or in some cancer cells, especially after adjuvant hormonal therapy, where sCLU may act as a prosurvival factor. On the other hand, studies recently published showed that in vivo CLU may indeed act as a tumor suppressor gene.
138
Saverio Bettuzzi
Consistently, CLU expression is downregulated during the early stages of cancer progression, but restored to high levels in TRAMP mice responding to chemoprevention with Green Tea Catechins (EGCG). Since no mutations of CLU have been found in human cancer yet, the likely mechanism of inactivation is epigenetic, as corroborated by the frequent observation of CpG island methylation or histone deacetylation in the proximity of the CLU gene in different cancers. A reconciliation with previous findings suggesting CLU to be a prosurvival oncogene probably relies in the fact that the process of tumorigenesis often exploits tumor suppressor genes to its own purposes in late stages of progression, when the tumor-suppressor factor is inactivated or acquiring improper activity. Therefore, early phase-associated events related to physiological action must be distinguished from late stages-associated ones. CLU may conduct a “double life”: inhibition of NF-B signaling may suppress tumorigenesis and metastatic spread. But since a proapoptotic NF-B signaling is often involved in replication stress induced by chemotherapeutic drugs, highly malignant cells could reactivate CLU expression to suppress NF-B and survive. The obsolete knowledge about CLU has prompt to some clinical trials. Women with metastatic breast cancer have been treated with CLU antisense in combination with docetaxel without experiencing any improvement in the disease progression. A similar approach did not result in exciting clinical improvement in prostate cancer patients so far. A possible novel alternative is targeting the early phases of prostate cancer by overexpressing CLU and nCLU with appropriate drugs and tools. This approach is probably more rewarding and likely to result in important clinical improvements in the next future. CLU and breast cancer: CLU expression has been studied in vitro and in vivo in breast cancer with the aim to identify new prognostic markers of cancer recurrence and to develop new and more effective therapies. Secreted (sCLU) and cytoplasmic CLU is believed to play a role in promoting breast tumorigenesis and progression, although its clinical utility as a survival prognostic factor remains to be clarified. The presence of many proteins forms with diverse and opposite effects constitutes a further obstacle to better define the utility of CLU as a biomarker or a possible target for breast cancer. Data suggest sCLU to be a cell survival factor induced following Tamoxifen therapy and chemotherapy, therefore positively involved in acquisition of drug resistance. Inhibition of CLU expression using antisense oligonucleotides and antibodies has been found to enhance the cytotoxic effects of chemotherapy agents, including Paclitaxel and Doxorubicin, in vitro. Based on these promising data, a phase II trial was conducted using the combination of CLU antisense oligonucleotides and Docexatel. The study is now closed and the result was disappointing: the therapy was
Conclusions
139
well tolerated, but there were an insufficient number of responses to meet the criteria for proceeding to the second stage of accrual. CLU and colon cancer: Screening is considered as an effective way to prevent colon cancer, leading to an important drop in the incidence of this cancer in recent years. Anyway, there is less consensus regarding optimal screening strategies. Putative approaches include proteomics-based testing, stool genetic testing, radiological imaging, and enhanced endoscopies. At present, colonoscopy is considered the gold standard for colon cancer diagnosis, with a sensitivity of 97% and a specificity of 98%, also offering the possibility to both diagnose and remove premalignant lesions, but is associated with patient discomfort, complications, variability in the sensitivity of detection by endoscopy, and high costs. In this view, a useful diagnostic assay must be sensitive, ideally detecting cancer at early onset, and highly specific to minimize false positives. Stool testing, unlike other conventional screening approaches, is noninvasive and requires no cathartic preparation. New stool tests for CRC diagnosis have been recently developed. In particular, a specificity of about 95% has been reported for tests based on detection of genetic mutations. On the other hand, some markers such as calprotectin do not show high sensitivity, leading to nearly 30% of false positive. Recently, a high specific serum testing for colon cancer-specific antigen 2and 4 (CCSA-2-4) has been proposed, but not all colon cancers do express this antigen. Colon cancer progression is characterized by loss of nCLU and increased sCLU production. The sCLU form is extracellularly released both in blood and stool. On this basis, a sensitive and specific method to detect sCLU as a new biomarker for noninvasive colon cancer screening has been designed. Furthermore, data on animal models showed that the increase in sCLU level and release correlates with tumor size. These data are in favor of the fact that a shift in the nCLU/sCLU production is an important event in colon cancer progression. Secreted CLU can be envisaged as a new biomarker for diagnosis and prognosis of colon cancer. Studies to unravel the molecular mechanisms regulating the activity of CLU promoter and CLU forms shifting are necessary to detect new molecular targets for specific therapies. CLU and lung cancer: There are several studies reported in the literature about the potential role of CLU in lung cancer, but results are often conflicting. Studies conducted in vitro using lung cancer cell lines showed high levels of CLU expression in lung cancer cells. In these cell lines, CLU seems to have a cytoprotective role. Particularly, cytoplasmic CLU or secreted CLU (sCLU) should be involved in cell resistance against chemo- and radiotherapy. Clinical studies may depict a different story. A recent study analyzed the prognostic role of CLU in patients with early nonsmall cell lung cancer who underwent pulmonary resection, showing that high CLU levels in lung tissue correlated with a better prognosis after 3 years from surgical
140
Saverio Bettuzzi
intervention. Various explanations may justify this discrepancy: (i) cancer cell lines may not be the best bench to challenge an hypothesis on CLU action, because they usually display many DNA mutations often related to resistance and positive selection during cell culturing, which may potently alter cell homeostasis (this may also apply to other cancer cell systems); (ii) the type of antibodies used in these studies performed in human lung biopsies do not distinguish different CLU protein forms. Data available on Oncomine database from three independent studies on lung adenocarcinoma, squamous cell lung carcinoma, and small cell lung cancer, showed that CLU was downregulated in tumors as compared to normal lung with a further reduction of its expression during the progression of the malignancy. CLU and chemoresistance: Resistance to anticancer agents is one of the main reasons for failure in effective cancer therapy. Drug resistance may rise by both intrinsic and acquired mechanisms which are still largely unknown. This is why any new knowledge in the field may be of great importance in the fight against cancer. Recently, CLU has been described as a key contributor to chemoresistance in several tumors, including prostate, melanoma, osteosarcoma, and ovarian cancer. In advanced stage and metastatic cancers (such as renal, bladder, breast, head and neck, colon, cervical, pancreatic, lung carcinomas, and lymphoma), CLU was found upregulated. Very importantly, only the cytoplasmic/secretory sCLU, and not nCLU, is expressed in aggressive late stage tumors. Thus, lack of nCLU and high expression of sCLU in these cells may cause acquisition of an overall antiapoptotic function. Most significantly, sCLU expression is documented to lead to broad-based resistance to several unrelated chemotherapeutic agents such as doxorubicin, cisplatin, etoposide and camphothecin, and also to targeted death-inducing factors among which are tumor necrosis factor, TRAIL, and histone deacetylase inhibitors. It is reasonable to believe that induction of sCLU expression may be an adaptive response to genotoxic and oxidative stress. The threat to malignant cells being treated with cytotoxic agents may result to “survival of the best” in an evolutionary fashion, therefore only cells with enhanced survival potential will escape death. Recruitment of sCLU in late stage of cancer may be instrumental to survival, although the actual mechanisms for sCLU induction and action are largely unknown. STAT1 is required for its constitutive upregulation in docetaxel-resistant tumor cells. sCLU, by stabilizing Ku70/Bax complexes, may sequester Bax and indirectly inhibit mitochondrial release of cytochrome c, the main trigger of apoptosis. These data, suggesting sCLU as a key element in preventing apoptosis induced by cytotoxic agents, also indicate that sCLU has the potential to be targeted for cancer therapy, but to this aim a specific tools correctly affecting the balance sCLU/nCLU is needed.
Conclusions
141
CLU and tumor microenvironment: Cancer cells need to interact synergistically with their surrounding microenvironment to survive, generate a neoplasm and progress further to colonize distant organs. The microenvironment exerts profound epigenetic effects on cancer cells. Tumor progression also requires immune escaping mimetism. In addition, the induction of a synergistic cooperation among transformed and stromal cells, competing for space and resources such as oxygen and nutrients, is required. Therefore, the extracellular milieu and tissue microenviroment heterotypic interactions cooperate to promote tumor growth, angiogenesis, and cancer cell motility through elevated secretion of pleiotropic cytokines and soluble factors. CLU represent one of the numerous cell factors sharing the inner intracellular information with the microenvironment. CLU is also systemically diffused at tight joining and involved in the “In and the Out” of the cell. Its role in this regard is still debated. CLU is still regarded as a multifunction factor. Alternatively, we still need to unravel a possible common function still masked. The knowledge accumulated so far focus on the importance of balanced expression of its different intra- and extracellular forms, most importantly nCLU/sCLU, a balance that most likely depends on the intraand extracellular microenvironment cross talk. The complicated balance of cytokines network and the regulation of CLU forms production in cancer and stromal cells undoubtedly represents an adaptative response linked to genomic stability and bystander effects induced by oxidative stresses and cell damage. Tumor–microenvironment interactions are supposed to be strictly involved in controlling local cancer growth and invasion, and play a major role in the regulation of production and release of different forms of CLU. In addition, the extracellular form of this protein, sCLU, seems to have pleiotropic actions that may play a crucial role in redirecting stromal functions and altering intracellular communications by binding cell surface receptors. In addition, sCLU may influence the secretion of pleiotropic chemokines in a paracrine and autocrine fashion. Further elucidation of the functions of CLU protein forms inside and outside the benign cell, as well as “in and out” of cancer cell, are clearly needed for a deeper understanding of the interplay between tumor and stroma during cancer progression. Regulation of CLU gene expression by oncogenes, epigenetic regulation and the role of CLU in tumorigenesis: These are exciting times for researchers studying the role of CLU in tumorigenesis. Regarded initially as an extracellular chaperone or a scavenging protein, we now know that CLU is also an important mediator of cell signaling, thanks to its documented ability to interfere with NF-B, PI3Kinase, or MAP kinase signaling. CLU is regarded now as a tumor modifier and an important player in human cancer, although its role is more complex than anticipated.
142
Saverio Bettuzzi
Cancer cells often hijack cellular signaling to their advantage and become “addicted” to a specific molecular pathway. By interfering in a negative or positive manner with such pathways, CLU could either promote or inhibit tumor onset and progression. Considering the complexity of tumorigenesis, which proceed through many pathways and steps, CLU appear to be mainly required to restrict the early stages of mammalian tumorigenesis and metastatic spread. Therefore, at initial stages of cancer progression CLU is regarded as a tumor suppressor gene, whose mechanism of action is not entirely clear. Its role as suppressor of the NF-B signaling is very likely a key factor in this scenario. Data shows that the main action of CLU in benign cells is to act as a proapoptotic factor (nCLU), a negative regulator of cell proliferation and migration, and very likely a differentiation agent, thus taking part in the complex mechanism of cell homeostasis. Therefore, downregulation of CLU expression through epigenetic mechanisms seems to be necessary for initial cancer progression. In late stages of tumorigenesis, data suggest that aggressive clones of cancer cells arising after chemotherapeutic drug treatments or natural selection could reactivate the expression of CLU. In this view, cancer cells inhibiting nuclear targeting and expressing CLU solely in the cytoplasm or as secreted form (sCLU) would be characterized by development of apoptosis-inhibitory properties and genetic instability. Under these conditions, high expression of sCLU could be a selective advantage by conferring increased resistance to drugs killing or enhance tumor cell survival in specific niches. The opposite role played by CLU in early versus advanced stages of tumorigenesis could also explain why epigenetic inactivation of CLU, but not gene rearrangements or mutations, is commonly detected in mammalian cancers.
III. GENERAL CONSENSUS CLU is present as a highly homolog single copy gene in mammalian genomes. The CLU gene codes for several distinct protein products. The proteomic profile of CLU is complex. CLU has to be considered a family of proteins and not a single protein. The protein products are detectable in different locations inside and outside the cell. In the many systems studied, CLU gene is generally found downregulated. With few exceptions, the basal level of CLU expression is moderate-low. The majority of CLU in the body is secreted in the extracellular compartment (blood and body fluids) as mature, fully processed, glycosylated form of
Conclusions
143
CLU (sCLU). The secreted CLU can be found both as heterodimer and tetramer. Several different stress stimuli can induce CLU gene expression. Besides quantitative regulation of CLU gene expression, CLU protein forms production and localization inside the cell is differentially regulated. Following specific stimuli, the localization of CLU inside the cell may change. The localization of CLU is fundamental for the fate of the cell. When secretion of CLU is inhibited, the cell is loaded with CLU and nuclear targeting of CLU is dominant (nCLU). nCLU is a prodeath protein form, capable to rapidly trigger anoikis cell death, characterized by early cell detachment followed by caspase-dependent apoptosis. When maturation and secretion of CLU is induced, sCLU is expelled in the extracellular compartment in high amounts. sCLU protects cells from death probably by binding a wide spectrum of hydrophobic molecules, misfolded proteins, and ligands. The general consensus among the researchers is that the ratio sCLU/nCLU drives the fate of the benign cell. This also applies to initial stages of transformation. During the cell transformation process, CLU expression is usually found: 1. downregulated in early steps; 2. upregulated in late, advanced cancer. Recent experimental data shows that CLU may act as a: 1. cell killer in the beginning, through the activity of proapoptotic nCLU (i.e., nCLU is a tumor suppressor factor); 2. prosurvival later, in chemoresistant cells, through the activity of sCLU as an extracellular chaperone (i.e., sCLU is a tumor inducer factor). Therefore, detection of CLU forms and localization in tissues and cells requires appropriate methods and tools (i.e., appropriate anti-CLU antibodies). Correct detection of the localization of CLU in intact cells is fundamental for the correct interpretation of the data. The tissue context and microenvironment is very important for the definition of the specific role of CLU in tumorigenesis. CLU is a short-lived protein(s) rapidly degraded inside the cancer cell through the proteasome following polyubiquitination. This may explain why the levels of CLU are usually very low in primary tumors. In addition, cancer cells seem to inhibit nuclear localization of CLU through still unknown mechanisms. Epigenetic regulation of CLU expression is very likely very important in tumorigenesis. Low levels or absence of CLU expression eventually found in cancer cells is not an irreversible condition. Thus, similarly to what happens
144
Saverio Bettuzzi
in the case of many other tumor suppressors, CLU may be recruited are late phases of tumorigenesis and during acquisition of chemo-resistance.
IV. OPEN ISSUES A. Regulation of CLU Gene Expression, Production of CLU Protein Forms, and Cell Trafficking The regulation of CLU gene expression is sketchy. More information is needed concerning the molecular mechanisms through which different CLU mRNA isoforms, as well as different protein forms, are produced. In this context, and considering the potential important role of CLU in tumorigenesis, it appears fundamental to gain more information on the epigenetic regulation of CLU expression.
B. How Important is the Tissue Context in CLU Action During Cell Transformation? The tissue context certainly plays an important role in CLU action. Oncomine (http://www.oncomine.org) is a research platform for online cancer gene expression analysis dedicated to the academic and nonprofit research community. It contains data from nearly 30,000 microarrays experiments performed in 41 cancer types. When the platform is queried for the level of expression of CLU in benign tissues in comparison to cancer, the answer is downregulation in all cases, with the only exceptions of brain and renal cancer. This independent data suggest that, although the tissue context is important, absence or reduced expression of CLU are widely diffused events in cell transformation.
C. How Critical is the Role of CLU Alone? There are growing evidences concerning the fact that CLU action requires several molecular partners. Due to the high propensity of CLU to bind other proteins/ligands, also depending on pH and local conditions, the difficult task would be to distinguish the physiological interactions from those due to the experimental conditions, mostly in cell-free systems. Several potential partners of CLU have been identified, both inside and outside the cell. The inventories of active molecules in the extracellular compartment, inside the cell and or on the outer membrane are very different, therefore it is rationale
Conclusions
145
to believe that sCLU and nCLU may interact with different targets and have diverse action.
D. Which are the other Pieces of the Puzzle? Evidences are showing that CLU can bind cell receptors (i.e., megalin) and affect signal transduction. In addition, CLU may act in an autocrine and a paracrine fashion. Unravelling these aspects will be fundamental in the next future.
V. POSSIBLE CLINICAL IMPLEMENTATIONS Some early reports have suggested that neoplasias have increased CLU expression. On the basis of this evidence, clinical trials based on CLU as target have been attempted. In particular, patients were injected with a CLU antisense oligonucleotide, called OGX-011 (Custirsen) to ablate CLU expression and to induce tumor regression. In vitro, OGX-011 has been used alone or in combination with other drugs or other antisense targeting different transcripts. No beneficial effects have ever been reported using OGX-011 alone. The clinical trials described on the NCI Web site (www.cancer.gov) are based on a combination of OGX-011 and other anticancer drugs. Trials on prostate cancer (NCT00054106), breast cancer (NCT00258375), and solid tumors (NCT00471432) have now been completed. The results are somehow disappointing. Side effects (dose-related chills and fever, for instance) have been observed as specific toxic effects. In the case of prostate cancer (Phase I) and solid tumors (Phase I), no major clinical advantage for the patients has been reported so far. Nevertheless, a Phase II study has been conducted for breast cancer. The trial is now closed and unfortunately there had been an insufficient number of responses to meet the criteria for proceeding to the second stage of accrual. In conclusion, at the moment none of the known clinical trials have resulted in clear clinical advantages for the patients. Two more studies have been recently closed and recruitment of patients has been stopped: lung cancer (NCT00138658) and hormone refractory prostate cancer (HRPC; NCT00327340). The final reports are expected at the end of 2009 for both studies. In the case of the lung, the positive correlation with survival that has been demonstrated using the high expression of CLU as a single predictor (see chapter “CLU and lung cancer,” this volume) may anticipate a negative result. All these information, publicly available, bring us to the temporary conclusion that clinical trials based on antisense strategy have been generally disappointing. A possible explanation is that too often
146
Saverio Bettuzzi
in vitro evidences judged as very promising have not been confirmed in vivo or in real clinical settings. The lesson that we can get out of this is a warning: CLU is an object much more complicated than we think. Starting points and rationales based on data obtained in vitro in cancer cell lines are often not reliable. Similarly to what was happened in many other cases in the past, an experimental confirmation in a suitable animal model of disease is necessary before attempting a clinical trial. Particular attention should be paid to studies in which CLU expression as been evaluated as single prognostic factor for cancer patient’s survival. As cited before, in the case of lung cancer (Albert et al., 2007) a better survival was associated to CLU expression in the cytoplasm of nonsmall cell lung cancer; in prostate cancer (Pins et al., 2004), CLU staining in cancer cells was negligible and not correlated to progression, while CLU staining in the stroma was a negative predictor of survival. This work is a confirmation of our previous data, and this conclusion has been anticipated in our previous work (Scaltriti et al., 2004). Therefore, considering that high expression of CLU in specific cell types, tissues or compartments can be a single predictor of survival, we may speculate that production of CLU (probably nCLU) is an obstacle to cancer progression, at least before drug resistance has occurred. Thus, strategy aimed at downregulating CLU expression aspecifically with regard to local compartments may be potentially harmful for patients (Rizzi and Bettuzzi, 2008). Detection of CLU by immunohistochemistry, Northern blot hybridization or real time quantitative PCR in prostate cancer specimens, alone or together with other markers, can be used as a biomarker for diagnosis (Bettuzzi et al., 2000; Rizzi et al., 2008) or prognosis (Bettuzzi et al., 2003). In particular, CLU was found significantly downregulated in patients with Gleason Score higher than 7 (Rizzi et al., 2008). Gleason Score higher than 7 is the most significant predictor of negative outcome in prostate cancer patient. These data have contributed to the validation of a novel gene signature for the diagnosis of prostate cancer (Rizzi et al., 2008). Therefore, in the next future detection of CLU gene expression, at transcriptional or translational level, in specific biological targets may be very useful as developing novel biomarkers of disease progression. Epigenetic regulation of CLU expression is now recognized as a key element for understanding CLU action (Suuronen et al., 2007). This issue may bear particular importance in tumorigenesis. In particular, CLU has been identified as an epigenetically silenced gene and found to be a negative regulator of endothelial cell growth and angiogenesis (Hellebrekers et al., 2007). Very importantly, CLU was found methylated in the mouse prostate cancer TRAMP-C2 and in the human prostate cancer LNCaP cell lines (Rauhala et al., 2008). In the same work, it was shown that CLU expression is significantly reduced in untreated and hormone-refractory human prostate
Conclusions
147
carcinomas. These recent results are in agreement with our findings in human prostate cancer (Scaltriti et al., 2004) and in the TRAMP mouse model, which displays in situ and invasive carcinoma of the prostate mimicking the whole spectrum of human prostate cancer progression from PIN to androgen-independent disease (Kaplan-Lefko et al., 2003). The TRAMPC2 cells have been derived from TRAMP mice. In the TRAMP model, we found (Caporali et al., 2004) that CLU is expressed at basal levels in young TRAMP mice before tumor onset. Then CLU is potently downregulated during prostate cancer progression. Administration of a purified extract of Catechins from Green Tea (GTCs) to TRAMP mice, very rich in EGCG, showed an overall 80% efficacy at preventing tumor progression. The activity of GTCs appeared to be mediated by induction of CLU expression and particularly nCLU in target cells. The changes in CLU expression detected in the TRAMP model during tumor progression and also following GTCs treatment have been confirmed both at transcriptional (Scaltriti et al., 2006) and translational levels (Caporali et al., 2004). The only rationale explanation for this result is that GTCs and EGCG are potent epigenetic regulator of CLU expression. The fact that potent anticancer agents actually under development such as EGCG (Caporali et al., 2004), Polyphenon E (Chow et al., 2001; McLarty et al., 2009), proteasome inhibitors (Rizzi et al., 2009), and DNA methyltransferase (DNMT) or histone deacetylase (HDAC) inhibitors such as 5-aza-2’-deoxycytidine and trichostatin A (Hellebrekers et al., 2007) can restore CLU expression in prostate cancer cells devoid of CLU suggest that low levels or absence of CLU expression in cancer cells is not an irreversible condition, not being a consequence of permanent genetic loss. On the other hand, epigenetic regulation of CLU expression may explain why, as it has been shown in the case of many other tumor suppressors, CLU may be recruited is late phases of tumorigenesis and during acquisition of chemoresistance. The effectiveness of a potential therapy aimed at knocking down CLU in drug-resistant cancer cells remains to be established in vivo in appropriate clinical settings. As suggested from the evidences brought in this book, an exciting alternative intervention could be to enhance CLU expression levels and blocking CLU secretion in order to trigger nCLU overload in the nucleus of target cancer cells, mostly in the early phases of tumorigenesis. EGCG seems to work exactly through this modality of action. It has been recently demonstrated that successful attack to the early tumor lesions (Bettuzzi et al., 2006) to achieve durable benefits for patients (Brausi et al., 2008) is actually possible in the clinical setting. This approach can be envisaged as an early anticancer therapy that should be effective before drug resistance has occurred, that is, in the vast majority of diagnosed clinical cases. This would open a novel exciting scenario in the
148
Saverio Bettuzzi
fight against cancer with a potential important impact in Western countries, where chemoprevention is supposed to be the new frontier in the next future. To get this important result, new tools based on the novel knowledge about CLU and consistent technical approaches are needed. Vectors driving tissuespecific expression of nCLU in target cells will have to be developed. Drugs (i.e., inhibitors of the proteasome) or natural compounds (i.e., EGCG and GTCs), used alone or in combination, which have been shown to induce nCLU probably by affecting the epigenetic regulation of CLU expression, will very likely play a major role in the next future for combating cancer. In conclusion, CLU is now seen as an important tumor modifier, mainly required to restrict the early stages of mammalian tumorigenesis and metastatic spread, while aggressive clones of cancer cells arising after chemotherapeutic drug treatments or natural selection may reactivate the expression of CLU to survive. Beside this role, CLU has been recently demonstrated involved in important pathological processes including deposition of protein and other harmful extracellular molecules, prevention of lipid oxidation and cell damage, autoimmune diseases, inflammatory conditions, and autoantibodies recognition. In addition, CLU has been proposed as a biosensor of generic environmental insults, a biomarker of increased body stress, and very importantly a marker of cancer recurrence. This knowledge will very likely bring important implementations for patients in the next future.
VI. FUTURE RESEARCH In the next year, a better definition of the followings issues is required: 1. The mechanisms through which the cell controls the production of the different CLU forms and their localization; 2. Identification of the specific stimuli affecting sCLU/nCLU expression ratio; 3. Definition of the actual structure of different CLU forms by purification and crystallization. Definition of possible active sites, docking sites, and potential molecular partners; 4. Identification of CLU form-specific epitopes and production of CLUform specific antibodies (which are not commercially available at the moment); 5. Definition of the diverse sites of intracellular action (membranes, organelles, cytoskeleton) and cognate receptors; 6. Elucidation of CLU autocrine and paracrine action; 7. Identification of the physiological molecular partners of CLU, both outside and inside the cell;
Conclusions
149
8. Assessing whether the different forms of CLU, detected with appropriate specific tools, can be used as biomarkers for specific diseases; 9. Design of vectors and tools for manipulating the expression of specific CLU forms in different tissue contexts/diseases; 10. Clinical trials against malignancies based on the novel knowledge about CLU. To this aim, we should take advantage of point 9 and use drugs or natural compounds specifically driving the epigenetic regulation of CLU expression in target cells.
ACKNOWLEDGMENTS Grant sponsor: FIL 2008 and FIL 2009, University of Parma, Italy; AICR (UK) Grant No. 06-711; Istituto Nazionale Biostrutture e Biosistemi (INBB), Roma, Italy.
REFERENCES Albert, J. M., et al. (2007). Cytoplasmic clusterin expression is associated with longer survival in patients with resected non small cell lung cancer. Cancer Epidemiol. Biomarkers Prev. 16, 1845–1851. Bettuzzi, S., et al. (2000). Tumor progression is accompanied by significant changes in the levels of expression of polyamine metabolism regulatory genes and clusterin (sulfated glycoprotein 2) in human prostate cancer specimens. Cancer Res. 60, 28–34. Bettuzzi, S., et al. (2003). Successful prediction of prostate cancer recurrence by gene profiling in combination with clinical data: A 5-year follow-up study. Cancer Res. 63, 3469–3472. Bettuzzi, S., et al. (2006). Chemoprevention of human prostate cancer by oral administration of green tea catechins in volunteers with high-grade prostate intraepithelial neoplasia: A preliminary report from a one-year proof-of-principle study. Cancer Res. 66, 1234–1240. Brausi, M., et al. (2008). Chemoprevention of human prostate cancer by green tea catechins: Two years later. A follow-up update. Eur. Urol. 54, 472–473. Caporali, A., et al. (2004). The chemopreventive action of catechins in the TRAMP mouse model of prostate carcinogenesis is accompanied by clusterin over-expression. Carcinogenesis 25, 2217–2224. Chow, H. H., et al. (2001). Phase I pharmacokinetic study of tea polyphenols following singledose administration of epigallocatechin gallate and polyphenon E. Cancer Epidemiol. Biomarkers Prev. 10, 53–58. Hellebrekers, D. M., et al. (2007). Identification of epigenetically silenced genes in tumor endothelial cells. Cancer Res. 67, 4138–4148. Kaplan-Lefko, P. J., et al. (2003). Pathobiology of autochthonous prostate cancer in a preclinical transgenic mouse model. Prostate 55, 219–237. McLarty, J., et al. (2009). Tea polyphenols decrease serum levels of prostate-specific antigen, hepatocyte growth factor, and vascular endothelial growth factor in prostate cancer patients and inhibit production of hepatocyte growth factor and vascular endothelial growth factor in vitro. Cancer Prev. Res. (Phila Pa) 2, 673–682.
150
Saverio Bettuzzi
Pins, M. R., et al. (2004). Clusterin as a possible predictor for biochemical recurrence of prostate cancer following radical prostatectomy with intermediate Gleason scores: A preliminary report. Prostate Cancer Prostatic Dis. 7, 243–248. Rauhala, H. E., et al. (2008). Clusterin is epigenetically regulated in prostate cancer. Int. J. Cancer 123, 1601–1609. Rizzi, F., and Bettuzzi, S. (2008). Targeting Clusterin in prostate cancer. J. Physiol. Pharmacol. 59(Suppl. 9), 265–274. Rizzi, F., et al. (2008). A novel gene signature for molecular diagnosis of human prostate cancer by RT-qPCR. PLoS ONE 3, e3617. Rizzi, F., et al. (2009). Clusterin is a short half-life, poly-ubiquitinated protein, which controls the fate of prostate cancer cells. J. Cell. Physiol. 219, 314–323. Scaltriti, M., et al. (2004). Clusterin (SGP-2, ApoJ) expression is downregulated in low- and high-grade human prostate cancer. Int. J. Cancer 108, 23–30. Scaltriti, M., et al. (2006). Molecular classification of green tea catechin-sensitive and green tea catechin-resistant prostate cancer in the TRAMP mice model by quantitative real-time PCR gene profiling. Carcinogenesis 27, 1047–1053. Suuronen, T., et al. (2007). Epigenetic regulation of clusterin/apolipoprotein J expression in retinal pigment epithelial cells. Biochem. Biophys. Res. Commun. 357, 397–401.
Index
A
Adenomatous polyposis coli (APC) gene, 46 Antisense oligonucleotide (ASO), 7 Antisense therapy, CLU cytotoxicity enhancement, 7 neoadjuvant hormone therapy, 9 OGX‐011, 8 resistance mechanism, 8 Apoptosis inhibition, CLU CLU expression, 49 Ku70, 49–50 Ku70/80–CLU–Bax interaction pathological interaction, 51 physiological interaction, 50
B
Bax, 106 B‐MYB, 4 Breast carcinomas, CLU cytoplasmic CLU and apoptosis, 24–25 distribution, 22 localization, 23–24 nuclear CLU (nCLU) function, 23 prognostic significance, 30–31 resistance antiestrogen, 34–35 chemotherapy, 31–34 dexamethasone, 35–36 secreted CLU (sCLU) function, 22–23 tumorigenesis and progression immunohistochemical analysis, 27 malignant epithelium, 26 MHC class I relation, 29–30 protective effect, 28 p53 suppression, 29 sCLU overexpression, 25 tumor suppressor gene, 25 BXL‐628, calcitriol analogue, 10
C
Caco‐2 cells, 57, 107 Castration‐resistant prostate cancer (CRPC), 7 Chemoresistance, sCLU androgen independence, 79 castration‐induced programmed cell death, 78 tumor necrosis factor (TNF) treatment, 78–79 Chemosensitization blockade strategies adaptive response, 89 phosphorothioate antisense oligonucleotide, 88 Chronic inflammation, 95–96 Clusterin (CLU) antisense therapy cytotoxicity enhancement, 7 OGX‐011, 8 resistance mechanism, 8 breast carcinoma resisitance antiestrogen, 34–35 chemotherapy, 31–34 dexamethasone, 35–36 Ca2þ fluxes, 135 caspase‐dependent apoptosis, 143 cell transformation process, 143 chaperone action, 136 chemoresistance, 140 chronic inflammation, and autoimmunity, 136 clinical implementations CLU expression, 146 epigenetic regulation, 146–147 novel biomarker, 146 OGX‐011 and drugs, 145 role, 148 colon cancer, 139 differential splicing forms, 95–96
151
152 Clusterin (CLU) (continued) epigenetic regulation, 143–144 gene expression regulation cancer genes, 117 cell signaling interference, 126–127 clonogenic toxicity, 127 CLU expression enhancement, 119 epigenetic regulation, 125–126 expression modulation, 116–117 genes categorization, 116 multifunctional gene, 116 Myc, 117–118 oncogenes, 119–121 primary tumors expression, 118 protooncogenic transcription factors, 121–124 role, 116 ionizing radiation (IR) and induction of, 96 isoform balance, 135 issues, 144–145 lung cancer, 139–140 microenvironment, in and out effects hypoxia inducible factor (HIF) regulation, 101–103 IL‐6 role, 103–105 sCLU Induction, 106–108 transforming growth factors‐ (TGF ), 98–101 up‐and downstreamsignals, 108–111 VEGF‐A activity, 105–106 novel biosensor, 136–137 nuclear CLU (nCLU), 135 oncogenes, gene expression regulation, 141–142 prostate cancer NF‐B signaling, 138 up and downregulation, 137–138 protein products, 142 secretory CLU (sCLU) protective functions, 136 stress stimuli, 143 transcriptional initiation, 134–135 tumor microenvironment, 141 Colon cancer, CLU apoptosis inhibition CLU expression, 49 Ku70, 49–50 Ku70/80–CLU–Bax interaction, 50–51 biomarker
Index diagnostic marker, 57 dot blot analysis, 57 early diagnosis protocols, 56 sCLU level, 57–58 cancer progression balanced expression, 50–51 CLU translocation, 53 cytoplasmic CLU overexpression, 55–56 elevated expression, 51–52 isoforms and tumor correlation, 52 multiple intestinal neoplasia (Min) mutation, 51 Myc‐transformed epithelial cells, 52 nCLU RNA and protein levels, 55 node‐negative and positive carcinomas, 53–54 p21, 52 sCLU overexpression, 56 somatostatin treatment, 55 triple immunostaining and confocal analysis, 54–55 tumor microenvironment, 55 Western blot analysis, 53 diagnostic assay, 59–60 genetic instability and DNA damage control checkpoint genes alteration, 47 Ku proteins interaction, 48 low penetrance genes, 47 microsatellite instability (MSI), 46 multistep progression, 46 nCLU, 58–59 neoplastic progression, 46–47 sCLU upregulation, 59 tumor‐specific modulation, 53
D
Dexamethasone, 35–36
E
Epithelial–mesenchymal transition (EMT), 67
F
Familial adenomatous polyposis syndrome (FAP), 46 Fecal occult blood test (FOBTs), 56
153
Index
G
Genotoxic and oxidative stresses, 85–86
H
HER2, 31 Her2, 85 Histone deacetylase (HDAC) inhibitors, 85 Hydrogen peroxide (H2O2), 82 Hypoxia inducible factor (HIF) regulation, 101–103
I
Interleukin‐6 (IL‐6) role acute‐phase inflammatory response, 104 aging, 104 neoplastic cell growth regulation, 104–105 tumor microenvironmental factors, 103 Irradiation and oxidative stress, with sCLU, 82–83
K
Ku70, 106–107
L
LNCAP tumor cells, 82 Lung cancer carcinogenesis isoform activity, 66–67 microarray technology, 67 tobacco smoke exposure, 65–66 clinical studies, 73–74 CLU expression and prognosis, 71–72 cytoprotective role, 72–73 Kaplan–Meier survival curve, 72 physiological processes, 65 progression/metastasis CLU gene, 68 CLU levels, 67–68 epithelial–mesenchymal transition (EMT), 67 survival rate, 64 treatment cell type dependence, 68–69 CLU ASO plus Paclitaxel, 69–70 CLU serum levels, 69 CLU silencing, OGX‐011, 70 immunostaining, 70 OGX‐011, 68 types, 64
M
Major histocompatibility complex (MHC) antigens, 29–30 Megalin, 109 Microecosystem, normal and cancer cells activated fibroblast effects, 97 afferent and efferent pathways, 98 epithelial–mesenchymal reciprocal interactions, 97 inflammatory signals, 97 neoangiogenesis, 97 Microenvironment effects, clusterin hypoxia inducible factor (HIF) regulation CLU production, 103 glycolysis, 101 PI3K and MAPK pathway activation, 102 STAT3 activation, 102–103 IL‐6 role acute‐phase inflammatory response, 104 aging, 104 neoplastic cell growth regulation, 104–105 tumor microenvironmental factors, 103 sCLU induction IL‐6 treatment, 107 Ku–Bax physical interactions, 106–108 transforming growth factors‐ (TGF ), 98–101 up‐and downstream signals antiapoptotic activity, 109 CLU‐megalin interaction, 109 EGFR activation requirement, 109 epithelial–mesenchymal transition (EMT)‐inducing effect, 110–111 MAPK and PI3K/AKT signaling activation, 110 neutrophil regulation, 108–109 PI3K–Akt pathway activation, 109 VEGF‐A activity, 105–106 Microsatellite instability (MSI), 46 Multidrug resistance, with sCLU antisense‐CLU, 81–82 cisplatin, 81 MYB, 123
N
Neuropilin‐1 (NP1), 105 NF‐6B, 123–124 Non‐small cell lung cancer (NSCLC), 64
154
O
OGX‐011, 33–34, 69, 88 Oncogenes, gene expression regulation c‐Fos expression, 120 c‐MYC and Ha‐RAS, 120–121 T64, 120 Oncomine, 6–7, 65 Oxidative stress, with sCLU, 82–83, 85–86
P
p21, 52 Prostate cancer chemoprevention carcinogenesis, 9 COX‐2 inhibitors, 9 definition, 9 finasteride, 10 green tea catechins (GTCs), 10–11 vitamin D production, 10 clinical relevance adenocarcinoma grading system, 3 prostate‐specific antigen (PSA) screening, 3 treatment, 3–4 CLU antisense therapy cytotoxicity enhancement, 7 OGX‐011, 8–9 resistance mechanism, 8 CLU expression, in human PCa localization, 5–6 metastatic stage, 6–7 negative tumor modulator, 6 oncomine, 6 CLU paradox, 2 CLU signaling, cell lines epigenetic modifications, 5 expression regulation, 4 oncogenes and oncoproteins downregulation, 4 physiological role, 4 mRNA level, 1–2 negative affects, 15 PNT1A cells, 13–14 prostate epithelium, 13 tumor modulator CLU expression loss, 11–12 MYCN protooncogene, 12 TRAMP mice, 12 Protooncogenic transcription factors Egr1, 124 GLI and TCF, 124
Index MYB, 123 MYC, 121–122 NF‐6B, 123–124 Stat1, 124 Protooncogenic transcription factors, CLU gene regulation Egr1, 124 GLI and TCF, 124 MYB, 123 MYC C‐MYC, 121 neuronal MYC (MYCN), 122 TGF‐ signaling pathway, 122 NF‐6B, 123–124 Stat1, 124
S
Secretory clusterin (sCLU) chemoresistance androgen independence, 79 castration‐induced programmed cell death, 78 taxanes, 80 tumor necrosis factor (TNF) treatment, 78–79 chemosensitization blockade strategies adaptive response, 89 phosphorothioate antisense oligonucleotide, 88 genotoxic and oxidative stresses, 85–86 induction and cytoprotection mechanism, 84–85 irradiation and oxidative stress, 82–83 multidrug resistance antisense‐CLU, 81–82 cisplatin, 81 progressive tumors benign and malignant tissues, 83 upregulation, 84 targeted therapy, 84–85 Small cell lung cancer (SCLC), 64 Stat1, 124
U
U‐2OS osteosarcoma cells, 82
V
VEGF‐A activity, 105–106