Nuclear Mechanics & Genome Regulation, Volume 98 (Methods in Cell Biology)

Series Editors Leslie Wilson Department of Molecular, Cellular and Developmental Biology University of California Santa...

Author: G.V. Shivashankar

20 downloads 481 Views 19MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

Series Editors Leslie Wilson Department of Molecular, Cellular and Developmental Biology University of California Santa Barbara, California

Paul Matsudaira Department of Biological Sciences National University of Singapore Singapore

Methods in Cell Biology VOLUME 98 Nuclear Mechanics & Genome Regulation

Edited by

G.V. Shivashankar Department of Biological Sciences and Research Center for Excellence in MechanoBiology, National University of Singapore, Singapore & National Center for Biological Sciences, TIFR-Bangalore, India

AMSTERDAM • BOSTON • HEIDELBERG • LONDON

NEW YORK • OXFORD • PARIS • SAN DIEGO

SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO

Academic Press is an imprint of Elsevier

Academic Press is an imprint of Elsevier 30 Corporate Drive, Suite 400, Burlington, MA 01803, USA 525 B Street, Suite 1900, San Diego, CA 92101-4495, USA 32, Jamestown Road, London NW1 7BY, UK Linacre House, Jordan Hill, Oxford OX2 8DP, UK First edition 2010 Copyright © 2010 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 web site 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–13: 978-0-12-381009-0 ISSN: 0091-679X For information on all Academic Press publications visit our website at elsevierdirect.com

Printed and bound in USA 10 11 12 13 10 9 8 7 6 5 4 3 2 1

Working together to grow libraries in developing countries www.elsevier.com | www.bookaid.org | www.sabre.org

CONTRIBUTORS

Numbers in parentheses indicate the pages on which the authors’ contributions begin.

Padra Ahmadi, (295) Mechanics and Genetics of Embryonic and Tumoral Develop ment group, UMR168 CNRS, Institut Curie, 11 rue Pierre et Marie Curie, F-75005, Paris, France Soneela Ankam, (241) Division of Bioengineering, National University of Singapore, Singapore Dipanjan Bhattacharya, (57) National Centre for Biological Sciences, Tata Institute of Fundamental Research, Bellary Road, Bangalore 560065, India Lesley Y. Chan, (241) Division of Bioengineering, National University of Singapore, Singapore; NUS Graduate School of Integrative Science and Engineering, National University of Singapore, Singapore; and Bioprocessing Technology Institute, A*Star, Singapore Matthew W. C. Chan, (179) Matrix Dynamics Group, Faculty of Dentistry, Univer sity of Toronto, Fitzgerald Building, Toronto, ON, Canada M5S 3E2 Kris Noel Dahl, (97) Department of Chemical Engineering and Department of Bio medical Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213 Rumi De, (143) Indian institute of Science Education and Research, Kolkata, Mohanpur 741252, Nadia, West Bengal, India Dennis E. Discher, (207) Biophysical Engineering Laboratory, School of Engineering and Applied Science, University of Pennsylvania, Philadelphia, Pennsylvania 19104 Emmanuel Farge, (295) Mechanics and Genetics of Embryonic and Tumoral Devel opment group, UMR168 CNRS, Institut Curie, 11 rue Pierre et Marie Curie, F-75005, Paris, France Maria-Elena Fernandez-Sanchez, (295) Mechanics and Genetics of Embryonic and Tumoral Development group, UMR168 CNRS, Institut Curie, 11 rue Pierre et Marie Curie, F-75005, Paris, France Ying-Hui Fu, (337) Department of Neurology, University of California, San Fran cisco, San Francisco, California 94158 2324 Sanjeev Galande, (35) National Centre for Cell Science, Ganeshkhind, Pune 411007, India and Indian Institute of Science Education and Research, Pashan, Pune 411021, India Soumya Gupta, (57) National Centre for Biological Sciences, Tata Institute of Funda mental Research, Bellary Road, Bangalore 560065, India and Department of Bio logical Sciences and Research Center for Excellence in MechanoBiology, National University of Singapore, Singapore 117543, Singapore Takamasa Harada, (207) Biophysical Engineering Laboratory, School of Engineering and Applied Science, University of Pennsylvania, Philadelphia, Pennsylvania 19104 ix

x

Contributors

Peter Hemmerich, (3) Leibniz Institute of Age Research, Fritz Lipman Institute, Beutenbergstr. 11, 07745 Jena, Germany Boris Hinz, (179) Matrix Dynamics Group, Faculty of Dentistry, University of Toronto, Fitzgerald Building, Toronto, ON, Canada M5S 3E2 Irena Ivanovska, (207) Biophysical Engineering Laboratory, School of Engineering and Applied Science, University of Pennsylvania, Philadelphia, Pennsylvania 19104 K. Venkatesan Iyer, (221) National Centre for Biological Sciences, Tata Institute of Fundamental Research, Bellary Road, Bangalore 560065, India and Department of Biological Sciences and Research Center for Excellence in MechanoBiology, National University of Singapore, Singapore 117543 Ranveer S. Jayani, (35) National Centre for Cell Science, Ganeshkhind, Pune 411007, India Benjamin Kim Kiat Teo, (241) Division of Bioengineering, National University of Singapore, Singapore and Mechanobiology Institute Singapore, National University of Singapore, Singapore Karolin Klement, (3) Leibniz Institute of Age Research, Fritz Lipman Institute, Beutenbergstr. 11, 07745 Jena, Germany Abhishek Kumar, (221) National Centre for Biological Sciences, Tata Institute of Fundamental Research, Bellary Road, Bangalore 560065, India and Department of Biological Sciences and Research Center for Excellence in MechanoBiology, National University of Singapore, Singapore 117543 Frank P.L. Lai, (323) Institute of Medical Biology, Immunos, 8A Biomedical Grove, Biopolis, Singapore 138648 Jan Lammerding, (121) Department of Medicine, Cardiovascular Division, Brigham and Women’s Hospital/Harvard Medical School, Boston, Massachusetts 02115 Thorsten Lenser, (3) Institute of Computer Science, Friedrich-Schiller-University, Ernst Abbe Platz 2, 07743 Jena, Germany and Carl Zeiss MicroImaging GmbH, Carl-Zeiss-Promenade 10, 07745 Jena, Germany Qingsen Li, (79) Mechanobiology Institute, National University of Singapore, Singapore 117411, Singapore Maria Lucia Lombardi, (121) Department of Medicine, Cardiovascular Division, Brigham and Women’s Hospital/Harvard Medical School, Boston, Massachusetts 02115 Shovamayee Maharana, (57) National Centre for Biological Sciences, Tata Institute of Fundamental Research, Bellary Road, Bangalore 560065, India and Department of Biological Sciences and Research Center for Excellence in MechanoBiology, National University of Singapore, Singapore 117543, Singapore Aprotim Mazumder, (221) National Centre for Biological Sciences, Tata Institute of Fundamental Research, Bellary Road, Bangalore 560065, India Christopher A. McCulloch, (179) Matrix Dynamics Group, Faculty of Dentistry, University of Toronto, Fitzgerald Building, Toronto, ON, Canada M5S 3E2 Radfidah A. Mutalif, (323) Institute of Medical Biology, Immunos, 8A Biomedical Grove, Biopolis, Singapore 138648 Quasar Saleem Padiath, (337) Department of Neurology, University of California, San Francisco, San Francisco, California 94158 2324 and Department of Human

Contributors

xi Genetics, University of Pittsburgh, Graduate School of Public Health, Pittsburgh, PA - 15261 J. David Pajerowski, (207) Biophysical Engineering Laboratory, School of Engineering and Applied Science, University of Pennsylvania, Philadelphia, Pennsylvania 19104 Siew Cheng Phua, (323) Institute of Medical Biology, Immunos, 8A Biomedical Grove, Biopolis, Singapore 138648 Nisha M. Ramdas, (221) National Centre for Biological Sciences, Tata Institute of Fundamental Research, Bellary Road, Bangalore 560065, India Praveena L. Ramanujam, (35) National Centre for Cell Science, Ganeshkhind, Pune 411007, India T. Roopa, (221) National Centre for Biological Sciences, Tata Institute of Fundamen tal Research, Bellary Road, Bangalore 560065, India Samuel A. Safran, (143) Department of Materials and Interfaces, Weizmann Institute of Science, Rehovot, 76100, Israel Fanny Serman, (295) Mechanics and Genetics of Embryonic and Tumoral Develop ment group, UMR168 CNRS, Institut Curie, 11 rue Pierre et Marie Curie, F-75005, Paris, France Bidisha Sinha, (57) National Centre for Biological Sciences, Tata Institute of Fundamental Research, Bellary Road, Bangalore 560065, India Deepak Kumar Sinha, (57) National Centre for Biological Sciences, Tata Institute of Fundamental Research, Bellary Road, Bangalore 560065, India G. V. Shivashankar, (57, 221) National Centre for Biological Sciences, Tata Institute of Fundamental Research, Bellary Road, Bangalore 560065, India and Department of Biological Sciences and Research Center for Excellence in MechanoBiology, National University of Singapore, Singapore 117543, Singapore Colin L. Stewart, (323) Institute of Medical Biology, Immunos, 8A Biomedical Grove, Biopolis, Singapore 138648 Joe Swift, (207) Biophysical Engineering Laboratory, School of Engineering and Applied Science, University of Pennsylvania, Philadelphia, Pennsylvania 19104 Swee Jin Tan, (79) NUS Graduate School for Integrative Sciences and Engineering, National University of Singapore, Singapore 117456 and Division of Bioengineer ing and Department of Mechanical Engineering, National University of Singapore, Singapore 117576, Singapore Shefali Talwar, (57) National Centre for Biological Sciences, Tata Institute of Funda mental Research, Bellary Road, Bangalore 560065, India and Department of Bio logical Sciences and Research Center for Excellence in MechanoBiology, National University of Singapore, Singapore 117543, Singapore Chwee Teck Lim, (79) NUS Graduate School for Integrative Sciences and Engineering; Division of Bioengineering and Department of Mechanical Engineering; and Mechan obiology Institute, National University of Singapore, Singapore 117411, Singapore Katherine L. Wilson, (97) Department of Cell Biology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205 Klaus Weisshart, (3) Carl Zeiss MicroImaging GmbH, Carl-Zeiss-Promenade 10, 07745 Jena, Germany

xii

Contributors

Tobias Ulbricht, (3) Leibniz Institute of Age Research, Fritz Lipman Institute, Beutenbergstr. 11, 07745 Jena, Germany Evelyn K.F. Yim, (241) Division of Bioengineering, National University of Singa pore, Singapore; Mechanobiology Institute Singapore, National University of Sin gapore, Singapore; Department of Surgery, National University of Singapore, Singapore Zhixia Zhong, (97) Department of Chemical Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213 Assaf Zemel, (143) Institute of Dental Sciences, Faculty of Dental Medicine, and Fritz Haber Center for Molecular Dynamics, Hebrew University-Hadassah Medical Center, Jerusalem, 91120, Israel

PREFACE

Biological cells are active mechanical systems sensing local microenvironments using specialized cell-surface receptors. Physicochemical signals from the extracellular matrix impinge on cellular geometry resulting in altered functional nuclear landscapes and gene function. These alterations regulate diverse biological processes including stem-cell differentiation, developmental genetic programs, and cellular homeostatic control systems. Although the cytoskeleton is a well-appreciated critical component of cellular morphology, emerging evidence suggests that it may also have important consequences for maintenance of nuclear architecture; its mechanical properties and genome function. Regulation of genetic programs in response to cellular geometric cues requires mechanisms that act at a distance. A number of signaling pathways are activated in response to mechanical signals converging on regulatory factors, which translocate to the nucleus via diffusive processes. Recent evidence also highlights the physical trans mission of active stresses via cytoplasmic-nucleus connections to remodel chromatin assembly. The physicochemical signals that arrive at the nucleus have to be further sorted to appropriate regulatory sequences within the 3D architecture of the cell nucleus to effect changes in genome function. Although the location of regulatory sequences on the 1D DNA polymer is known from genome sequencing, its 3D location when folded into chromatin via histone and nonhistone proteins within the nucleus is largely unknown. In addition, a number of essential posttranslational modifications of histone proteins deter mine both specificity and accessibility to regulatory sequences on the genome. We are just beginning to appreciate the impact of cellular geometry on nuclear mechanics and genome regulation. In addition, the mechanical integrity of the cell nucleus and nuclear mechanical signaling are found to profoundly influence cellular homeostatic controls: driving cells toward differentiation, proliferation, or apoptosis. Further, diseases such as cancer are conjectured to originate at a single-cell level in its local mechanical environment, within tissue contexts. The chapters in this book describe both methods and advances in our understanding of the spatio-temporal organization of genome assembly, its integration to mechanical properties of the cell nucleus and how mechanoregulation of gene function may be defined in interphase cells and during their differentiation and development. The last section discusses the growing number of diseases associated with altered nuclear organization. Clearly, understanding the mechan ical aspects of the cell nucleus and how it impinges on genome function in living cells has become a central theme in modern cell and developmental biology and biophysics. With the advent of new methods and approaches, some of which are described in this book, there exists now a promising future in this emerging research frontier. G.V. Shivashankar xiii

SECTION A

Cell Nucleus: Organization & MechanoBiology

CHAPTER 1

Fluorescence Fluctuation Microscopy to Reveal 3D Architecture and Function in the Cell Nucleus Thorsten Lenser*,†, Klaus Weisshart†, Tobias Ulbricht‡, Karolin Klement‡, and Peter Hemmerich‡ *

Institute of Computer Science, Friedrich-Schiller-University, Ernst Abbe Platz 2, 07743 Jena, Germany

†

Carl Zeiss MicroImaging GmbH, Carl-Zeiss-Promenade 10, 07745 Jena, Germany

‡

Leibniz Institute of Age Research, Fritz Lipman Institute, Beutenbergstr. 11, 07745 Jena, Germany

Abstract I. Introduction A. Three-Dimensional Organization of the Cell Nucleus B. Assembly Mechanisms of Nuclear Structures C. Fluorescence Fluctuation Microscopy Techniques II. Rationale III. Materials and Methods A. Cell Culture and Transfection B. Raster Image Correlation Spectroscopy C. FRAP and Mathematical Modeling IV. Results and Discussion A. FCS Measurements in the Cell Nucleus B. FRAP Measurements in the Cell Nucleus and Modeling C. RICS Measurements in the Nucleus V. Conclusions A. FRAP and FCS in Nuclear Cell Biology B. Benefits of the RICS Technique C. Lessons for the Mathematical Modeling of FRAP D. Kinetic Analysis of Fluorescence Microscopy Experiments References

METHODS IN CELL BIOLOGY, VOL. 98 Copyright 2010 Elsevier Inc. All rights reserved.

3

978-0-12-381009-0

DOI: 10.1016/S0091-679X(10)98001-1

4

Thorsten Lenser et al.

Abstract The three-dimensional (3D) architecture of the cell nucleus is determined not only by the presence of subnuclear domains, such as the nuclear envelope, chromosome territories, and nuclear bodies, but also by smaller domains which form in response to specific functions, such as RNA transcription, DNA replication, and DNA repair. Since both stable and dynamic structures contribute to nuclear morphology, it is important to study the biophysical principles of the formation of macromolecular assemblies within the nucleus. For this purpose, a variety of fluorescence fluctuation microscopy techniques can be applied. Here, we summarize our current knowledge on the 3D architecture of the mammalian cell nucleus and describe in detail how the assembly of functional nuclear protein complexes can be analyzed in living cells using fluorescence bleaching techniques, fluorescence correlation spectroscopy, raster image correlation spectroscopy, and mathematical modeling. In conclusion, the appli cation of all these techniques in combination is a powerful tool to assess the full spectrum of nuclear protein dynamics and to understand the biophysical principles underlying nuclear structure and function.

I. Introduction A. Three-Dimensional Organization of the Cell Nucleus The cell nucleus is responsible for the storage, propagation, maintenance, and expres sion of the genetic material it contains (Diekmann and Hemmerich, 2005). These duties are executed by biochemical activities, namely DNA compaction/decompaction, DNA replication and segregation, DNA repair, and RNA transcription/processing, respec tively. The corresponding machineries are highly structured, yet dynamic macromole cular assemblies (Misteli, 2007) which must work on chromatin with high fidelity in a crowded nuclear environment (Richter et al., 2007). In addition, the mammalian cell nucleus contains a variety of subnuclear domains, nuclear bodies, or subnuclear com partments (Fig. 1). DNA in the form of chromatin is easily visualized as individual chromosomes in mitotic cells. In the interphase cells, chromosomes decondense into socalled chromosome territories (CTs), which occupy distinct volume regions (Fig. 1) (Cremer et al., 2006; Heard and Bickmore, 2007). Staining of interphase chromatin using DNA dyes does not reveal CT structures but allows the discrimination between transcriptionally active euchromatin and transcriptionally silent heterochromatin. Con stitutive heterochromatin is mainly composed of pericentromeric DNA, and in this case, the chromosome’s centromere/kinetochore complex can be found embedded within this chromatin region (Fig. 1) (Probst and Almouzni, 2008). The nucleus obtains structural support through the nuclear lamina, which is attached to the nuclear double membrane, together forming the nuclear envelope (Fig. 1) The nuclear envelope controls traffic of molecules between the cytoplasm and the nucleoplasm but has also emerged as a critical determinant in genome architecture (Starr, 2009). As a consequence of this important

Heterochromatin

Nuclear lamina

Nuclear membrane

Nuclear pore complex Nucleolus

Preinucleolar compartment

Speckle

Pericentromeric

heterochromatin

with centromere

Paraspeckle Cajal body

PML body

RNA transcription site

Chromosome

territory

intermingling

DNA replication site

Interchromatin space

DNA

repair site

Chromatin loops with shared

transcription site

Nuclear envelope

1. Fluorescence Fluctuation Microscopy

Chromosome territory

Nucleolus

Speckles

Cajal bodies

PML bodies

Heterochromatin

Centromere

Fig. 1 Nuclear architecture.The mammalian cell nucleus contains chromatin in the form of chromosome territories (CTs). CTs may overlap at their touching

5

borders (intermingling) or create the so-called interchromatin space (white). Constitutive heterochromatin (dark gray in the center cartoon, blue in the bottom right panel) is mainly found as pericentromeric chromatin in patches throughout the nuclear volume, at the nuclear periphery, as well as around nucleoli. Structural hallmarks in the periphery of the nucleus include nuclear pore complexes, the nuclear membrane (dark green), and the meshwork-like nuclear lamina. Chromatin loops with associated transcription factories may extrude out of CTs within the nucleolus as well as throughout the nucleoplasm. Transcription (orange), replication (yellow), and DNA repair processes (light blue) usually occur in small domains with a diameter below 100 nm. A diverse set of nuclear bodies, such as speckles, paraspeckles, perinucleolar compartment, Cajal bodies, or promyelocytic leukemia (PML) bodies can be visualized by confocal immunofluorescence analysis (bottom panels). (See Plate no. 1 in the Color Plate Section.)

6

Thorsten Lenser et al.

function, mutations in a subset of nuclear envelope components are associated with a variety of diseases (Dauer and Worman, 2009). The most prominent subnuclear domains include the nucleolus, perinucleolar com partment, speckles, paraspeckles, Cajal bodies (CBs), and promyelocytic leukemia (PML) bodies (Fig. 1). In addition, a variety of other nuclear bodies have been identified such as polycomb group (PcG) bodies, Gemini bodies (Gems), Oct1/PTF/transcription (OPT) domain, cleavage bodies, and SAM68 nuclear body (Handwerger and Gall, 2006; Spector, 2001). Subnuclear structures are macromolecular complexes that consist of membrane-less accumulations of specific sets of functionally related molecules. For example, components of the ribosome biogenesis pathway are predominantly confined to the nucleolus. First thought to be exclusively devoted to the synthesis of ribosomal RNA and assembly of ribosomal subunits, it has become clear that the nucleolus serves a variety of additional functions, including regulation of mitosis, cell cycle progression, proliferation, and various stress responses (Raska et al., 2006; Sirri et al., 2008). The biochemical function(s) of the other subnuclear domains are less clear or unknown. PML bodies attract a limited and selected set of nuclear proteins which are functionally quiet promiscuous. Therefore, PML bodies have been implicated in the regulation of diverse cellular functions, such as the induction of apoptosis and cellular senescence, inhibition of proliferation, maintenance of genomic stability, and antiviral responses (Bernardi and Pandolfi, 2007). PML bodies are structurally rela tively stable structures at which controlled molecule traffic and post-translational modifications may regulate the activity of specific proteins throughout the genome and the epigenome in response to various cellular stresses (Bernardi and Pandolfi, 2007; Torok et al., 2009). Speckles, also referred to as interchromatin granule clusters (IGCs), are enriched in pre-mRNA splicing factors. At the microscopic level, speckles appear as irregular, punctate domains varying in size and shape (Fig. 1). They are considered to be the main sites for storage, assembly, and/or recycling of the essential spliceosome components (Lamond and Spector, 2003). Because highly transcribed genes are found in the periphery of speckles and also other subnuclear domains, they may also serve to efficiently integrate and regulate mRNA transcription and mRNA processing machineries (Zhao et al., 2009). CBs are involved in the biogenesis of several classes of small nuclear ribonucleoprotein particles (snRNPs) as well as their modification (Gall, 2000; Matera et al., 2009). Resembling the speckles/gene associa tion mentioned above, CBs associate specifically with histone and snRNA genes. This colocalization is transcription dependent, requires expression of snRNA coding regions, and is probably based on an energy-driven motor activity in the nucleus (Dundr et al., 2007; Frey and Matera, 2001). There might even exist a functional interplay between speckles and CBs as integrated entities, sharing functional features of both structures (Bogolyubov et al., 2009; Pandit et al., 2008). In mammalian cell nuclei, DNA replication, RNA transcription, and repair of damaged DNA occur in dot-like structures with a mean diameter of 100 nm (Fig. 2). With respect to transcription and replication, these focal sites have been coined “factories” as each site contains all of the enzymatic activity required (Cook, 1999). A general model was recently suggested for the organization of all genomes in

(D)

30 min

60 min

120 min

180 min

210 min

240 min

300 min

330 min

360 min

420 min

450 min

480 min

540 min

570 min

600 min

E rly Ea

mRNA

DN A RNA Pol II

1. Fluorescence Fluctuation Microscopy

(A) FI-U

δ

(B) BrdU

SSB

Helicase

Mid

Po l

PCNA

RPA

γ -H2AX

MDC1 B ATR BRCA1

Ku

Ku

MRE11 Rad50 0 NBS1 53BP1 1 ATM

γ -H2AX

Late

(C)

7

Fig. 2 Factory dynamics. (A) Transcription can be visualized after brief exposure of living cells to the nucleotide analog on fluorouridine (Fl-U), followed by immunodetection of the Fl-U epitope. (B) Bromodeoxyuridine (BrdU) can be used in a similar incorporation assay to visualize nascent DNA during replication. (C) Sites of DNA double-strand break repair are detected using antibodies against a phosphorylated form of histone H2AX (-H2AX). (D) The changing pattern of replication factory distribution throughout S phase can be visualized by time-lapse microscopy of GFP-tagged PCNA. For technical details, we refer to WeidtkampPeters et al. (2006).

8

Thorsten Lenser et al.

which the transcription factories play a central role (Cook, 2010). Notably, the model proposes that active RNA polymerases do not move along their templates during elongation but are bound to a factory acting as both motors that reel in their templates and a fixed structural entity that holds active chromatin loops in place (Cook, 2010). DNA replication also occurs at similarly specialized subnuclear sites where the factors directly or indirectly involved in replication are concentrated (Fig. 2) (Leonhart et al., 2000). Finally, the repair of damaged DNA at focal sites throughout the genome is also a dynamic process that requires careful orchestration of a multitude of enzymes, adaptor proteins, and chromatin constituents (Fig. 2) (Lukas et al., 2005). B. Assembly Mechanisms of Nuclear Structures Fluorescence recovery after photobleaching (FRAP) analyses of subnuclear domains such as nucleoli, speckles, and CBs have revealed that their component parts rapidly exchange with nucleoplasmic pools (Misteli, 2008). Typical residence times of proteins at these compartments are in the seconds range. Similarly, factors acting at transcription, replication, and repair foci show rapid exchange at chromatin (Fig. 3) (Misteli, 2007, 2008). These observations have led to the conclusion that nuclear body proteins undergo repeated and rapid cycles of association and dissocia tion between the nuclear body and the nucleoplasm. As a consequence, nuclear bodies and factories are in perpetual flux. Their structure is determined by the ratio of on-rate versus off-rate of its components, clearly suggesting self-organization as the mechan ism of their assembly (Matera et al., 2009; Misteli, 2008). In contrast, the centromere/ kinetochore complex is assembled in a cell-cycle-dependent manner containing fast exchanging components and very tight binding proteins (Fig. 3) (Hemmerich et al., 2008). A similar observation has been documented for PML bodies, at which some component parts have residence times of up to 1 h (Brand et al., 2010; WeidtkampPeters et al., 2008). Therefore, not only self-organization but also self-assembly mechanisms may contribute to the 3D architecture of the nucleus. C. Fluorescence Fluctuation Microscopy Techniques The development of in vivo microscopy techniques using genetically encoded fluor escent tags, such as the green fluorescent protein (GFP), has opened the door to probe nuclear architecture and function in living cells. By analyzing macroscopic relaxation after disturbing the equilibrium state, fluorescence intensity images can be used to assess diffusion times, interactions, and binding constants of molecules. Fluorescence fluctua tion microscopy (FFM) approaches have been developed to investigate few molecules in small regions of a cell providing dynamic information in dependence of time and space by creating cellular diffusion and concentration maps (Fig. 4). A major consideration is the accessible resolution as nuclear processes can take place in a time scale ranging from microseconds to hours (Fig. 3) and single molecules or huge macromolecular assemblies in well-defined stoichiometries can be involved. FFM allowed for the first time not only to visualize protein dynamics and interactions but also to quantitatively determine biophysical properties of proteins

9

1. Fluorescence Fluctuation Microscopy

Dynamic Residence times: seconds

• Basal transcription factors • Chromatin remodelers • Histone modifiers • Structural proteins

Semistable Residence times: minutes

• Linker histones • Engaged polymerases • DNA repair and replication

Stable Residence times: hours

• Core histones • Centromeric proteins

components

(HMGs, HP1)

Fig. 3

Dynamics of chromatin-binding proteins.Depending on the biochemical duty, chromatin-binding proteins exchange at their DNA-binding sites with different kinetics. Typically, the residence time of bona fide transcription factors is within the seconds range. Active RNA or DNA polymerases stay attached to their template nucleic acids in the minute range. The nucleosome, and in particular the chromosomal centromere complex, is extremely stable. Their component parts exchange with soluble nucleoplasmic pools only very slowly.

in intact cells. FFM approaches include time-lapse microscopy (Heun et al., 2001), fluorescence recovery after photobleaching (FRAP), fluorescence loss in photobleaching (FLIP), inverse fluorescence recovery after photobleaching (iFRAP), etc. (Bancaud et al., 2009; van Royen et al., 2009), fluorescence correlation spectro scopy (FCS) (Haustein and Schwille, 2003), continuous fluorescence photobleach ing (CP) (Weidemann et al., 2003), raster image correlation spectroscopy (RICS) (Digman et al., 2005), single particle tracking (SPT) (Levi and Gratton, 2008), fluorescence resonance energy transfer and fluorescence lifetime imaging (Wallrabe and Periasamy, 2005), and multiparameter fluorescence image spectroscopy (Weidtkamp-Peters et al., 2009a). Figure 4 summarizes the potential of these techniques to study nuclear protein dynamics.

1. Fluorescence Recovery after Photobleaching When subjected to repeated cycles of excitation and emission, fluorescent molecules eventually lose their ability to emit fluorescence, enabling the creation of photobleached spots by the repeated application of a strong laser beam. Since nearly all proteins in the nucleus are highly mobile, fluorescence eventually returns to these dark

10

Thorsten Lenser et al.

s p s

f i p

p

a

r c c t r l c

e

t

i c f

s i

s

Fig. 4 Fluorescence fluctuation microscopy techniques to analyze nuclear protein mobility. For live-cell analyses, nuclear proteins are expressed in fusion with enhanced green fluorescent protein (EGFP) or monomeric red fluorescent protein (RFP). Depending on its current function, the nuclear protein(s) may occur diffusely in the nucleoplasm or in a nuclear structure such as a factory or a nuclear body. Mobility and interactions can then be assessed by the indicated techniques.

areas. Measuring the flux of fluorescence into this region then yields the FRAP recovery curve, which can be analyzed using mathematical models to yield kinetic parameters of the proteins under study (Fig. 5) (Bancaud et al., 2009; Carrero et al., 2003). The original description of FRAP was coined continuous fluorescence microphotolysis, which itself has been established for more than three decades (Peters et al., 1974,1981). In the first mathematical analysis of FRAP, Axelrod et al. (1976) devel oped an effective diffusion model (without binding and unbinding processes) based on a two-dimensional (2D) photobleach with a Gaussian intensity profile. This provided an explicit solution for the FRAP curve, which can be fitted to the measurement in

11

1. Fluorescence Fluctuation Microscopy

(A) Prebleach

Postbleach

(B)

Fig. 5 Fluorescence recovery after photobleaching. (A) Typical FRAP experiment performed on U-2 OS cells that express GFP-tagged PML. Fluorescence was bleached within two circled areas containing one nuclear body each (white circles) and fluorescence recovery into this area was monitored over time. Scale bars = 5 µm. (B) Quantification of a FRAP experiment. Such graphs typically show mean values from at least 20 FRAP experiments as relative fluorescence intensity (RFI) after normalization to prebleach levels. The FRAP curve immediately delivers information on mobile and less mobile populations. Mathematical modeling can be performed to extract biophysical parameters, such as the diffusion coefficient and binding constants.

order to obtain the effective diffusion constant Deff. Using this approach, Phair and Misteli (2000) measured effective diffusion coefficients for several molecules involved in the cell nucleus. Today, many FRAP models of processes in the cell nucleus assume that the proteins undergo diffusion as well as binding/unbinding events, both contri buting to their spatial dynamics (Beaudouin et al., 2006; Sprague et al., 2004).

2. Fluorescence Correlation Spectroscopy FCS is a method to analyze diffusing particles in solution or in living cells. This technique was introduced in the early 1970s (Magde et al., 1974). In FCS, fluorescent molecules or particles diffuse by Brownian motion in and out of a space-limited detection volume. This detection volume represents a diffraction-limited smallillumination ellipsoid created by a laser beam that is focused through a high numerical aperture objective. Photons emitted from the fluorescent particles are counted con tinuously through the same optics over time. The intensity fluctuations reflect the photophysical and hydrodynamic properties of the diffusing particles. While the fluctuation amplitude depends on particle concentration and brightness, its frequency

12

Thorsten Lenser et al.

contains information on the diffusion times of the fluorescent particles. For quantitative evaluation, the fluctuation frequency is correlated with a time-shifted replica of itself (autocorrelation) at different time values. By fitting theoretical model functions to the measured autocorrelation curves, the diffusion coefficient and the concentration of the diffusing species can be extracted (Fig. 6). (A)

(B)

(D) 1.4 Concentration: 37 nM

18

1.3

16

G (τ)

Count rate (kHz)

(C) 20

14 12

1.1

10

1.0 0

1

2

3

4 5 6 Time (s)

7

8

9

1

10

(E)

10

100 1000 10000 1e+005 1e+006 Time (μs)

(F) 3.5 3.0 2.5 2.0 1.5 1.0 0.5

1500

G (τ)

Count rate (kHz)

D = 2.35 μm2/s

1.2

1000 500 0 0

1

2

3

4 5 6 Time (μs)

7

8

9

10

1

10

100 1000 100001e+0051e+006 Time (μs)

Fig. 6 Assessing nuclear protein mobility using FCS.(A) Confocal image of a living HEp-2 cell expressing a GFP-tagged nuclear body protein (green) overlayed with the corresponding differential interference contrast image (gray). (B) GFP fluorescence from the image shown (A). Scale bars = 10 µm. FCS measurements were performed at two different positions of the nucleus (red crosses). (C) Count rate trace and respective autocorrelation curve (D) of FCS measurements within the nucleoplasm. The red line represents the fit using a 3D-1-component anomalous diffusion model for which the diffusion coefficient (D) and the number of fluorescent particles (N) was extracted. (E) Count rate trace and respective autocorrelation curve (F) of FCS measurements within a PML body. The decay of the count rate trace (red-dotted line) contains information on the binding behavior of the GFP-tagged protein at nuclear bodies. Evaluation, however, requires mathematical models developed for continuous photobleaching Weidemann et al., 2003. (See Plate no. 2 in the Color Plate Section.)

1. Fluorescence Fluctuation Microscopy

13

Microscope developments provided reduction of the detection volume into the femto liter range and short measurement times. Considering a particle concentration of 1 nM, the FCS detection volume contains less than one particle at any given time, thus providing single-molecule resolution. This advantage of FCS is at the same time its drawback: it works only properly within a limited concentration range, typically between 10 nM and 1 µM. Due to its high sensitivity, FCS is subject to certain artifacts that must be carefully controlled, such as photobleaching, cellular autofluorescence, intramolecular dynamics of the fluorophore, laser beam polarization effects, refractive index of the objective’s immersion medium, pinhole misadjustment, cover-slide thickness, and optical saturation (Bacia and Schwille, 2007; Enderlein et al., 2004). In recently developed commercial FCS devices, these potential pitfalls are mostly eliminated. The time scale of resolution in FCS is in the nanoseconds to seconds range. This makes FCS a powerful tool to study biological processes, particularly in living cells, and complements related techniques, such as FRAP and SPT. Application of FRAP and FCS on the same molecules combined with mathematical modeling allows the determination of all biophysical parameters of a nuclear protein (Müller et al., 2009; Weidtkamp-Peters et al., 2008). Typical FCS measurements performed in the cell nucleus are shown in Fig. 6. Detailed protocols for FCS applications in the nucleus have been documented recently (Weidtkamp-Peters et al., 2009b).

3. Raster Image Correlation Spectroscopy As mentioned above, the most widely used approach to study dynamic cellular processes so far uses fluorescently tagged molecules in their in situ environment and far-field fluorescence microscopy techniques. FRAP is able to access average dynamics as all diffusing molecules will contribute to the signal from whatever region of the cell they will come from. Ideally, however, one wants to know the spaceresolved behavior of single molecular entities in terms of their kinetics and interactions and without the disturbance of the equilibrium state. All these parameters are provided by image correlation spectroscopy (ICS) (Petersen et al., 1993). ICS data are computed from the power spectrum of the spatial autocorrelation function that is obtained from the intensity images by 2D fast Fourier transformation algorithms. Due to the point-scanning process, images taken with a laser-scanning microscope contain hidden time information, which can be exploited to analyze fast to slow dynamic processes as well as concentrations of molecules within a cellular environ ment. The scanning process itself is used in a relatively new addition to the correlation techniques named RICS. RICS is able to bridge the gap between the accessible timescales of FCS and ICS as it can resolve dynamics in the range of microseconds to seconds with still a sufficient spatial resolution (Digman et al., 2005). Data in cells are most conveniently acquired as a time series stack by raster scanning of images of larger cell areas. Due to its broad dynamic access by analyzing the fluctuations between neighboring pixels in the x- and y-direction, nearly all diffusion processes that take place in cellular subregions can be studied (Digman et al., 2005). A major

14

Thorsten Lenser et al.

advantage of the RICS technology is that it can be used in principle on any commercial confocal microscope with analog detection (Brown et al., 2008). These instruments are generally of high quality, automated, and show excellent performance.

II. Rationale The definition of specific biochemical interactions among nuclear proteins in dis tinct compartments has led to an image of structural continuity and functional stability within the nucleus. For years, nuclear pathways have been exhaustively examined using biochemical and molecular approaches without much consideration of the special restrictions presented by the nuclear architecture. To understand nuclear func tion, it is important to study the mechanisms of nuclear substructure formation, maintenance, and assembly by quantitative assessment of the mobility of their compo nent parts in living cells (Fig. 4). “The eukaryotic nucleus and the control of chromatin function pose greater chal lenges for a systems biology modeling approach than many (of the) other cellular networks…” (Visser and Fell, 2007). “The ultimate goal of a systems biology view of the cell nucleus is to understand genome function within the architectural framework of the nucleus. […] Simulation is becoming indispensable for the analysis of the kinetics of nuclear processes.” (Gorski and Misteli, 2005). These two quotes exemplify the emerging consensus in the community that a further understanding of nuclear pro cesses in the cell is only possible through a combination of in vivo experimental techniques and computer modeling. A kinetic model is essentially a mathematical description of the hypothesized biological processes. The model is characterized by biophysical parameters, such as binding and release constants, residence times, and diffusion coefficients. Using dynamic fluorescence microscopy data, the set of para meters is determined that results in the best model fit to the data, which serves as a test of the model as well as quantitative information about the parameters under study (Dinant et al., 2009; Phair and Misteli, 2001). This approach has been employed, for example, to set up a kinetic framework model for the RNA polymerase I transcription machinery in the nucleolus (Dundr et al., 2002) or the ordered recruitment of DNA repair factors (Politi et al., 2005). In this chapter, we provide methodological con siderations for the combined use of FFM and mathematical modeling.

III. Materials and Methods A. Cell Culture and Transfection Human HEp-2 cervix carcinoma epithelial cells and NIH-3T3 cells (ATCC, Mana ssas, VI, USA, CCL23; Brand et al., 2010) were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal calf serum in a 10% CO2 atmosphere at 37°C. For live cell-imaging experiments, cells were seeded on 42-mm

15

1. Fluorescence Fluctuation Microscopy

glass dishes (Saur Laborbedarf, Reutlingen, Germany) and transfected with plasmid DNA 1–2 days before observation using FuGENE-HD Transfection Reagent (Roche, Basel, Switzerland) according to the manufacturer’s protocol. B. Raster Image Correlation Spectroscopy

1. Theory of RICS If we consider isotropic movement, the probability to encounter a particle by a scanned laser beam at a certain position r and time t is proportional to the molecular concentration C(r,t), and depends in addition on the diffusion coefficient D of the molecule: Pðr; tÞ / Cðr; tÞ =

1 ð4DtÞ

3=2

er

2

=4Dt

ð1Þ

Equation (1) describes the change in concentration as a function of time and location. The equation consists of two factors: a temporal polynomial term that depends on the diffusion coefficient of the particle and the observation time, and a spatial exponential Gaussian term that in addition depends also on the location. If the particle was at the origin (r = 0) at t = 0, it will be found at the distance r from the origin at a later time with a probability that is characterized by a Gauss distribution, where the variance depends on the time and the diffusion coefficient. The fluorescence signal or intensity at a given measurement time I(t) is directly proportional to the concentration C(r,t): Z I ðtÞ = Q

W ðrÞ C ðr; tÞ dr

ð2Þ

where stands for the instrument sensitivity, Q for the quantum yield of the fluorescent dye, and W(r) for the point spread function (PSF) of the microscope system. The integral is calculated over the entire excitation volume. In Eq. (2) the concentration of particles as a function of time is convoluted with the PSF or the illumination profile. For a continuous sampling at one position over time, the temporal autocorrela tion function of the fluorescence intensity fluctuations decays with a characteristic time (the diffusion time D) that depends on the diffusion coefficient of the particle and the size of the illumination volume, which is defined by the radial and axial e2 extensions !r and !a of the laser beam as defined in Eq. (3). Gt ðÞ =

hI ðtÞ I ðt þ Þi

= N

hI ðtÞi2

1 1 þ D

!

!1=2

1 1þ

1 S2

D

with

D =

!r2 4D

and S =

!a !r

ð3Þ

16

Thorsten Lenser et al.

The angular brackets indicate the average over time. is hereby the geometric correction factor for the PSF and S the structural parameter describing the ratio between the axial and radial e2 extensions of the illumination beam. By inser tion of Eq. (2) in the first term of Eq. (3), analytical expressions for the temporal and spatial autocorrelation functions can be derived, which depend on the diffusion process and to which the experimental data can be fitted. In Eq. (3), the second term expresses the free 3D diffusion of one molecule species. The function decays faster for faster molecules since the probability to see a molecule once it entered the volume at a later lag or correlation time is less the faster the molecule transverses the volume element. If the beam is not parked but scanned, then data sampling at different spatial positions will be possible. If the PSF is substantially larger than the pixel size, then the fluctuations of adjacent points will correlate as the scanner moves along due to superposition of the PSFs and there will be autocorrelation amplitude even for immo bile or very slow-moving molecules. For immobile molecules, the 2D spatial autocorrelation function displays the extent of the overlapping PSFs of the scanned object. For moving molecules, the spatial autocorrelation function of the fluorescence inten sity fluctuations decays with a characteristic length that again depends on the diffusion coefficient of the particle and the size of the illumination volume as defined in Eq. (4): Gs ðξ; cÞ =

hIðx; yÞ Iðx þ ξ; y þ cÞi hIðx; yÞi2

= Sðξ; cÞ Gðξ; cÞ

=e

ðξr =!r Þ2 þðcr =!r Þ2 = 1þ

with

D =

!r2 4D

and

ξ p þc l D

S=

0

!a !r

N

1 0

11=2

B C B C 1 B 1 C B C B C B C 1 A @1 þ A @ 1þ 2 D S D ð4Þ

where ξ and c denote the spatial lag increments. The spatial autocorrelation function will depend on the spatial overlap of the PSF as well as the time distance between neighboring pixels. If the diffusion of the particle is random, longer time intervals between the pixels corresponding to slower scanning will lead to a decreased correla tion at shorter distances, but increased correlation at longer ones. Related to this, the 2D autocorrelation function narrows faster but will be kept up for longer distances until it decays to zero amplitude for fast-moving molecules (Digman et al., 2005). Hence the shape of the 2D spatial autocorrelation function contains the information on the diffusion time. Along the scan direction, adjacent volumes are sampled rapidly. In contrast adjacent volumes in consecutive lines are sampled at a much slower rate. These different sampling rates enable the method to analyze diffusion processes ranging from the microsecond to the millisecond or even second timescales.

17

1. Fluorescence Fluctuation Microscopy

It follows from Eq. (4) that the spatial correlation function is the product of two terms: a spatial Gauss term S(ξ,c) and a temporal polynomial term G(ξ,c). The analytical expression for the spatial term is determined by the scanning regime. The scanning term in Eq. (3) represents the situation for a raster scan that is defined by the pixel size r (typically between 0.05 and 0.2 µm) as well as the pixel residence time p (typically between 1 and 100 µs) and the line repetition rate l (typically in the order of 1–100 ms). The term has to be accordingly adjusted for circular or line scans (Digman et al., 2005). The temporal term can be derived from the one obtained for the temporal correlation function where the lag time is represented by the spatial increments, the relation of which is as described in Eq. (5): = ξ p þ c l

ð5Þ

In any fluctuation analysis, the proper data-sampling procedure is of uttermost importance in order to provide statistical significant signals clearly exceeding the noise contribution. For correlation analysis, one should register at least 10,000 transi tions for a reliable determination of the diffusion time, which can be fitted from the decay characteristics of the function (Brown et al., 2008). The number of molecules can be computed from the amplitude as both are inverse proportional to each other. Normally this would require measurement times in the seconds range per pixel, which would lead to prohibitive long acquisition times. If, however, one sacrifices on pixel resolution and the average dynamics across a larger area is analyzed, then enough data points are generated for sufficient statistics. In solution measurements with a higher signal-to-noise ratio (SNR), one image can be enough. However, in the noisier environment of a cell, statistics will have to be further improved by taken a time series of images and average the correlation functions. In the majority of cases, the time series should encompass around 50–100 images.

2. RICS Measurements: General Considerations One important consideration when using PMT detectors is their characterization in terms of their contribution to the correlation. Because of the analog–digital conversion PMTs can produce per se a correlation signal. This signal can be estimated by analyzing the dark current or shot noise, where the microscope is set up in a way that no light reaches the detector (laser switched off, no sample, beam path set to go to the oculars). The shot noise will depend in this way merely on the scan speed (digitization rate) and will be independent from the image resolution or optical zoom. To eliminate pixel-to-pixel noise contribution, the first one or two pixel shifts should be skipped for fitting the data. The removal is especially important to analyze fast diffusion, since the autocorrelation function decays faster for short lag spaces. One should keep in mind, however, that omitting the first two pixels for data analysis will cut down on the spatial and temporal resolution. This restriction does not apply for photon count detectors such as APDs. The zero-lag amplitude Gs(0,0) should always be avoided as its error will be huge due to the shot-noise peak.

18

Thorsten Lenser et al.

The need for sufficient statistics will limit the obtainable spatial resolution for RICS and will in consequence restrict the image area that can still be analyzed. As the image sizes drop, noise contribution rises for longer lag distances resulting in correlation functions that do not decay to zero any more. Such offsets are characteristic for undersampling and lead to an overestimation of the diffusion coefficient, yet they influence the amplitude only to a minor extent (Brown et al., 2008). Statistics will also be improved as correlations from many images are averaged. With decreasing frames analyzed, the accuracy of the fit result decreases as the noise in each data point increases. Peaks at longer lag spaces, where the central correlation should have dropped already to zero, are due to random correlation events as images are shifted in x- and y-directions and these will be less averaged out the fewer the number of frames is that are used for averaging. For cell imaging one has therefore to compromise and go for fewer frames in order to keep bleaching low or increase temporal resolution at the cost of precision. If one cuts down on spatial resolution, larger frame sizes can be analyzed keeping the temporal resolution equal. To set RICS to work, the PSFs should heavily overlap as the scanner moves from pixel to pixel. That means that oversampling higher than the Nyquist criterion is required. A pixel should be represented in at least 10 data points for analog detection, where often the first two space lags might have to be avoided. This in turn demands that the pixel size is at least three- to fivefold smaller than the PSF of the respective objective. If no pixel-to-pixel noise takes place, the requirements for the pixel size are more relaxed and can be in the order of two-third of the PSF. With scan and optical zooms available, these pixel sizes can be realized with a variety of image formats ranging from 128 128 to 2048 2048 pixels. However, the larger the image size, the more bleaching might be introduced if cells are the object of investigation. Correct settings can be verified directly by the visualization of correlation images. These should have as low correlation contributions at long lag distances as possible. Choosing the correct acquisition speed depends on the diffusion coefficient of the particle under investigation. If the scan speed is too fast in relation to the molecule movement, the Gauss term in the correlation function will dominate and obscure the diffusion time of the molecule. On the other hand, if scanning is too slow, the molecule will have moved out before the scanner moves on so that correlation rapidly drops to zero, again making it impossible to fit any diffusion time. There might, however, be more than one speed that match the experimental conditions and could be applicable. In general, the slower the scanning speed associated with higher integration times, the higher the SNR, but the less data points are obtained before the autocorrelation function decays to zero. Especially in the y-direction, the autocorrelation function will drop rapidly for slow scan speeds. It will therefore again depend on the experiment to decide on the preferred scan speed. In solution, for example, with fast diffusion times, where the y-direction does not contain any information, it might be better to scan faster with less SNR to obtain more data points that determine the shape in the xdirection to a better extent. For slower molecules in a noisier cellular environment, slower scan speeds with increased SNR will be more advantageous. Albeit SNR can be increased with more laser power, a cell might create limits due to bleaching issues and

1. Fluorescence Fluctuation Microscopy

19

ideally one wants to stay as low as possible with the light load onto the cell. For a 100 mW laser, 0.1–0.2% of the power should be preferentially used. Bleaching itself, as long as no phototoxic processes are evoked, does not pose a restriction to retrieve diffusion times, as the single correlations are computed from one frame before aver aging and the gradient of bleaching within one frame can be neglected in most of the cases. Another way to increase sensitivity is increasing the PMT gain. Since the gain also amplifies the noise, images might occur very noisy; however, thermal noise from the PMT due to high detector gain is random and does not contribute to the shape of the correlation. Speeds of approximately 4–6 and 8–32 µs can be used as a rule of thumb to investigate proteins diffusing in solution (D > 100 µm2/s) and cytosol (D = 10– 100 µm2/s), respectively. For binding and extreme slow movement within membranes or in the nucleus (D = 0.1–10 µm2/s), optimal scan speeds can decrease to as low as 32–100 µs per pixel.

3. RICS in Living Cells: Determination of Diffusion Coefficients and Concentrations If transient transfections are performed, low-expressing cells with high enough SNR should be preferred. The laser power can often not be adjusted to the best SNR ratio, since bleaching might be prohibitive. In addition, cells represent a heterogeneous environment and immobile structures will dominate the auto correlation function (ACF) at long-range distances, which will obscure the correlations due to the faster moving molecules. Removal of immobile structures requires a time series stack and is accomplished by computing the pixel-wise average image from all frames and sub tracting from each frame. To avoid negative pixels, a scalar is added to each frame that is conveniently calculated from the average intensity of the average frame image or the single frames (Digman et al., 2005). By doing so, fluctuations become more pro nounced as the background does not contribute to the fluctuations but simply decreases their extent. Another complication may arise by cellular drift or by slowly moving structures such as organelles within the cells that will lead to a broadening of the spatial autocorrelation function even after subtraction of the immobile fraction. Instead, a moving (running, gliding) average has to be applied. To this end, the average of a few frames is computed, whereby the frames taken into account will overlap from average frame to average frame. For example, if a moving average of four frames is built, then the averages of frames 1–4, 2–5, 3–6, etc., are computed and subtracted from the original stack frames. Since the first and last frames do not have a corresponding moving average frame, the stack size after the procedure will be reduced. If a gliding average of n frames is applied to an original stack of m frames, the resulting stack after subtraction will have m n þ 1 frames. In this way again, the small fluctuations due to the mobile molecules are revealed. The moving average algorithm does not affect the diffusion coefficient as the dynamics are not dependent on the intensity, but rely solely on the fluctuations.

20

Thorsten Lenser et al.

In contrast to diffusion, the moving average algorithm alters the amplitude of the ACF and hence concentration measurements are hampered. The concentration is dependent on the intensity of the image since the ACF is normalized by division with the squared mean intensity value. For a correct determination of the concentration, the moving average algorithm has to be adapted. Normally, if the average intensity is added back to avoid negative pixels, concentrations in regions with little bright mobile fractions will be overestimated. Therefore, it is mandatory to add back the average of a region without immobile bright structures since then the ACF will be properly normal ized and concentration measurements become possible. However, one should keep in mind that this method will only work for the mobile fraction. Since there is a lot of variation within a cell, it is best to choose a region without immobile structures near the region of interest (ROI) to be analyzed within the time frame of the observation. To estimate the concentration within areas containing immobile structures, calibra tion measurements to generate standard curves will be needed in addition. Those standard curves should be created with the free fluorescent tag of interest at different expression levels. Expressing the tag alone guarantees that all protein is soluble and not attached to immobile structures. If the instrumental settings are kept constant, then the intensity can be used to determine concentrations by comparing to the standard curve. If the quality of the measurement is sufficient to analyze subregions in the image, then diffusion and concentration maps can be created. The smaller the region, the higher the resolution will be, but the worse is the SNR. In this chapter, RICS was performed on an LSM 710 (Carl Zeiss MicroImaging GmbH, Jena, Germany) using a C-Apochromat infinity-corrected 1.2 NA 40 water objective. During RICS measurements, cells were maintained in N-2-Hydroxyethylpiper azine-N0 -2-ethanesulfonic Acid (HEPES)-buffered medium without phenol red to mini mize background fluorescence. A time series of at least 50 images was recorded at a frame size of 512 512 pixels, a pixel size of 0.03 µm, and a scan speed of 6.4 µs per pixel. After subtracting a moving average to remove slow-moving structures and cellular movement, the average spatial correlation was computed and fitted to a 3D free diffusion model provided by the inbuilt software module. Diffusion maps were created using a ROI size of 64 64 pixels with 32 pixel shifts in each direction and the same model for fitting. C. FRAP and Mathematical Modeling Fluorescence recovery after photobleaching (FRAP) experiments were carried out on a Zeiss LSM 510/Meta confocal microscope (Carl Zeiss, Jena, Germany) essentially as described before (Hemmerich et al., 2008; Weistkamp-Peters et al., 2008). About 5– 10 images were taken before the bleach pulse and 50–200 images after bleaching of “ROIs” containing one nuclear structure each at 0.05% laser transmission to minimize scan bleaching. Image acquisition frequency was adapted to the recovery rate of the respective GFP fusion protein. The pinhole was adjusted to 1 airy unit. Quantitation of relative fluorescence intensities was done according to Schmiedeberg et al. (2004) using Excel (Microsoft, Redmond, WA, USA) and Origin software (OriginLab, North hampton, MA, USA).

1. Fluorescence Fluctuation Microscopy

21

1. Initial Conditions for the Mathematical Model For the initial conditions of the model, we assume that at some time t = 0, the fluorescence intensity inside the ROI is photobleached to 0. Due to inevitable diffusion during the bleach process, the actual fluorescence in the ROI certainly does not reach 0. However, the raw data from the microscope has been normalized by subtracting back ground fluorescence, leaving an initial value of 0 as our best guess. Starting from this assumption, we numerically solve the model equations until the sum of fluorescence from freely diffusing, loosely bound, and tightly bound molecules in the ROI reaches the value of the first FRAP measurement. Due to a temporary “blinding” of the detector by the photobleach, the first FRAP measurement is always taken some time after the bleach and thus is always truly positive. For this first point, we now have an approximation of the fluorescence distribution between the three fractions of molecules included in the model. Standard numerical integration continues from here to yield the rest of the estimated FRAP recovery curve. Recently, the criticism has been raised that the initial fluorescence distribution is Gaussian in shape, not constant (Mueller et al., 2008). Even though this might principally enhance our results, we do not have enough data on this at the moment in order to get a more accurate postbleach estimation. Thus, we usually apply the commonly used argument that when the bleached area is small enough, the Gaussian profile is approximated by a constant (Carrero et al., 2003; McGrath et al., 1998). Braeckmans et al. (2003) call this the uniform disc model, which can be applied in cases where the bleached disc is large in comparison to the laser beam’s point spread resolution, but not too large so that diffusion during bleach does not play a significant role. In order to avoid errors in the parameter estimation, no significant recovery should occur during the bleach. In general, bleach time should be at least 15 times shorter than the characteristic recovery time (Braga et al., 2004). This is the case for the measure ments performed here, as the binding/unbinding processes yield a very slow effective diffusion constant and thus a high characteristic recovery time. Therefore, we can safely assume that bleaching is instantaneous and no significant diffusion takes place during the bleach process. In contrast to other approaches reported in the literature, we do not use FRAP data to estimate the diffusion coefficient D, but rather utilize FCS to measure D and subsequently include this measurement into our FRAP model. Since the first phase of the recovery curve is diffusion dominated, this protocol reduces the sensitivity to the initial FRAP measurements. These may be more imprecise than the later ones due to interference effects of the bleach beam, diffusion during the bleach process, and approx imation of the initial fluorescence distribution.

2. Relation of Local Diffusion Coefficient to Net Flux into ROI In this section, we show how the diffusion coefficient D in Fick’s law (measured by FCS here) is related to the net flux (exchange rate) between compartments, which is needed to quantify the inflow of fluorescent material into the ROI of the FRAP model. We assume a circular ROI with radius r0 in 2D (a disk), which is embedded in an infinite space with constant concentration (or relative fluorescence intensity, RFI) of c0.

22

Thorsten Lenser et al.

Our model starts from the diffusion equation (more commonly called heat equation or Fick’s second law): ∂Cðx; y; tÞ = DDCðx; y; tÞ ∂t

ð6Þ

Using polar coordinates and with the assumption that the bleach pulse annihilates all fluorescence in the ROI, we can solve Eq. (6) to yield: Cðr; ; tÞ = Cðr; tÞ = c0 2c0

1 X J0 ðn r=r0 Þ n=1

n J1 ðn Þ

exp

2 n Dt r02

ð7Þ

In Eq. (7), J0 and J1 are Bessel functions of the first kind and n is the n-th positive root of J0(x). If we assume that no further reactions involving the considered molecules take place inside the ROI, we can derive an expression for the average concentration inside the ROI at time t: C ðtÞ = c0 4c0

1 X expðð2 Dt=r2 ÞÞ n

n=1

n J1 ðn Þ

0

ð8Þ

Since we have a solution for C(r,t), we can also calculate the first derivative at the boundary of the disk, which describes the flux perpendicular to the boundary: 2 1 ∂C 2c0 X n Dt ðr0 ; tÞ = exp ∂r r0 n = 1 r02

ð9Þ

For values of t such that Dt/r02 > 0.2, the infinite sums in Eqs. (8) and (9) are closely approximated by their first term. Putting them together, we get: ∂C 2 ðr0 ; tÞ» 1 ðc0 C ðtÞÞ ∂r 2r0

ð10Þ

It is important to remember that the approximation is almost exact for Dt/r02 > 0.2, but is increasingly incorrect for t ! 0. This is expected, since for small t, the derivative naturally tends to infinity, while the average concentration converges to zero. Using Green’s theorem, we can now derive an approximate expression for the net flux into the disk: 1 r02

Z2r0 D 0

∂ D2 Cðr0 ; tÞ ds » 2 1 ðc0 C ðtÞÞ ∂r r0

ð11Þ

The expected error between the approximation (Eq. (10)) and the analytical solution of Eq. (11) can be computed, and a correction factor can be introduced into the diffusion coefficient. For the given setting, the diffusion coefficient used in the

1. Fluorescence Fluctuation Microscopy

23

model is thus given by D0 = 1.3856 D, where D is the original diffusion coefficient determined by FCS. It is interesting to note that this correction factor is independent of c0, r0, and D, although it is only valid to disk-shaped compartments. Of course, with modern computing power it would be possible to use a much more accurate approximation in the compartmental model, involving more than just the first components of the sums of Bessel functions. However, there are two reasons not to do so. Firstly, when fitting data to an ODE model rather than simulating it once, computa tional constraints do play a role indeed. When hundreds or thousands of fitness evaluations—and thus model simulations—are needed, every possible speedup is welcome. Secondly, an FRAP model involves reaction processes inside the ROI, in addition to diffusion. With these processes, the analytical solution derived above becomes invalid and has to be replaced by a differential equation description. The linear formulation is both obviously straightforward and—following the reasoning above—mathematically justified. When comparing the approximate solution to the analytical one, it is important to recognize that the actual error might be smaller than discussed above. Rather than bleaching the ROI 100% and not changing the directly adjacent area, the bleach profile of a laser pulse is better approximated by the exponential of a Gaussian function (Braeckmans et al., 2003; Braga et al., 2004; Mazza et al., 2007) and thus does not have a sharp transition between zero and one at the boundary of the ROI. Therefore, the instantaneous diffusional inflow after the bleach pulse is not as sharp as the analytical solution describes it.

3. Numerical Reaction–diffusion Models for FRAP Analysis To analyze FRAP recovery curves and to relate them to kinetic parameters of the proteins under study, mathematical models of the hypothesized binding behavior have been fitted to the FRAP curves. For these purposes, we used three model structures: a simple one-step binding/unbinding process, a two-step process with sequential binding steps, and a two-step process with parallel, independent binding steps (Fig. 7). Differ ent simplifications for FRAP models, such as reaction-dominant and diffusiondominant models, are discussed in the literature (Sprague et al., 2004). However, the vast array of different proteins under study here leads to the expectation that only the full model can adequately describe their dynamics, and thus no reduced model variant is used. Erroneously neglecting diffusion can lead to parameter estimation errors of several orders of magnitude (Sprague et al., 2006). The experimental setup provided that bleached ROI and FRAP ROI are similar and, so, are assumed to be the same for modeling purposes. Diffusion inside and out of the ROI was modeled as a linear twoway process with a modified diffusion constant D0 based on the diffusion constant D measured by FCS (see above). The model system describes three variables represent ing the different fractions of fluorescent protein: free diffusion (xf), bound in the first binding state (xb1), and bound in the second binding state (xb2). The one-step model only contains xf and xb1. Since these different fractions cannot be distinguished directly in FRAP, the observable amount of fluorescence is given by xobs(t) = xf(t) þ xb1(t) or xobs(t) = xf(t) þ xb1(t) þ xb2(t), respectively.

24

Thorsten Lenser et al.

(A) xf

(B)

xb1

xb1

(C)

xf

xf

xb2

xb1

xb2

Fig. 7 Three models used to analyze FRAP data.(A) One-step binding/unbinding process. (B) Two-step process with sequential steps. (C) Two-step process with parallel, independent steps.

The ratio between background fluorescence and fluorescence inside the PML body (p) is determined individually for each type of molecule by using confocal microscopy and pixel intensity evaluation with MetaMorph software (Molecular Devices, Sunny vale, CA, USA). Since the FRAP values inside the ROI are normalized to 1.0 in steady state, the p values can be used to compute the relative concentration of free protein outside the ROI, namely, xf outside = 1/p. The binding/unbinding processes are represented with mass-action kinetics, which leads to the following ODE representation (exemplarily for the sequential two-step model, Fig. 7): dxf = koff xb1 kon1 xf þ D0 ðxSS f xf Þ dt dxb1 = kon1 xf koff 1 xb1 þ koff 2 xb2 kon2 xb1 dt

dxb2

= kon2 xb1 koff 2 xb2 dt An important quantity to describe the binding behavior in the FRAP models is the residence time, that is, the average time until a newly bound molecule returns to the state of free diffusion. For the one-step, the parallel, and the sequential models, the residence time is: Rt = 1=koff 1 1 kon2 1 kon1 Rt = þ ; kon1 þ kon2 koff 1 kon1 þ kon2 koff 2 0 1 1 @ kon2 A Rt = 1þ koff 1 koff 2

and

respectively.

4. Numerical Integration, Initial Conditions, and Parameter Fitting The mathematical model is numerically solved using the ode45 method in MATLAB (The MathWorks, Natick, MA, USA). Assuming an ideal bleach, all three

1. Fluorescence Fluctuation Microscopy

25

variables are initially set to zero. The timelines of the recorded data and the model predictions are then synchronized by aligning the time at which the sum of all three variables in the model prediction reaches the first measured fluorescence value. Parameter fitting is done by minimizing the sum (over time) of squared deviations between the fluorescence measurements and the predictions provided by the model, that is, the sum of all three variables. Even though the model is not too large, parameter fitting is difficult due to potential local optima and the nonlinear relation between parameters and model predictions. Therefore, an evolution strategy with covariance matrix adaptation (CMA-ES) was employed (Hansen and Ostermeier, 2001), using the MATLAB implementation by Nikolaus Hansen with default parameters.

IV. Results and Discussion A. FCS Measurements in the Cell Nucleus Figure 6 illustrates the typical FCS measurements in a cell nucleus expressing a GFP–PML fusion protein (Weidtkamp-Peters et al., 2008). PML is the major building subunit of PML nuclear bodies and therefore localizes diffusely throughout the nucleoplasm as well as accumulated in nuclear bodies (Fig. 6B). When the FCS laser beam is positioned in the nucleoplasm, a photon count rate trace can be recorded over time (Fig. 6C). Subsequent autocorrelation and fitting of the autocorrelation curve to an appropriate diffusion model allow one to determine the diffusion coefficient and the number of particles in the confocal volume (Fig. 6D). From the latter number, we can determine the concentration of GFP–PML to be 37 nM the soluble pool throughout the nucleus in this experiment. When the laser beam is positioned within a nuclear body, the count rate trace becomes a bleaching curve (Fig. 6E) and the corresponding autocorrelation curve is shifted to very long diffusion times (Fig. 6F). The FCS bleaching curve, however, also contains details about the binding process of the bleached molecules which can be further evaluated as described elsewhere (Müller et al., 2009; Weidemann et al., 2003). B. FRAP Measurements in the Cell Nucleus and Modeling Typical FRAP applications on nuclear proteins are shown in Fig. 8. Heterochromatin protein 1 (HP1) is a structural component of constitutive heterochromatin, yet its exchange rate at chromatin appears to be in the seconds range on superficial inspection of FRAP data (Fig. 8A). However, kinetic modeling of such FRAP data has revealed the presence of differently mobile fractions of HP1 in heterochromatin (Cheutin et al., 2004; Schmiedeberg et al., 2004), the slowest of which has a residence time of > 2 min (Müller et al., 2009). In contrast, the exchange rate of structural components at nuclear bodies is usually slow when observed by FRAP (Fig. 8B). Mathematical modeling of component exchange revealed that some proteins can have a residence time at PML bodies of up to 1 h (Brand et al. 2010; Weidtkamp-Peters et al., 2008). Interestingly, a subpopulation of coilin, the building subunit of CBs, exhibits a FRAP

26

Thorsten Lenser et al.

2 sec

10 sec

20 sec

(B) Pre

Post

2 min

4 min

6 min

(C) Pre

Post

0.5 hrs

1.5 hrs

3 hrs

(D) PML body kon/off ut

PML

PML

/o

k in PML

Nucleoplasm

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0

Normalized fluorescence

Post

Normalized fluorescence

(A) Pre

GFP-PML I

200

400 600 800 1000 Time (sec)

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0

GFP-PML II

200

400 600 800 1000 Time (sec)

(E) Protein

Cells +/+

GFP-PML I

PML

GFP-PML I

PML–/–

kon (s–1)

koff (s–1)

–1 kin (s )

kout (s–1)

R.t. (s)

bndout

bndin

0.0633 0.0015

0.0051 0.0015

0.0002 >0.0001

0.0004 >0.001

300 648

0.65 1.00

0.35 0.00

Fig. 8 Applying FRAP and mathematic modeling in the nucleus. Examples of FRAP experiments of GFPlabeled nuclear proteins showing rapid (A, GFP–HP1), slow (B, GFP–PML VI), and very slow (C, GFP CENP-H) exchange dynamics. Images were taken before (pre) and after (post) the bleach pulse and at different time points during fluorescence redistribution monitoring. Scale bar, 10 µm. In (C), cells coexpressed the replication factor PCNA fused to red fluorescent protein (red), allowing one to apply FRAP in cells during different stages of the cell cycle (Hemmerich et al., 2008). Details on HP1, PML, and centromere protein dynamics in living cells can be found in Schmiedberg et al. (2004), WeidtkampPeters et al. (2008), and Hemmerich et al. (2008), respectively. (D, left panel) Kinetic modeling of PML nuclear body assembly according to a diffusion-binding model. Molecules with the potential to accumulate at PML nuclear bodies move by diffusion (D) in the nucleoplasm outside nuclear bodies. Upon contact, molecules associate and dissociate from the periphery of the nuclear body with rate constants kon and koff, respectively. From the periphery, the nondissociating pool may penetrate into and out of the core of the nuclear body with rate constants kin and kout, respectively. (D, middle and right panels) Fitting of FRAP data with the diffusion-binding model. FRAP curves for the indicated GFP-tagged proteins (blue dots) were fitted using the diffusion-binding model (red solid lines). (E) Biophysical parameters extracted from the mathematical modeling approach for GFP–PML I in NIH-3T3 cells expressing endogenous PML (PMLþ/ þ ) or in NIH-3T3 cell obtained from a PML knockout mouse (PML/). R.t., mean residence time at nuclear bodies; bndout, fraction of molecules residing at the surface of the nuclear body; bndin, fraction of molecules residing in the core of the nuclear body. (See Plate no. 3 in the Color Plate Section.)

27

1. Fluorescence Fluctuation Microscopy

recovery halftime of 31 min in Xenopus oocytes, also indicating very long residence times (Hanwerger et al., 2003). Therefore, stably binding molecules may also play a critical role for the sustained architecture of a nuclear structure. A stable binding mechanism certainly underlies the structural integrity of the centromere, because many of its component parts exchange very slowly (i.e., CENP-H, Fig. 8C). Some

B A C (A1)

(A2)

(D)

Fluorescence intensity

255

0 4 × 10–4

2.7 2.2

D = 3.13 μm–2s–1

(B1)

(B2)

1.6

0,9 × 10–3

1.1

0

0.5 0.0

Diffusion coefficient (μm2/s–1)

0

D = 1.41 μm–2s–1

(C1)

(C2)

1.4 × 10–2 0

0.0040 0.0035 0.0030 0.0025 Correlation 0.0020 0.0015 0.0010 0.0005 0.0000

–250 –150 –50 50 –250

D = 0.29 μm–2s–1

Y

150 –150

–50

50 X

250 150

250

Fig. 9 Spatial mapping of nuclear protein mobility in the nucleus by RICS. A PML body component fused to GFP was expressed in HEp-2 cells. For RICS analysis, a time series of GFP fluorescence images was acquired in a subnuclear region of the nucleus (top image) by confocal microscopy (frame size: 512 512 pixels; pixel size: 0.03 µm; scan speed: 6.4 µs/pixel). Scale bar = 2 µm. Subregions with a size of 64 64 pixels within this time series containing nucleoplasm only (A1), a nucleolus (B), or a PML body (C) were extracted. Correlation spectra were assessed from these subregions (A2, B2, and C2, respectively) and the diffusion coefficient (D) determined from fitting of the correlation functions with a three-dimensional free diffusion model. (D) RICS analysis of a GFP-tagged chromatin-binding protein. The upper panel shows one image of the time series stack acquired as described in (A). Using a region of interest (ROI) analysis, which scans subregions within the original image, maps for relative molecule numbers and diffusion coefficients can be generated. Note that the diffusion map allows to highly resolve the diffusion behavior of the GFPtagged nuclear protein at different locations within the nucleus (red circles). The correlation can be visualized as a 2.5-dimensional map (bottom image).

28

Thorsten Lenser et al.

of them, such as CENP-A and CENP-I, do not turn over at centromeres at all (Hemmerich et al., 2008). Modeling of FRAP data for PML isoforms provides a deeper understanding of protein turnover at nuclear bodies. Modeling of FRAP curves obtained with GFPtagged isoforms at PML bodies has revealed the presence of at least two differently mobile populations (Fig. 8D and E) (Weidtkamp-Peters et al., 2008). Applying this model on FRAP curves for GFP–PML isoform I in cells that express endogenous PML protein (PMLþ/þ cells) or not (PML/ cells) immediately delivers insights into the assembly mechanism of PML bodies (Brand et al., 2010). For example, while there are two differently mobile populations of PML I at PML bodies in PMLþ/þ cells, only one exists in PML/ cells (Fig. 8E). The more than twofold increase in the residence time in PML/ cells also indicates the formation of a more immobile, aggregate-like structure. This in turn suggests that additional PML isoforms are required for “normal” exchange dynamics of PML I at nuclear bodies. All these quantitative information cannot be obtained by visual inspection of FRAP curves. C. RICS Measurements in the Nucleus RICS was performed in HEp-2 cell nuclei expressing a PML body component fused to GFP (Fig. 9A–C) or GFP–HP1 (Fig. 9D). Within the LSM images, subregions with a size of 64 64 pixels can be selected and subjected to RICS. This approach reveals different diffusion coefficients of the GFP-tagged nuclear proteins in different regions of the nucleus (Fig. 9). Alternatively, a diffusion coefficient map can be generated by scanning of 64 64 pixel overlapping subregions, as exemplified for GFP–HP1 (Fig. 9B). Such an analysis also provides a spatial determination of molecule dynamics in the nucleus. The RICS diffusion coefficient data obtained here are in good agreement with FCS data (Müller et al., 2009; Schmiedeberg et al., 2004; Weidtkamp-Peters et al., 2008).

V. Conclusions A. FRAP and FCS in Nuclear Cell Biology FRAP is the most commonly used technique to assess protein mobility in the nucleus, as it is feasible to most researchers through acquisition of a state-of-the-art confocal or spinning disc microscope equipped with the appropriate hardware and software. Bleaching experiments can be performed with most if not all currently available confocal microscopes using established protocols (Bancaud et al., 2009). Beginners will be satisfied with retrieving the recovery halftime of their proteins of interest, although even this simple approach requires thorough considerations, such as bleach pulse intensities and size and shape of bleach regions (Trembecka et al., 2010). Soon, the researcher desires to know more about the protein dynamics, foremost the binding constants, and this requires mathematical modeling of the FRAP data. If protein dynamics are in the microseconds to seconds range, FCS might also become

1. Fluorescence Fluctuation Microscopy

29

a powerful tool to extract binding parameters, since the autocorrelation data can also be modeled to include both free diffusion and binding events (Michelman-Ribeiro et al., 2009). For very slow processes (several minutes to hours), FRAP is the only method available because such time scales are inaccessible by FCS or RICS. B. Benefits of the RICS Technique RICS is a powerful tool to study dynamic processes in living cells covering a wide range of diffusion processes. Measurements can be executed even in the presence of large immobile or slowly moving structures and still give reliable diffusion coefficients of mobile molecules. Moreover, by extending the models, RICS is also able to characterize binding kinetics (Digman and Gratton, 2009). In principle, the method works on any confocal or even bright field microscope (Gaborski et al., 2010), given that instrumental settings are chosen appropriately and if analog detector filtering is taken into account. Since RICS uses intensity fluctuations, it can be used to cover a broad range of dynamics, investigating flow direction or the aggregation state of the molecules (Dignam and Gratton, 2009; Dignam et al., 2005). In this respect, RICS adds to the biologist’s tool kit to study cellular processes in a true quantitative way. C. Lessons for the Mathematical Modeling of FRAP FRAP modeling seems simple, but it is not! At the onset, the mathematical modeling of FRAP experiments seems straightforward: the models involved are relatively low dimensional and simple, and the relation between data and model seems clear. Digging deeper, it becomes obvious that things may not be as clear as they first seem to be. For example, the range of models that can be considered is actually quite wide: from the simple exponential function, via a compartmental model with multiple binding states (as in most cases), to spatially explicit models containing a detailed representation of the nuclear architecture (Beaudouin et al., 2006). From a different perspective, many potential problems with the data are unearthed: How is the fluorescence decay caused by laser scanning taken into account? Is there a “blinding” effect of the photobleach pulse on the first image acquisitions? How much noise is there in the data? And how much variation is there between repeated measurements? Taking together the uncer tainties about model and data, one is forced to conclude that the mathematical analysis of FRAP experiments is limited to rough estimations of the kinetic parameters (Mueller et al., 2008). Specifically, there is currently no possibility to arbitrarily increase the resolution and reliability of these estimations, no matter how much time and money is invested. Fundamental advances in the methodology, both in vivo and in silico, are needed for this. In a recent publication, Mueller et al. (2008) describe two principal sources of systematic error found in many approaches to FRAP analysis: a neglect of the role of diffusion and an incorrect approximation of the initial fluorescence distribution after the photobleach. They show that when applying different FRAP approaches that do not correct for these errors, the choice of the FRAP analysis method determines the kinetic

30

Thorsten Lenser et al.

parameters, not the type of protein under study (i.e., Sprague et al., 2004). Studies performed in our labs are designed from the onset to take diffusion into account. Diffusion is measured separately (by FCS or RICS) and then included into the FRAP model, which is used to estimate the binding coefficients. Moreover, the problem of the initial photobleach distribution does not seem too severe here, even though improvements in this matter are likely to add more precision to the results. There cannot be enough emphasis on the argument that in the current state, FRAP analysis by mathematical modeling can only yield predictions of the magnitude of kinetic parameters, not precise values (Mueller et al., 2008; Sprague and McNally, 2005). The main reason for this is that there is no “gold standard” for FRAP experiments, that is, no experimental setup where the “correct” values are precisely known. Such a standard could be used to calibrate and test FRAP analysis proce dures. Without it, systematic errors that underlie the different methods can easily go undetected, and the only possible safeguard is the comparison of results from different methodologies (Mueller et al., 2008).

D. Kinetic Analysis of Fluorescence Microscopy Experiments In contrast to dead matter under study in physics or chemistry, living entities always come with a high degree of noise, variability, and sensitivity to laboratory conditions. Thus, there is always a certain level of uncertainty implied in biological measurements. In analogy, mathematical models of biological systems always have to be more abstract than the real thing, and thus simplifications and approximations are unavoidable. In spite of all this, the strongest argument for the kind of scientific work presented here is that it is simply the best we have. When systems are inherently noisy and variable, measuring them can never be done without uncertainty. When systems are complex, modeling has to simplify and approximate. Therefore, we believe that with the methodologies currently available, the approach presented here is well worth the effort. The kinetic quantification of fluorescence measurements, in this detail only possible through mathematical modeling, has led to significant insights that would be unavail able by pure inspection of the measurement curves.

References Axelrod, D., Koppel, D. E., Schlessinger, J., Elson, E., and Webb, W. W. (1976). Mobility measurement by analysis of fluorescence photobleaching recovery kinetics. Biophys. J. 16 (9), 1055–1069. Bacia, K., and Schwille, P. (2007). Fluorescence correlation spectroscopy. Methods Mol. Biol. 398, 73–84. Bancaud, A., Huet, S., Rabut, G., and Ellenberg, J. (2009). Fluorescence perturbation techniques to study mobility and molecular dynamics of proteins in live cells: FRAP, photoactivation, photoconversion, and FLIP. In “Live Cell Imaging: A Laboratory Manual” (J. Swedlow, R. Goldman and D. Spector, eds.) Cold Spring Harbor Press, New York, USA. Beaudouin, J., Mora-Bermúdez, F., Klee, T., Daigle, N., and Ellenberg J. (2006). Dissecting the contribution of diffusion and interactions to the mobility of nuclear proteins. Biophys. J. 90(6), 1878–1894. Bernardi, R., and Pandolfi, P. P. (2007). Structure, dynamics and functions of promyelocytic leukaemia nuclear bodies. Nat Rev. Mol. Cell Biol. 8(12), 1006–1016.

1. Fluorescence Fluctuation Microscopy

31

Bogolyubov, D., Stepanova, I., and Parfenov, V. (2009). Universal nuclear domains of somatic and germ cells: Some lessons from oocyte interchromatin granule cluster and Cajal body structure and molecular composition. Bioessays 31(4), 400–409. Braeckmans, K., Peeters, L., Sanders, N. N., De Smedt, S. C., and Demeester J. (2003). Three-dimensional fluorescence recovery after photobleaching with the confocal scanning laser microscope. Biophys. J. 85(4), 2240–2252. Braga, J., Desterro, J. M., and Carmo-Fonseca, M. (2004). Intracellular macromolecular mobility measured by fluorescence recovery after photobleaching with confocal laser scanning microscopes. Mol. Biol. Cell 15(10), 4749–4760. Brand, P., Lenser, T., and Hemmerich, P. (2010). Assembly dynamics of PML nuclear bodies in living cells. PMC Biophys. 3(1), 3. Brown, C. M., Dalal, R. B., Hebert, B., Digman, M. A., Horwitz, A. R., and Gratton, E. (2008). Raster image correlation spectroscopy (RICS) for measuring fast protein dynamics and concentrations with a commer cial laser scanning confocal microscope. J. Microsc. 229(Pt 1), 78–91. Carrero, G., McDonald, D., Crawford, E., de Vries, G., and Hendzel, M. J. (2003). Using FRAP and mathematical modeling to determine the in vivo kinetics of nuclear proteins. Methods 29(1), 14–28. Cheutin, T., Gorski, S. A., May, K. M., Singh, P. B., and Misteli, T. (2004). In vivo dynamics of Swi6 in yeast: Evidence for a stochastic model of heterochromatin. Mol. Cell Biol. 24(8), 3157–3167. Cook, P. R. (1999). The organization of replication and transcription. Science 284(5421), 175–1790. Cook, P. R. (2010). A model for all genomes: The role of transcription factories. J. Mol. Biol. 395(1), 1–10. Cremer, T., Cremer, M., Dietzel, S., Müller, S., Solovei, I., and Fakan, S. (2006). Chromosome territories–a functional nuclear landscape. Curr. Opin. Cell Biol. 18(3), 307–316. Dauer, W. T., and Worman, H. J. (2009). The nuclear envelope as a signaling node in development and disease. Dev. Cell. 17(5), 626–638. Diekmann, S., and Hemmerich, P. (2005). “Visions of the Cell Nucleus.” American Scientific Publishers, Los Angeles, CA, USA. Digman, M. A., Sengupta, P., Wiseman, P. W., Brown, C. M., Horwitz, A. R., and Gratton, E. (2005). Fluctuation correlation spectroscopy with a laser-scanning microscope: Exploiting the hidden time struc ture. Biophys J. 88(5), L33–L66. Digman, M. A., and Gratton, E. (2009). Analysis of diffusion and binding in cells using the RICS approach. Microsc. Res. Tech. 72(4), 323–332. Dinant, C., Luijsterburg, M. S., Höfer, T., von Bornstaedt, G., Vermeulen, W., Houtsmuller, A. B., and van Driel, R. (2009). Assembly of multiprotein complexes that control genome function. J. Cell Biol. 185(1), 21–26. Dundr, M., Hoffmann-Rohrer, U., Hu, Q., Grummt, I., Rothblum, L. I., Phair, R. D., and Misteli, T. (2002). A kinetic framework for a mammalian RNA polymerase in vivo. Science 298(5598), 1623–1626. Dundr, M., Ospina, J. K., Sung, M. H., John, S., Upender, M., Ried, T., Hager, G. L., and Matera, A. G. (2007) Actin-dependent intranuclear repositioning of an active gene locus in vivo. J. Cell Biol. 179(6), 1095–1103. Enderlein, J., Gregor, I., Patra, D., and Fitter, J. (2004). Art and artefacts of fluorescence correlation spectroscopy. Curr. Pharm. Biotechnol. 5(2), 155–161. Frey, M. R. and Matera, A. G. (2001). RNA-mediated interaction of Cajal bodies and U2 snRNA genes. J. Cell Biol. 154(3), 499–509. Gaborski, T. R., Sealander, M. N., Ehrenberg, M., Waugh, R. E., and McGrath, J. L. (2010). Image correlation microscopy for uniform illumination. J Microsc. 237(1), 39–50. Gall, J. G. (2000). Cajal bodies: The first 100 years. Annu Rev Cell Dev Biol. 16, 273–300. Gorski, S., and Misteli, T. (2005). Systems biology in the cell nucleus. J. Cell Sci. 118(Pt 18), 4083–4092. Handwerger, K. E., and Gall, J. G. (2006). Subnuclear organelles: New insights into form and function. Trends Cell Biol. 16(1), 19–26. Handwerger, K. E., Murphy, C., and Gall, J. G. (2003). Steady-state dynamics of Cajal body components in the Xenopus germinal vesicle. J. Cell Biol. 160(4), 495–504. Hansen, N., and Ostermeier, A. (2001). Completely derandomized self-adaptation in evolution strategies. Evol. Comput. 9, 159–195.

32

Thorsten Lenser et al. Haustein, E., and Schwille, P. (2003). Ultrasensitive investigations of biological systems by fluorescence correlation spectroscopy. Methods 29(2), 153–166. Heard, E., and Bickmore, W. (2007). The ins and outs of gene regulation and chromosome territory organisation. Curr. Opin. Cell Biol. 19(3), 311–316. Hemmerich, P., Weidtkamp-Peters, S., Hoischen, C., Schmiedeberg, L., Erliandri, I., and Diekmann, S. (2008).

Dynamics of inner kinetochore assembly and maintenance in living cells. J Cell Biol. 180(6), 1101–1114.

Heun, P., Taddei, A., and Gasser, S. M. (2001). From snapshots to moving pictures: New perspectives on

nuclear organization. Trends Cell Biol. 11(12), 519–525. Lamond, A. I., and Spector, D. L. (2003). Nuclear speckles: A model for nuclear organelles. Nat. Rev. Mol. Cell Biol. 4(8), 605–612. Leonhardt, H., Sporbert, A., and Cardoso, M. C. (2000). Targeting regulatory factors to intranuclear replication sites. Crit. Rev. Eukaryot. Gene Expr. 10(2)., 127–133. Levi, V., and Gratton, E. (2008). Chromatin dynamics during interphase explored by single-particle tracking. Chromosome Res. 16(3), 439–449. Lukas, C., Bartek, J., and Lukas, J. (2005) Imaging of protein movement induced by chromosomal breakage: Tiny “local” lesions pose great “global” challenges. Chromosoma 114(3), 146–154. Magde, D., Elson, E. L., and Webb, W. W. (1974). Fluorescence correlation spectroscopy. II. An experi mental realization. Biopolymers 13(1), 29–61. Matera, A. G., Izaguire-Sierra, M., Praveen, K., and Rajendra, T. K. (2009). Nuclear bodies: Random aggregates of sticky proteins or crucibles of macromolecular assembly? Dev. Cell 17(5), 639–647. Mazza, D., Cella, F., Vicidomini, G., Krol, S., and Diaspro, A. (2007). Role of three-dimensional bleach distribution in confocal and two-photon fluorescence recovery after photobleaching experiments. Appl. Opt. 46(30), 7401–7411. McGrath, J. L., Tardy, Y., Dewey, C. F. Jr, Meister, J. J., and Hartwig, J. H. (1998). Simultaneous measurements of actin filament turnover, filament fraction, and monomer diffusion in endothelial cells. Biophys. J. 75(4), 2070–2078. Michelman-Ribeiro, A., Mazza, D., Rosales, T., Stasevich, T. J., Boukari, H., Rishi, V., Vinson, C., Knutson, J. R., and McNally, J. G. (2009). Direct measurement of association and dissociation rates of DNA binding in live cells by fluorescence correlation spectroscopy. Biophys. J. 97(1), 337–346. Misteli T. (2007). Beyond the sequence: Cellular organization of genome function. Cell 128(4), 787–800. Misteli T. (2008). Physiological importance of RNA and protein mobility in the cell nucleus. Histochem. Cell Biol. 129(1), 5–11. Mueller, F., Wach, P., and McNally, J. G. (2008). Evidence for a common mode of transcription factor interaction with chromatin as revealed by improved quantitative fluorescence recovery after photobleach ing. Biophys. J. 94(8), 3323–3339. Müller, K. P., Erdel, F., Caudron-Herger, M., Marth, C., Fodor, B. D., Richter, M., Scaranaro, M., Beaudouin, J., Wachsmuth, M., and Rippe, K. (2009). Multiscale analysis of dynamics and interactions of hetero chromatin protein 1 by fluorescence fluctuation microscopy. Biophys. J. 97(11), 2876–2885. Pandit, S., Wang, D., and Fu, X. D. (2008). Functional integration of transcriptional and RNA processing machineries. Curr. Opin. Cell Biol. 20, 260–265. Peters, R., Peters, J., Tews, K. H., Bähr, W. (1974). A microfluorimetric study of translational diffusion in erythrocyte membranes. Biochim. Biophys. Acta. 367(3), 282–294. Peters, R., Brünger, A., and Schulten, K. (1981). Continuous fluorescence microphotolysis: A sensitive method for study of diffusion processes in single cells. Proc. Natl. Acad. Sci. USA. 78(2), 962–966. Petersen, N. O., Höddelius, P. L., Wiseman, P. W., Seger, O., Magnusson, K. E. (1993). Quantitation of membrane receptor distributions by image correlation spectroscopy: Concept and application. Biophys. J. 65(3), 1135–1146. Phair, R. D., and Misteli, T. (2000). High mobility of proteins in the mammalian cell nucleus. Nature 404 (6778), 604–609. Phair, R. D., and Misteli, T. (2001). Kinetic modelling approaches to in vivo imaging. Nat. Rev. Mol. Cell Biol. 2(12), 898–907.

1. Fluorescence Fluctuation Microscopy

33

Politi, A., Moné, M.J., Houtsmuller, A.B., Hoogstraten, D., Vermeulen, W., Heinrich, R., and van Driel, R. (2005). Mathematical modeling of nucleotide excision repair reveals efficiency of sequential assembly strategies. Mol. Cell 19(5), 679–690. Probst, A. V., and Almouzni, G. (2008). Pericentric heterochromatin: Dynamic organization during early development in mammals. Differentiation 76(1), 15–23. Raska, I., Shaw, P. J., and Cmarko, D. (2006). Structure and function of the nucleolus in the spotlight. Curr. Opin. Cell Biol. 18(3), 325–334. Richter, K., Nessling, M., and Lichter, P. (2007). Experimental evidence for the influence of molecular crowding on nuclear architecture. J. Cell Sci. 120(Pt 9), 1673–1680. Schmiedeberg, L., Weisshart, K., Diekmann, S., Meyer. Zu Hoerste, G., and Hemmerich, P. (2004). Highand low-mobility populations of HP1 in heterochromatin of mammalian cells. Mol. Biol. Cell 15(6), 2819– 2833. Sirri, V., Urcuqui-Inchima, S., Roussel, P., and Hernandez-Verdun, D. (2008). Nucleolus: The fascinating nuclear body. Histochem. Cell Biol. 129(1), 13–31. Spector, D. L (2001). Nuclear domains. J. Cell Sci. 114(16), 2891–2893. Sprague, B. L., Pego, R. L., Stavreva, D. A., and McNally, J. G. (2004). Analysis of binding reactions by fluorescence recovery after photobleaching. Biophys. J. 86(6), 3473–3495. Sprague, B. L. and McNally, J. G. (2005). FRAP analysis of binding: Proper and fitting. Trends Cell Biol. 15 (2), 84–91. Sprague, B. L., Müller, F., Pego, R. L., Bungay, P. M., Stavreva, D. A., and McNally, J. G. (2006). Analysis of binding at a single spatially localized cluster of binding sites by fluorescence recovery after photobleaching. Biophys. J. 91(4), 1169–1191. Starr, D. A. (2009). A nuclear-envelope bridge positions nuclei and moves chromosomes. J. Cell Sci. 122(Pt 5), 577–86. Torok, D., Ching, R. W., and Bazett-Jones, D. P. (2009). PML nuclear bodies as sites of epigenetic regulation. Front Biosci. 14, 1325–1336. Trembecka, D. O., Kuzak, M., and Dobrucki, J. W. (2010). Conditions for using FRAP as a quantitative technique-Influence of the bleaching protocol. Cytometry A 77(4), 366–370. van Royen, M. E., Farla, P., Mattern, K. A., Geverts, B., Trapman, J., and Houtsmuller, A. B. (2009). Fluorescence recovery after photobleaching (FRAP) to study nuclear protein dynamics in living cells. Methods Mol. Biol. 464, 363–385 Visser, A. E., and Fell, D. A. (2007). Systems biology meets chromatin function. Workshop on Nuclear Organization. EMBO Rep. 8(5), 446–450. Wallrabe, H., and Periasamy, A. (2005). Imaging protein molecules using FRET and FLIM microscopy. Curr. Opin. Biotechnol. 16(1), 19–27. Weidemann, T., Wachsmuth, M., Knoch, T. A., Müller, G., Waldeck, W., and Langowski, J. (2003). Counting nucleosomes in living cells with a combination of fluorescence correlation spectroscopy and confocal imaging. J. Mol. Biol. 334(2), 229–240. Weidtkamp-Peters, S., Lenser, T., Negorev, D., Gerstner, N., Hofmann, T. G., Schwanitz, G., Hoischen, C., Maul, G., Dittrich, P., and Hemmerich, P. (2008). Dynamics of component exchange at PML nuclear bodies. J. Cell Sci. 121(Pt 16), 2731–2743. Weidtkamp-Peters, S., Felekyan, S., Bleckmann, A., Simon, R., Becker, W., Kühnemuth, R., and Seidel, C. A. (2009) Multiparameter fluorescence image spectroscopy to study molecular interactions. Photochem. Photobiol. Sci. 8(4), 470–480. Weidtkamp-Peters, S., Rahn, H. P., Cardoso, M.C., and Hemmerich, P. (2006). Replication of centromeric heterochromatin in mouse fibroblasts takes place in early, middle, and late S phase. Histochem Cell Biol. 125(1–2), 91–102. Weidtkamp-Peters, S., Weisshart, K., Schmiedeberg, L., and Hemmerich, P. (2009). Fluorescence correlation spectroscopy to assess the mobility of nuclear proteins. Methods Mol. Biol. 464, 321–341. Zhao, R., Bodnar, M. S., and Spector, D. L. (2009). Nuclear neighborhoods and gene expression. Curr. Opin. Genet. Dev. 19(2), 172–179.

CHAPTER 2

Studying Histone Modifications and Their Genomic Functions by Employing Chromatin Immunoprecipitation and Immunoblotting Ranveer S. Jayani*, Praveena L. Ramanujam*, and Sanjeev Galande *,† *

National Centre for Cell Science, Ganeshkhind, Pune 411007, India

†

Indian Institute of Science Education and Research, Pashan, Pune 411021, India

Abstract I. Introduction II. Rationale III. Chromatin Immunoprecipitation A. Materials B. Method IV. Immunoblotting and Coomassie Staining A. Materials B. Method V. Discussion VI. Summary

Acknowledgments

References

Abstract Histones are one of the most abundant and highly conserved proteins in eukaryotes. Apart from serving as structural entities for orderly compaction of genomes, they play an instrumental role in the regulation of many important biological processes involving DNA such as transcription, DNA repair, and the cell cycle. Histone modifications have been implicated in maintaining the “transcriptionally poised” state of important genesin embryonic stem cells. Histone modifications are believed to be responsible for METHODS IN CELL BIOLOGY, VOL. 98 Copyright 2010 Elsevier Inc. All rights reserved.

35

978-0-12-381009-0 DOI: 10.1016/S0091-679X(10)98002-3

36

Ranveer S. Jayani et al.

compartmentalization of chromatin into active and inactive domains. Hence, the tools and techniques required for studying these proteins are of utmost importance to biologists. This chapter provides a brief review of the posttranslational modifications of the N-terminal tails of histones and their biological roles, followed by step-by-step protocols for the most common techniques employed to study them. Here, we describe chromatin immunoprecipitation (ChIP) for studying the genomic functions of the most widely studied histone modifications, namely, histone H3 lysine 9 acetylation and histone H3 lysine 9 trimethylation that are typically associated with transcriptional activation and repression, respectively. Special emphasis has been given on the method of preparation of sonicated chromatin prior to immunoprecipitation since this single step affects the success of ChIP greatly and is often poorly described in published protocols. Protocol for histone isolation by acid-extraction and detection by Coomassie staining has also been described. We also describe the protocol for immunoblot analysis of histones using antibodies against key histone modifications. This chapter will serve as a useful resource in the study of histones and their posttranslational modifications.

I. Introduction In eukaryotes, genomic DNA is packaged within the nucleus in a structural form called chromatin. Chromatin is a complex of DNA and proteins that is present in highly condensed form in the nucleus. Such orderly structural organization enables packaging of DNA into the limited space of the nucleus. Inside nuclei, chromatin is segregated into distinct regions of transcriptionally inactive, condensed region, called the heterochromatin and transcriptionally active, “open” chromatin called the euchromatin (Cam et al., 2009). Recent studies have revealed that the demar cation of chromatin is highly dynamic and the chromatin “loopscape” of the nucleus is the direct manifestation of the physiological condition of the cell (Galande et al., 2007). This condensation of chromatin is brought about by its association with various proteins, majority of them being the histones. Nucleosome forms the basic repeat unit of the chromatin. It consists of two molecules of each core histone, namely, H3, H4, H2A, and H2B, which form a “histone octamer,” around which are wrapped 146 base pairs of double-stranded DNA in 1.65 left-handed superhelical turns (Jhunjhunwala et al., 2009). High-resolution crystal structure of histone octamer revealed the com pactness of the DNA–histone complex (Burlingame et al., 1985; Luger et al., 1997). Histone H1, the most diverse histone among all histones, acts as the “linker” histone and contacts the DNA at exit/entry of DNA strand on the nucleosome (Davey et al., 2002; Widom, 1998). Until the last decade of the previous century, histones were thought to play a packaging role with no bearing on gene regulation (Grunstein, 1992). A large number of studies in the last two decades have implicated histones in epigenetic gene regulation, DNA damage and repair, recombination, transcription, and replication (Lennartsson and Ekwall, 2009).

2. Linking Histone Modifications with Genome Organization

37

The core histones have their N-terminal “histone tails” protruding out of the globular domain and serve as the molecular memory systems (Strahl and Allis, 2000). The histone tails are important motifs of histones, which contribute to the integrity of the nucleosome and serve as docking sites for various effector proteins (Brower-Toland et al., 2005). The N-terminal histone tails are decorated with various histone modifications, namely, acetylation, methylation, phosphorylation, sumoylation (Shiio and Elsenman, 2003), and ADP ribosylation (Kouzarides, 2007) (Fig. 1) whereas ubiquitination occurs at the C-terminus (Jason et al., 2002; Robzyk et al., 2000; Thorne et al., 1987). These modifications play an important role in orchestrating diverse range of cellular pro cesses, which have been summarized in Fig. 2. Various studies have also shown that the histone modifications in the globular domains of core histones also play an important role in many physiological processes (Rogakou et al., 1998; Schneider et al.,

Fig. 1 A cartoon of important posttranslational modifications of core histones. The globular domains of the core histone are depicted in the center as solid circles with N-terminal tails protruding out. The amino acid residues in the tails have been numbered according to their positions in human histones and the respective modifications have been shown on the residues. A few important globular residues with corresponding modifications have also been shown. A color code for different amino acids has been placed on the left side. Standard one-letter abbreviations have been used for each amino acid (A, alanine; R, arginine; E, glutamic acid; Q, glutamine; G, glycine; H, histidine; L, leucine; K, lysine; P, proline; S, serine; T, threonine; V, valine). Me, methylation; ac, acetylation; P, phosphorylation; Ub, ubiquitination; and N, N-terminal end of the histones.

38

Ranveer S. Jayani et al. Chromatin condensation (T3, T11, T32) Transcriptional activation (S10, S28)

Phosphorylation

Transcription activation (K9, K14, K18, K23, K27) DNA repair (K14, K18, K23)

Acetylation

Transcription activation (K4, R17, K79) Transcription repression (K9, K27)

Phosphorylation

Chromatin condensation (S1)

Acetylation

Transcription activation (K5, K8, K12, K16) DNA repair (K5, K8, K12, K16)

Methylation

Methylation

Transcription activation (R3) Transcription repression (K20, K59)

Mitosis (S1, T119) Chromosomal stability (S121) DNA repair (S129, S139)

Phosphorylation

Phosphorylation

Transcriptional activation (K4, K5, K7)

Acetylation

Spermatogenesis (K119)

Ubiquitination

H3

H2A

H4

H2B

Chromatin condensation (S14) Transcription activation (S33)

Acetylation

Transcription activation (K5, K11, K12, K15, K20)

Ubiquitination

Meiosis (K120) Euchromatin, spermatogenesis (K123)

Fig. 2 Biological roles of various histone modifications. The figure represents the major modifications undergone by various histones and the biological role associated with those modifications. The residues involved in the biological processes are mentioned in parenthesis.

2006; Schulze et al., 2009; Vempati et al., 2010). The histone modifications constitute a “Histone code” by acting sequentially or in combination, which in turn act as a signal for various downstream processes (Peterson and Laniel, 2004; Strahl and Allis, 2000). Acetylation of histone tails is the most studied modification (Grunstein, 1997). Hebbes et al. (1988) provided the first clue for the relation between the acetylation of histones and a particular functional state of chromatin with the demonstration that active chromatin fractions are enriched in acetylated histones. Acetylation alters the conformation of chromatin fiber and facilitates the binding of transcription factors to DNA. It is known that acetylated chromatin is more susceptible to digestion by nucleases as compared to control chromatin (Krajewski and Becker, 1998; Simpson, 1978) and does not follow the condensation pattern of control chromatin under high salt conditions (Annunziato et al., 1988). These results indicate that acetylation of histones at various lysine residues plays an important role in facilitating active tran scription. Discovery of the enzymes that bring about acetylation (histone acetyl transferases (HATs)) and deacetylation (Histone deacetylases (HDACs)) of histones has greatly increased our understanding of the underlying mechanisms involved in the manifestation of the in vivo roles played by these modifications (Struhl, 1998). Methylation of histones was first described in 1964 (Murray, 1964). Methylation of lysine and arginine residues was among the less studied posttranslational modifications affecting histones (Strahl and Allis, 2000). However, recent studies have revealed that methylation is one of the more important histone modifications as histone methylation at lysine and arginine residues are relatively stable modifications and may thus play an important role in transferring the epigenetic information during cell division

2. Linking Histone Modifications with Genome Organization

39

(Barski et al., 2007). In addition, they can force their regulatory role at multiple levels as the �-amino group of lysine residue can be mono-, di-, and tri methylated and the side chain guanidino moiety of arginine residues can be methylated symmetrically or asymmetrically (Zhang and Reinberg, 2001). Histone methylation can function as a repressive or activation mark depending on the residue modified. H3K9 methylation mark is associated with silencing of the interferon-� locus (Gyory et al., 2004); H3K27 methylation is a repressive mark associated with transcriptional silencing of the HOX gene cluster in Drosophila (Muller et al., 2002) and human cells (Cao et al., 2002). High-throughput whole genome analyses have shown H3K4(me)3 mark to be found on the nucleosomes flanking the transcription start sites (TSSs) of actively transcribed genes in all eukaryotes examined till date (Guenther et al., 2007; Heintzman and Ren, 2007; Heintzman et al., 2007; Hon et al., 2009; Ozsolak et al., 2007; Schones and Zhao, 2008). Also, H3K4me mark is enriched on enhancers (Heintzman et al., 2007). H3K36(me)3 is associated with gene bodies of actively transcribed genes. Studies in Caenorhabditis elegans, mouse, and human showed that H3K36(me)3 is highly enriched at exons as compared to introns (Kolasinska-Zwierz et al., 2009; Hon et al., 2009). This observation strengthens the hypothesis that transcription and spli cing are coupled events (Hon et al., 2009; Sims and Reinberg, 2009). Methylation of histones is catalyzed by three distinct families of proteins: the SET domain-containing family of proteins for modifying lysine residues; the non-SET domain family of methylases consisting of a single member—Dot1 (Disruptor of Telomeric silencing 1), which methylates H3K79 residue located within the globular domain of histone H3 (Lu et al., 2008; Schulze et al., 2009); and the protein arginine methyltransferases (PRMT) family for modifying arginine residues (Trievel, 2004). Although the enzymes that bring about histone methylation have been known for long, the discovery of histone demethylases has highlighted the importance of methylation in recent years (Cloos et al., 2008; Shen et al., 2009). Two classes of histone lysine demethylases (KDMs) have been identified: the amine oxidase domain-containing LSD1/KDM1 (Shi et al., 2004) and the jumonji-domain-containing family (Nottke et al., 2009). LSD1/KDM1 demethylates mono- and dimethylated H3K4 and H3K9 residues (Metzger and Schule, 2007; Nottke et al., 2009). The jmc-domain- containing proteins modify di- and trimethylated H3K9 (Yamane et al., 2006), H3K27 and H3K36 residues (Nottke et al., 2009). There is evidence linking demethylases with methyltransferases both at the level of physical interaction and at the functional level (Shen et al., 2009). Phosphorylation of serine residues in histone tails also plays an important role in transcription, DNA repair, apoptosis, and chromosome condensation (Baker et al., 2010; Cheung et al., 2000). Phosphorylation of histone H3 has also been linked to transcriptional regulation during interphase. Mass spectrometric analyses of human histone H3 have revealed that phosphorylation and methylation are found on adjacent residues of the same H3 molecule such as Lys9/Ser10 and Lys27/Ser28 (Bonenfant et al., 2007; Garcia et al., 2005) during mitosis providing support for the “methyla tion/phosphorylation” binary switch hypothesis according to which, posttranslational modifications on adjacent amino acids, such as methylation of H3K9 and

40

Ranveer S. Jayani et al.

phosphorylation of H3S10, could modulate the binding of effector molecules to histones (Cerutti and Casas-Mollano, 2009; Fischle et al., 2003). Reports suggest the existence of cross talk between phosphorylation and methylation in mammalian cells. Silenced genes are generally present in regions rich in heterochromatin. Hetero chromatin is characterized by the presence of Heterochromatin protein (HP1) anchored to the chromatin by the H3K9(me)3 mark (Cerutti and Casas-Mollano, 2009; Ebert et al., 2006; Kouzarides, 2007). Upon initiation of mitosis, H3S10 is phosphorylated leading to dissociation of HP1 from chromosomes without an altera tion in H3K9(me)3 levels (Cerutti and Casas-Mollano, 2009; Fischle et al., 2005; Hirota et al., 2005) whereas phosphorylation of H3T11 potentiates demethylation of H3K9 (Metzger et al., 2008) Histone ubiquitination, like other histone modifications, plays an important role in gene regulation. Although all four core histones undergo ubiquitination, only that of H2A and H2B have been well studied. Both H2A and H2B can undergo mono- as well as polyubiquitination (Weake and Workman, 2008). The relationship between tran scription and histone ubiquitination is a complex phenomenon as ubiquitination can be associated with both active and inactive loci. For instance, histones present at the active immunoglobulin �-chain loci are deubiquitinated, whereas the transcriptionally inac tive Tetrhymena thermophila micronuclei and mouse spermatid sex body carry ubiqui tinated histones (Huang et al., 1986; Zhang, 2003). Simultaneously, several early studies showed an abundance of ubiquitinated histones at transcriptionally active gene loci. For example, nucleosomes of the transcriptionally poised HSP70 gene contain up to 50% ubiquitinated histone H2A whereas nucleosomes associated with untranscribed satellite DNA contain only one ubiquitinated H2A for every 25 nucleo somes (Bhaumik et al., 2007; Levinger and Varshavsky, 1982; Zhang, 2003). Further more, inhibiting transcription abolishes ubiquitinated H2B (Davie and Murphy, 1990; Ericsson et al., 1986; Zhang, 2003) suggesting that maintenance of ubiquitinated H2B is dependent on the ongoing transcription. Thus, the outcome of histone ubiquitination at a particular gene locus in terms of its transcriptional status is context-dependent, based on gene location or possibly the presence of other histone covalent modifications such as acetylation and methylation. Research findings that the C-terminal of HDAC6 can directly bind to ubiquitin (Seigneurin-Berny et al., 2001; Zhang, 2003), suggest a potential link between histone ubiquitination and histone acetylation. However, sub stantial evidence is now available that functionally links histone ubiquitination to histone methylation. Recent studies have shown that interaction of functional Rad6 with intact H3K123 is required for COMPASS-mediated H3K4 methylation (Bhaumik et al., 2007; Hon et al., 2009; Kim et al., 2009; Lee et al., 2007) and Dot1-mediated H3K79 methylation (Shilatifard, 2006; Schulze et al., 2009). In the absence of H2B ubiquitination, COMPASS lacks its Cps35 subunit and can only monomethylate H3K4. Ubiquitination of H2B by Rad6 (which is a component of the BUR complex) enables Csp35 recruitment to COMPASS leading to COMPASS mediated di- and trimethyla tion of H3K4 (Lee et al., 2007; Weake and Workman, 2008). Furthermore, mutations in the BUR complex that reduce H2B ubiquitination, specifically affect only H3K4 trimethylation (Laribee et al., 2005; Weake and Workman, 2008). Thus, from these

2. Linking Histone Modifications with Genome Organization

41

studies, it is clear that prior ubiquitination of H2B specifically affects di-and trimethy lation of H3K4 (Dehe et al., 2005; Schneider et al., 2005; Shahbazian et al., 2005; Weake and Workman, 2008). Histone H2B ubiquitination, plays an important role in trimethylation of H3K79 at the HOXA9 promoter region (Krivtsov et al., 2008; Schulze et al., 2009). In contrast, methylation of H3K36 requires deubiquitination of H2B (Zhu et al., 2005). DNA methylation and various histone posttranslational modifications also seem to cross talk. During the turning off of pluripotentcy genes in embryonic stem cells, pluripoten genes such as Oct3/4 and Nanog have unmethylated CpG islands and have promoters enriched in H3K9ac, H3K27ac, and H3K4me (Cedar and Bergman, 2009; Fuhrmann et al., 2001). At the onset of differentiation, lysine residues are deacetylated by HDAC and H3K4 is demethylated. This is followed by methylation of H3K9 that serves as a docking site for heterochromatin protein 1 (HP1) (Cedar and Bergman, 2009; Feldman et al., 2006) resulting in the formation of local heterochromatin. Finally, the underlying DNA is methylated de novo by the methylases DNMT3A and DNMT3B (Cedar and Bergman, 2009; Epsztejn-Litman et al., 2008; Feldman et al., 2006), thereby resulting in turning off of pluripotency genes. A bivalent chromatin mark consisting of regions harboring stretches of both H3 lysine 27 trimethylation (associated with repression) and H3 lysine 4 trimethylation (associated with activation) modifications serves to poise key developmental genes for lineage-specific activation or repression in ES cells (Bernstein et al., 2006). In ES cells, almost all High CpG Promoters (HCPs) are marked by either H3K4(me)3 alone or H3K4(me)3 in combination with H3K27(me)3. Majority of HCPs having bivalent marks in stem cells resolve to a monovalent status in committed cells. Those HCPs that resolve to H3K4(me)3 state alone show increased activation, whereas those that resolve to H3K27(me)3 state alone are silenced. Notably, few HCPs remain bivalent even after differentiation and continue to be repressed (Bernstein et al., 2006; Mikkelsen et al., 2007). Thus these HCPs having the bivalent chromatin mark are poised for lineage-specific activation or repression (Pietersen and van Lohuizen, 2008). Chromatin structure is a dynamic entity that constantly changes in response to external stimuli. Considerable evidence has accumulated suggesting that cellular state is closely related to chromatin state—modifications of proteins, especially histones involved in genome packaging. In addition to defining and controlling gene expression patterns in spatial and temporal manner, chromatin modifications also determine a cells’ response to environmental or developmental cues in terms of its transcriptional output. The ‘poised’ phenomenon has been well documented at promoters where a bivalent chromatin state ensures a poised transcriptional state critical for development (Bernstein et al., 2006; Mikkelsen et al., 2007), and most likely also applies to enhancers (Heintzman et al., 2009; Hon et al., 2009; Lupien et al., 2008) and, by extension, to other regulatory elements as well. These poised elements may be most critical in defining the cellular response. Although the histones bring about their cellular effects through the modification of their tail and globular domain residues, recently, a new mechanism of “histone tail clipping” has been described in mouse embryonic stem cells (Duncan et al., 2008) and

42

Ranveer S. Jayani et al.

Saccharomyces cerevisiae (Santos-Rosa et al., 2009), where the authors describe the mechanism of proteolytic processing of N-terminal tail of histone H3 by specific enzymes. In light of the above discussion and the plethora of literature available on the importance of histones and histone modifications in regulating the chromatin structure and gene regulation, described here are a few methods that can be used to study these modifications and their role in diverse cellular processes. The authors describe a method for chromatin immunoprecipitation (ChIP) for representative histone modifi cations (histone H3 lysine 9 acetylation and histone H3 lysine 9 trimethylation) on c-myc promoter upon PMA/ionomycin activation (Brunner et al., 2000) and LiCl treatment (Notani et al., 2010). Also described is a method for acid extraction of histones and protocols for Coomassie staining of histones and study of the histone modifications by immunoblot analysis using antibodies against key histone modifications.

II. Rationale Histones play an important role in maintaining the genomic landscape within the nucleus. The spatial and temporal changes in the levels of posttranslational modifica tions in histone tails as well as in the core globular domains dictate the active, inactive or poised status of a particular genomic locus. Hence, the study of these modifications in different physiological conditions is an important tool for understanding various cellular mechanisms. In this chapter, we describe two important tools which are typically employed toward accomplishing the above-said goal. These include ChIP analysis using antibodies to various histone modifications and immunoblot analysis of histone modifications. The chromatin obtained by ChIP can be further used for the ChIP-on-chip or ChIP sequencing (ChIP-seq) analysis to get a better insight into the target genomic regions, which harbor various histone modifications during a specific physiological state of the cell.

III. Chromatin Immunoprecipitation A. Materials 1. Jurkat cells (1 107 108) grown in RPMI-1640 (Rosewell Park Memorial Institute-1640) culture medium supplemented with 10% FBS (fetal bovine serum) and penicillin/streptomycin, under 5% CO2 atmosphere. 2. PMA (phorbol 12-myristate 13-acetate) (Sigma-Aldrich, St. Louis, MO, USA). 3. Ionomycin (Sigma-Aldrich, St. Louis, MO, USA). 4. Lithium chloride (LiCl, Sigma-Aldrich, St. Louis, MO, USA). 5. Formaldehyde (37%, Sigma-Aldrich, St. Louis, MO, USA). 6. Glycine (2M, USB, Cleveland, OH, USA).

2. Linking Histone Modifications with Genome Organization

43

7. PBS, 1 (Phosphate-buffered saline)-10 mM phosphate buffer (pH 7.4), and 138 mM sodium chloride (Sigma-Aldrich, St. Louis, MO, USA). 8. Wash buffer-1 (WB-1): 0.25% Triton X-100, 10 mM EDTA (pH 8.0), 0.5 mM EGTA (pH 8.0), 10 mM HEPES (pH 7.5), 10 mM sodium butyrate, and 1 protease inhibitor cocktail (Roche, Indianapolis, IN, USA). 9. Wash buffer-2 (WB-2): 0.2 M NaCl, 1 mM EDTA (pH 8.0), 0.5 mM EGTA (pH 8.0), 10 mM HEPES (pH 7.5), 10 mM sodium butyrate, and 1 protease inhibitor cocktail (Roche, Indianapolis, IN, USA). 10. Lysis buffer: 15 mM NaCl, 25 mM Tris-HCl (pH 7.5), 5 mM EDTA (pH 8.0), 1% Triton X-100, 0.1% sodium dodecyl sulfate (SDS), 0.5% sodium deoxycholate, 10 mM sodium butyrate, and 1 protease inhibitor cocktail (Roche Indianapolis, IN, USA). 11. Sonicator (Bioruptor XL, Diagenode, Belgium). 12. Thermomixer (Eppendorf AG, Hamburg, Germany). 13. Protein A/G plus ultralink resin (Thermo Scientific, Rockford, IL, USA). 14. Salmon sperm DNA (20 mg/ml, USB). 15. Bovine serum albumin (BSA, 10 mg/ml, New England Biolabs, Ipswich, MA, USA). 16. End-to-end IP rocker (VWR Scientific, San Francisco, CA, USA). 17. Histone modification antibodies (anti-pan H3, anti-H3K4(me)3, anti-H3K9(me)3, H3K9(ac), and anti-H3K27(me)3) and control antibodies (normal rabbit IgG) (all from Millipore, Bedford, MA, USA). 18. Radio immunoprecipitation assay (RIPA) buffer: 0.1% SDS, 1% sodium deoxycholate, 150 mM NaCl, 2 mM EDTA (pH 8.0), 0.2 mM sodium orthovandate, 1% IGEPAL CA-630, 10 mM sodium phosphate (pH 7.2). 19. Tris-EDTA buffer (TE buffer): 10 mM Tris–HCl (pH 8.0), 1 mM EDTA (pH 8.0). 20. SDS (10%, USB, Cleveland, OH, USA). 21. Sodium bicarbonate (NaHCO3, 1M). 22. Dithiothreitol (DTT, 1M). 23. Sodium chloride (NaCl, 4M). 24. Ethylene diamine tetraacetate (EDTA, pH 8.0). 25. Tris–HCl (1M, pH 6.5). 26. Proteinase K (10 mg/ml, USB, Cleveland, OH, USA). 27. Phenol–chloroform–isoamylalcohol (25:24:1). 28. Chloroform (CHCl3). 29. Sodium acetate (CH3COONa, 3 M, pH 5.4). 30. Ethanol (100%). 31. Taq DNA polymerase (5 U/µl, Promega, Madison, WI, USA). 32. Taq DNA polymerase buffer (10, Promega): Tris–HCl (200 mM, pH 8.8), 100 mM KCl, 100 mM (NH4)2SO4, 20 mM MgSO4, 1.0% Triton X-100 and 1 mg/ml BSA. 33. dNTP mix (2.5 mM each, GE Healthcare, Piscataway, NJ, USA). 34. Tris–borate–EDTA (TBE) buffer (5): 445 mM Tris base, 445 mM boric acid, and 20 mM EDTA, pH 8.0). 35. Ethidium bromide (1 mg/ml).

44

Ranveer S. Jayani et al.

36. DNA loading dye (6): 10 mm Tris–Cl (pH 7.6), 0.03% bromophenol blue, 0.03% xylene cyanol, 60 mM EDTA (pH 8.0), and 30% glycerol. 37. Agarose (Sigma, Sigma-Aldrich). 38. Human c-Myc locus-specific oligonucleotide primers (forward primer: GTGAATACACGTTTGCGGGTTAC and reverse primer: AGAGACCCTTGTGAAAAAAACCG). 39. SYBR green IQ real-time PCR reagent mix (Bio-Rad, Hercules, CA, USA). 40. Gel documentation system (Chemigenius, Syngene, Cambridge, UK). 41. Ethchinmate (Nippon Gene, Toyama, Japan).

B. Method The method for chromatin immunoprecipitation has been adapted from Weinmann and Farnham (2002) with some modifications. It can be divided into two parts: I. Preparation of cross-linked chromatin: 1. Treat Jurkat cells with PMA (0.02 µM)/Ionomycin (0.04 µM) or 10 mM LiCl for 24 h. Cross-link 1 107 – 108 Jurkat cells by adding formaldehyde (37%) to a final concentration of 1%, directly to the culture medium in the plate/flask. Incubate on rocker for 10 min at room temperature. 2. Stop cross-linking by adding 2 M glycine to a final concentration of 125 mM. 3. Wash cells twice with ice-cold 1 PBS (pH 7.4). Collect cells by scraping (for adherent cells) or centrifugation (for suspended cells). 4. Wash the cell pellet sequentially with WB-1 and WB-2. Collect cells by centrifugation for 5 min at 2000 g, 4°C. The cells can be stored in 80°C deep freezer for up to 4 weeks at this stage. 5. Gently resuspend the cell pellet in lysis buffer and incubate on ice for 10 min. 6. Sonicate the suspension using a Bioruptor XL, for 7 min using a pulse of 10 s “on” and 10 s “off”. 7. At this stage, remove 200 µl of each sample to check the extent of chromatin shearing. Centrifuge at 12,000 g, 10 min at 4°C and collect the supernatant in a fresh tube. 8. Add 0.05 volume of 4M NaCl to the sample removed and decrosslink at 65°C for 2–4 h. The decrosslinked sample can be directly checked for sonication efficiency at this point. However, one can also treat it with RNase and subsequently with proteinase K to reveal proper DNA fragmentation in the form of a tight band true to its size (depicted in Fig. 3) 9. Run 25 µl of this sample on a 0.8% agarose gel. The fragment size of sonicated DNA should typically range between 400 and 600 bp (Fig. 3). If the chromatin is not sheared to this size, sonicate again, repeat decrosslinking, and check on gel for ensuring optimal fragmentation. 10. Centrifuge the remaining chromatin at 13,000 g in a microcentrifuge, 4oC, 10 min. Collect supernatant, this is the input-soluble cross-linked chromatin. This may be diluted 10-fold with lysis buffer before immunoprecipitation. (Fig. 4)

45

bp

100 bp ladder

1 Kbp ladder

2. Linking Histone Modifications with Genome Organization

1

2

3

4

3000 1500 1000 500

100

Fig. 3 Optimal shearing of chromatin prior to ChIP. A small portion of chromatin was checked for extent of shearing after sonication. Equal amounts of non-decrosslinked (lane 1); decrosslinked (lane 2); decrosslinked and RNase-treated (lane 3); and decrosslinked, RNase-treated, and proteinase K-treated (lane 4) chromatin samples were resolved on a 1% agarose gel. DNA size markers (1 kbp and 100 bp ladders) as indicated.

II. Immunoprecipitation 1. Preclear the chromatin by adding 20 µl/ml of protein A/G-plus bead cocktail (50% slurry, 100 µg of salmon sperm DNA/ml, 500 µg of BSA/ml) and rocking at 4oC for 1–4 h. 2. Centrifuge at 1,000 g at 4oC for 5 min and collect supernatant. 3. Divide the supernatant into aliquots. Incubate each aliquot with 2 µg each of specific antibodies (anti-H3K9 ac and anti-H3K9(me)3) and isotype control (normal rabbit IgG). (At this stage, one part should be stored as input at 4oC.) 4. Maintain at 4oC for overnight on an end-to-end rocker. 5. Add 20 µl of protein A/G-plus bead cocktail, and continue rocking at 4oC for 2–4 h. 6. Harvest beads by centrifugation at 1,000 g at 4oC for 5 min. 7. Wash beads twice with RIPA buffer and twice with TE. (The samples should be kept at room temeprature on end-to-end rocker for 5 min between each wash.) 8. Elute the chromatin antibody complexes by adding 2% SDS, 0.1 M NaHCO3, and 10 mM DTT to the beads and incubating at 37oC for 1h at maximum shaking on a thermomixer. 9. Reverse cross-link by addition of 0.05 volume of 4M NaCl and incubation for 4h at 65oC (input should also be included from this step.) 10. Treat the chromatin with RNase (100 µg/ml) at 37oC for 1h to remove any RNA contaminants.

46

Ranveer S. Jayani et al.

11. Add 0.025 volume of 0.5M EDTA (pH 8.0), 0.05 volume of 1M Tris–HCl (pH 6.5) and proteinase K (100 µg/ml). Incubate for 1 h at 45oC. 12. Recover DNA by extraction using phenol–chloroform–isoamylalcohol. Precipitate by addition of 0.1 volume of 3 M sodium acetate (pH 5.2), 1 µl/ 100 µl ethchinmate and 2.5 volumes of ethanol. Ethchinmate can be replaced by 20 µg of glycogen per ml. 13. Wash the pellet with 70% ethanol and air dry. 14. Dissolve precipitated DNA in water and analyze by PCR using gene-specific primers (c-Myc as an example in this article) using one cycle of 95°C for 5 min and 35 cycles of 95°C for 1 min, 58°C for 30s, and 72°C for 1 min and analyze the amplicons by resolving on a 1% agarose gel. 15. To get a better approximation of change in fold expression, perform quantitative PCRs using SYBR green IQ supermix (Bio-Rad) and the ICycler IQ real-time thermal cycler (Bio-Rad). Changes in threshold cycle (CT) values can be calculated using the following formula: ΔCT=CT(IP)- CT(IgG). Fold change in occupancy is calculated as 2ΔCT.

IV. Immunoblotting and Coomassie Staining A. Materials 1. Jurkat cells (1 107 108) grown in RPMI-1640 supplemented with 10% FBS and penicillin/streptomycin, under 5% CO2 atmosphere. 2. PBS, 1 (Phosphate buffered saline): 10 mM phosphate buffer (pH 7.4), 138 mM sodium chloride (Sigma-Aldrich, St. Louis, MO, USA). 3. Lysis buffer: 10 mM Tris–Cl (pH 8.0), 1 mM KCl, 1.5 mM MgCl2, 1 mM DTT, 1 protease inhibitor cocktail, 10 mM sodium butyrate, and 1 mM sodium orthovanadate. 4. Sulfuric acid (H2SO4, 0.4 N) 5. Trichloroacetic acid (TCA, 100%): dissolve 22.0 g TCA in 10 ml distilled water. 6. Bio-Rad Dc protein assay kit (Bio-Rad, Hercules, CA, USA). 7. Bio-Rad Mini Protean-3 assembly (Bio-Rad, Hercules, CA, USA). 8. Sigmacote (Sigma-Aldrich, St. Louis, MO, USA). 9. Resolving gel (15%), for 6 ml: 1875 µl of acrylamide (40%), 500 µl bisacrylamide (2%), 1250 µl lower buffer (4), 1375 µl distilled water, 50 µl ammonium persulfate, and 5 µl N,N,N0 ,N0 -tetramethyethylenediamine (TEMED). 10. Stacking gel (4.5%), for 2 ml: 200 µl of acrylamide (40%), 100 µl bisacrylamide (2%), 500 µl lower buffer (4), 1180 µl distilled water, 20 µl ammonium persulfate, and 2 µl N,N,N0 ,N0 -tetramethyethylenediamine (TEMED). 11. SDS-PAGE running buffer (5), for 1l: 15.1 g Tris base, 72 g glycine and 5 g SDS. 12. SDS sample buffer (6): 0.28 M Tris–Cl (pH 6.8), 10% SDS, 0.5 M DTT, 30% glycerol, and 0.002% bromophenol blue. 13. Whatman chromatography paper, 3 MM Chr (Whatman Plc, Kent, UK).

2. Linking Histone Modifications with Genome Organization

47

14. Transfer buffer (10): 100 mM monobasic sodium phosphate (NaH2PO4H2O) and 100 mM dibasic sodium phosphate (Na2HPO47H2O). 15. Protein transfer assembly (Bio-Rad, Hercules, CA, USA). 16. PVDF membrane (Polyvinylidene fluoride, Millipore, Billerica, MA, USA). 17. TST (Tris–Saline–Tween, 1): 20 mM Tris–Cl (pH 7.4), 0.5 M NaCl and 0.05% Tween-20. 18. Blocking buffer: 5% nonfat dry milk in 1TST. 19. Primary antibodies against various histone posttranslational modifications: antipan H3 (Abcam plc, Cambridge, UK), anti-H3K4(me)3, anti-H3K9(me)3, H3K9 (ac), and anti-H3K27(me)3 (Millipore, Billerica, MA, USA). 20. Anti-rabbit IgG-HRP conjugate (Bio-Rad, Hercules, CA, USA). 21. Immobilon Western Chemiluminescent HRP Substrate (Millipore, Billerica, MA, USA). 22. Fixer solution: 10% acetic acid and 50% methanol. 23. Staining solution: 10% acetic acid, 50% methanol, and 0.25% Coomassie Brilliant Blue (CBB) R-250. 24. Destaining solution: 10% acetic acid and 30% methanol. B. Method I. Acid Extraction of Histones Acid extraction was performed as described by Shechter et al. (2007), with a few modifications. The steps involved are described below. 1. Take 1 106 Jurkat cells and collect them in an eppendorf by centrifugation at 1000 g for 5 min at 4oC. Wash the cells with 1 PBS. 2. Gently resuspend the cells in 1 ml of lysis buffer and incubate on an end-to-end rocker for 1 h to rupture the cells and release the nuclei in solution. 3. Collect the nuclei by centrifugation (12,000 g, 10 min, 4oC). 4. Discard the supernatant, add 600 µl of 0.4 N H2SO4, and resuspend the nuclei completely (Nuclei form compact clumps and may need to be vortexed to completely resuspend them). 5. Incubate on the rocker for 2–4 h at 4oC. 6. Centrifuge at 15,000 g for 10 min, 4oC and collect the supernatant containing nuclear proteins. 7. Add 100% trichloro acetic acid drop-by-drop to a final concentration of 33% and incubate on ice for 1–2 h. 8. Collect histones by centrifugation at 15,000 g, 10 min at 4oC and discard the supernatant. 9. Wash the histone pellet twice with chilled acetone to remove the residual acid. Spin at 15,000 g, 10 min at 4oC. 10. Air-dry the pellet and resuspend in 200 µl PBS. 11. Quantify the amount of histones using the Bio-Rad Dc protein assay kit. 12. Separate 1, 2, 4, and 8 µg of histones (for Coomassie staining) and 0.25, 0.5, 1, and 2 µg (for immunoblotting) on a 15% SDS-PAGE gel and proceed for immunoblotting and CBB staining as described in subsequent sections.

48

Ranveer S. Jayani et al.

II. Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis We use the Bio-Rad Mini Protean-3 assembly in our laboratory. However, the described method can be adapted to any equivalent vertical gel running apparatus. 1. Clean the large glass plate (with 1 mm spacers) and the short plate with distilled water and coat them with Sigmacote (Sigma, St. Louis, MO, USA) so that the gel does not stick to the plate. Rinse the plates with water and wipe them dry with lint-free tissue paper. 2. Cast 15% resolving gel using the cocktail described in “Materials” and gently overlay it with 1 ml of isopropanol. 3. After 10 min, decant the isopropanol and pour the stacking gel and insert the comb carefully such that it does not pierce the resolving gel and leave it for 10 min. 4. Remove the gel from the casting stand and wash it with running water to clean the wells. Assemble the plates in electrode assembly, according to manufacturer's instructions. 5. Pour 1 running buffer (diluted from 5 buffer) in the buffer tank and submerge the electrode gel assembly in it. Pour more buffer between the plates. 6. Prepare samples by addition of 6 SDS sample buffer to the afore mentioned amounts of histones. Boil the mixture for 10 min at 95oC and load on the gel. 7. Separate the samples by running the gel at 30 mA until the dye front reaches the lower edge of the gel. At this stage, the gel can be stained with CBB to visualize the proteins or it can be immunoblotted with various histone modification antibodies, as described in subse quent sections. III. Immunoblotting The procedure described here uses the Bio-Rad transfer apparatus. 1. Lift the plates from the gel tank, remove the stacking gel and transfer the resolving gel carefully on to a Whatman filter paper. 2. Immerse a transfer cassette in pre-chilled 1 transfer buffer with its cathode side (black) towards bottom. 3. Place a fiber pad on the cathode and gently place the filter paper (along with the gel) on it. Bubbles, if any should be removed thoroughly. 4. Place a PVDF membrane (pre-wetted with methanol for 5 min, rinsed with water and equilibrated with transfer buffer) on top of the gel. 5. Cover it with a filter paper sheet and fiber pad and close the cassette. 6. Insert the cassette in the holding stand and transfer the whole assembly to transfer tank. 7. Fill the tank with pre-chilled 1 transfer buffer and place the lid carefully. Place the whole assembly inside a refrigerator (4oC). 8. Transfer histones to the PVDF membrane by applying 400 mA constant current for 3 h. 9. Remove the PVDF membrane after transfer and incubate it in blocking buffer for 1 h on a rocking platform.

2. Linking Histone Modifications with Genome Organization

49

10. Wash once with 1 TST (5 min) and add histone modification antibody at a dilution of 1:2000. Incubate on a rocking platform for 3 h. 11. Wash the membrane with 1 TST thrice for 5 min each on a rocking platform. 12. Incubate it with 1:10,000 dilution of HRP-conjugated rabbit antibody for 1 h. 13. Wash thrice with 1 TST for 5 min on a rocking platform. 14. Incubate the membrane with the luminescence detection reagent (500 µl of each of HRP substrate luminol reagent and HRP substrate peroxide reagent for a standard membrane size of 6 9 cm) and expose to X-ray film for different time intervals. IV. Coomassie Brilliant Blue Staining Apart from immunoblotting, the gel can be stained with CBB R-250 to visualize the protein bands on gel. 1. Remove the gel from plates and wash it thoroughly with distilled water. 2. Place the gel in 20 ml fixer solution for 30 min. 3. Decant the fixer and add 20 ml CBB stain solution and incubate the gel on rocking platform for 1 h. 4. Remove the staining solution and wash the gel with destaining solution (30 min each wash) until the background turns transparent and the protein bands are visible distinctly. 5. The gel can be scanned or dried between the sheets of gelatin paper for longterm preservation. If the protein amounts are too low to be detected with Coomassie staining then silver staining of the gels can be performed as described elsewhere (Chevallet et al., 2006).

V. Discussion Chromatin immunoprecipitation has been a popular method over number of years for studying the association of transcription factors and other modulating enzymes with the DNA in living systems. The procedure involves in vivo cross-linking of the proteins to the DNA in the physical state in which they are present by addition of a cross-linking agent such as formaldehyde. This is analogous to clicking a “snapshot” of chromatin. The chromatin (DNA–protein complex) is then sheared into smaller fragments (typically 400–600 bp) by sonication. This is followed by immunoprecipita tion of the sequences bound to the protein of interest by employing specific antibodies against that protein. The DNA is then eluted and is used for locus-specific PCR reactions to study the binding sites of the protein in question. Recent developments such as ChIP-on-chip and ChIP sequencing provide high-throughput data enabling analysis of genome-wide occupancy. These high-throughput methods deliver the complete repertoire of the DNA sequences bound by the DNA-binding protein in question. The authors have elaborated a highly optimized protocol suitable for

50

Ranveer S. Jayani et al.

Activation 3

4

(C)

2

3

4

(D) 6 Fold change in occupancy

12 10 8 6 4 2

4 3 2 1 0 K9

(m

e)

3

c) (a K9

H3

H3

3

e) (m

K9

H3

H3

K9

(a

c)

IgG

0

5

IgG

Fold change in occupancy

1

IgG H3 K9 (a H3 K9 c) (m e) 3

2

IgG H3 K9 (a H3 K9 c) (m e) 3

Inp ut

1

LiCI treatment (B)

Inp ut

(A)

Fig. 4 Chromatin immunoprecipitation (ChIP) for c-Myc locus. (A) Jurkat cells were activated with PMA/Ionomycin (see “Methods” for details). ChIP was performed using normal rabbit IgG (lane 2), anti H3K9(ac) (lane 3) and anti-H3K9(me)3 (lane 4) as described in “Methods” followed by real-time PCR using primers specific to c-Myc. (B) Jurkat cells were treated with LiCl to activate Wnt signaling. ChIP and realtime PCR were performed as in (A). (C) and (D) CT values were calculated for the experiments performed in (A) and (B) and represented as relative fold occupancy of anti-H3K9(ac) and anti-H3K9(me)3 as compared with ChIP using IgG. The CT values were normalized with that of input chromatin. All experiments were performed thrice, error bars represent standard deviation.

studying the roles of histone posttranslational modifications in various physiological processes (Fig. 4). As evident from the method described above, proper sonication of the sample to obtain appropriate size fragments is critical for the success of the protocol (Fig. 3), we have used an improved probeless sonication instrument (Bioruptor XL, Diagenode, Belgium) that serves multiple advantages. First, it can handle a large number of samples simultaneously, ensuring uniform sonication efficiency for all samples. Second, it can be used for a range of volumes, thus enabling choice of performing analytical as well as quantitative ChIP. Third, it minimizes the chances of cross-contamination because the samples are sonicated by sonic waves traveling through chilled water and there is no probe involved. The time of sonication required may vary from cell line to cell line and tissue to tissue. Different primary cells are known to respond differently to sonication. The sonication pulse will also have to be optimized for different cell densities.

51

2. Linking Histone Modifications with Genome Organization

(A)

kDa 200 150 100 75 50 37

M

1

2

3

4

H1

25 20

H3 H2B H2A H4

15

(B) 1

2

3

4

Anti-pan H3

Anti-H3K4(me)3

Anti-H3K9(me)3

Anti-H3K9(ac)

Anti-H3K27(me)3

Fig. 5 Coomassie staining and immunoblot analysis for histones. (A) Histones extracted from Jurkat cells were resolved on a 15% SDS-polyacrylamide gel. The amounts of histones loaded were 1 µg (lane 1), 2 µg (lane 2), 4 µg (lane 3), and 8 µg (lane 4). Respective positions of all core histones are indicated on right. Molecular weight markers are as indicated (lane M, kDa). (B) Immunoblotting was performed with anti-pan H3, anti-H3K4(me)3, anti-H3K9(me)3, H3K9(ac), and anti-H3K27(me)3, after resolving 0.25 µg (lane 1), 0.5 µg (lane 2), 1 µg (lane 3), and 2 µg (lane 4) of acid-extracted histones from Jurkat cells on a 15% SDS-polyacrylamide gel.

The DNA quantity recovered during ChIP is typically very low, hence the recovery of sample can be compromised. To circumvent this, we have used ethchinmate (Nippon Gene, Toyama, Japan), a neutral polyacrylamide polymer solution that can be used for recovering extremely small quantity of nucleic acids without any affect on its downstream processing. Not all Histone modifications occur abundantly in cells. Thus, quantitative detection of such low abundance modifications could pose problems. The affinity of antibodies to various modifications also varies as evidenced by differential detection of same amount of histones by various antibodies (Fig. 5B). To facilitate detection of low abundance modifications, we suggest enrichment of histones by acid extraction. ChIP and immunoblot methods provided here have been used extensively in our

52

Ranveer S. Jayani et al.

laboratory and one can comfortably complete the whole procedure in approximately two working days.

VI. Summary This chapter provides an overview of chromatin structure, histones, and their organiza tion in the nucleosome, the various posttranslational modifications of the tails, and the globular domains of histones and the diverse roles played by these modifications in maintaining cellular integrity. We also discuss various histone-modifying enzymes, which include histone acetyl transferases, histone deacetylases, and histone demethylases. Histones are a very important group of nuclear proteins which profoundly influence the physiological state of a cell. Hence, the methods to study these proteins and the effects of various histone modifications are of great importance. We have provided detailed proto cols for comprehensive study of histone modifications using chromatin immunoprecipi tation and immunoblotting that would serve as a ready reference. Optimization of key steps such as sonication of chromatin constitute important features of this protocol. For quantitative detection of all histone modifications, we suggest a strategy including the enrichment of histones by acid extraction followed by immunoblot analysis. Acknowledgments This Work was supported by grants from the Department of Biotechnology, Government of India, and the Wellcome Trust, UK. R.S.J. is supported by fellowship from the Council of Scientific and Industrial Research, India. P.L.R. is supported by fellowship from the University Grants Commission, India. Work in Galande laboratory is supported by grants from the Department of Biotechnology, Government of India, and the Wellcome Trust, UK. S.G. is an international senior research fellow of the Wellcome Trust, UK. The authors declare no conflict of interest.

References Annunziato, A. T., Frado, L. L., Seale, R. L., and Woodcock, C. L. (1988). Treatment with sodium butyrate inhibits the complete condensation of interphase chromatin. Chromosoma 96, 132–138. Baker, S. P., Phillips, J., Anderson, S., Qiu, Q., Shabanowitz, J., Smith, M. M., Yates, J. R., Hunt, D. F., and Grant, P. A. (2010). Histone H3 Thr 45 phosphorylation is a replication associated post-translational modification in S. cerevisiae. Nat. Cell. Biol. 12, 294–298. Barski, A., Cuddapah, S., Cui, K., Roh, T.-Y., Schones, D. E., Wang, Z., Wei, G., Chepelev, I., and Zhao, K. (2007). High-resolution profiling of histone methylations in the human genome. Cell 129, 823–837. Bernstein, B. E., Mikkelsen, T. S., Xie, X., Kamal, M., Huebert, D. J., Cuff, J., Fry, B., Meissner, A., Wernig, M., Plath, K., Jaenisch, R., and Wagschal, A., et al. (2006). A bivalent chromatin structure marks key developmental genes in embryonic stem cells. Cell 125, 315–326. Bhaumik, S. R., Smith, E., and Shilatifard, A. (2007). Covalent modifications of histones during develop ment and disease pathogenesis. Nat. Struct. Mol. Biol. 14, 1008–1016. Bonenfant, D., Towbin, H., Coulot, M., Schindler, P., Mueller, D. R., and van Oostrum, J. (2007). Analysis of dynamic changes in post-translational modifications of human histones during cell cycle by mass spectro metry. Mol. Cell. Proteomics 6, 1008–1016. Brower-Toland, B., Wacker, D. A., Fulbright, R. M., Lis, J. T., Kraus, W. L., and Wang, M. D. (2005) Specific contributions of histone tails and their acetylation to the mechanical stability of nucleosomes. J. Mol. Biol. 346, 135–146.

2. Linking Histone Modifications with Genome Organization

53

Brunner, T., Kasibhatla, S., Pinkoski, M. J., Frutschi, C., Yoo, N. J., Echeverri, F., Mahboubi, A., and Green, D. R. (2000). Expression of fas ligand in activated T cells is regulated by c-Myc. J. Biol. Chem. 275, 9767–9772. Burlingame, R. W., Love, W. E., Wang, B. C., Hamlin, R., Nguyen, H. X., and Moudrianakis, E. N. (1985). Crystallographic structure of the octameric histone core of the nucleosome at a resolution of 3.3 A. Science 228, 546–553. Cam, H. P., Chen, E. S., and Grewal, S. I. (2009). Transcriptional scaffolds for heterochromatin assembly. Cell 136, 610–614. Cao, R., Wang, L., Wang, H., Xia, L., Erdjument-Bromage, H., Tempst, P., Jones, R. S., and Zhang, Y. (2002). Role of histone H3 lysine 27 methylation in Polycombgroup silencing. Science 298, 1039–1043. Cedar, H., and Bergman, Y. (2009). Linking DNA methylation and histone modification: patterns and paradigms. Nat. Rev. Genet. 10, 295–304. Cerutti, H., and Casas-Mollano, J. A. (2009). Histone H3 phosphorylation: universal code or lineage specific dialects? Epigenetics 4, 71–75. Cheung, P., Tanner, K. G., Cheung, W. L., Sassone-Corsi, P., Denu, J. M., and Allis, C. D. (2000). Synergistic coupling of histone H3 phosphorylation and acetylation in response to epidermal growth factor stimula tion. Mol. Cell. 5, 905–915. Chevallet, M., Luche, S., and Rabilloud, T. (2006). Silver staining of proteins in polyacrylamide gels. Nat. Protoc. 4, 1852–1858. Cloos, P. A., Christensen, J., Agger, K., and Helin, K. (2008) Erasing the methyl mark: histone demethylases at the center of cellular differentiation and disease. Genes Dev., 22, 1115–1140. Davey, C. A., Sargent, D. F., Luger, K., Maeder, A. W., and Richmond, T. J. (2002). Solvent mediated interactions in the structure of the nucleosome core particle at 1.9 Ao resolution. J. Mol. Biol. 319, 1097–1113. Davie, J. R., and Murphy, L. C. (1990). Level of ubiquitinated histone H2B in chromatin is coupled to ongoing transcription. Biochemistry 29, 4752–4757. Dehé, P. M., Pamblanco, M., Luciano, P., Lebrun, R., Moinier, D., Sendra, R., Verreault, A., Tordera, V., and Géli, V. (2005). Histone H3 lysine 4 mono-methylation does not require ubiquitination of histone H2B. J. Mol. Biol. 353, 477–484. Duncan, E. M., Muratore-Schroeder, T. L., Cook, R. G., Garcia, B. A., Shabanowitz, J., Hunt, D. F., and Allis, C. D. (2008). Cathepsin L proteolytically processes histone H3 during mouse embryonic stem cell differentiation. Cell 135, 284–294. Ebert, A., Lein, S., Schotta, G., and Reuter, G. (2006). Histone modification and the control of heterochro matic gene silencing in Drosophila. Chromosome Res 14, 377–392. Epsztejn-Litman, S., Feldman, N., Abu-Remaileh, M., Shufaro, Y., Gerson, A., Ueda, J., Deplus, R., Fuks, F., Shinkai, Y., Cedar, H., and Bergman, Y. (2008). De novo DNA methylation promoted by G9a prevents reprogramming of embryonically silenced genes. Nat. Struct. Mol. Biol. 15, 1176–1183. Ericsson, C., Goldknopf, I. L., and Daneholt, B. (1986). Inhibition of transcription does not affect the total amount of ubiquitinated histone 2A in chromatin. Exp. Cell. Res. 167, 127–134. Feldman, N., Gerson, A., Fang, J., Li, E., Zhang, Y., Shinkai, Y., Cedar, H., and Bergman, Y. (2006). G9a-mediated irreversible epigenetic inactivation of Oct-3/4 during early embryogenesis. Nat. Cell. Biol. 8, 188–194. Fischle, W., Wang, Y., and Allis, C. D. (2003). Binary switches and modification cassettes in histone biology and beyond. Nature 425, 475–479. Fischle, W., Tseng, B. S., Dormann, H. L., Ueberheide, B. M., Garcia, B. A., Shabanowitz, J., Hunt, D. F., Funabiki, H., and Allis, C. D. (2005). Regulation of HP1chromatin binding by histone H3 methylation and phosphorylation. Nature 438, 1116–1122. Fuhrmann, G., Chung, A. C., Jackson, K. J., Hummelke, G., Baniahmad, A., Sutter, J., Sylvester, I., Scholer, H. R., and Cooney, A. J. (2001). Mouse germline restriction of Oct4 expression by germ cell nuclear factor. Dev. Cell. 1, 377–387. Galande, S., Purbey, P. K., Notani, D., and Kumar, P. P. (2007). The third dimension of gene regulation: organization of dynamic chromatin loopscape by SATB1. Curr. Opin. Genet. Dev. 17, 408–414. Garcia, B. A., Shabanowitz, J., and Hunt, D. F. (2005). Analysis of protein phosphorylation by mass spectrometry. Methods 35, 256–264.

54

Ranveer S. Jayani et al. Grunstein, M. (1997). Histone acetylation in chromatin structure and transcription. Nature 389:349–352.

Grunstein, M. (1992). Histones as regulators of genes. Sci. Am. 267, 68–74B.

Guenther, M. G., Levine, S. S., Boyer, L. A., Jaenisch, R., and Young, R. A. (2007). A chromatin landmark

and transcription initiation at most promoters in human cells. Cell 130, 77–88. Gyory, I., Wu, J., Fejér, G., Seto, E., Wright, K. L. (2004). PRDI-BF1 recruits the histone H3 methyltrans ferase G9a in transcriptional silencing. Nat. Immunol. 5, 299–308. Hebbes, T. R., Thorne, A. W., and Crane-Robinson, C. (1988). A direct link between core histone acetylation and transcriptionally active chromatin. EMBO J. 7, 1395–1402. Heintzman, N. D., Hon, G. C., Hawkins, R. D., Kheradpour, P., Stark, A., Harp, L. F., Ye, Z., Lee, L. K., Stuart, R. K., Ching, C. W., Ching, K. A., Antosiewicz-Bourget, J. A., et al. (2009). Histone modifications at human enhancers reflect global cell-type-specific gene expression. Nature 459, 108–112. Heintzman, N. D., and Ren, B. (2007). The gateway to transcription: identifying, characterizing and under standing promoters in the eukaryotic genome. Cell. Mol. Life Sci. 64, 386–400. Heintzman, N. D., Stuart, R. K., Hon, G., Fu, Y., Ching, C. W., Hawkins, R. D., Barrera, L. O., Calcar, S. V., Qu, C., Ching, K. A. Wang, W., Weng, Z., et al.., (2007). Distinct and predictive chromatin signatures of transcriptional promoters and enhancers in the human genome. Nat. Genet. 39, 311–318. Hirota, T., Lipp, J. J., and Toh, B. H., Peters, J. M. (2005). Histone H3 serine 10 phosphorylation by Aurora B causes HP1 dissociation from heterochromatin. Nature 438, 1176–1180. Hon, G. C., Hawkins, R. D., and Ren, B. (2009). Predictive chromatin signatures in the mammalian genome. Hum. Mol. Genet. 18, R195–201. Huang, S. Y., Barnard, M. B., Xu, M., Matsui, S., Rose, S. M., and Garrard, W. T. (1986). The active immunoglobulin kappa chain gene is packaged by non-ubiquitinconjugated nucleosomes. Proc. Natl. Acad. Sci. USA 83, 3738–3742. Jason, L. J. M., Moore, S. C., Lewis, J. D., Lindsey, G., and Ausió, J. (2002). Histone ubiquitination: a tagging tail unfolds? Bioessays 24, 166–174. Jhunjhunwala, S., van Zelm, M. C., Peak, M. M., and Murre, C. (2009). Chromatin architecture and the generation of antigen receptor diversity. Cell 138, 435–448. Kim, J., Guermah, M., McGinty, R. K., Lee, J. S., Tang, Z., Milne, T. A., Shilatifard, A., Muir, T. W., and Roeder, R. G. (2009). RAD6-Mediated transcription-coupled H2B ubiquitylation directly stimulates H3K4 methylation in human cells. Cell 137, 459-471. Kolasinska-Zwierz, P., Down, T., Latorre, I., Liu, T., Liu, X. S., and Ahringer, J. (2009). Differential chromatin marking of introns and expressed exons by H3K36me3. Nat. Genet. 41, 376–381. Kouzarides, T. (2007). Chromatin modifications and their function. Cell 128, 693–705. Krajewski, W. A., and Becker, P. B. (1998). Reconstitution of hyperacetylated, DNase I-sensitive chromatin characterized by high conformational flexibility of nucleosomal DNA. Proc. Natl. Acad. Sci. USA 95, 1540–1545. Krivtsov, A. V., Feng, Z., Lemieux, M. E., Faber, J., Vempati, S., Sinha, A. U., Xia, X., Jesneck, J., Bracken, A. P., Silverman, L. B., Kutok, J. L., and Kung, A. L., (2008). H3K79 methylation profiles define murine and human MLL-AF4 leukemias. Cancer Cell 14, 355–368. Laribee, R. N., Krogan, N. J., Xiao, T., Shibata, Y., Hughes, T. R., Greenblatt, J. F., and Strahl, B. D. (2005). BUR kinase selectively regulates H3 K4 trimethylation and H2B ubiquitylation through recruitment of the PAF elongation complex. Curr. Biol. 15, 1487–1493. Lee, J. S., Shukla, A., Schneider, J., Swanson, S. K., Washburn, M. P., Florens, L., Bhaumik, S. R., and Shilatifard, A. (2007). Histone crosstalk between H2B monoubiquitination and H3 methylation mediated by COMPASS. Cell 131, 1084–1096. Lennartsson, A., and Ekwall, K. (2009). Histone modification patterns and epigenetic codes. Biochim. Biophys. Acta 1790, 863–868. Levinger, L., and Varshavsky, A. (1982). Selective arrangement of ubiquitinated and D1 protein-containing nucleosomes within the Drosophila genome. Cell 28, 375–385. Lu, X., Simon, M. D., Chodaparambil, J. V., Hansen, J. C., Shokat, K. M., and Luger, K. (2008). The effect of H3K79 dimethylation and H4K20 trimethylation on nucleosome and chromatin structure. Nat. Struct. Mol. Biol. 15, 1122–1124.

2. Linking Histone Modifications with Genome Organization

55

Luger, K., Mader, A. W., Richmond, R. K., Sargent, D. F., and Richmond, T. J. (1997). Crystal structure of the nucleosome core particle at 2.8 Ao resolution. Nature 389, 251–260. Lupien, M., Eeckhoute, J., Meyer, C. A., Wang, Q., Zhang, Y., Li, W., Carroll, J. S., Liu, X. S., and Brown, M. (2008). FoxA1 translates epigenetic signatures into enhancer-driven lineage-specific transcription. Cell 132, 958–970. Metzger, E., Schule, R. (2007). The expanding world of histone lysine demethylases. Nat. Struct. Mol. Biol. 14, 252–254. Metzger, E., Yin, N., Wissmann, M., Kunowska, N., Fischer, K., Friedrichs, N., Patnaik, D., Higgins, J. M., Potier, N., Scheidtmann, K. H., Buettner, R., and Schüle, R. (2008). Phosphorylation of histone H3 at threonine 11 establishes a novel chromatin mark for transcriptional regulation. Nat. Cell. Biol. 10, 53–60. Mikkelsen, T. S., Ku, M., Jaffe, D. B., Issac, B., Lieberman, E., Giannoukos, G., Alvarez, P., Brockman, W., Kim, T. K., Koche, R. P., Lee, W., and Mendenhall, E., et al. (2007). Genome-wide maps of chromatin state in pluripotent and lineage-committed cells. Nature 448, 553–560. Müller, J., Hart, C. M., Francis, N. J., Vargas, M. L., Sengupta, A., Wild, B., Miller, E. L., O’Connor, M. B., Kingston, R. E., and Simon, J. A. (2002). Histone methyltransferase activity of a Drosophila Polycomb group repressor complex. Cell 111, 197–208. Murray, K. (1964). The occurrence of N-methyl lysine in histones. Biochemistry 3, 10–15. Notani, D., Gottimukkala, K. P., Jayani, R. S., Limaye, A. S., Damle, M. V., Mehta, S., Purbey, P. K., Joseph, J., and Galande, S. (2010). Global regulator SATB1 recruits bcatenin and regulates TH2 differentiation in Wnt-dependent manner. PLoS Biol. 8, e1000296. Nottke, A., Colaiacovo, M. P., and Shi, Y. (2009). Developmental roles of the histone lysine demethylases. Development 136, 879–889. Ozsolak, F., Song, J. S., Liu, X. S., and Fisher, D. E. (2007). High-throughput mapping of the chromatin structure of human promoters. Nat. Biotechnol. 25, 244–248. Peterson, C. L., and Laniel, M. A. (2004). Histones and histone modifications. Curr. Biol. 14, R546–R551. Pietersen, A. M., and van Lohuizen, M. (2008). Stem cell regulation by polycomb repressors: postponing commitment. Curr. Opin. Cell Biol. 20, 201–207. Robzyk, K., Recht, J., and Osley, M. A. (2000). Rad6-dependent ubiquitination of histone H2B in yeast. Science 287, 501–504. Rogakou, E. P., Pilch, D. R., Orr, A. H., Ivanova, V. S., and Bonner, W. M. (1998). DNA double-stranded breaks induce histone H2AX phosphorylation on serine 139. J. Biol. Chem. 273, 5858–5868. Santos-Rosa, H., Kirmizis, A., Nelson, C., Bartke, T., Saksouk, N., Cote, J., and Kouzarides, T. (2009). Histone H3 tail clipping regulates gene expression. Nat. Struct. Mol. Biol. 16, 17–22. Schneider, J., Bajwa, P., Johnson, F. C., Bhaumik, S. R., and Shilatifard, A. (2006). Rtt109 is required for proper H3K56 acetylation: a chromatin mark associated with the elongating RNA polymerase II. J. Biol. Chem. 281, 37270–37274. Schneider, J., Wood, A., Lee, J. S., Schuster, R., Dueker, J., Maguire, C., Swanson, S. K., Florens, L., Washburn, M. P., and Shilatifard, A. (2005). Molecular regulation of histone H3 trimethylation by COMPASS and the regulation of gene expression. Mol. Cell 19, 849–856. Schones, D. E., and Zhao, K. (2008). Genome-wide approaches to studying chromatin modifications. Nat. Rev. Genet. 9, 179–191. Schulze, J. M., Jackson, J., Nakanishi, S., Gardner, J. M., Hentrich, T., Haug, J., Johnston, M., Jaspersen, S. L., Kobor, M. S., and Shilatifard, A. (2009). Linking cell cycle to histone modifications: SBF and H2B monoubiquitination machinery and cell-cycle regulation of H3K79 dimethylation. Mol. Cell 35, 626–641. Seigneurin-Berny, D., Verdel, A., Curtet, S., Lemercier, C., Garin, J., Rousseaux, S., and Khochbin, S. (2001). Identification of components of the murine histone deacetylase 6 complex: link between acetyla tion and ubiquitination signaling pathways. Mol. Cell. Biol. 21, 8035–8044. Shahbazian, M. D., Zhang, K., and Grunstein, M. (2005). Histone H2B ubiquitylation controls processive methylation but not monomethylation by Dot1 and Set1. Mol. Cell 19, 271–277. Shechter, D., Dormann, H. L., Allis, C. D., and Hake, S. B. (2007). Extraction, purification and analysis of histones. Nat. Protoc. 2, 1445–1457

56

Ranveer S. Jayani et al. Shen, X., Kim, W., Fujiwara, Y., Simon, M. D., Liu, Y., Mysliwiec, M. R., Yuan, G., Lee, Y., and Orkin, S. H. (2009). Jumonji modulates polycomb activity and self-renewal versus differentiation of stem cells. Cell 139, 1303–1314. Shi, Y., Lan, F., Matson, C., Mulligan, P., Whetstine, J. R., Cole, P. A., and Casero, R. A. (2004). Histone demethylation mediated by the nuclear amine oxidase homolog LSD1. Cell 119, 941–953. Shiio, Y., and Eisenman, R. N. (2003). Histone sumoylation is associated with transcriptional repression. Proc. Natl. Acad. Sci. USA 100, 13225–13230. Shilatifard, A. (2006). Chromatin modifications by methylation and ubiquitination: implications in the regulation of gene expression. Annu. Rev. Biochem. 75, 243–269. Simpson, R. T. (1978). Structure of chromatin containing extensively acetylated H3 and H4. Cell 13, 691–699. Sims, R. J., and Reinberg, D. (2009). Processing the H3K3me3 singnature. Nat. Genet. 41, 270–271. Strahl, B. D., and Allis, C. D. (2000). The language of covalent histone modifications. Nature 403, 41–45. Struhl, K. (1998). Histone acetylation and transcriptional regulatory mechanisms. Genes Dev. 12, 599–606. Thorne, A. W., Sautiere, P., Briand, G., and Crane-Robinson, C. (1987). The Structure of ubiquitinated histone H2B. EMBO J. 6, 1005–1010. Trievel, R. C., (2004). Structure and function of histone methytransferases. Crit. Rev. Eukaryot. Gene Expt. 14, 147–170. Vempati, R. K., Jayani, R. S., Notani, D., Sengupta, A., Galande, S., and Haldar, D. (2010). p300 mediated acetylation of histone H3 lysine 56 functions in DNA damage response in mammals. J. Biol. Chem. (Epub ahead of print). Weake, V. M., and Workman, J. L. (2008). Histone ubiquitination: Triggering gene activity. Mol. Cell 29, 653–663. Weinmann, A. S., and Farnham, P. J. (2002). Identification of unknown target genes of human transcription factors using chromatin immunopercipitation. Methods 26, 37–47. Widom, J. (1998). Chromatin structure: Linking structure to function with histone H1. Curr. Biol. 8, R788–791. Yarmane, K., Toumazou, C., Tsukada, Y., Erdjument-Bromage, H., Tempst, P., Wong, J., and Zhang, Y. (2006). JHDM2A, a JmjC-containing H3K9 demethylase, facilitates transcription activation by androgen receptor. Cell 125, 483–495. Zhang, Y., and Reinberg, D. (2001). Transcription regulation by histone methylation: interplay between different covalent modifications of the core histone tails. Genes Dev. 15, 2343–2360. Zhang, Y. (2003). Transcriptional regulation by histone ubiquitination and deubiquitination. Genes Dev. 17, 2733–2740. Zhu, B., Zheng, Y., Pham, A. D., Mandal, S. S., Erdjument-Bromage, H., Tempst, P., and Reinberg, D. (2005). Monoubiquitination of human histone H2B: the factors involved and their roles in HOX gene regulation. Mol. Cell 20, 601–611.

CHAPTER 3

Dynamic Organization of Chromatin Assembly and Transcription Factories in Living Cells Bidisha Sinha*, Dipanjan Bhattacharya*, Deepak Kumar Sinha*, Shefali Talwar*,†, Shovamayee Maharana*,†, Soumya Gupta*,†, and G. V. Shivashankar*,† * National Centre for Biological Sciences, Tata Institute of Fundamental Research, Bellary Road, Bangalore 560065, India † Department of Biological Sciences and Research Center for Excellence in MechanoBiology, National University of Singapore, Singapore 117543, Singapore

Abstract I. Introduction II. Chromatin Dynamics A. Cell Culture and Tagged Proteins B. Fluorescence Recovery After Photobleaching C. Fluorescence Correlation Spectroscopy D. Results III. Higher-Order Chromosome Compaction A. Cell Culture B. Fluorescence Anisotropy C. Results IV. Nuclear Plasticity A. Cell Culture B. Fly Lines C. Results V. Spatio-temporal Organization of Transcription Factories and Gene Loci A. Visualization of Transcription Factories by Fluorescent UTPs in Live Cells B. Single Particle Tracking and Analysis C. Labeling Strategies of Gene Loci in Live Cells D. Results VI. Conclusions

Acknowledgments

References

METHODS IN CELL BIOLOGY, VOL. 98 Copyright 2010 Elsevier Inc. All rights reserved.

57

978-0-12-381009-0

DOI: 10.1016/S0091-679X(10)98003-5

58

Bidisha Sinha et al.

Abstract The interphase nucleus is an active organelle involved in processing genetic infor mation. In higher order eukaryotes, information control is compartmentalized – for example at the scale of inter-chromosome territories and nuclear bodies. Regulatory proteins, nuclear bodies and chromatin assembly are found to be highly dynamic within the nucleus of primary cells and through cellular differentiation programs. In this chapter we describe live-cell fluorescence based techniques and single particle tracking analysis, to probe the spatio-temporal dimension in nuclear function.

I. Introduction The nucleus of a cell was first observed by light microscopy by Robert Brown in 1831. However, the initial ideas of nuclear organization into heterochromatin and euchromatin came into picture with the availability of DNA-binding dyes. Advances in methods for visualizing proteins and DNA have aided in exploring the subtle organizing principles of chromatin assembly in interphase nucleus. In a mammalian cell, the 2 m long DNA is compacted into a micron-sized nucleus, without compromis ing the accessibility for various nuclear proteins to carry out different nuclear functions (Misteli, 2007). Interestingly, the experiments marking chromosomes in live and fixed interphase cells suggested that nuclear architecture is compartmentalized into chromo some territories and interchromatin spaces (Ferreira et al., 1997; Spector, 2003). Further, the chromatin is also organized into a less dense transcription permissive euchromatin and a more dense transcriptionally less permissive heterochromatin. The organization of DNA into euchromatin and heterochromatin depends on binding of chromatin-remodeling proteins and the nucleosome stability. While compaction is required for efficient packaging of the chromatin, nucleosomal movement is also crucial for making the DNA accessible to various protein complexes in order to carry out functions like transcription, replication, and repair. Hence these nuclear functions clearly depend on the structural and dynamic organization of the chromatin. The histone tail modifications like acetylation, methylation, phosphorylation, and ubiquitination are believed to be an important intermediate between chromatin struc ture and function, which can modulate electrostatic interactions between histone octamer and the DNA. These modifications in turn recruit various chromatin remodel ing enzymes which can actively reposition nucleosomes, thus activating or repressing transcription. For example, the acetylation of Histone H3 and H4, usually cause transcriptional activation whereas methylation is generally associated with repression of transcription (Khorasanizadeh, 2004). The combinatorial code of the histone mod ification which impinges on gene regulation is yet to be fully understood. At the level of a gene locus, most of these modifications are associated with the 50 and 30 end of the open reading frame and with the core promoter, where they regulate the interaction of chromatin with other DNA-binding proteins. Gene expression can also be regulated by incorporation of certain histone variants (Talbert and Henikoff, 2010). On the linear

3. Organization of Chromatin Assembly and TFs in Living Cells

59

sequence of DNA the co-regulated genes are frequently found to be together-like odorant receptor, MHC complex, immunoglobulin, and hox genes (Sproul et al., 2005). Genome wide Chromatin Immunoprecipitation (ChIP) experiments have revealed that similar histone modifications are established around clusters of genes in linear sequence (Schones and Zhao, 2008). The gene locus also interacts with the underlying nuclear scaffold (Dechat et al., 2008) to form active chromatin loops or to interact with DNA sequences like Locus Control Regions (LCRs) or enhancers (Heng et al., 2004). Molecular biology techniques like Chromosome Conformation Capture (3C) (Dekker et al., 2002) have revealed these clustering of DNA sequences (Palstra et al., 2003). Hence even at gene locus, the chromatin organization and the 3D architecture form important components of genome regulation. In addition to spatial organization, nuclear functions are also compartmentalized, where proteins required for similar processes are found to be co-clustered. Many examples of such co-clustering are seen in the form of nuclear bodies like repair, replication and transcription factories (TFs), nuclear speckles, Promyelocytic Leukemia (PML), and Cajal bodies (Spector, 2001). As nuclear functions are carried out by complexes of many proteins, co-clustering is believed to reduce the search time by individual components by increasing the local concentrations at regulatory sites (Sutherland and Bickmore, 2009). In line with this postulate, co-regulated genes have been shown to spatially cluster together perhaps to share TFs, which are enriched in NTPs, and other machinery required for transcription (Osborne et al., 2004; Zhou et al., 2006). Besides, it has also been demonstrated for a few genes that they are localized away from the chromosome territory to have easier access to TFs (Cremer and Cremer, 2001; Francastel et al., 2000). Molecular biology tools like ChIP, 3C, and immunocytochemistry of various nuclear components are valuable for getting static information of nuclear processes at DNA and nuclear level. To obtain dynamic insights into these processes, modern fluorescencebased tools like Fluorescence Correlation Spectroscopy (FCS), Fluorescence Anisotropy, and Fluorescence Recovery after Photobleaching (FRAP) have been employed. In this chapter we describe these methods to study chromatin assembly and TFs in living cells.

II. Chromatin Dynamics The fundamental unit of chromatin is a nucleosome; an octameric unit wound around by 146 base pairs of DNA. Core and linker histone proteins form the major class of nuclear proteins that condense the genome into a highly organized chromatin assembly. The octamer contains two copies each of the four conserved core histones—H2A, H2B, H3, and H4 (Bruno et al., 2003; Luger and Richmond, 1998a, b; Luger et al., 1997; Muthurajan et al., 2003). The nucleosomal units are separated by a region of linker DNA, which is associated with a less conserved histone, usually referred to as the linker histone H1 (Bustin et al., 2005; Happel and Doenecke, 2009). The tight packaging of the nucleosome complex creates a barrier to the regulatory machinery to access DNA. The chromatin is therefore required to be continually remodeled and the histone–DNA inter actions in the nucleosome to be relaxed, to allow polymerases and other proteins to access

60

Bidisha Sinha et al.

the DNA template (Workman and Kingston, 1998). The functional roles of core and linker histones are found to be distinct, arising due to the differential nature of their interaction and association with the DNA. While the core histones with their epigenetic modifications play a central role in regulating access to DNA, the linker histones (with other nonhistone proteins) are thought to participate in maintaining the dynamic higher order chromatin structure, as it clamps the entry and exit sites of DNA around the core histone octamer. Recent progress in fluorescence-based live-cell monitoring techniques using histone proteins tagged with Enhanced Green Fluorescent Protein (EGFP) (Gasser, 2002) have revealed that the core and linker histones that package the genome are highly dynamic within living cells. Core histones are found to exchange with a t1/2 2 h whereas the linker histones have a t1/2 few minutes (Kimura and Cook, 2001; Lever et al., 2000; Misteli et al., 2000). Single cell photobleaching experiments also reveal that H2B-GFP exchanged more rapidly than H3-GFP and H4-GFP (Kimura and Cook, 2001). In the following section, the differences in dynamics of core and linker histones are described by studying their underlying diffusion mechanisms using FRAP and FCS (Bhattacharya et al., 2006). A. Cell Culture and Tagged Proteins In order to investigate chromatin organization in live cells, fluorescently tagged protein(s) of interest are expressed in cells. Core histone proteins like H2B, H3, and H4 and linker histones like H1.1 and H1.5 were tagged on C-terminus with EGFP or monomeric Red Fluorescent Protein (mRFP) in plasmid vectors driven by CMV or EF1 () promoters. These modified proteins are expressed in mammalian cells by transfect ing the plasmids using Lipofectamine 2000 and Opti-MEM (Gibco). Cancerous cell line HeLa was cultured in Dulbeco’s Modified Eagle Medium supplemented with 10% Fetal Bovine Serum and penicillin–streptomycin (Gibco). Cells were maintained at 37°C in a 5% CO2 incubator. Cells were grown on glass bottom coverslip dishes for one day, before being transfected with 500 ng–1 µg of DNA and imaged 24 h later for FRAP and FCS experiments. Stable cell lines of HeLa-H2B-EGFP and HeLa-EGFP were prepared by antibiotic selection using Blasticidin (1 µg/ml, Sigma) and G418 (400 µg/ml, Sigma), respectively. Before imaging, the medium is replaced with M1 medium (150 mM NaCl, 20 mM HEPES, 1 mM MgCl2) supplemented with 1% glucose. B. Fluorescence Recovery After Photobleaching Photobleaching of fluorescently tagged protein is used to probe their translational diffusion in various compartments within the cell and nucleus. GFP or EGFP is the preferred fluorescent tag to target protein of interest which is localized to various cellular compartments. Fluorescent chimeric protein within a small region of interest (ROI) is irreversibly bleached with very high intensity laser, and its exchange with the surrounding unbleached populations of fluorophore is then monitored by recovery in fluorescence intensity in the ROI over time. Photobleaching can be carried out using a confocal laser scanning microscope having precise positioning of the bleach spot and equipped to modulate fast switching between low-intensity imaging settings and higher

3. Organization of Chromatin Assembly and TFs in Living Cells

61

intensity for bleaching. EGFP protein is imaged with a 30 mW, 488 nm laser line at low power (2–5%) and bleached at high intensity (up to 100%, with multiple bleach iterations). FRAP studies provide information on the mobile fraction of protein and the rate of exchange, which is related to the diffusion time D. Upon bleaching the initial fluorescence intensity (Fi) of the bleach spot drops to F0 which over time plateaus to a constant value F1. The mobile fraction (R) of the protein is defined as R = (F1 – F0)/(Fi – F0). These mobile fraction and diffusion timescales of protein depend on their interaction with other cellular components such as macromolecular complexes, matrices, and membranes. In the absence of any active transport, the intracellular protein dynamics is primarily driven by Brownian motion. For Brownian motion, the expected diffusion constant, D, of proteins can be calculated from the equation D = KT/6Rh, where T is the temperature, is the viscosity of the solution (cytoplasm/nucleus), K is the Boltzmann constant, and Rh is the hydrodynamic radius of the protein of interest. Temperature and viscosity vary little across the cell and are therefore assumed to be constant (to be noted effective viscosity may depend on macromolecular crowding). Thus, the size of the macromolecule Rh in the cell under study is the most important determinant of the diffusion constant and hence the diffusion time D. For FRAP studies described here, a Zeiss Confocor (model LSM510-Meta/Confocor2) fluorescence microscope equipped with a C-Apochromat 40 /1.2 NA water corrected objective was used. 512 512 pixels, 12 bit confocal images were acquired with a pinhole aperture of 1 airy unit. EGFP fusion proteins were excited with the 488 nm line of an argon-ion laser (Lasos, Jena, Germany) and the emission collected with a 500–530 nm bandpass filter.

C. Fluorescence Correlation Spectroscopy FCS is based on the temporal fluctuations of fluorescence intensity, occurring in small volume (femtoliters) of observation (Rigler, 2001). Rate of fluctuations in fluorescence depend on the rate of diffusion of fluorophores through the observation volume, as the fluorescent molecule enters confocal volume, gets excited, and con tributes to fluorescence emission till it exits. Smaller molecules diffuse faster than a larger molecule, reflecting in their correlation timescales, calculated from the intensity time series using the following autocorrelation function:

GðÞ ¼

hIðt þ ÞIðtÞi hI 2 ðtÞi hI 2 ðtÞi

where is the correlation time. The intensity time series was collected from diffraction limited confocal volume over a period of 10 s intervals and averaged over ten runs to get the autocorrelation function and the corresponding fits. The pin-hole size was kept at 70 µm for 488 nm laser line (confocal diameter of 300 nm) and 78 µm for 543 nm laser line (confocal diameter of 360 nm). The following function was used to fit the

62

Bidisha Sinha et al.

experimentally obtained autocorrelation curves for unhindered three-dimensional diffusion: GðÞ ¼

0 2 1 A þ A expð=CÞ B 1 6 @ 4h 1A N

31 1þ

ih D

1 1 þ ð1=s2 Þ

7C i1=2 5A D

where N is the number concentration of the fluorescent species in the confocal volume, D is the diffusion timescale, such that diffusion constant D = !2/4 D, where ! is the XY spread of the confocal spot, and s (!z/!x) is the structure parameter; A and C are the triplet fraction and triplet timescales respectively. However, diffusion for larger mole cules like core histones inside crowded cell nucleus cannot be adequately described by this model. In order to understand these results, the autocorrelation function was modified with an anomalous subdiffusion term , which describes the underlying heterogeneity of the matrix: 0 2 6 1 A þ A expð=CÞ B B 1 6 GðÞ ¼ B 6 @ N 4 1A

31

1þ

D

1 1þ

s2 D

7C 7C 7C 1=2 5A

The data obtained for the linker histone dynamics was fitted to a sum of two diffusing species using the maximum entropy method––MEMFCS (Periasamy and Verkman, 1998): GðÞ ¼

2

00

1 A þ A expð=CÞ BB 6 @@ðN 1 N 2Þ4h 1A 0

2

B 6 þ@ N 1ð1 N 2Þ4h

1þ

D2

ih

1 1þ

s2 D2

31

ih

1

1 þ D1 1þ 311

s2 D1

7C i1=2 5A

7CC

i1=2 5AA

where N1 is the inverse of the total number of bright molecules (EGFP-tagged linker histones) in the confocal volume and N2 corresponds to the fraction of the species having correlation timescale D1.

D. Results The mobility of core histone H2B-EGFP within the HeLa cell nucleus (Fig. 1A) was measured using FCS. Since the mobility of the histone proteins is a measure of their size and interaction(s) with themselves or underlying mesh, FCS curves associated with differently sized, noninteracting particle were measured. EGFP (RH 1.5 nm) is

63

3. Organization of Chromatin Assembly and TFs in Living Cells

(A)

HeLa

Polytene

Polytene 600 mM NaCl

D (μm2/s)

(B)

0.8

20 10 0

G(τ)

o alt FP nucl -Cyt ly s 2B H2B Po

EG

0.4

EGFP H2B-EGFP H2B cytoplasm H2B polytene H2B polytene salt

0.0 101

102 103 104 Correlation time (μs)

105

Fig. 1 Translational diffusion of passive molecules probe the chromatin mesh and core histones reveal a multimeric form.(A) Fluorescence images of H2B EGFP-transfected HeLa nucleus, Drosophila salivary gland polytene chromosomes, and polytene chromosome with 600 mM NaCl (scale bar 5 µm). (B) Autocorrelation curves of EGFP, core histone H2B-EGFP in cytoplasm and nucleus and H2B-EGFP of salivary gland nucleus, and polytene chromosome with 600 mM NaCl. Inset––Mean diffusion constant-D for the above is plotted.

insensitive to the architecture and undergoes unhindered 3D diffusion with a D of 220 µs and diffusion constant D of 26.4 ± 2.7 µm2/s as shown in Fig. 1B. The typical correlation timescale D for H2B-EGFP is 830.5 ± 232 µs (the corresponding D = 7.3 ± 1.9 µm2/s) under normal physiological conditions. The mean correlation timescale observed for H2B-EGFP is much higher than expected according to its molecular size (41 kDa), indicating that it may exist in a multimeric state. The typical diffusion timescale estimation of purified H2B-EGFP monomers in PBS solution is 120 µs, whereas in viscous environment of nucleus or cytoplasm the correlation time is expected to be 330 µs. Correlation timescale of H2B-EGFP in the cytoplasm of a cell overexpressing the protein is estimated to be 268 ± 65.2 µs (with = 0.67), comparable to its monomeric form; however, higher D (830.5 ± 232 µs) in the nucleus indicates multi meric associations (Fig. 1B). The standard deviation in the timescales also reflects the heterogeneity in the chromatin architecture. The correlation timescales for a different core histone H4-EGFP was similar to that for H2B-EGFP (871 ± 272 µs). However, the polytene chromosome from salivary gland of Drosophila larvae revealed both the timescales corresponding to monomeric and multimeric forms (Fig. 1A and B). In order to probe the effect of heterogeneous chromatin structure on core histone mobility, disassembly of nucleosomes was induced under high salt condition (600 mM

64

Bidisha Sinha et al.

NaCl) (Mangenot et al., 2002). The altered chromatin architecture is reflected in the shift of value to unity (Bhattacharya et al., 2006). Here the correlation curve fits perfectly with single species normal 3D unhindered diffusion with a single mean correlation timescale of 301 ± 104 µs with the corresponding D of 20.9 ± 6.9 µm2/s as shown in Fig. 1B. The lower correlation time of H2B-EGFP upon addition of high salt is possibly due to the dissociation of both free and bound multimeric H2B-EGFP into monomers. The decrease in the standard deviation of correlation timescales suggests that diffusion of core histone proteins is dependent on the spatial heterogeneity of chromatin architecture within the nucleus. Autocorrelation curves of the linker histone protein H1.1-EGFP have significantly different profile than that of the core histone proteins, which do not fit with single species 3D unhindered diffusion or with the anomalous diffusion. There is a second distinct timescale in the autocorrelation function, which may be attributed to the dynamic interactions of the linker histones with the chromatin fiber, as suggested by FRAP experiments. To obtain the underlying diffusion timescales, the data was fitted with a two species diffusion model described in the methods section. The fits to the data show two distinct timescales ( D1 = 298.3 ± 58.8 µs) with diffusion constant D1 = 19.5 ± 3.5 µm2/s commensurate with 3D diffusion, and ( D2 = 26.5 ± 12.8 ms) D2 = 0.3 ± 0.1 µm2/s possibly arising due to H1.1-EGFP interaction with DNA (Fig. 2B). (A)

Control H1.1-EGFP

Tailless H1.1-EGFP

(B) Intensity

0.6

G(τ)

0.8

Tailless H1.1 H1.1 EGFP H2B H4

0.3

0.0 0

0.4

40

80

120

160

Time (s)

H1.1 H1.1 tailless H1.5 cytoplasm H1.5 nucleus

0.0 101

103 Correlation time (μs)

105

Fig. 2 Linker histone diffusion reveals two distinct timescales. (A) Fluorescence images of H1.1-EGFP and tailless H1.1-EGFP transfected HeLa nucleus (scale bar 5 µm). (B) The associated autocorrelation function curves for H1.1-EGFP, tailless H1.1-EGFP and H1.5-EGFP in cytoplasm and cell nucleus. Inset—Fluorescence recovery curves for tailless H1.1-EGFP, H1.1-EGFP, EGFP, and core histones, H2B-EGFP and H4-EGFP.

3. Organization of Chromatin Assembly and TFs in Living Cells

65

The amino acid residues (1–40 and 121–216) corresponding to the N and C terminal tail sequences of H1.1 were deleted to explore the origin of the second timescale ( D2) (Th’ng et al., 2005). H1.1 tails could interact both with DNA and with adjacent histones on the chromatin assembly. Deleting the tail residues of H1.1-EGFP histones abolished the second diffusion timescale and the resultant FCS curves fit well with single species unhindered 3D diffusion with D = 20.2 ± 5.2 µm2/s, suggesting that these interactions are the source of the second timescale. FRAP experiments for H1.1-EGFP, tail-less H1.1-EGFP in comparison with free EGFP, H4-EGFP, and H2B-EGFP indicate loss of interaction in tail-less H1.1 (inset to Fig. 2B). Further, the correlation timescale of H1.5-EGFP in the cytoplasm of overexpressing cells fits with single species subdiffusive autocorrelation behavior having mean correlation timescale of 376.6 ± 132.4 µs (Fig. 2B). This indicates that the second timescale ( D2) of the linker histones within the nucleus arises primarily due to its interaction with the chromatin assembly. In addition, these methods allow a quantitative analysis of histone protein dynamics under differential functional perturbation of cellular processes such as ATP depletion, histone deacetylase inhibition or inducing cellular apoptosis by staurosporine (Bhattacharya et al., 2006). Using FCS and FRAP, we show a number of applications to study the translational dynamics of histone proteins in living cells. In the following section, we describe methods to quantify the rotational diffusion of histone proteins.

III. Higher-Order Chromosome Compaction Higher-order chromatin structure is aided by histone tail–tail interactions and their associations with nonhistone proteins like HMGs and HP1 result in the assembly of 30 nm fiber into loops of several kilo- to mega-basepairs (Grewal and Jia, 2007). The packing ratio is around 1000 for cells in interphase and reaches a highest of 7000 during mitosis. Such levels of packaging of DNA provide hindrance to its accessibility to processes such as gene transcription and DNA replication. However, this packaging of chromatin is nonuniform, distinct between decondensed, transcriptionally active euchromatin and more condensed, mostly silent heterochromatin. Though it is fairly established that heterochromatin has tighter folding, it is not clear how distinctly different they are from each other in terms of packing and the spatial gradients in compaction that separate them. Since nuclear organization is a dynamic process, the modulation of chromatin compaction in live cells, integrating various physical, che mical, and biological cues in temporal domain needs to be mapped. In vitro methods like Atomic Force Microscopy studies and X-Ray crystallography have given important insights to the various levels of chromatin structure. However, the heterogeneity in spatial organization is provided by fluorescence-based methods. This section describes fluorescence polarization–based methods to measure compac tion states of chromatin assembly in living HeLa cells. Further, compaction profiles are monitored by perturbing the chromatin structure as well as during functional alterations accompanying cell cycle.

66

Bidisha Sinha et al.

A. Cell Culture Stable lines of HeLa-EGFP or HeLa-H2BEGFP were used for the experiment. H1.1-EGFP was transiently transfected 12–16 h before the experiment. Cells were cultured as mentioned in the previous section. Cells were treated with 50, 100, or 150 ng/ml TSA (Sigma) in DMEM with 5% FBS for 4 h before imaging. For ATP depletion, cells at mid-log phase were washed with PBS and then treated with 10 mM Sodium Azide (Sigma) and 6 mM 2-deoxy-D-glucose (Sigma) in M1 without glucose for 1 h after which their medium was replaced by the imaging medium and cells were imaged. For inducing apoptosis, cells were treated with 10 µM Staurosporine (Sigma) for 4 h followed by replacing medium with fresh M1 for imaging.

B. Fluorescence Anisotropy Fluorescence anisotropy measures the rotational mobility of the fluorophores that are excited with polarized light. On exposure to polarized light, the fluorophores absorb photons with cos2 probability where is the angle between the incident polarization and the absorption dipole of the fluorophore, causing the photo selection of excited fluorophore population. Small molecules have a typical rotational diffusion () timescale of around 50–100 ps whereas the lifetime of fluorescence decay is around 10 ns. Thus, the rotational diffusion of the fluorophore brings about depolarization of the emitted light. Hence, anisotropy depicts average angular displacement in the time between absorption and emission of photon. For anisotropy images, the parallel (III) and perpen dicular (I?) components of the emitted light (with respect to polarized excitation) are simultaneously acquired and used to calculate the steady state fluorescence anisotropy as r ¼ ðIII I? Þ= ðIII þ 2I? Þ. In steady state measurements, anisotropy follows the Perrin’s equation, r = r0/(1 þ RT/V), where r0 is the value of anisotropy at t = 0 after short pulse excitation, is the fluorescence life time of the fluorophore, is the local viscosity of the solution, and V is the hydrodynamic radius, indicating the size and shape of molecule. In these experiments, the dependence of r as a function of (the viscosity) gives a measure of the average local fluidity of the fluorophores. Anisotropy values were calculated from wide-field fluorescence images acquired on a NIKON/OLYMPUS microscope with 100/1.4 NA objective and images captured with ICCD cameras (Roper Scientific). Mercury arc lamp was used for the excitation light which is then selected for vertical polarization using a sheet polarizer (Melles Griot). The collected emission is split into its parallel and perpendicular polarization components using a polarizing beam splitter (Melles Griot) or by swapping between two polarizers (Melles Griot) parallel and perpendicular to the excitation. Images were captured using Vþþ Digital Optics software and analyzed using LabVIEW.

C. Results Fluorescence anisotropy measures the rotational mobility of the fluorophores, which in this case is free EGFP or EGFP tagged to H2B or H1.1 proteins, reflecting the local

67

3. Organization of Chromatin Assembly and TFs in Living Cells

(A)

WT

ATPdep

Stau

0.15

chromatin compaction. Steady state anisotropy imaging of HeLa cells at interphase reveals appreciable spatial heterogeneity in compaction (Banerjee et al., 2006). Aniso tropy line scan across the nucleus indicates that the packing is indeed nonuniform at micrometer length scales. Gaussian fits to these structures show their sizes range from 0.5 to 2 µm. The mean and standard deviation in intensity for single HeLa cells expressing plain EGFP, H1.1-EGFP, or H2B-EGFP, were similar, thus ruling out that heterogeneity in compaction is a result of variation in intensity. However, the mean anisotropy (r) as well as standard deviation (sd) in anisotropy are different and reflect the interaction of the probe with the chromatin since the histones show greater mean and sd in anisotropy values. The noninteracting EGFP that diffuses freely in the nucleus shows negligible heterogeneity as compared to the histones (EGFPsdWT = 0.005). H2B-EGFP shows various rotational mobilities in the nucleus with a higher standard deviation of aniso tropy H2BEGFPsdWT = 0.013 . In contrast, the map of the linker histone, H1.1-EGFP, showed a less heterogeneous environment (H1.1EGFPsdWT = 0.010) (Fig. 3A and B). The lower sd for linker proteins is consistent with the dynamic interaction model of the linker histones (Lever et al., 2000; Misteli et al., 2000) in which, while the core histones are mostly immobile, a large fraction of linker histones are under continuous transit from one binding site to another.

(B) 0.20

EGFP 0.021

0.18

0.172

EGFP

0.014

Sd

r

0.19

0.007

0.17

0.194

0.20

H2B-EGFP

0.021

0.19

r

0.014

Sd

0.18

0.216

H1.1-EGFP

Control -- ATPdep -- Stau -

0.007

0.17

0.238

0.20

H1.1-EGFP

0.020 0.19

r

0.015 0.18

0.26

0.025

Sd

H2B-EGFP

Control -- ATPdep -- Stau -

0.010

0.17

0.005 Control -- ATPdep -- Stau -

Fig. 3 Chromatin is heterogeneously packed in the nucleus. (A) Color-coded anisotropy map of HeLaEGFP, HeLa-H1.1-EGFP, and HeLa-H2BEGFP cells in untreated (WT), ATP-depleted (ATP dep), or staurosporine (Stau)-treated cells (scale bar 10 µm). (B) Associated bar graph representation of mean anisotropy (r) and standard deviation of anisotropy, sd of all pixels inside a single nucleus (N = 45 cells in each case).

68

Bidisha Sinha et al.

The chromatin compaction heterogeneity in the presence and absence of ATP was studied to understand if it is actively maintained. Upon depleting the cells of ATP, the micron scale substructures as observed previously form larger clumps, which show a broader standard deviation H2BEGFPsdATPdep = 0.019 indicating an increase in com paction heterogeneity with an overall loosening of the chromatin. The anisotropy map observed by labeling the linker histone, H1.1-EGFP, also shows an increase in the heterogeneity H1.1EGFPsdATPdep = 0.012 . However for plain EGFP, the heterogeneity still remains at a much lower level EGFPsdATPdep = 0.006 (Fig. 4A and B) with no specific structures emerging. Apoptosis is yet another process where the state of chromatin is drastically altered and the chromatin is known to both fragment and aggregate (Rogalinska, 2002). When apoptosis is induced (10 µM Staurosporine), the heterogeneity is altered to a state where regions either low or high H2BEGFPon the chromatin show anisotropy values that are distinctly sdStau = 0.022 . H1.1-EGFP showed a less increase H1.1EGFPsdStau = 0.016 and for plain EGFP EGFPsdStau = 0.008 the values remain low though higher than the untreated population (Fig. 3A and B). It thus indicates that the maintenance of the normal chromatin heterogeneity requires active cellular processes that keep a balance between opened up and closed structures by creating a large number of intermediate structures.

(A)

0.03

00:00

(B)

0.00

P (r )

0.03

00:32

0.00 0.03

05:21

0.00 0.03

15:11

0.00 0.1

0.2 0.3 0.15 Anisotropy (r )

0.26

Fig. 4 Transitions in local chromatin fluidity during cell cycle. (A) Anisotropy distributions of H2B-EGFP from metaphase to interphase in a single HeLa cell nucleus. Anaphase (t = 0:17 min) onward the anisotropy distributions of the left daughter cell has been plotted. (B) Panel shows color-coded anisotropy snapshots at corresponding time points (scale bar 10 µm).

3. Organization of Chromatin Assembly and TFs in Living Cells

69

Steady state time-lapse anisotropy in a single HeLa cell expressing H2BEGFP from prometaphase or late anaphase into G1 interphase was captured to study large-scale chromatin reorganization accompanying various stages of the cell cycle (Tremethick, 2007). The compaction levels are much higher initially, and also the heterogeneity in the compaction which increases during anaphase decreases as interphase is attained. The high heterogeneity observed is an evidence for large conformational freedom possible for the mitotic chromatin and the increase in heterogeneity during late M phase is consistent with the increased exchange rates observed for core and linker histones (Chen et al., 2005). The two distinct compaction states at early G1 represent an unfolding intermediate (second peak) and the partially decondensed chromatin (first peak) consistent with the chromonema model for interphase chromatin. Measuring rotational mobility of fluoro phores thus gives compaction details of the nucleus providing the handle to be able to directly study differential chromatin dynamics across the nucleus in time (Fig. 4A and B). The techniques described so far can be used to probe a functional process such as cellular differentiation which is marked by large-scale chromatin reorganization.

IV. Nuclear Plasticity Cellular differentiation results in entire repertoire of distinct cell types that are derived from programmed changes in the genome wide expression patterns. The mechanism for uncommitted embryonic stem (ES) cells to transit into cell types with distinct transcrip tional profiles is not clearly understood. Such transitions in gene expression with lineage commitment might be facilitated by changes in the higher-order chromatin assembly. A number of studies to date have indicated unique epigenetic signatures of ES cells compared to differentiated cells (Azuara et al., 2006; Bernstein et al., 2006; Jenuwein and Allis, 2001). It is also shown that major architectural chromatin proteins are hyperdynamic and bind loosely to chromatin in ES cells (Meshorer et al., 2006), thus contributing to a hypothesis that chromatin is retained in a globally relatively open, plastic state which is important for the maintenance of pluripotency. While plasticity at the level of higher-order chromatin assembly is functionally important, how it relates to the structural dynamics of nuclear architecture is poorly understood. In this section, the dynamics of lamin protein in mouse ES cells is described and is contrasted with lineage-restricted primary fibroblast cells using livecell fluorescence imaging. Further, the temporal evolution of translational and rota tional dynamics of histone proteins, using FRAP and anisotropy, during development of the Drosophila embryo is discussed. A. Cell Culture R1 ES cells were cultured on a layer of feeder cells (primary mouse embryonic fibroblasts—PMEF) with DMEM-F12 supplemented with 15% fetal bovine serum (HyClone), 1 mM sodium pyruvate, 0.1 mM nonessential amino acids, 2 mM L-Glutamine, 0.1 mM -mercaptoethanol (Sigma), 500 U/ml leukemia inhibitory factor

70

Bidisha Sinha et al.

(Chemicon), and penicillin–streptomycin. PMEF were cultured with DMEM-F12 sup plemented with 5% FBS, and penicillin–streptomycin. Cells were maintained at 37°C in a 5% CO2 incubator. PMEF cells up to third passage were used in experiments. Cells were cultured for 24 h before being transfected with the required plasmid constructs— pBOS H2B-EGFP, pC1 EGFP-LaminB1, or pN1 HP1-EGFP. B. Fly Lines Early embryo of transgenic fruit fly Drosophila melanogaster, where core histone H2B is tagged to EGFP (H2B-EGFP), was used for some of the experiments. Flies were kept for 1 h on a sucrose plate for egg laying. The embryo was then mounted on a No. 1 coverslip and covered with halocarbon oil 700. Alexa488 (Molecular Probes, Eugene, OR)-labeled linker histones (H1-Alexa488) were microinjected into wild-type Canton-S (CS) embryos before the 11th nuclear division. C. Results Time-lapse fluorescence imaging of the EGFP-LaminB1 marked nucleus showed a highly dynamic nuclear lamina in the ES cells (Bhattacharya et al., 2009). Here EGFP LaminB1 shows significant nucleoplasmic intensity in PMEF cells as compared to ES cells where it is restricted to the envelope. The time series of the mean-square fluctuation [h(r)2i = (ri)2/N] of the nuclear radius, computed over all angles from the centroid position, reveals large fluctuations in the ES cells compared to the PMEF cells, indicative of a physically more plastic organization of the lamina architecture in the ES cells (Fig. 5A and B). Further, histone dynamics with progression of differentiation can be studied in vivo in a developing model organism. This was done in an early embryo of transgenic Drosophila melanogaster line expressing the core histone H2B-EGFP (Fig. 6A) Before cellularization, during syncytial blastoderm stage, for a whole nucleus photo-bleaching experiment, recovery fraction was 94% after the 11th division, 65% after the 12th division, 25% after the 13th division, and 8% 1 h postcellu larization in 400 s (Fig. 6B). It was observed that the histone proteins exchanged rapidly through nuclear membrane in and out of the nuclei, before onset of celluraliza tion. Small area photobleaching, within nuclei, indicates hyper dynamic plasticity in early stage of development. After cell membrane formation, core histones are very dynamic within nuclei but cease within 5 h from 13th nuclear division, indicating that the chromatin structure is undergoing compaction postcellularization. Alexa488 labeled linker histones microinjected into the embryo exchanged rapidly during 12th and 13th division, showing similar dynamics as core histones. Fluorescence-anisotropy imaging to map the rotational freedom of the bound H2B-EGFP in these early nuclei reflects their chromatin compactions. Color-coded anisotropy images of H2B-EGFP in developing Drosophila embryos showed that concomitant with the impeded mobility of core histone proteins, there was an overall rise in chromatin rigidity (Fig. 6C). This is also reflected as increase in the standard deviation over pixels of anisotropy, indicating the emergence of heterogeneity in chromatin rigidity states (Bhattacharya

71

(A)

PMEFs ES cells

3. Organization of Chromatin Assembly and TFs in Living Cells

B 0 sec

282 sec

445 sec

553 sec

646 sec

(B) (δr )2(μm2)

0.16

<(δr )2> (μm2)

0.4

0.08

0.00 ES cell

0.2

PMEF

ES PMEF

0.0 0

325 Time (s)

650

Fig. 5 Plasticity in the nuclear scaffold in undifferentiated ES cells. (A) Time-lapse images of ES and PMEF cells expressing LaminB1-EGFP are shown for the time points as indicated. (B) Time series of mean square fluctuations of the nuclear membrane in ES and PMEF cells. Inset—Mean and standard error of mean square fluctuations of nuclear membrane is plotted.

et al., 2009). Thus, these fluorescence-based methods (FRAP, time-lapse imaging and anisotropy) were able to probe the translational and rotational dynamics in interphase cells and during cellular differentiation and development. In the next section, we describe the methods to study the dynamic organization of functional compartments within the nucleus of living cells.

V. Spatio-temporal Organization of Transcription Factories and Gene Loci Cell biology experiments of visualizing the sites of transcription show that it occurs at discrete sites in the nucleus called TFs and that they are enriched in RNA polymerase. The clustering of RNA polymerases in TFs leads to their enrichment perhaps facilitating transcription. By immunostaining techniques accompanied with confocal imaging and image deconvolution, it has been shown that these TFs in the eukaryotic nucleus are much fewer in number as compared to number of genes being transcribed at any given time. This disparity in the number is resolved by studies that imply the formation of active chromatin hub where gene loops and TFs co-cluster to bring about transcription of co-regulated genes (Osborne et al., 2004; Xu and Cook, 2008; Zhou et al., 2006).

72

Bidisha Sinha et al.

AnteriorPosterior

(A)

(B)

Normalized intensity

1.6 B

1.2

0s

416 s

16 s

After 11th After 12th After 13th 0 hr After 13th 1 hr

0.8

0.4

0.0 0

(C)

100

Anterior

200 Time (s)

300

Posterior

400

0.20

0 min 0.16

70 min 0.12

Fig. 6 Hyperdynamic mobilities of histone- and chromatin-binding proteins in uncommitted systems. (A) H2BEGFP-labeled Drosophila embryo before cellularization (scale bar 50 µm). (B) Normalized fluorescence intensities of typical nuclei at different time points (after the 11th, 12th, and 13th nuclear divisions and 1 h from the 13th nuclear division). Inset—A nucleus before photobleaching and subsequent frames after photobleaching (0, 200, and 400 s) and before cellularization (after the 12th nuclear division) is shown (scale bar 5 µm). (C) Color-coded anisotropy maps are shown for typical nuclei in the anterior and posterior of the embryo at 0 min and at 70 min. The bar to the right shows anisotropy values corresponding to specific colors.

In this section we describe time-lapse imaging methods to probe the dynamics of TFs and its association with gene loci. A. Visualization of Transcription Factories by Fluorescent UTPs in Live Cells To visualize TFs in live cells (Sinha et al., 2008), Texas-Red labeled UTP (Mole cular Probes) or CY5-UTP (NeN) were used and UTP molecules were incorporated into cells by a brief hypotonic shock with KHB buffer (10 mM HEPES at pH7.4, 30 mM KCl) containing 10 µM UTP from 5 to 10 min (depending on cell type). Similar

73

3. Organization of Chromatin Assembly and TFs in Living Cells

(B)

TC 21 trajectory

60

30

Y (μm)

H2BEGFP TF

TF (%)

(A)

0

Normal

Sub

11 11

Super

21 X (μm)

MSD (μm)2

(C)

(D)

(E)

0.02

Sub α < 1 20

0.5

0.01

10

0.0

0.00

0

Normal α=1

1.0

0

25

50

75

0

25

50

75

Super α>1

0

25

50

75

Fig. 7 Organization of transcription into factories and its dynamic coupling to chromatin.(A) H2BEGFP HeLa cells with Alexa546 UTP-labeled TFs. (B) XY trajectories of various TCs from the same nucleus are shown in black. Inset—Percentage of TFs showing different regimes of diffusion. Typical MSD vs. plots for determination of for (C) normal diffusion, (D) subdiffusion, and (E) superdiffusion.

procedure was followed for serial two-color labeling where the second UTP was incorporated after 2 h. Cells were then washed with DMEM and incubated with 5% FBS at 37°C in 5% CO2 for 5–10 min before imaging. The imaging medium used throughout is 5% FBS in DMEM without phenol red. Cells were then imaged at 37°C for the UTPs in the nucleus as seen in Fig. 7A.

B. Single Particle Tracking and Analysis The advancement in the field of digital image processing has enabled the automation of object detection and analysis. To obtain the trajectories for individual transcription compartments, as described above, IMAQ Vision and LabVIEW were used. The automation of TFs detection and localization was achieved by appropriate iterative intensity thresholding and calculation of the center of mass ((! x ðtÞor (x(t),y(t)). The trajectories describe their mobility through the nuclear mesh qualitatively (Fig. 7B) suggesting that the nature of mobility of TFs vary from being confined to transiently confined to completely mobile. Further quantitative analysis was used to compare the changes in the TFs mobility caused by the perturbation of nuclear architecture. Velocity and mean square displacement (MSD) distributions of about 30 TFs in each

74

Bidisha Sinha et al.

condition were calculated. The trajectories were also used to calculate the MSD for 2 each TF using MSDðÞ ¼ hf! x ðt þ Þ ! x ðtÞg it ¼ 4Dt , where ! x ðtÞ represents a position of the TF at time t, is arbitrary time window, D is the diffusion coefficient, and an exponent describing the nature of motion. To estimate D, the calculated MSD was fit to which was then used to characterize the nature of TF dynamics, where is a direct readout for different mode of motions; 1 for pure diffusion, < 1 for subdiffusion and > 1, for superdiffusion. C. Labeling Strategies of Gene Loci in Live Cells Gene loci in live cells are made visible by insertion of protein-binding DNA sequences flanking it which are bound by co-expressed fluorescent DNA-binding proteins. This could be either used as a transient transfection system or incorporated in the genome by making stable cell lines. The most extensively used system for marking gene loci uses Lac repressor (LacI)-binding sequences and Lac Operator site (LacO) inserted upstream of the gene of interest (Bystricky et al., 2004; Chuang et al., 2006; Kumaran and Spector, 2008). A LacI-RFP containing Nuclear Localizing Signal (NLS) was co-transfected with the plasmid having EGFP reporter and 96 LacO sequences. The binding of the fluorescent LacI-NLS enabled visualization of the gene locus as bright punctae in the background haze of its nonspecific binding. To control the transcriptional status of a reporter gene (EGFP), a hormone response element, Tet response element was used (Gossen and Bujard, 1992). The transcription of this regulatory element can be controlled by adding doxycycline (Sinha et al., 2008). D. Results Live cell imaging of the TFs showed that these nuclear bodies are dynamic in nature (Fig. 7B). Single particle tracking of large number of TFs in HeLa cells showed that they undergo three regimes of diffusion: (a) normal diffusion where of TFs were between 0.8 and 1.0 (Fig. 7C); (b) subdiffusion where a is less than 0.8 (Fig. 7D), and (c) superdiffusive motion, where is greater than 1 (Fig. 7E). Quantification of these TF trajectories revealed that a statistically significant fraction of TFs (18%) undergo superdiffusion or normal diffusion (23%), while a larger number of TFs exhibit subdiffusive (59%) transport (Fig. 7B). The mean 0.6 and velocities Vavg 71 nm/s of TFs were computed from their trajectories. The dynamics of these compartments is sensitive to the chromatin architecture and the accompanying perturbations (Sinha et al., 2008). Using Fluorescence In Situ Hybridization (FISH) for gene locus and corresponding chromosome territories, it has been shown that genes like Hox B loop out of its chromosome territories when activated (Chambeyron et al., 2005). Similar experi ments for other gene like CFTR (Zink et al., 2004), globin (Zhou et al., 2006), and INO-1 (Brickner and Walter, 2004) have shown that these loci move toward the nuclear periphery or near to the constitutive heterochromatin when inactive and move toward the nuclear center or away from the repressive constitutive heterochromatin when active. Though these methods revealed changes in their positions, these are

75

3. Organization of Chromatin Assembly and TFs in Living Cells

mRFP-LacI-NLS (B)

(A)

Gene locus and TC

EGFP TRE pCMVmin 96lacO repeats (7tetO)

(C)

0.9

Gene locus trajectories

Y (μm)

0.6

0.3 Expressing

0.0 Repressed

0.0

0.3 X (μm)

0.6

0.9

Fig. 8 Gene locus dynamics.(A) Image of HeLa cell expressing mRFP-lacI-NLS (scale bar 5 µm). Arrow indicates position of the EGFP gene locus. (B) Schematic of mRFP lacI binding to 96lacO repeat site upstream of pCMVmin-driven EGFP gene. (C) Typical trajectories of gene loci in repressed and activated state.

insufficient in addressing the issue of dynamics and timescales of these movements. The strategy for this involves the use of fluorescent DNA-binding proteins that can specifically label gene loci and that could be coupled to regulate gene expression systems as described before. Live cell imaging studies revealed transcription depen dent dynamics of these gene loci (Bystricky et al., 2004). In our experiments, gene locus showed constrained diffusion within the radius of 0.5 µm. In the expressing state the locus showed more mobility which reduced in the nonexpressing state induced by repressor protein tTS or DRB (transcriptional inhibitor) (Fig. 8C). In the expressing state of the gene locus, the labeled TFs and the gene locus were seen to co-localize and move in synchrony (Sinha et al., 2008).

VI. Conclusions Nuclear architecture is as yet poorly defined domain and its impact on genome functions like transcription, replication, and repair is beginning to be understood. With the extended tool box of labeling and imaging techniques, some of which are discussed here, the functional importance of chromatin organization are coming to light. Tech niques like FCS, FRAP, and anisotropy have shown that histones are dynamic and thus

76

Bidisha Sinha et al.

regulate the accessibility of DNA for various nuclear processes. These techniques were able to unravel differential interaction timescales of histone proteins with DNA in living cells and developing Drosophila embryo. Both the chromatin and the nuclear lamina are kept in a plastic state in the undifferentiated state as revealed by FRAP experiments. In addition they reveal dynamic organization of gene locus and TFs. The advent of these methods and rapid progress in visualizing the spatio-temporal aspects of functional nuclear architecture has opened up new avenues to understand in processing of genetic information, close to single nucleosome resolution in living systems.

Acknowledgments We thank the Nanoscience Initiative of Department of Science and Technology (DST) for funding and the NCBS Common Imaging and Flow Facility (CIFF). SM and ST thank Council for Scientific and Industrial Research (CSIR) for their graduate research fellowships.

References Azuara, V., Perry, P., Sauer, S., Spivakov, M., Jorgensen, H. F., John, R. M., Gouti, M., Casanova, M., Warnes, G., Merkenschlager, M., and Fisher, A. G. (2006). Chromatin signatures of pluripotent cell lines. Nat. Cell Biol. 8(5), 532–538. Banerjee, B., Bhattacharya, D., and Shivashankar, G. V. (2006). Chromatin structure exhibits spatio-temporal heterogeneity within the cell nucleus. Biophys. J. 91(6), 2297–2303. Bernstein, B. E., Mikkelsen, T. S., Xie, X., Kamal, M., Huebert, D. J., Cuff, J., Fry, B., Meissner, A., Wernig, M., Plath, K., et al. (2006). A bivalent chromatin structure marks key developmental genes in embryonic stem cells. Cell 125(2), 315–326. Bhattacharya, D., Mazumder, A., Miriam, S. A., and Shivashankar, G. V. (2006). EGFP-tagged core and linker histones diffuse via distinct mechanisms within living cells. Biophys. J. 91(6), 2326–2336. Bhattacharya, D., Talwar, S., Mazumder, A., and Shivashankar, G. V. (2009). Spatio-temporal plasticity in chromatin organization in mouse cell differentiation and during Drosophila embryogenesis. Biophys. J. 96(9), 3832–3839. Brickner, J. H., and Walter, P. (2004). Gene recruitment of the activated INO1 locus to the nuclear membrane. PLoS Biol. 2(11), e342. Bruno, M., Flaus, A., Stockdale, C., Rencurel, C., Ferreira, H., and Owen-Hughes, T. (2003). Histone H2A/H2B dimer exchange by ATP-dependent chromatin remodeling activities. Mol. Cell 12(6), 1599–1606. Bustin, M., Catez, F., and Lim, J. H. (2005). The dynamics of histone H1 function in chromatin. Mol. Cell 17 (5), 617–620. Bystricky, K., Heun, P., Gehlen, L., Langowski, J., and Gasser, S. M. (2004). Long-range compaction and flexibility of interphase chromatin in budding yeast analyzed by high-resolution imaging techniques. Proc. Natl. Acad. Sci. USA 101(47), 16495–16500. Chambeyron, S., Da Silva, N. R., Lawson, K. A., and Bickmore, W. A. (2005). Nuclear re-organisation of the Hoxb complex during mouse embryonic development. Development 132(9), 2215–2223. Chen, D., Dundr, M., Wang, C., Leung, A., Lamond, A., Misteli, T., and Huang, S. (2005). Condensed mitotic chromatin is accessible to transcription factors and chromatin structural proteins. J. Cell Biol. 168(1), 41–54. Chuang, C. H., Carpenter, A. E., Fuchsova, B., Johnson, T., de Lanerolle, P., and Belmont, A. S. (2006). Long-range directional movement of an interphase chromosome site. Curr. Biol. 16(8), 825–831. Cremer, T., and Cremer, C. (2001). Chromosome territories, nuclear architecture and gene regulation in mammalian cells. Nat. Rev. Genet. 2(4), 292–301.

3. Organization of Chromatin Assembly and TFs in Living Cells

77

Dechat, T., Pfleghaar, K., Sengupta, K., Shimi, T., Shumaker, D. K., Solimando, L., and Goldman, R. D. (2008). Nuclear lamins: Major factors in the structural organization and function of the nucleus and chromatin. Genes. Dev. 22(7), 832–853. Dekker, J., Rippe, K., Dekker, M., and Kleckner, N. (2002). Capturing chromosome conformation. Science 295(5558), 1306–1311. Ferreira, J., Paolella, G., Ramos, C., and Lamond, A. I. (1997). Spatial organization of large-scale chromatin domains in the nucleus: A magnified view of single chromosome territories. J. Cell Biol. 139(7), 1597–1610. Francastel, C., Schubeler, D., Martin, D. I., and Groudine, M. (2000). Nuclear compartmentalization and gene activity. Nat. Rev. Mol. Cell Biol. 1(2), 137–143. Gasser, S. M. (2002). Visualizing chromatin dynamics in interphase nuclei. Science 296(5572), 1412–1416. Gossen, M., and Bujard, H. (1992). Tight control of gene expression in mammalian cells by tetracyclineresponsive promoters. Proc. Natl. Acad. Sci. USA 89(12), 5547–5551. Grewal, S. I., and Jia, S. (2007). Heterochromatin revisited. Nat. Rev. Genet. 8(1), 35–46. Happel, N., and Doenecke, D. (2009). Histone H1 and its isoforms: Contribution to chromatin structure and function. Gene 431(1–2), 1–12. Heng, H. H., Goetze, S., Ye, C. J., Liu, G., Stevens, J. B., Bremer, S. W., Wykes, S. M., Bode, J., and Krawetz, S. A. (2004). Chromatin loops are selectively anchored using scaffold/matrix-attachment regions. J. Cell Sci. 117(Pt 7), 999–1008. Jenuwein, T., and Allis, C. D. (2001). Translating the histone code. Science 293(5532), 1074–1080. Khorasanizadeh, S. (2004). The nucleosome: From genomic organization to genomic regulation. Cell 116(2), 259–272. Kimura, H., and Cook, P. R. (2001). Kinetics of core histones in living human cells: Little exchange of H3 and H4 and some rapid exchange of H2B.J. Cell Biol. 153(7), 1341–1353. Kumaran, R. I., and Spector, D. L. (2008). A genetic locus targeted to the nuclear periphery in living cells maintains its transcriptional competence. J. Cell Biol. 180(1), 51–65. Lever, M. A., Th’ng, J. P., Sun, X., and Hendzel, M. J. (2000). Rapid exchange of histone H1.1 on chromatin in living human cells. Nature 408(6814), 873–876. Luger, K., and Richmond, T. J. (1998a). DNA binding within the nucleosome core. Curr. Opin. Struct. Biol. 8(1), 33–40. Luger, K., and Richmond, T. J. (1998b). The histone tails of the nucleosome. Curr. Opin. Genet. Dev. 8(2), 140–146. Luger, K., Mader, A. W., Richmond, R. K., Sargent, D. F., and Richmond, T. J. (1997). Crystal structure of the nucleosome core particle at 2.8 A resolution. Nature 389(6648), 251–260. Mangenot, S., Leforestier, A., Vachette, P., Durand, D., and Livolant, F. (2002). Salt-induced conformation and interaction changes of nucleosome core particles. Biophys. J. 82(1 Pt 1), 345–356. Meshorer, E., Yellajoshula, D., George, E., Scambler, P. J., Brown, D. T., and Misteli, T. (2006). Hyperdynamic plasticity of chromatin proteins in pluripotent embryonic stem cells. Dev. Cell 10(1), 105–116. Misteli, T. (2007). Beyond the sequence: Cellular organization of genome function. Cell 128(4), 787–800. Misteli, T., Gunjan, A., Hock, R., Bustin, M., and Brown, D. T. (2000). Dynamic binding of histone H1 to chromatin in living cells. Nature 408(6814), 877–881. Muthurajan, U. M., Park, Y. J., Edayathumangalam, R. S., Suto, R. K., Chakravarthy, S., Dyer, P. N., and Luger, K. (2003). Structure and dynamics of nucleosomal DNA. Biopolymers 68(4), 547–556. Osborne, C. S., Chakalova, L., Brown, K. E., Carter, D., Horton, A., Debrand, E., Goyenechea, B., Mitchell, J. A., Lopes, S., Reik, W., and Fraser, P. (2004). Active genes dynamically colocalize to shared sites of ongoing transcription. Nat. Genet. 36(10), 1065–1071. Palstra, R. J., Tolhuis, B., Splinter, E., Nijmeijer, R., Grosveld, F., and de Laat, W. (2003). The beta-globin nuclear compartment in development and erythroid differentiation. Nat. Genet. 35(2), 190–194. Periasamy, N., and Verkman, A. S. (1998). Analysis of fluorophore diffusion by continuous distributions of diffusion coefficients: Application to photobleaching measurements of multicomponent and anomalous diffusion. Biophys. J. 75(1), 557–567.

78

Bidisha Sinha et al. Rigler., R. (2001). Fluorescence Correlation Spectroscopy—Theory and Applications.Springer, New York. Rogalinska, M. (2002). Alterations in cell nuclei during apoptosis. Cell Mol. Biol. Lett. 7(4), 995–1018. Schones, D. E., and Zhao, K. (2008). Genome-wide approaches to studying chromatin modifications. Nat. Rev. Genet. 9(3), 179–191. Sinha, D. K., Banerjee, B., Maharana, S., and Shivashankar, G. V. (2008). Probing the dynamic organization of transcription compartments and gene loci within the nucleus of living cells. Biophys. J. 95(11), 5432–5438. Spector, D. L. (2001). Nuclear domains. J. Cell Sci. 114(Pt 16), 2891–2893. Spector, D. L. (2003). The dynamics of chromosome organization and gene regulation. Annu. Rev. Biochem. 72573–72608. Sproul, D., Gilbert, N., and Bickmore, W. A. (2005). The role of chromatin structure in regulating the expression of clustered genes. Nat. Rev. Genet. 6(10), 775–781. Sutherland, H., and Bickmore, W. A. (2009). Transcription factories: Gene expression in unions? Nat. Rev. Genet. 10(7), 457–466. Talbert, P. B., and Henikoff, S. (2010). Histone variants—ancient wrap artists of the epigenome. Nat. Rev. Mol. Cell Biol. 11(4), 264–275. Th’ng, J. P., Sung, R., Ye, M., and Hendzel, M. J. (2005). H1 family histones in the nucleus. Control of binding and localization by the C-terminal domain.J. Biol. Chem. 280(30), 27809–27814. Tremethick, D. J. (2007). Higher-order structures of chromatin: The elusive 30 nm fiber. Cell 128(4), 651–654. Workman, J. L., and Kingston, R. E. (1998). Alteration of nucleosome structure as a mechanism of transcriptional regulation. Annu. Rev. Biochem. 67545–67579. Xu, M., and Cook, P. R. (2008). Similar active genes cluster in specialized transcription factories. J. Cell Biol. 181(4), 615–623. Zhou, G. L., Xin, L., Song, W., Di, L. J., Liu, G., Wu, X. S., Liu, D. P., and Liang, C. C. (2006). Active chromatin hub of the mouse alpha-globin locus forms in a transcription factory of clustered housekeeping genes. Mol. Cell Biol. 26(13), 5096–5105. Zink, D., Amaral, M. D., Englmann, A., Lang, S., Clarke, L. A., Rudolph, C., Alt, F., Luther, K., Braz, C., Sadoni, N., et al. (2004). Transcription-dependent spatial arrangements of CFTR and adjacent genes in human cell nuclei. J. Cell Biol. 166(6), 815–825.

CHAPTER 4

Manipulation and Isolation of Single Cells and Nuclei Swee Jin Tan*,†, Qingsen Li‡, and Chwee Teck Lim*,†,‡ *

NUS Graduate School for Integrative Sciences and Engineering, National University of Singapore, Singapore 117456, Singapore

† Division of Bioengineering and Department of Mechanical Engineering, National University of Singapore, Singapore 117576, Singapore ‡

Mechanobiology Institute, National University of Singapore, Singapore 117411, Singapore

Abstract I. Introduction II. Techniques for Single-Cell Manipulation A. Microfluidics B. Dielectrophoresis C. Optical-Based Techniques D. Micropipette Aspiration III. Nuclear Isolation and Manipulation A. Nuclear Isolation Method B. Application of Nucleus Study IV. Discussion and Future Implications

References

Abstract The heterogeneous behavior of cells within a cell population makes measurements at the multicellular level insensitive to changes in single cells. Single-cell and single-nucleus analyses are therefore important to address this deficiency which will aid in the under standing of fundamental biology at both the cellular and subcellular levels. Recent technological advancements have enabled the development of new methodologies capable METHODS IN CELL BIOLOGY, VOL. 98 Copyright � 2010 Elsevier Inc. All rights reserved.

79

978-0-12-381009-0 DOI: 10.1016/S0091-679X(10)98004-7

80

Swee Jin Tan et al.

of handling these new challenges. This review highlights various techniques used in single-cell and single-nucleus manipulation and isolation. In particular, the applications related to microfluidics, electrical, optical, and physical methods will be discussed. Ultimately, it is hoped that these techniques will enable fundamental tests to be conducted on single cells and nuclei. One important potential outcome is that this will contribute not only towards detection and isolation of diseased cells but also more accurate diagnosis and prognosis of human diseases.

I. Introduction Despite being the basic units of life, cells are extremely complex and dynamic in nature. In fact the study of biological cells is important as the condition of the human body is closely related to the state and function of these entities (Alberts et al., 2002). Various biochemical processes within the cell accounts for its different roles and are critical in cell migration, growth, and apoptosis during development (Heyder et al., 2006; Papadaki and Eskin, 1997; Stossel, 1993). Aberrations in these cellular func tions tend to lead to disease manifestation which is detrimental to the health and well being of the human body. Cancer, for example, is caused by single-cell malfunctions at the genetic level due to heredity or exposure to external stimulus such as ionizing radiation and chemicals, which has dire consequences. Manipulation and study of single cells at the cellular and genetic level will aid in better understanding the pathophysiology of the disease, and assist in pursuing enhanced treatments and diagnostic methods (Pantel et al., 2008; Zhong et al., 2008). Also, in stem cell research, these cells hold important therapeutic possibilities (McNeish, 2004) and the ability to manipulate stem cells will be useful to fully utilize its potential. Also, the manipulation of single cell has helped to increase the success rate of assisted reproduction via various chemical and physical means in in vitro fertilization (Raty et al., 2004). Thus, given the vast potential applications, there is interest to manip ulate single cells, and understand the heterogeneous behavior of cells at the cellular, nuclear, and genetic levels. Biological cells typically in the range of several microns to tens of microns are extremely hard to handle for its small size. Precise control and instrumentation are required to work at such resolutions to correctly position the cell to the desired location for measurements to be taken (Van Vliet et al., 2003). The advent of various breakthroughs in micro and nanotechnology has aided the development of numerous manipulation and analytical methods for the qualitative and quantitative analysis of single cells. Techniques that are built upon engineering principles such as microelec tronics as well as cell and nuclear mechanics have enabled the handling of micron and submicron size objects or samples more precisely. These methodologies also offer a high-throughput analysis using small sample volumes (Bashir, 2004; Whitesides, 2006) and make these technologies suitable to handle the challenges involved in single-cell and single-nucleus manipulations, and analyze the molecular components such as DNA and RNA. For instance, microfluidic devices offer numerous advantages for such

4. Manipulation and Isolation of Single Cells and Nuclei

81

activities, providing a platform for both cell separation and single-cell analysis. By utilizing size, density, or affinity-based methods, single-cell operations can be effectively achieved. Micropipette aspiration, which involves the use of glass pipettes, is a straight forward means to position single cells. Subsequently, analyses can then be applied such as investigating their electrical and mechanical properties of these single cells. Working with single cells also requires a suitable environment as excessive perturbations to cells may affect their integrity. Hydrodynamic forces have been reported to directly activate certain biochemical pathways, affecting important events such as cytoskeletal arrangement, cell motility, and proliferation (Chang et al., 2008; Ding et al., 2001; Papadi mitriou et al., 1999). Larger perturbations can potentially lyze and damage the cells (Weiss, 1991; Weiss and Dimitrov, 1986; Weiss et al., 1985). These should be minimized during cell handling so that downstream analysis can be free from false-positive or false-negative results. In addition, ultrasensitive detection systems are needed to accurately measure minute changes within a single cell (Heath and Davis, 2008) and have to take into consideration the throughput of the system where a significant number of measurements are required to derive at a sound deduction (Lekka and Laidler, 2009). Therefore, there are challenges when dealing with single-cell manipulations that have to be considered. Traditional benchtop tools in cell and molecular biology are not sufficient to address the needs for sensitive and accurate measurements of single cells. The technologies described in this chapter aims to complement current methods used and will highlight with clear examples how they can be applied. In particular, various techniques utilized in single-cell manipulation pertaining to the physical, optical, and electrical aspects for cellular analysis will be introduced. The focus will also be on the applications of these technologies to analyze cells at the cellular and molecular levels.

II. Techniques for Single-Cell Manipulation A. Microfluidics Microfluidics involves the miniaturization of systems for the handling and manipulation of small quantities of fluids. With microchannels in the dimensions of a few to hundreds of micrometers, it is well suited for single-cell handling which is of comparable dimensions. The flow characteristics of such systems will further aid in the precise control of single cells. The small size in microfluidic devices ensures laminar flow characteristics (low Reynolds number) which make the fluid flow predictable and controllable. A vast number of applications are being developed based on this technology which includes analytical systems in biochemistry, biomedical devices for disease detection, and tools used in systems biology (Martini et al., 2007; Ohno et al., 2008; Smith and Figeys, 2006; Whitesides et al., 2001). Besides being suitable to handle single-cell analyses, other motivations for using microfluidic platforms are abundant. It provides a quick means to test out designs due to the fast turnover time and offers the ability to integrate several devices with various functions to form a complete integrated laboratory on chip (Melin and Quake, 2007). Furthermore, the miniaturized platform allows for minimal use of expensive reagents compared to similar conventional biological benchtop methods, thereby saving cost.

82

Swee Jin Tan et al.

The technology fundamentally introduces myriad possibilities to enhance and bring about new capabilities in a variety of analyses (Sorger, 2008). There are numerous methods for fabricating microfluidic devices which also involve choices of various materials, such as glass, silicon, and polymers. For exam ple, the use of a polymer, polydimethlysiloxane (PDMS) (Sylgard 184, Dow Corning Corp., Midland, MI, USA), in the manufacturing of microfluidic devices have been well accepted for its ease to work with and being inexpensive. The material is optically transparent after curing, which is advantageous for various detection schemes such as fluorescence microscopy. Being biocompatible also allows for applications in cell assays to probe and manipulate cells. Fabrication of microdevices using this polymer is done using soft lithography and procedural details are well reported (Duffy et al., 1998; McDonald and Whitesides, 2002; McDonald et al., 2000; Voldman et al., 1999; Xia and Whitesides, 1998). The uses of microfluidic devices in the manipulation and study of single cells have been successfully applied in various areas. Skelley et al. (2009) made use of

(a)

(c)

(b) t = 0.00s

t = 0.178s

(d) t = 0.356s

t = 0.535s

(B) Back-side capture cup

d

Embryo insertion inlet

Front-side capture cup

Embryos

Support pillar Control channels

Inlet valve

Culture chamber 10 kV

× 450

50 µm 0000

33 37

s

(C) Outlet valve

MCF-7 breast

cancer cells

(A)

Fig. 1

(D)

(E)

Applications for single-cell manipulation using microfluidics. (A) Microfluidic control for cell pairing using weir-based structures (Skelley et al., 2009). (B) Deformation study of benign (MCF-10A) and malignant (MCF-7) breast epithelial cells (Hou et al., 2009) through a microchannel. (C) CTC isolation using affinity-based separation from peripheral blood via anti-EpCAM-coated micropillars (Nagrath et al., 2007). (D) CTC isolation using size and deformability of cancer cells via semicircular microtraps (Tan et al., 2009.) (E) Microembryo culture platform (Melin et al., 2009) with mechanical membrane valves

4. Manipulation and Isolation of Single Cells and Nuclei

83

microfluidic control to pair two different single cells in a physical trap to achieve highthroughput cell fusions as shown in Fig. 1A. Fusion of cells allows the combination of genetic materials and is the key to generate hybridomas and somatic cells reprogram ming (Miller and Ruddle, 1976; Tada et al., 1997). Using a passive weir-based cell trap, thousands of single cells can be immobilized onto the structures and pairing of cells can be achieved with good accuracy. Cell fusion is then effected with either chemical or electrical means. Efficient paring of up to 70% is attained with the densely packed array of traps and successful fused cells of greater than 50% are obtained. Comparing with conventional methods, the yields from employing fluidic control to manipulate cells are significantly higher. Hence, with careful control of flow para meters in microfluidic devices, the handling of single cells with high precision and high throughput is possible. Microfluidic devices of dimensions similar to biological cell sizes offer an excellent in vitro model to mimic capillary-like microenvironment. Controlling the movement of single cells within such devices can provide a means to quantify the physical para meters of cells. Hou et al. (2009) used a constricted microdevice which allows only single-cell passage at any instance to measure the deformability of benign and meta static breast epithelial cells. Due to the laminar flow profile and size (10 µm � 10 µm cross section) of the microchannel, single cells aligned one after another toward the constricted channel (Fig. 1B). By keeping the pressure settings constant across the microdevice, single cancer cell deformability measurements can be performed in a comparable physiological setting. The platform permits rapid and high-throughput processing of single cancer cells and the study concluded that benign breast epithelial cells are found to be stiffer than the metastatic breast cancer cells, by quantitatively analyzing each single cell entering and traversing through the microchannel. This may be correlated to intracellular changes during disease transformation that manifest into structural and functional irregularities in cells (Discher et al., 2009). Similar studies through manipulating and analyzing single blood cells in microchannels are also reported in understanding the disease malaria (Shelby et al., 2003). Thus, by directing single cells through microchannels that mimics the physiological state, the mechanical characteristics of cells can be obtained. Microfluidic devices can also be especially useful in investigating rare cell events like the detection of circulating tumor cells (CTCs) in peripheral blood. CTC numbers can reach as low as 1 tumor cell per milliliter of blood (Pantel et al., 1999), which contains approximately 5–6 billion blood cells. CTCs have been proven to be directly correlated to disease development and progression, as well as provide a measure of cancer treatment efficacy (Cristofanilli et al., 2004; Pantel and Alix-Panabieres, 2007). Thus, platforms with the capability to handle single cells are required to isolate them. Nagrath et al. (2007) showed the ability to retain cancer cells in peripheral blood using affinity-based interactions of cancer cells to anti-Epithelial Cell Adhesion Mole cule (anti-EpCAM). As demonstrated in Fig. 1C, isolation of CTCs is achieved when each of these cells bind to the anti-EpCAM coated microposts. Through careful design of the microposts placements and controlled flow conditions, it maximizes the prob ability for CTCs to attach to the structures and yet limits the shear stresses acting on

84

Swee Jin Tan et al.

these cells to maintain its viability. This passive control of small numbers of cells for CTC detection has been shown to be successful in patients with different cancer types (Maheswaran et al., 2008; Nagrath et al., 2007). An alternative to detecting CTCs is to use physical property differences of cancer cells from blood cells (Tan et al., 2009). Tan et al. provided a platform that temporarily immobilizes single CTC in individual traps based on the assumption that cancer cells are generally larger and stiffer than blood cells. The uniform array as shown in Fig. 1D facilitates cell enumeration while gaps in each of the traps allow blood components to pass through with ease. Only CTCs remain and these cells can be effectively recovered by changing the flow conditions within the microdevice so that downstream analyses such as molecular detection and cytogenetic analysis can be carried out. The use of active control of fluid flow in microfluidic systems with inbuilt valves also aid in the manipulation of single cells, which is made possible with microfabrication technologies (Thorsen et al., 2002). This is shown to be effective in a number of applications. For instance, the technology has been applied to study the external influence on blastocyst development of mouse embryos (Melin et al., 2009) and has applications in clinical infertility treatment by in vitro fertilization (IVF). The challenges of IVF are that in vitro human embryo cultures have a relatively high attrition rate, attributed by various genetic and environmental factors (Behr, 1999; Jun et al., 2008). In order to improve infertility treatment, this needs to be better understood. Melin et al. created microliter chambers to hold embryo culture (Fig. 1E) which is effectively controlled with micromechanical membrane valves. The membrane valves made from thin layers of PDMS are activated by exerting positive pressures of 9 psi in the control lines that closes the main channels as shown in Fig. 1E, effectively creating a closed environment. With the setup, it allows the movement and placement of single cells into individual culture chambers as well as concurrent parallel analyses. This tool allows the study of embryo development under closely resembled in vivo condition. B. Dielectrophoresis The controlled motion of particles or cells in the presence of a nonuniform electric field is attributed by dielectrophoretic forces, which was first described by Pohl (1951). The magnitude of forces that are exerted is a function of numerous parameters such as the dielectric properties of the particle and surrounding medium; electric field fre quency; and the dimensions of the particles. Detailed analyses of the relationships are established in various reports (Gascoyne and Vykoukal, 2002; Pohl, 1978) to calculate the forces acting in the path of the nonuniform electric field and the technique is potentially useful in single-cell manipulation for its precision. The experimental setup has also been improved from using tungsten filaments in the early 1950s (Pohl, 1951) to inbuilt microelectrodes on substrates using photolithography and microfabrication (Pethig and Markx, 1997). These advancements have scaled down the platforms for dielectrophoresis manipulations to dimensions suitable for single-cell analysis. The advantages of dielectrophoresis are the label-free and contactless approaches in single-cell handling. Michael and Hywel (1998) used a stable negative

85

4. Manipulation and Isolation of Single Cells and Nuclei

dielectrophoretic trap to hold submicron biological species against the actions of Brownian motion in between the electrodes. The electrode configuration and confine ment of a fluorescent particle are shown in Fig. 2A. Such techniques are useful to restrict the movement of cells. This is particularly important in monitoring the drug response of single cells and cell separation. Furthermore, single cells can be isolated without chemical or physical means which otherwise may damage the cells. The system also allows multiplexing for high throughput. Voldman et al. (2002) demon strated the ability to separate and confine cells in a microfluidic flow chamber using a (D)

(A)

a ITO glass

b

Projected image

b a.c.

a 0

20 µm 0

(B)

Interconnects

10 µm

Gold cylindrical electrodes

Glass substrate Nitride Undoped a-Si:H

Light-emitting diode 10x

n+ a-Si:H ITO

DMD microdisplay

(C)

100 µm

Fig. 2 Manipulation of single cells using dielectrophoretic forces. (A) A quadrupole electrode configuration for a single negative dielectrophoretic trap (Michael and Hywel, 1998). (B) Scanning Electron Microscope (SEM) image of an electrode array consisting of trapezoidal arranged gold electrodes (Voldman et al., 2002). (C) Isolation of single particles (left) and HeLa cells (right) against a continuous flow (Thomas et al., 2009). (D) Schematic for creating photo definable electrode designs for dielectrophoresis. An array of 15,000 particle traps for parallel handling of single particles across a 1.3 mm � 1.0 mm area. (Chiou et al., 2005)

86

Swee Jin Tan et al.

series of electrodes in similar quadrupole configurations as shown in Fig. 2B. The system is also flexible to allow the controlled release or trapping of HL-60 cells in the electrode arrays. Using a different electrode design, Thomas et al. (2009) produced an array of cell traps that are easily scalable to trap large number of single cells. The ring electrode design also enables lesser electrical connections comparing with traps using quadru pole formation. The platform was successfully tested on HeLa cells against a moving fluid. The electrode fabrication and design is crucial to the manipulation of single cells using dielectrophoresis. Chiou et al. (2005) demonstrated that using optical images projected onto a photoconductive surface, the desired electrode configuration can be obtained as shown in Fig. 2D. The technique offers flexibility to alter the designs for different purposes in single-cell manipulation and allow an easily scaled up platform for high throughput analyses. Sorting live and dead cells as well as a 15,000 cell trap for parallel manipulation of single particles have been demonstrated successfully using this system.

C. Optical-Based Techniques The optical-based techniques for the control, isolation, and handling of single cells involve the use of lasers (Pantel et al., 1999; Papadaki and Eskin, 1997; Papadimitriou et al., 1999; Parnaik and Manju, 2006; Pethig and Markx, 1997; Pohl, 1951, 1978). One such technique is the optical stretcher. Here, two identical divergent laser beams which produce opposing forces allow cells that enter the path to be trapped along the laser axis as shown in Fig. 3A. Forces are induced at the surface of the trapped cell due

(B) Single-mode fiber

Single-mode

fiber

0.1 - 0.3 mm 6 - 30 µm

Divergent laser beam (1064 nm, up to 2 W)

Trapped cell

15 mW

30 mW

50 mW

80 mW

110 mW

σ0 = 0.19 Nm–2

σ0 = 0.38 Nm–2

σ0 = 0.64 Nm–2

σ0 = 1.02 Nm–2

σ0 = 1.47 Nm–2

ch

Derfomed

Cells

ss

St re

Stre

ed

Ce l

l

ce ll

Deformed cell

e ofil Pr

Cells

cell fiber l

fiber

I

Fibers

(A)

flow direction

Ce l

I

(C)

Fig. 3 Optical stretcher for single-cell manipulation. (A) Overview of optical trapping process (Lincoln et al., 2004). (B) Optical stretching of human red blood cells with increasing laser intensities (Guck et al., 2001). (C) Schematic presentation of the cell stretcher and cells being deformed in an integrated microfluidic device to stretch breast epithelial cells (Lincoln et al., 2004).

4. Manipulation and Isolation of Single Cells and Nuclei

87

to momentum transfer and larger laser intensities will result in higher net forces acting on the cells. As a result, a single cell under continuous flow is confined within the opposing laser beams. The laser-induced forces deformed the trapped single cell which can then be used to determine its mechanical properties (Guck et al., 2001, 2005; Lincoln et al., 2004, 2007; Wottawah et al., 2005). Small perturbations from these forces can be exerted onto the surface of a single cell, translating into physical deformation which can be optically measured (Guck et al., 2005). Guck et al. (2001) showed that using the optical stretcher setup, precise measurements about the stress profiles and forces acting on human red blood cells can be determined. Figure 3B depicts the deformation of red blood cells using different laser intensities, highlighting that radiation damage is minimized and no external bead attachment is required to exert forces to stretch the cells. In order to manipulate and measure cells at a high throughput using the optical stretcher, Lincoln et al. (2004) incorporated the setup with a simple microfluidic flow chamber. The advantage of a polymeric-based microfluidic device is that it allows various systems to be easily integrated at the microscale and for the optical fibers to be inserted. The system provides a means to probe the elasticity of cells at a high throughput. The deformability measurements of benign and malignant breast epithelial cells (Lincoln et al., 2004) are collated and a typical experimental setup is shown in Fig. 3C. The changes in deformability of cancer cells can be traced back to the altered cytoskeleton which are determined and studied during disease transformation. The differences in deformability can be the cutoff criteria to separate benign and malignant cells. Käs et al. (2005) fabricated a microdevice that permits single cells to be probed for the ability to deform and sorted after analysis by using the optical forces generated from the divergent laser beams. The entire system consists of a cell centering setup that focuses and aligns single cells to the center of the flow in the 80- to 100-µm-wide microchannel, an optical stretcher that performs the deformability tests and another single divergent laser beam at the downstream of the microchannel that pushes the selected cells into a side channel. Single cells are first probed individually by optical deformation and later sorted according to the measured elasticity by directing the cell to an alternate microchannel using optical forces. This study which examined normal as well as cancerous fibroblast and breast cells, shows the distinctive optical deforma tion variance in these different cell types. Without the need for any molecular tag, the methodology is attractive for probing and manipulating single cells for cancer cell recognition and enrichment. D. Micropipette Aspiration Micropipette aspiration is a widely used experimental method to manipulate and study the mechanical properties of individual cells (Hochmuth, 2000). The application of the physical technique is straightforward and involves a microsized glass capillary which is used to apply suction to the whole cell under specific pressure conditions. The aspirated length into the capillary is a function of the deformability of the cell and can be quantified using various theoretical or computational models (Lim et al., 2006). Many

88

Swee Jin Tan et al.

studies have been carried out using the micropipette aspiration technique to investigate the overall mechanical properties of different types of cells such as erythrocytes (Jiao et al., 2009), leukocytes, chondrocytes, endothelial cells (Chien et al., 1978; Jones et al., 1999; Sato et al., 1987; Schmid-Schonbein et al., 1981), and of isolated cell nuclei (Dahl et al., 2004; Rowat et al., 2006). For single-cell handling, the micropipette aspiration technique provides a direct means to manipulate and extract mechanical properties from single cells. In terms of clinical relevance, this is useful to bridge the links of possible structural–property–function relationship which will aid in the understanding of diseases at the cellular level. Jiao et al. developed computational models comparing with experimental data (Fig. 4A ) depicting the micropipette aspiration of malaria-infected red blood cells (Jiao et al., 2009). The stiffness and adhesion properties of infected red blood cells increase with the progression of the disease stage and micropipette aspiration is suitable to quantify the changes with a wide range of applied force. In cancer studies, the viscoelastic properties and elasticity changes of the whole cell can be attributed to the disease transformation.

Increasing suction pressure (A)

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(B)

Fig. 4 Micropipette aspiration for the manipulation and mechanical characterization of single cells. (A) Probing the mechanical properties of a malaria infected red blood cell in the mid-Trophozoite stage (Jiao et al., 2009). (B) A dual pipette setup to measure separation forces in cell–cell-mediated adhesion (a-d) Two cells are caught by two pipettes and brought in contact with each other to form adhesion; (e-h) One pipette tries to separate the two cells using a gradually increased force with the other pipette holding the two cells at one side.(Vedula et al., 2009).

4. Manipulation and Isolation of Single Cells and Nuclei

89

Wu et al. (Wu et al., 2000; Zhang et al., 2002) measured the mechanical properties of hepatocellular carcinoma cells, showing significant differences in the elastic coefficients comparing with hepatocytes. In addition, further characterizations were performed on both the normal and cancerous cells under the influence of various cytoskeletal disrupt ing agents. The variations seen in the mechanical properties as well as the varying effects of the disrupting agents are believed to be a direct reflection of the phenotypes exhibited by the different cell types. It is further hypothesized that these property changes may affect tumor cell invasion and metastasis. There is also significant interest in using micropipette aspiration for the manipula tion of single cells to measure forces involved in cell–cell adhesion via a dual micropipette setup. The technique allows flexibility to pick up cells and bring them into contact so that cellular adhesions are formed. With one pipette holding two adhering cells at one side, the other pipette is brought in contact with the other end of the cells and tries to separate them using a gradually increased pressure setting. The amount of pressure needed to separate these two cells is then used to calculate the deadhesion force. Vedula et al. (2009) quantified the separation forces required by tight junction proteins using the technique as shown in Fig. 4B. The method is versatile and allows further parametric studies of cell–cell adhesions in the presence of chemicals, disrupting agents and drugs.

III. Nuclear Isolation and Manipulation The cell nucleus is separated from the cytoplasm by a nuclear envelope which consists of an inner nuclear membrane (INM), outer nuclear membrane, an extension of rough endoplasmic reticulum (ER), and nuclear lamina (Prokocimer et al., 2006). The nuclear lamina, which is the major structural component of nuclear envelope, is a dense network of lamins plus lamin-associated proteins lying beneath the INM. Lamins are part of the intermediate filament (IF) gene family and are thought to be the evolutionary progenitors of IF proteins. The IF gene super family comprises five groups with approximately 60 members. Group I–IV are cytoplasmic IF, and lamins belongs to the group V IF family. Lamins are classified into types A and B according to their difference in biochemical properties, expression pattern, and beha vior during mitosis (Stuurman et al., 1998). A-type lamins, which include lamin A and C, are products of alternative splicing from the LMNA gene, and B-type lamins are encoded by two separate genes, LMNB1 and LMNB2. Type B lamins are present in all mammalian cells as they are essential for cell viability, but type A lamins are developmentally regulated. Type A lamins are absent in human embryonic stem cells, but are expressed only after cells differentiate and generally increased during terminal differentiation and growth arrest (Prokocimer et al., 2006). Lamins are very important for their contribution to the nuclear structure. They determine the nuclear integrity and are also involved in numerous nuclear functions. Specifically, A-type lamins play a major role in the preservation of the nuclear shape (Dahl et al., 2006; Lammerding et al., 2006; Scaffidi and Misteli, 2006), stability

90

Swee Jin Tan et al.

(Broers et al., 2005; Dahl et al., 2004; Lammerding et al., 2004), and structural integrity (Broers et al., 2005; Lammerding et al., 2006; Stewart et al., 2007). Moreover, lamins regulate and support protein complexes involved in gene expression, nuclear position ing (Meaburn and Misteli, 2007), DNA replication, transcription, and repair (Parnaik and Manju, 2006), and aging (Scaffidi and Misteli, 2006). In order to examine the molecular aspects in nuclear studies, the extraction methods of the nuclei from indivi dual cells are important, especially to prevent damage to miniscule nuclear structures. A. Nuclear Isolation Method There are several methods to isolate the nuclei from cells, which include mechanical, chemical, or a combination of both techniques. However, most nuclear isolation methods apply only to mechanically stable interphase nuclei.

1. Mechanical Isolation Method Nuclei of different cell types can be isolated by single-cell mechanical extraction method (Guilak et al., 2000) and bulk method (Dahl et al., 2005). Micropipette with a smaller diameter compared to the cell is used to repeatedly aspirate the cell to break the cell membrane (Fig. 5A and B). The nucleus is then collected using the same micropipette to separate it from the cellular debris (Fig. 5C and D). Guilak et al. (2000) isolated cell nuclei mechanically and the isolated nuclei are further used in the micropipette aspiration tests to characterize their viscoelastic properties. Alternatively, for bulk extraction, a Dounce homogenizer is used to stroke and break the suspended cells, and the cell lysate is then centrifuged to separate the isolated nuclei from the cellular debris. The collected nuclei are used later in the micropipette aspiration and Atomic Force Microscopy (AFM) indentation tests (Dahl et al., 2005).

(A)

(B)

(C)

(D)

Fig. 5 Mechanical method for cell nucleus isolation using micropipette aspiration. Arrowheads indicate the cell nucleus. (A–D) Step-by-step processes for isolating the cell nucleus (Guilak et al., 2000).

4. Manipulation and Isolation of Single Cells and Nuclei

91

2. Chemical Method Nucleus can also be isolated using chemical techniques. Confluent cells were treated with low-ionic-strength extraction solution and then detergent (NP-40). The top layer of cells would break away and the nuclei would pop out by shaking the dish (Deguchi et al., 2005). Other non-ionic detergent (Igepal CA-630) can also be used for nuclear isolation (Caille et al., 2002; Thoumine et al., 1999). The isolated nucleus was further used in the mechanical tests.

3. Mixed Method Chemical and mechanical methods can also be combined to isolate nuclei. Cells were first treated with Triton X-100 and then they were sheared by passing through a syringe needle. The cell nucleus were collected after centrifuging the cell lysate and used for chromatin organization (Mazumder and Shivashankar, 2007). B. Application of Nucleus Study As the importance of nuclear structure and its function in cells become apparent, especially in understanding mechanotransduction and gene expression, an increasing number of studies have been carried out to investigate the structure and mechanical properties of the nucleus. This is especially important in cancer research (Bernhard and Granboulan, 1963; Koller, 1963), given defects at the genetic level is solely responsible for the cell aberration. Breast cancer cell (MCF-7) nucleus has been shown to be twice softer than nonmalignant breast cells (MCF-10A) (Li, 2009). Micropipette aspiration technique is thus suitable to probe the mechanical characteristics of the cell nucleus which can be easily isolated (Dounce, 1963). For example, Guilak et al. (Guilak, 2000; Guilak et al., 2000) studied the viscoelastic properties of mechanically and chemically isolated nuclei of articular chondrocytes. It is observed that isolated nucleus has fluid-like characteristics similar to the cytoplasm and can be treated as a viscoelastic solid material. However, the nucleus is 3–4 times stiffer than and twice as viscous as the cytoplasm. These observations suggest that the nucleus is the main contributor to the heterogeneity of the apparent mechanical properties of the whole cell. In addition, more information on nuclear mechanics can be obtained with the technique. This will aid in understanding the force-induced changes in gene expres sion, and the subsequent remodeling of the nuclear architecture in the context of cell and disease development. Dahl et al. used micropipette aspiration combined with immunofluorescence to study the structure of protein–lamin B in an isolated nucleus under different swelling conditions, which are shown to contribute to the viscoelastic properties. Their investigations showed that chromatin is a primary force-bearing element in unswollen nuclei, whereas lamina sustained much of the load for the swollen nuclei (Dahl et al., 2005). Mutations on the gene encoding A type lamins, LMNA, can cause a premature aging disease known as Hutchinson–Gilford progeria syndrome. The results thus demonstrated that reduced deformability of the nucleus can cause misregulation of mechanosensitive gene expression (Dahl et al., 2006).

92

Swee Jin Tan et al.

With nanometer-scale resolution capability, the AFM is also suited to measure various physical characteristics of intracellular organelles such as the cell nucleus. Vaziri et al. (2006) presented a systematic analysis of AFM indentation of isolated nuclei from mouse embryo fibroblast cells. Together with various computational models, the study examined the role of major nuclear elements.

IV. Discussion and Future Implications It is evident that cells are dissimilar entities, showing disparities in both the genetic and phenotypic expressions. Even for cells of the same origin, there is diversity within the population as well as variability from cell to cell. Therefore, single-cell analyses are likely to provide greater insights to understand this biological diversity. Conventional benchtop assays have its limitations to manipulate single cells and rapid advances in technological development are offering new and novel methodologies. These technol ogies present numerous innovative possibilities to control and probe various targets at the cellular and nuclear levels. The experimental study on single cells and single nuclei has its advantages and will likely have an impact on sample preparation, disease diagnosis, and addressing fundamental cell biology questions. These new technologies will continue to have an important role to aid in the understanding of the pathophy siological and molecular aspects of various diseases. The rapid responses and high sensitivities of the platforms are ideal to complement current benchtop methodologies to manipulate single cells and nuclei for measurements and other downstream applica tions. Ultimately, the development of these novel techniques will enhance the under standing of diseases and allow more precision studies of cellular and subcellular events, with the aim of producing new treatments and novel medical diagnostics that will benefit patients suffering from various diseases. References Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K., and Watson, J. D. (2002). In “Molecular Biology of the Cell” (S. Gibbs, ed.). Garland Publishers, New York. Bashir, R. (2004). BioMEMS: State-of-the-art in detection, opportunities and prospects. Adv. Drug Delivery Rev. 56, 1565–1586. Behr, B. (1999). Blastocyst culture and transfer. Hum. Reprod. 14, 5–6. Bernhard, W., and Granboulan, N. (1963). The fine structure of the cancer cell nucleus. Exp. Cell Res. 24(9), 19–53. Broers, J. L., Kuijpers, H. J., Ostlund, C., Worman, H. J., Endert, J., and Ramaekers, F. C. (2005). Both lamin A and lamin C mutations cause lamina instability as well as loss of internal nuclear lamin organization. Exp. Cell Res. 304, 582–592. Caille, N., Thoumine, O., Tardy, Y., and Meister, J. J. (2002). Contribution of the nucleus to the mechanical properties of endothelial cells. J. Biomech. 35, 177–187. Chang, S. F., Chang, C. A., Lee, D. Y., Lee, P. L., Yeh, Y. M., Yeh, C. R., Cheng, C. K., Chien, S., and Chiu, J. J. (2008). Tumor cell cycle arrest induced by shear stress: Roles of integrins and Smad. Proc. Natl. Acad. Sci. USA 105, 3927–3932.

4. Manipulation and Isolation of Single Cells and Nuclei

93

Chien, S., Sung, K. L., Skalak, R., Usami, S., and Tozeren, A. (1978). Theoretical and experimental studies on viscoelastic properties of erythrocyte membrane. Biophys. J. 24, 463–487. Chiou, P. Y., Ohta, A. T., and Wu, M. C. (2005). Massively parallel manipulation of single cells and microparticles using optical images. Nature 436, 370–372. Cristofanilli, M., Budd, G. T., Ellis, M. J., Stopeck, A., Matera, J., Miller, M. C., Reuben, J. M., Doyle, G.V., Allard, W. J., Terstappen, L. W., and Hayes, D. F. (2004). Circulating tumor cells, disease progression, and survival in metastatic breast cancer. N. Engl. J. Med. 351, 781–791. Dahl, K. N., Kahn, S. M., Wilson, K. L., and Discher, D. E. (2004). The nuclear envelope lamina network has elasticity and a compressibility limit suggestive of a molecular shock absorber. J. Cell Sci. 117, 4779–4786. Dahl, K. N., Engler, A. J., Pajerowski, J. D., and Discher, D. E. (2005). Power-law rheology of isolated nuclei with deformation mapping of nuclear substructures. Biophys. J. 89, 2855–2864. Dahl, K. N., Scaffidi, P., Islam, M. F., Yodh, A. G., Wilson, K. L., and Misteli, T. (2006). Distinct structural and mechanical properties of the nuclear lamina in Hutchinson-Gilford progeria syndrome. Proc. Natl. Acad. Sci. USA 103, 10271–10276. Deguchi, S., Maeda, K., Ohashi, T., and Sato, M. (2005). Flow-induced hardening of endothelial nucleus as an intracellular stress-bearing organelle. J. Biomech. 38, 1751–1759. Ding, L., Sunamura, M., Kodama, T., Yamauchi, J., Duda, D. G., Shimamura, H., Shibuya, K., Takeda, K., and Matsuno, S. (2001). In vivo evaluation of the early events associated with liver metastasis of circulating cancer cells. Br. J. Cancer. 85, 431–438. Discher, D., Dong, C., Fredberg, J. J., Guilak, F., Ingber, D., Janmey, P., Kamm, R. D., Schmid-Schonbein, G. W., and Weinbaum, S. (2009). Biomechanics: Cell research and applications for the next decade. Ann. Biomed. Eng. 37, 847–859. Dounce, A. L. (1963). The isolation of nuclei from tumor cells. Exp. Cell Res. 24(9), 126–143. Duffy, D. C., McDonald, J. C., Schueller, O. J. A., and Whitesides, G. M. (1998). Rapid prototyping of microfluidic systems in Poly(dimethylsiloxane). Anal. Chem. 70, 4974–4984. Gascoyne, P. R., and Vykoukal, J. (2002). Particle separation by dielectrophoresis. Electrophoresis 23, 1973–1983. Guck, J., Ananthakrishnan, R., Mahmood, H., Moon, T. J., Cunningham, C. C., and Kas, J. (2001). The optical stretcher: A novel laser tool to micromanipulate cells. Biophys. J. 81, 767–784. Guck, J., Schinkinger, S., Lincoln, B., Wottawah, F., Ebert, S., Romeyke, M., Lenz, D., Erickson, H. M., Ananthakrishnan, R., Mitchell, D., Kas, J., Ulvick, S., et al. (2005). Optical deformability as an inherent cell marker for testing malignant transformation and metastatic competence. Biophys. J. 88, 3689–3698. Guilak, F. (2000). The deformation behavior and viscoelastic properties of chondrocytes in articular cartilage. Biorheology 37, 27–44. Guilak, F., Tedrow, J. R., and Burgkart, R. (2000). Viscoelastic properties of the cell nucleus. Biochem. Biophys. Res. Commun. 269, 781–786. Heath, J. R., and Davis, M. E. (2008). Nanotechnology and cancer. Annu. Rev. Med. 59, 251–265. Heyder, C., Gloria-Maercker, E., Hatzmann, W., Zaenker, K. S., and Dittmar, T. (2006). Visualization of tumor cell extravasation. Contrib. Microbiol. 13, 200–208. Hochmuth, R. M. (2000). Micropipette aspiration of living cells. J. Biomech. 33, 15–22. Hou, H. W., Li, Q. S., Lee, G. Y., Kumar, A. P., Ong, C. N., and Lim, C. T. (2009). Deformability study of breast cancer cells using microfluidics. Biomed. Microdevices 11, 557–564. Jiao, G. Y., Tan, K. S.W., Sow, C. H., Dao, M., Suresh, S., and Lim, C. T. (2009). Computational modeling of the micropipette aspiration of malaria infected erythrocytes. In “13th International Conference on Biome dical Engineering” (C. T. Lim, and J. C. H. Goh, eds.) pp. 1788–1791. Springer, Singapore. Jones, W. R., Ting-Beall, H. P., Lee, G. M., Kelley, S. S., Hochmuth, R. M., and Guilak, F. (1999). Alterations in the Young’s modulus and volumetric properties of chondrocytes isolated from normal and osteoarthritic human cartilage. J. Biomech. 32, 119–127. Jun, S. H., Choi, B., Shahine, L., Westphal, L. M., Behr, B., Reijo Pera, R. A., Wong, W. H., and Yao, M. W. (2008). Defining human embryo phenotypes by cohort-specific prognostic factors. PLoS ONE 3, e2562.

94

Swee Jin Tan et al. Käs, J. A., Martin, M., Mueller, K., Wottawah, F., Schinkinger, S., Lincoln, B., and Romeyke, M. (2005). Feeling with light for cancer. In “Advanced Biomedical and Clinical Diagnostic Systems IV” (G. E. Cohn, W. S. Grundfest, D. A. Benaron, and T. Vo-Dinh, eds.) SPIE, San Jose, CA, USA. Koller, P. C. (1963). The nucleus of the cancer cell. A historical review. Exp. Cell Res. 24(Suppl. 9), 3–14. Lammerding, J., Schulze, P. C., Takahashi, T., Kozlov, S., Sullivan, T., Kamm, R. D., Stewart, C. L., and Lee, R. T. (2004). Lamin A/C deficiency causes defective nuclear mechanics and mechanotransduction. J. Clin. Investig. 113, 370–378. Lammerding, J., Fong, L. G., Ji, J. Y., Reue, K., Stewart, C. L., Young, S. G., and Lee, R. T. (2006). Lamins A and C but not lamin B1 regulate nuclear mechanics. J. Biol. Chem. 281, 25768–25780. Lekka, M., and Laidler, P. (2009). Applicability of AFM in cancer detection. Nat. Nanotechnol. 4, 72; author reply 72–73. Li, Q. S. (2009). Understanding structure-mechanical property relationship of breast cancer cells PhD, Department of Mechanical Engineering, National University of Singapore, Singapore. Lim, C. T., Zhou, E. H., Li, A., Vedula, S. R.K., and Fu, H. X. (2006). Experimental techniques for single cell and single molecule biomechanics. Mater. Sci. Eng. C 26, 1278–1288. Lincoln, B., Erickson, H. M., Schinkinger, S., Wottawah, F., Mitchell, D., Ulvick, S., Bilby, C., and Guck, J. (2004). Deformability-based flow cytometry. Cytometry A 59, 203–209. Lincoln, B., Schinkinger, S., Travis, K., Wottawah, F., Ebert, S., Sauer, F., and Guck, J. (2007). Reconfigur able microfluidic integration of a dual-beam laser trap with biomedical applications. Biomed. Microdevices 9, 703–710. Maheswaran, S., Sequist, L. V., Nagrath, S., Ulkus, L., Brannigan, B., Collura, C. V., Inserra, E., Diederichs, S., Iafrate, A. J., Bell, D. W., Digumarthy, S., Muzikansky, A., et al. (2008). Detection of mutations in EGFR in circulating lung-cancer cells. N. Engl. J. Med. 359, 366–377. Martini, J., Hellmich, W., Greif, D., Becker, A., Merkle, T., Ros, R., Ros, A., Toensing, K., and Anselmetti, D. (2007). Systems nanobiology: From quantitative single molecule biophysics to microfluidic-based single cell analysis. Subcell Biochem. 43, 301–321. Mazumder, A., and Shivashankar, G. V. (2007). Gold-nanoparticle-assisted laser perturbation of chromatin assembly reveals unusual aspects of nuclear architecture within living cells. Biophys. J. 93, 2209–2216. McDonald, J. C., and Whitesides, G. M. (2002). Poly(dimethylsiloxane) as a material for fabricating microfluidic devices. Acc. Chem. Res. 35, 491–499. McDonald, J. C., Duffy, D. C., Anderson, J. R., Chiu, D. T., Wu, H., Schueller, O. J., and Whitesides, G. M. (2000). Fabrication of microfluidic systems in poly(dimethylsiloxane). Electrophoresis 21, 27–40. McNeish, J. (2004). Embryonic stem cells in drug discovery. Nat. Rev. Drug Discov. 3, 70–80. Meaburn, K. J., and Misteli, T. (2007). Cell biology: Chromosome territories. Nature 445, 379–381. Melin, J., and Quake, S. R. (2007). Microfluidic large-scale integration: The evolution of design rules for biological automation. Annu. Rev. Biophys. Biomol. Struct. 36, 213–231. Melin, J., Lee, A., Foygel, K., Leong, D. E., Quake, S. R., and Yao, M. W. (2009). In vitro embryo culture in defined, sub-microliter volumes. Dev. Dyn. 238, 950–955. Michael, P. H., and Hywel, M. (1998). Dielectrophoretic trapping of single sub-micrometre scale bioparticles. J. Phys. D Appl. Phys. 31, 2205–2210. Miller, R. A., and Ruddle, F. H. (1976). Pluripotent teratocarcinoma-thymus somatic cell hybrids. Cell 9, 45–55. Nagrath, S., Sequist, L. V., Maheswaran, S., Bell, D. W., Irimia, D., Ulkus, L., Smith, M. R., Kwak, E. L., Digumarthy, S., Muzikansky, A., Ryan, P., Balis, U. J., et al. (2007). Isolation of rare circulating tumour cells in cancer patients by microchip technology. Nature 450, 1235–1239. Ohno, K., Tachikawa, K., and Manz, A. (2008). Microfluidics: Applications for analytical purposes in chemistry and biochemistry. Electrophoresis 29, 4443–4453. Pantel, K., and Alix-Panabieres, C. (2007). The clinical significance of circulating tumor cells. Nat. Clin. Pract. Oncol. 4, 62–63. Pantel, K., Cote, R. J., and Fodstad, O. (1999). Detection and clinical importance of micrometastatic disease. J. Natl. Cancer Inst. 91, 1113–1124. Pantel, K., Brakenhoff, R. H., and Brandt, B. (2008). Detection, clinical relevance and specific biological properties of disseminating tumour cells. Nat. Rev. Cancer. 8, 329–340.

4. Manipulation and Isolation of Single Cells and Nuclei

95

Papadaki, M., and Eskin, S. G. (1997). Effects of fluid shear stress on gene regulation of vascular cells. Biotechnol. Prog. 13, 209–221. Papadimitriou, M. N., Menter, D. G., Konstantopoulos, K., Nicolson, G. L., and McIntire, L. V. (1999). Integrin alpha4beta1/VCAM-1 pathway mediates primary adhesion of RAW117 lymphoma cells to hepatic sinusoidal endothelial cells under flow. Clin. Exp. Metastasis 17, 669–676. Parnaik, V. K., and Manju, K. (2006). Laminopathies: Multiple disorders arising from defects in nuclear architecture. J. Biosci. 31, 405–421. Pethig, R., and Markx, G. H. (1997). Applications of dielectrophoresis in biotechnology. Trends Biotechnol. 15, 426–432. Pohl, H. A. (1951). The motion and precipitation of suspensoids in divergent electric fields. J. Appl. Phys. 22, 869–871. Pohl, H. A. (1978). “Dielectrophoresis: The Behavior of Neutral Matter in Nonuniform Electric Field.” Cambridge University Press, Cambridge, New York. Prokocimer, M., Margalit, A., and Gruenbaum, Y. (2006). The nuclear lamina and its proposed roles in tumorigenesis: Projection on the hematologic malignancies and future targeted therapy. J. Struct. Biol. 155, 351–360. Raty, S., Walters, E. M., Davis, J., Zeringue, H., Beebe, D. J., Rodriguez-Zas, S. L., and Wheeler, M. B. (2004). Embryonic development in the mouse is enhanced via microchannel culture. Lab Chip 4, 186–190. Rowat, A. C., Lammerding, J., and Ipsen, J. H. (2006). Mechanical properties of the cell nucleus and the effect of emerin deficiency. Biophys. J. 91, 4649–4664. Sato, M., Levesque, M. J., and Nerem, R. M. (1987). An application of the micropipette technique to the measurement of the mechanical properties of cultured bovine aortic endothelial cells. J. Biomech. Eng. 109, 27–34. Scaffidi, P., and Misteli, T. (2006). Lamin A-dependent nuclear defects in human aging. Science 312, 1059–1063. Schmid-Schonbein, G. W., Sung, K. L., Tozeren, H., Skalak, R., and Chien, S. (1981). Passive mechanical properties of human leukocytes. Biophys. J. 36, 243–256. Shelby, J. P., White, J., Ganesan, K., Rathod, P. K., and Chiu, D. T. (2003). A microfluidic model for singlecell capillary obstruction by Plasmodium falciparum-infected erythrocytes. Proc. Natl. Acad. Sci. USA 100, 14618–14622. Skelley, A. M., Kirak, O., Suh, H., Jaenisch, R., and Voldman, J. (2009). Microfluidic control of cell pairing and fusion. Nat. Methods 6, 147–152. Smith, J. C., and Figeys, D. (2006). Proteomics technology in systems biology. Mol. Biosyst. 2, 364–370. Sorger, P. K. (2008). Microfluidics closes in on point-of-care assays. Nat. Biotechnol. 26, 1345–1346. Stewart, C. L., Roux, K. J., and Burke, B. (2007). Blurring the boundary: The nuclear envelope extends its reach. Science 318, 1408–1412. Stossel, T. P. (1993). On the crawling of animal cells. Science 260, 1086–1094. Stuurman, N., Heins, S., and Aebi, U. (1998). Nuclear lamins: Their structure, assembly, and interactions. J. Struct. Biol. 122, 42–66. Tada, M., Tada, T., Lefebvre, L., Barton, S. C., and Surani, M. A. (1997). Embryonic germ cells induce epigenetic reprogramming of somatic nucleus in hybrid cells. EMBO J. 16, 6510–6520. Tan, S. J., Yobas, L., Lee, G. Y., Ong, C. N., and Lim, C. T. (2009). Microdevice for the isolation and enumeration of cancer cells from blood. Biomed. Microdevices 11, 883–892. Thomas, R. S., Morgan, H., and Green, N. G. (2009). Negative DEP traps for single cell immobilisation. Lab Chip 9, 1534–1540. Thorsen, T., Maerkl, S. J., and Quake, S. R. (2002). Microfluidic large-scale integration. Science 298, 580–584. Thoumine, O., Ott, A., Cardoso, O., and Meister, J. J. (1999). Microplates: A new tool for manipulation and mechanical perturbation of individual cells. J. Biochem. Biophys. Methods 39, 47–62. Van Vliet, K. J., Bao, G., and Suresh, S. (2003). The biomechanics toolbox: Experimental approaches for living cells and biomolecules. Acta Mater. 51, 5881–5905. Vaziri, A., Lee, H., and Mofrad, M. R. K. (2006). Deformation of the cell nucleus under indentation: Mechanics and mechanisms. J. Mater. Res. 21, 2126–2135.

96

Swee Jin Tan et al. Vedula, S., Lim, T., Kausalya, P., Lane, E., Rajagopal, G., Hunziker, W., and Lim, C. (2009). Quantifying forces mediated by integral tight junction proteins in cell–cell adhesion. Exp. Mech. 49, 3–9. Voldman, J., Gray, M. L., and Schmidt, M. A. (1999). Microfabrication in biology and medicine. Annu. Rev. Biomed. Eng. 1, 401–425. Voldman, J., Gray, M. L., Toner, M., and Schmidt, M. A. (2002). A microfabrication-based dynamic array cytometer. Anal. Chem. 74, 3984–3990. Weiss, L. (1991). Deformation-driven, lethal damage to cancer cells. Its contribution to metastatic inefficiency. Cell Biophys. 18, 73–79. Weiss, L., and Dimitrov, D. S. (1986). Mechanical aspects of the lungs as cancer cell-killing organs during hematogenous metastasis. J. Theor. Biol. 121, 307–321. Weiss, L., Dimitrov, D. S., and Angelova, M. (1985). The hemodynamic destruction of intravascular cancer cells in relation to myocardial metastasis. Proc. Natl. Acad. Sci. USA 82, 5737–5741. Whitesides, G. M. (2006). The origins and the future of microfluidics. Nature 442, 368–373. Whitesides, G. M., Ostuni, E., Takayama, S., Jiang, X. Y., and Ingber, D. E. (2001). Soft lithography in biology and biochemistry. Annu. Rev. Biomed. Eng. 3, 335–373. Wottawah, F., Schinkinger, S., Lincoln, B., Ebert, S., Muller, K., Sauer, F., Travis, K., and Guck, J. (2005). Characterizing single suspended cells by optorheology. Acta Biomater. 1, 263–271. Wu, Z. Z., Zhang, G., Long, M., Wang, H. B., Song, G. B., and Cai, S. X. (2000). Comparison of the viscoelastic properties of normal hepatocytes and hepatocellular carcinoma cells under cytoskeletal perturbation. Biorheology 37, 279–290. Xia, Y., and Whitesides, G. M. (1998). Soft lithography. Annu. Rev. Mater. Sci. 28, 153–184. Zhang, G., Long, M., Wu, Z. Z., and Yu, W. Q. (2002). Mechanical properties of hepatocellular carcinoma cells. World J. Gastroenterol. 8, 243–246. Zhong, J. F., Feng, Y., and Taylor, C. R. (2008). Microfluidic devices for investigating stem cell gene regulation via single-cell analysis. Curr. Med. Chem. 15, 2897–2900.

CHAPTER 5

Beyond Lamins: Other Structural Components of the Nucleoskeleton Zhixia Zhong*, Katherine L. Wilson†, and Kris Noel Dahl*,‡ *

Department of Chemical Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213

†

Department of Cell Biology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205

‡

Department of Biomedical Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213

Abstract I. Introduction A. Actin in the Nucleus B. Nuclear Spectrin C. Nuclear Titin D. LINC Complexes at the Nuclear Envelope E. Roles for Nucleoskeletal Proteins in Mitosis and Nuclear Assembly II. Methods A. Localization of Endogenous Nucleoskeletal Proteins by Indirect Immunofluorescence Microscopy B. Endogenous Protein Localization at the INM Versus the ONM C. Immunoprecipitation of Nucleoskeletal Proteins D. Micropipette Aspiration and Recoil III. Discussion and Prospects

References

Abstract The nucleus is bordered by a double bilayer nuclear envelope, communicates with the cytoplasm via embedded nuclear pore complexes, and is structurally supported by an underlying nucleoskeleton. The nucleoskeleton includes nuclear intermediate fila ments formed by lamin proteins, which provide major structural and mechanical support to the nucleus. However, other structural proteins also contribute to the function of the METHODS IN CELL BIOLOGY, VOL. 98 Copyright 2010 Elsevier Inc. All rights reserved.

97

978-0-12-381009-0 DOI: 10.1016/S0091-679X(10)98005-9

98

Zhixia Zhong et al.

nucleoskeleton and help connect it to the cytoskeleton. This chapter reviews nucleo skeletal components beyond lamins and summarizes specific methods and strategies useful for analyzing nuclear structural proteins including actin, spectrin, titin, linker of nucleoskeleton and cytoskeleton (LINC) complex proteins, and nuclear spindle matrix proteins. These components can localize to highly specific functional subdomains at the nuclear envelope or nuclear interior and can interact either stably or dynamically with a variety of partners. These components confer upon the nucleoskeleton a functional diversity and mechanical resilience that appears to rival the cytoskeleton. To facilitate the exploration of this understudied area of biology, we summarize methods useful for localizing, solubilizing, and immunoprecipitating nuclear structural proteins, and a state-of-the-art method to measure a newly-recognized mechanical property of nucleus.

I. Introduction The nucleus houses the genome and is the largest organelle in eukaryotic cells. Its best-known architectural components include the nuclear envelope, nuclear pore complexes (NPCs), and the nucleoskeleton, which is formed primarily by separate networks of nuclear intermediate filaments formed by A- or B-type lamins. The nucleoskeleton is concentrated near the nuclear envelope (“peripheral” nucleoskeleton) but also extends throughout the interior (“internal” nucleoskeleton) with loosely distributed lamins and associated proteins. Chromosomes and chromatin also associate with lamins (Guelen et al., 2008; Wen et al., 2009), as do most characterized inner nuclear membrane (INM) proteins, suggesting a variety of structures contribute to nuclear architecture (Zastrow et al., 2004). Lamin networks resist deformation and force transmission and are major mechanical elements of the nucleus (Dahl et al., 2008). Nuclei reconstituted in lamin-deficient Xenopus egg extracts are extremely fragile (Newport et al., 1990). Similarly mamma lian cells depleted of lamins, particularly A-type lamins, are significantly weaker than their wildtype counterparts (Broers et al., 2004; Lammerding et al., 2004). Nuclear A- and B-type lamin networks also contribute, mechanically or nonmechanically, to many other functions including chromatin organization, transcription, replication, differentiation, and signaling (Dechat et al., 2008; Gruenbaum et al., 2005). Numerous diseases (“laminopathies”) are caused either by perturbed expression of B-type lamins or by mutations in LMNA (encoding A-type lamins) or other genes encoding nuclear envelope membrane proteins (Capell and Collins, 2006; Gruenbaum et al., 2005). In many cases, these mutations alter nuclear mechanics and clinically affect load-bearing tissues (Dahl et al., 2008). The spectrum of known laminopathies includes muscular dystrophy, lipodystrophy and diabetes, skeletal dysplasia, skin disorders, neuropathy, leukodystrophy, and progeria (premature aging) (Capell and Collins, 2006). It remains unclear how mutations in these proteins, particularly A-type lamins, can produce such widely different diseases. Current evidence points to multiple and varied mechanisms, including perturbed regulation of gene expression and altered

99

5. The nucleoskeleton beyond lamins

nuclear mechanics (Worman and Courvalin, 2004). To understand the etiology of laminopathies, we must first understand the complexities of nuclear architecture and mechanics, an understudied area of biology. It is naive to consider the architecture of the nucleoskeleton as a function of lamins only, since nuclei have many other structural proteins. The cytoskeleton includes multiple “skeletal” elements, each of which contributes uniquely to the structure, dynamics, molecular mechanics, and rheological properties of the cytoplasm (Wang et al., 1993). For example, cytoskeletal actin filaments can be cross-linked either rigidly or flexibly (Gardel et al., 2004), and actin filaments can interact with microtubules or cytoplasmic intermediate filaments (Stricker et al., 2010). This chapter summarizes evidence that similar interactions are relevant in the nucleoskeleton. Many proteins with known structural significance in the cytoskeleton are known to either localize specifically in the nucleus or shuttle in and out of the nucleus. These include b- and g-(nonmuscle) actin (Gieni and Hendzel, 2009), and specific isoforms of spectrins, protein 4.1, nesprins (spectrin-repeat proteins), and titins, each of which has one or more demonstrated roles in the nucleus (Table I). Most of these “nonlamin” nucleoskeletal proteins interact with lamins and are likely to confer complementary mechanical properties to the nucleoskeleton. Lamins contribute significantly to the viscoelastic stiffness of the nucleus, as shown by several well-characterized methods Table I Nonlamin structural proteins in nucleus and their partners and functions Structural protein Associated proteins

Associated functions

Actin

CH domain proteins: aII-spectrin, a actinin-4, bII-spectrin, filamin A, dystrophin, Nucleoskeleton, chromatin remodeling, vav, nesprin-2, L-plastin; Chromatin remodeling complexes: BAF, SWI-SNF, BAP, NuA4, TIP60, PBAF, p400, SWR1, INO80 complexes; Ribonucleoprotein RNA transcription and complexes proteins: hrp36, hrp65, DBP40; Other proteins: emerin, titin, protein 4.1, processing, nuclear lamin A, MAL, profilin, capG, exportin-6, zyxin, myopodin, Nrf2, NDHII, DNase I export

aII-spectrin

Structural proteins: actin, protein 4.1, bSpIV�5, lamin A, emerin; DNA repair DNA repair, fanconi proteins: hHR23B, XPA, RPA32, RPA70, XPB, XPD, XPG, XPF, ERCC1, anemia, nucleoskeleton MRE11, RAD50, RAD51, XRCC2, Ku70,Ku80; Chromatin remodeling proteins: actin, FANCA, BRG1, hBRM, CSB; Fanconi anemia proteins: FANCA, FANCC FANCD2, FANCF, FANCG, FANCJ; Transcription and RNA processing proteins: p40, PML (hnRNP) A2/B1

Titin

Lamin A, lamin B, actin, nuclear myosin I

Spindle matrix

Nesprin

Lamin A, lamin B, actin, emerin, SUN proteins

LINC complex

Protein 4.1

Actin, aII-spectrin, NuMA, U2AF, SC35

Nucleoskeleton, pre-mRNA processing

NuMA

Dynein, dynactin, protein 4.1, lamins, Arp1, GAS41, INI1, LGN, tubulin

Nucleoskeleton, spindle matrix

EAST

Actin, CP60, megator

Endonucleoskeleton

Megator

EAST

Spindle matrix

Skeletor

Chromator

Spindle matrix

Chromator

Skeletor

Spindle matrix

100

Zhixia Zhong et al.

(Dahl et al., 2004, 2005; Lammerding et al., 2004; Rowat et al., 2004). Here, we summarize the “nonlamin” components of the nucleoskeleton and describe tools to investigate their contributions to nuclear structure and function. Notably, the nucleos keleton and cytoskeleton are linked—directly and mechanically—by multiprotein “LINC” complexes that span the outer nuclear membrane (ONM) and INM of the nuclear envelope (Crisp et al., 2005; Dahl et al., 2008). These connections in living cells and the existence of both cytoskeletal and nucleoskeletal isoforms of key struc tural proteins, as well as the structural contributions of chromatin, collectively pose ongoing challenges to studying and understanding the superstructure and function of the nucleoskeleton. A. Actin in the Nucleus Actin, a major component of the cytoskeleton, actively exchanges between the nucleus and the cytoplasm and has numerous functions within the nucleus (Pederson and Aebi, 2005). In the nucleus, actin does not polymerize as phalloidin-stainable F-actin, but instead assembles nuclear-specific or other “unconventional” short poly mers (Bettinger et al., 2004; Pederson and Aebi, 2002). Also, a large fraction of actin in the nucleus is found as G-actin. Fluorescence recovery after photobleaching (FRAP) experiments in cells that transiently express green fluorescence protein (GFP)-fused b-actin revealed a dynamic equilibrium between low-mobility versus rapidly diffusing populations of actin in the nucleoplasm, strongly suggesting the existence of struc tured/polymeric actin in nucleus (McDonald et al., 2006). Under the very specialized conditions of the developing Xenopus oocyte nuclei, which are huge and have only nuclear envelope associated (not internal) lamins, nuclear export of actin is blocked to deliberately accumulate actin to enhance structural support for the oocyte nucleus (Clark and Merriam, 1977; Parfenov et al., 1995). Actin has many diverse roles in the nucleus including chromatin remodeling and the transcription, processing, and export of mRNAs (Bettinger et al., 2004). Actin and actin-related proteins (ARPs) are core components of switch/sucrose nonfermentable (SWI/SNF) chromatin remodeling complexes (Olave et al., 2002). Actin is also a component of transcription preinitiation complexes and stimulates transcription by RNA polymerase II (Hofmann et al., 2004; Percipalle et al., 2001). Actin and nuclear myosin 1c are reported to both associate with and stimulate RNA polymerase I, and actin is required for transcription by RNA polymerase III (Fomproix and Percipalle, 2004; Lanerolle et al., 2005; Philimonenko et al., 2004). However, it remains unclear exactly how actin and myosin function during transcription. Actin polymers are also required for the integrity of filaments (not lamins) that connect NPC baskets to nucleosomes and Cajal bodies in the Xenopus oocyte nucleus interior (Fig. 1). These filaments have been visualized by transmission electron microscopy (Arlucea et al., 1998) and in three dimensions by field emission scanning electron microscopy (feSEM) (Kiseleva et al., 2001, 2004; Ris, 1997). These “pore linked filaments” (PLFs) were destroyed by the actin-depolymerizing drug latrunculin A and altered by treatment with jasplakinolide, which stabilizes actin polymers. With

101

5. The nucleoskeleton beyond lamins

(A)

1.25 μm (B)

167 n

Fig. 1

Actin-based pore-linked filaments in the interior of Xenopus oocyte nuclei. (A) Nuclear content fixed and visualized by feSEM after “peeling” away the nuclear envelope. Intranuclear filaments associated with spherical bodies are seen in the figure. (B) The immunogold feSEM identifies the nuclear actin in filamentous network (pseudocolored yellow). Adapted from Kiseleva et al. (2004).

jasplakinolide the filaments became more open and lacy with regularly spaced short “struts.” Actin and protein 4.1 each localized on PLFs, as visualized by immunogold labeling and feSEM. The protein 4.1-gold epitopes were spaced at 120-nm intervals and were often located within 60–80 nm of filament “forks” (Kiseleva et al., 2004). The “backbone” protein of PLFs is thought to be related to Tpr, the NPC basket protein. Actin interacts functionally with lamin complexes. The INM protein emerin, which binds lamins directly, is also a pointed-end F-actin capping protein that enhances actin polymerization in vitro (Holaska et al., 2004). This observation, and associations between emerin, nuclear myosin 1c, and spectrin (discussed below), led to a proposal that the nuclear envelope, like the red blood cell membrane, might be supported by a “cortical” network of membrane (emerin)-anchored spectrin and actin filaments (Gruenbaum et al., 2005; Holaska and Wilson, 2007). Actin also binds lamin A directly (Simon et al., 2010), at two reported sites within the tail domain (Sasseville and Langelier, 1998; Zastrow et al., 2004). Lamin A is unique in having two actinbinding sites, compared to one site in lamin C and one relatively weak site in lamin B, and lamin A can “bundle” F-actin in vitro (Simon et al., 2010). Whether this actinbundling activity is relevant in living cells remains unknown. The emerging number and variety of structural and nonstructural roles for actin in the nucleus complicates the interpretation of studies in which the “cytoskeleton” is manipulated, for example, by actin-polymerizing drugs. For example, in many studies measuring nuclear mechanics within a cell, the cytoskeleton is depolymerized so that it does not dissipate applied force. In addition to disrupting cell structure and signaling,

102

Zhixia Zhong et al.

perturbing actin may cause unexpected (hence, untested) phenotypes in the nucleus that contribute to the composite phenotype. Better knowledge and new tools are required to understand the structure, function, and regulation of actin in the nucleus. B. Nuclear Spectrin Spectrins, first identified as a major component of the membrane skeleton of red blood cells (erythrocytes), are a well-characterized class of cytoskeletal proteins that line the plasma membrane. Human spectrins are encoded by seven genes, with two encoding a spectrins (aI and aII) and five encoding b-spectrins (bI through bV) (Bennett and Baines, 2001). Spectrins form a structural unit consisting of ab tetramers, for example, (aIbI)2 in erythrocytes. The aI-spectrin gene is expressed in erythrocytes, whereas aII spectrin is highly expressed in vertebrate nucleated cells (Cianci et al., 1999; Young and Kothary, 2005). Importantly, aII-spectrin localizes in both the cytoplasm and the nucleus; in the nucleus it is best known for its links to Fanconi anemia (FA) since aII-spectrin provides a scaffold that helps recruit DNA repair proteins to sites of DNA damage (McMahon et al., 1999, 2001). Cells from FA patients have decreased levels (35–40% of normal) of nuclear aII-spectrin (McMahon et al., 2001). aII-Spectrin binds directly to Fanconi anemia complementary group G (FANCG) (a component of the FANCA, FANCC, FANCF, FANCG complex) and colocalizes with the cross-link repair protein, XPF, in damage-induced nuclear foci after treatment of cells with DNA interstrand cross-linking agents (Lefferts et al., 2009; Sridharana et al., 2003). The phenotypes caused by aII-spectrin depletion support the idea that aII-spectrin functions in both the cytoskeleton and the nuclei. For example, siRNA-downregulated lymphoblastoid cells and HeLa cells show chromosomal instability and hypersensitivity to DNA interstrand cross-linking agents (McMahon et al., 2009). The aII-spectrin deficiency phenotype in WM-266 human melanoma cells included loss of stress fibers, cell adhesion defects, and reduced density of focal adhesions, as well as reduced proliferation with cell cycle arrest at the G1 phase, Rb phosphorylation, and elevated expression of the cyclin-dependent kinase inhibitor, p21Cip (Metral et al., 2009). Other nuclear proteins are known to associate with nuclear aII-spectrin, but whether they bind directly or indirectly is unknown. Among the nuclear proteins that coimmu noprecipitated with aII-spectrin are actin, protein 4.1B, b-spectrin (bSpIV�5), lamin A, and emerin (Sridharana et al., 2006). Independently, chromatographic purification of six proposed emerin-containing complexes from HeLa cell nuclei revealed a proposed 1.5 MDa complex that included emerin, aII-spectrin, actin, nuclear myosin Ic, lamins, and SUN2 (Holaska and Wilson, 2007). These findings implicate spectrins as significant components of the nucleoskeleton. In erythrocytes the spectrin–actin–protein 4.1 network is responsible for elasticity and mechanical recovery after deformation (Bennett and Gilligan, 1993). Their combined presence in the nucleus suggests similar, possibly essential, contributions to nuclear elasticity that will be important to test. Interestingly, protein 4.1 is essential to assemble nuclei in vitro in cell-free extracts of Xenopus eggs (Krauss et al., 2003). In addition to the structural proteins that coimmunoprecipitated with aII-spectrin

5. The nucleoskeleton beyond lamins

103

(Sridharana et al., 2006), there were also proteins involved in DNA repair (e.g., XPA, XPB, XPF, ERCC1), Fanconi anemia (e.g., FANCA, FANCC, FANCD2), chromatin remodeling (e.g., actin, BRG1, hBRM, and CSB), or transcription and RNA proces sing (e.g., P40, hnRNP A2/B1, PML). These findings suggest aII-spectrin has multiple functions in the nucleus, both at the peripheral nucleoskeleton and in internal struc tures, most of which await investigation. C. Nuclear Titin Titins, also known as connectins, are a family of giant elastic proteins first found in vertebrate striated muscle. Titin is the largest known human protein, with a mass of 3 MDa and contour length greater than 1 µm (Cola et al., 2005). Titins have more than 240 tandem repeats of predominantly immunoglobulin(Ig)-like and fibronectin (FN3)-like domains, each of which has a persistence length of 10 nm (Lee et al., 2007). Skeletal muscle titin localizes in the sarcomere; it has a kinase domain for signaling and also functions mechanically as a molecular spring to provide muscle with elasticity, allow postcontraction recovery, and prevent overextension (Machado and Andrew, 2000). Mutations in titin cause dilated cardiomyopathy (Gerull et al., 2002); cardiomyopathy is also a clinical phenotype in many lamin-linked syndromes (Capell and Collins, 2006). Interestingly, most eukaryotic cells also have a nuclear isoform of titin, which associates with chromosomes and is essential for mitotic chromosome condensation (Machado et al., 1998). In Drosophila, loss of titin is lethal during mitosis: chromosomes fail to condense properly during prophase, have defects in sister chromatin cohesion, and missegregate during mitosis (Machado and Andrew, 2000). Titin, which has Ig-folds itself, binds directly to the Ig-fold domain of both A- and B-type lamins in vitro, with a slight preference for lamin A (Zastrow et al., 2006). Titin binding was mildly but selectively sensitive to specific laminopathy-causing missense mutations in the Ig-fold domain of lamin A, suggesting these mutations might perturb lamin–titin connections. Altering titin–lamin interactions leads to highly dismorphic nuclei (Fig. 2). In Caenorhabditis elegans embryos, titin colocalized with lamins at the nuclear envelope, and this localization required lamins, suggesting nuclear titin is anchored or organized by lamin filaments (Zastrow et al., 2006). Titin is also slightly enriched at the nuclear envelope in human cells (Zastrow et al., 2006). These findings, and the presence of other titin-binding proteins (notably, actin) in the nucleus, implicate titin in the long-range organization, stability, elasticity, and mechanics of both the nuclear envelope and the chromosomes during interphase, as well as in mitosis. Further studies, sorely needed to test predicted mechanical func tions, will be challenging given the enormous size of titin and potential functional overlap with other titin-related genes. D. LINC Complexes at the Nuclear Envelope There are direct mechanical connections between the cytoskeleton and the nucleo skeleton, mediated by LINC complexes (Dahl et al., 2008; Stewart-Hutchinson et al.,

104

Zhixia Zhong et al.

GFP

Lamin B

Merge

GFP-M-is7

GFP-M-is7

Control

DNA

Fig. 2 Indirect immunofluorescence microscopy indicates that cells expressing GFP–NLS (nuclear localization signal) fused titin fragment M-is7 had a high frequency of aberrantly shaped nuclei and nuclear envelope herniations (arrowheads) that contained GFP-is7, but not chromatin or lamin B. Negative control uses cells transfected with GFP–NLS fused to pyruvate kinase. Scale bar: 2 mm. Adapted from Zastrow et al. (2006).

2008). The basic components of these complexes are Klarsicht-Anc1-syne1 homology (KASH)-domain proteins (e.g., mammalian nesprins, aka SYNE) and Sad1-UNC-84 homology (SUN)-domain proteins, which are embedded in the ONM and INM and interact (via the SUN and KASH domains) within the lumenal space of the nuclear envelope (Razafsky and Hodzic, 2009). The discovery of these complexes has “coupled” the disparate research worlds of the cytoskeleton and nucleoskeleton (Hale et al., 2008). LINC complexes explain how the nucleus is positioned within cells (Tzur et al., 2006), how chromosomes can be moved in the plane of the nuclear envelope (Razafsky and Hodzic, 2009), and how mechanical forces can be transduced directly from the cytoskeleton to the nuclear interior (Dahl et al., 2008). The human genome encodes four nesprin genes (SYNE-1, SYNE-2, Nesprin-3, and Nesprin-4). SYNE-1 and -2 are alternatively transcribed and spliced to yield protein isoforms with different sizes, functions, and locations, including specific localization to either the ONM, INM, nuclear interior, or cytoplasmic organelles including the Golgi complex (Zhang et al., 2002, 2005). Nesprin-3 (isoforms a and b) and Nesprin-4 localize at the ONM (Roux et al., 2009; Wilhelmsen et al., 2005). To localize such proteins specifically requires both epitope-specific tags or antibodies and methods that can distinguish between the INM and the ONM. One classic method, digitonin permeabilization (described below), is still a powerful approach to localize specific proteins on the nuclear envelope. The largest (“giant”) isoforms of nesprin-1, nesprin-2, and nesprin-3 bind either actin (nesprin-1g, nesprin-2g) or plectin (nesprin-3a and nesprin-3b) in the cytoplasm

5. The nucleoskeleton beyond lamins

105

and are embedded in the ONM as key components of the LINC complex (Crisp et al., 2005; Dahl et al., 2008). Other, smaller nesprin isoforms such as nesprin-1a and nesprin-2b localize at the INM and directly bind lamins and emerin. The nesprin (nuclear envelope spectrin repeats) polypeptide structure can include an N-terminal actin-binding “calponin homology” (CH) domain (present on “giant” isoforms only), variable numbers of tandem spectrin-like repeats (SLRs), and near the C-terminus a unique “adaptive” domain followed by several more SLRs and the C-terminal membrane-anchoring KASH domain (Zhang et al., 2002). The SLRs in nesprins are similar to typical spectrin repeats, suggesting nesprins (like other spectrin repeat superfamily proteins) have elastic properties, and the “adaptive” domain is evolutio narily conserved and structurally stabilizes the SLRs in nesprins (Zhong et al., 2010). The KASH domain comprises both a transmembrane domain and an evolutionarily conserved 30-residue motif that is located in the nuclear envelope lumen and directly binds SUN proteins (Razafsky and Hodzic, 2009). The 200-residue SUN domain is encoded by four mammalian genes (SUN1, 2, 3, and SPAG4). SUN1 and SUN2 are both INM proteins (Tzur et al., 2006); their SUN domain is located in the nuclear envelope lumen, and their N-terminal domain is nucleoplasmic and binds lamin filaments. SUN proteins function as dimers. The unique “adaptive” domain of nesprins may also mediate homodimerization, thereby increasing the mechanical strength and stability of NE-spanning SUN/KASH-mediated LINC complexes (Zhong et al., 2010). E. Roles for Nucleoskeletal Proteins in Mitosis and Nuclear Assembly Nearly all mechanical studies of nucleoskeletal proteins have been done in inter phase cells, since mitosis involves very rapid, large-scale structural changes that can challenge biophysical measurement. However, many nucleoskeletal proteins appear to have dual roles: they also contribute structurally during mitosis, by organizing the mitotic spindle. The idea of a “spindle matrix,” first proposed in the 1960s (Smetana et al., 1963), was that a stiff nonmicrotubule filamentous structure might help anchor and move chromosomes during mitosis. This idea is supported by recent evidence that lamin B and other nucleoskeletal proteins, including titin, can localize to spindle-like structures during mitosis (Fabian et al., 2007; Tsai et al., 2006). For example, by immunofluor escence, titin colocalizes with microtubules and also with other spindle matrix candi date proteins including (in Drosophila) the enhanced adult sensory threshold (EAST), skeletor, megator, and chromator proteins (Fabian et al., 2007). Proposed spindle matrix proteins in vertebrates include nuclear mitotic apparatus protein (NuMA), a conserved 200–240-kDa coiled-coil microtubule-binding protein (Dionne et al., 1999; Zeng et al., 1994). NuMA was first identified as a mitotic polar component that originated from the nucleus. In Xenopus, NuMA associates with dynein–dynactin com plexes in centrosomes to anchor spindle microtubules at the poles (Dionne et al., 1999). The interphase functions of NuMA are not well understood. However, NuMA is known to interact with both structural and regulatory proteins including lamins, protein 4.1,

106

Zhixia Zhong et al.

microtubules, and the INI1 (chromatin remodeling) complex, suggesting roles in both nucleoskeleton and chromatin organization (Barboro et al., 2001). EAST is a 250-kDa Drosophila protein with seven potential nuclear localization sequences (NLSs) and twelve potential PEST proteolytic signals (Wasser and Chia, 2000). EAST is an essential, ubiquitous nuclear protein that forms a network throughout the nucleus and colocalizes with actin. During interphase in Drosophila, megator surrounds chromosomes and is enriched at the nuclear envelope, whereas skelator and chromator (which bind each other directly) localize on chromosomes. During mitosis, these three proteins collectively form a “spindle matrix” that underlies the microtubule spindle (Qi et al., 2005; Rath et al., 2004; Walker et al., 2000). Interestingly, megator is orthologous to the mammalian NPCbasket protein Tpr; megator interacts with EAST and is proposed to form a nuclear “endoskeleton” with EAST. We speculate that this interior nucleoskeleton might be analogous to the PLFs visualized in Xenopus oocyte nuclei (Kiseleva et al., 2004).

II. Methods A. Localization of Endogenous Nucleoskeletal Proteins by Indirect Immunofluorescence Microscopy Historically, for proteins now known to function in both the cytoskeleton and the nucleoskeleton, nuclear signals were often either disregarded or assumed to be non specific. The seemingly simple task of localizing a proposed endogenous nonlamin nucleoskeletal protein in interphase nuclei, by indirect immunofluorescence, is a litmus test that can pose frustrating challenges. For example, staining may be negative because epitopes are inaccessible in assembled structures, accessible only at certain stages of the cell cycle, blocked by association with chromatin or lamin filaments, or masked by posttranslational modifications. Conversely, positive signals provide key information and can be used to assess the behavior of epitope-tagged (e.g., GFPtagged) exogenous versions of the same protein, since tags and overexpression can cause mislocalization or other artifacts. However, agreement between these methods supports further experiments using the more easily detected tagged protein. Below we outline a “typical” indirect immunofluorescence protocol that has been used, with minor modifications, to localize lamins (Dahl et al., 2006; Lammerding et al., 2006) and several nonlamin nucleoskeletal proteins including LINC proteins, titins, and spectrins (Haque et al., 2006; Roux et al., 2009; Sridharana et al., 2003; Zastrow et al., 2006) in cultured cells and in some cases also in C. elegans, Drosophila, or tissue samples. We highlight important caveats and describe alternative approaches that might improve antibody-based detection of endogenous nucleoskeletal proteins. All steps are done at room temperature (22–25°C) unless otherwise noted. 1. Culture cells on coverslips (22 22 mm); wash gently 3 times with phosphatebuffered saline (PBS), then fix 15 min in PBS containing 3.7% formaldehyde (stock solution is 37%, prepare fresh). Caveat and alternative: some epitopes are sensitive to fixation. Try other fixation methods, for example, (a) Paraformaldehyde fixation:

5. The nucleoskeleton beyond lamins

2.

3.

4.

5. 6. 7.

107

fix with 3.5% paraformaldehyde at 4°C for 5 min followed by 10 min at room temperature (DiDonato and Brasaemle, 2003). (b) Cold methanol fixation: fix with cold methanol (store at –20°C before use) for 5 min in a prechilled glass tray kept on dry ice (Jiang and Serrero, 1992). (c) Cold acetone fixation: fix with cold acetone (store at –20°C before use) for 10 min in a prechilled glass tray kept on dry ice during fixation (Hammond and Glick, 2000). (d) TCA fixation: fix the cells in 10% trichloroacetic acid in water cooled to 4°C for 15 min, then wash with 30 mM glycine in PBS (Haraguchi et al., 2004). Wash fixed cells 3 times with PBS, then permeabilize (all membranes) by incubating 30 min in PBS/0.2% Triton X-100. Permeabilization is not required for samples fixed using organic solvents (e.g., methanol or acetone). Optimal permeabilization is very important for detecting proteins within the nucleus, since intact INM and ONM block antibody access, and potential cytoskeletal signals will overwhelm. Excess or harsh permeabilization can risk solubilizing or destroying endogenous structures. Wash 3 times with PBS, then block 1 h in PBS plus a nonspecific blocking protein, such as 2% bovine serum albumin (BSA) or nonspecific serum that does not conflict with your species-specific primary or secondary antibodies. Dilute primary antibodies (dilution is antibody dependent; test dilution series) in small volume (50–100 ml) of PBS/2% BSA, add to fixed cells, and incubate 1 h (room temperature) or longer at 4°C. Keep coverslips in a moist chamber or invert onto a 50-ml drop of antibody solution on parafilm. Do not let coverslips dry. Caveat and alternative: epitope(s) can be blocked by posttranslational modifications, bound partners, or nearby structures (e.g., lamins, actin, chromatin). Try a different antibody that targets a different region of your protein or a polyclonal antibody (likely to recognize a variety of epitopes on your protein). Different antibodies can preferentially detect either the nuclear or cytoplasmic isoforms of a given protein (e.g., aII-spectrin (McMahon et al., 1999; Metral et al., 2009; Sridharana et al., 2003)). Wash 3 times with PBS, then incubate 1 h with diluted species-specific conjugated fluorescent secondary antibody. Wash twice with PBS, then once with PBS/DAPI (1:3000 dilution of 1 mg/ml stock). Invert coverslips onto a slide with 4-ml Vectashield or similar antifade reagent. Seal periphery with clear fingernail polish and visualize by fluorescence microscopy. Confocal microscopy is recommended for high Z-axis resolution and threedimensional image reconstruction; this is particularly useful to distinguish subnuclear localizations and visualize nuclear envelope enrichment. Care must be taken to avoid touching or pushing coverslips, which can grossly distort cell and nuclear height. This issue is avoided by immunolabeling cells cultured on glassbottom dishes (described next).

1. Alternative: Immunolabel Cells Cultured on Glass-Bottom Dishes To protect cells from mechanical pressure (and avoid air bubbles that can be introduced when mounting coverslips), consider culturing and labeling cells in a

108

Zhixia Zhong et al.

glass-bottom dish. Note that these fixed preparations can become contaminated and should be imaged without delay. 1. Culture cells in 35-mm glass-bottom dish (e.g., from Mattek or Warner) and then fix, permeabilize, block, and label as described above (A, steps 1–6), taking care to keep cells wet; larger volumes of all reagents, including antibodies, will be required. 2. Fill the dish with 1-ml PBS, and then image fluorescence using an inverted microscope. Dishes wrapped with parafilm can be stored at 4°C for up to 1 month.

2. Alternative: Isolate Nuclei Prior to Immunolabeling If the cytoskeletal signal overwhelms the nuclear signal, consider labeling isolated nuclei. Nuclei can be isolated using a kit (several are available from Sigma), or the method described here. Kits are easy but may retain cytoplasmic remnants. The method described here is more labor intensive and gives significantly lower yields than kits, but nuclei are relatively free of cytoskeleton and endoplasmic reticulum. To isolate nuclei (adapted from (Dahl et al., 2005; Dean and Kasamatsu, 1994): 1. Start with 107 cells; rinse twice with PBS and once with cold (4°C) 10 mM HEPES pH 7.5, 1 mM dithiothreitol (DTT). 2. Scrape cells into minimal volume (<1 ml) HEPES/DTT and incubate 10 min on ice. 3. Lyse cells by 10–25 strokes in Dounce homogenizer or a small bore needle. 4. Add lysed cells to one-fifth volume of 5 sucrose–salt solution (1.25 M sucrose, 250 mM Tris pH 7.6, 125 mM KCl, 15 mM MgCl2, 15 mM CaCl2, and protease inhibitors) and incubate 10 min on ice. Salt condenses DNA; sucrose will produce a gradient for ultracentrifugation. 5. Add 2.3 M sucrose–salt solution (2.3 M sucrose, 50 mM Tris pH 7.6, 25 mM KCl, 3 mM MgCl2, 3 mM CaCl2 and protease inhibitor) to achieve a final concentration of 1.6 M sucrose. 6. Layer this 1.6 M sucrose–salt–lysate suspension onto a 150-ml cushion of 2.3 M sucrose–salt and centrifuge at 4°C for 1 h at 166,000g. Debris will collect at the interface; nuclei are dense and will pellet. 7. Resuspend nuclei in PBS. Alternatively, to dilate nuclei, resuspend in 50 mM Tris pH 7.5. Once nuclei are isolated (by either method): 1. Prepare polylysine-coated coverslips by adding a 0.1% (w/v) solution of polyL-lysine (premade, sterile solutions are available from Sigma) directly to clean, dry coverslips. Allow lysine to adsorb 20 min, then aspirate and dry. (Lysine solution can be reused several times). 2. Dilute isolated nuclei in PBS (or 50 mM Tris pH 7.5) to approximately 106 nuclei/ml and place on coverslip. 3. Fix, permeabilize, block, and label as described above (A, steps 1–6).

5. The nucleoskeleton beyond lamins

109

Caveat and alternative: epitopes can be blocked by chromatin or soluble nuclear components. Before fixing, consider the following strategy to remove most DNA and nonnucleoskeletal components from isolated nuclei (adapted from Graham and Rickwood (1997)): 1. Suspend isolated nuclei in 55 ml 0.25 M sucrose, 5 mM MgCl2, 50 mM Tris–HCl pH 7.5, then add DNAse and RNAse to final concentrations of 250 mg/ml each. 2. Incubate 60 min at 4°C, then pellet nuclei (1000g, 10 min, 4°C). 3. Resuspend in 50 ml 10 mM Tris–HCl pH 7.4, 0.2 mM MgCl2 and then slowly add four volumes of 2 M NaCl, 0.2 mM MgCl2, 10 mM Tris–HCl pH 7.4. 4. Add 2-mercaptoethanol to final concentration of 1% and incubate 15 min at 4°C. 5. Pellet nuclei by centrifuging 30 min at 1,600g. 6. Repeat steps 3 and 5 (skip 4) and resuspend in 50 ml 10 mM Tris–HCl pH 7.5.

B. Endogenous Protein Localization at the INM Versus the ONM Nuclear envelope membranes and their enclosed lumenal space are continuous with the endoplasmic reticulum, but nevertheless represent three highly specialized functional domains: the ONM domain, the INM domain, and the “pore membrane” domain around each NPC. Proteins that localize at the pore membrane domain are readily distinguished by punctate fluorescent staining that overlaps with NPC markers, as recently (and surprisingly) shown for SUN1 in HeLa cells (Liu et al., 2007). One can also deduce the subnuclear envelope localization of candidate proteins by double or triple labeling with antibodies against NPC proteins located on either the cytoplas mic (e.g., Nup358 or Nup214) or nucleoplasmic (e.g., Nup153 or Nup98) side of the NPC, or marker proteins specific to the INM (e.g., emerin, MAN1) or nuclear envelope lumen (Schermelleh et al., 2008). However, these methods require very high spatial sensitivity in imaging. Localization can be determined unambiguously by other methods including trans mission electron microscopy of immuno-gold-labeled samples (not described here), or the following classic method that distinguishes ONM and INM localization based on indirect immunofluorescence staining of cells made permeable using either Triton X100 (dissolves all membranes) or Digitonin (affects the plasma membrane but not interior or nuclear membranes). Digitonin is a steroid glycoside that binds cholesterol and other b-hydroxysterols that are highly enriched in the plasma membrane, relative to intracellular membranes (Fiskum et al., 1980). Antibodies incubated with digitonin-permeabilized cells have access only to epitopes that face the cytoplasm, including proteins on the ONM. Proteins located at the INM, nuclear envelope lumen, or nucleoplasm are detected only in Triton-permeabilized cells (see Fig. 3). To apply the digitonin method to fixed cells, incubate a second set of fixed cells 15 min (on ice) in PBS/0.004% digitonin (reported range is 0.002–0.005%), rather than PBS/Triton, and then continue as described in Section I.

110

Zhixia Zhong et al.

(A)

Digitonin-treated cell

(B)

Triton X-100-treated cell

INM protein

(C)

(D)

ONM protein

Lamin A/C

A′

Lamin A/C

B′

Nesp4

C′

Nesp4

D′

Fig. 3 Altered membrane permeablization allows visualization of INM versus ONM proteins.Digitonin permeabilizes the plasma membrane but not the nuclear membrane, whereas Triton X-100 permeabilizes all membranes. During immunocytochemistry, this allows for altered antibody accessibility to different regions of the nucleus. INM proteins cannot be detected with digitonin permeabilization (A and A0 ) but can be detected with Triton X-100 treatment (B and B0 ). ONM protein can be detected with both digitonin (C and C0 ) and Triton X-100 (D and D0 ) treatment. This has been shown functionally with the nucleoskeleton protein lamin A/C which is inside the nucleus, and the ONM protein Nesprin-4 (A0 through D0 ). Adapted from Roux et al. (2009).

5. The nucleoskeleton beyond lamins

111

C. Immunoprecipitation of Nucleoskeletal Proteins Coimmunoprecipitation (co-IP) is an excellent way to detect protein–protein associa tions in cells. This procedure can validate biochemical experiments and, when coupled to other methods such as mass spectrometry, can be used to identify novel partners or components. The main challenge for lamins, and probably most other nucleoskeletal proteins, is that one must first dissociate preexisting (potentially highly stable) structures, in order to solubilize and study them. Summarized below are examples of methods used to solubilize one or more specific nucleoskeletal proteins. These methods focus specifi cally on two key steps: cell/nuclear lysis and protein solubilization.

1. Solubilization Conditions Suitable for Recovery of Lamins, Other Nucleoskeletal Components, and Other Proteins Including Nuclear Membrane Protein Emerin, from Isolated HeLa Cell Nuclei First, isolate nuclei by hypotonic lysis of cells (Offterdinger et al., 2002): Wash cells twice in PBS, harvest by scraping, pellet cells (1850g, 10 min, 4°C), resuspend in five packed cell volumes of hypotonic buffer (10 mM HEPES pH 7.9, 1 .5 mM MgCl2, 10 mM KCl, 0.2 mM PMSF, 0.5 mM DTT, and protease inhibitors). Then recentrifuge, suspend in three packed cell volumes of hypotonic buffer, incubate on ice 10 min, homogenize (Dounce, type B pestle, 25 strokes), and centrifuge (3300g, 15 min, 4°C) to pellet nuclei. Wash the nuclear pellet twice with 10-ml hypotonic buffer. Then, to solubilize nucleoskeletal and other components, switch to the following protocol, which allowed subsequent immunoaffinity and chromatographic purification of com plexes that included emerin plus specific combinations of other proteins including lamin A, lamin B, II-spectrin, actin, nuclear myosin 1c, SUN2, histones, or transcrip tion regulators (Holaska and Wilson, 2007). Resuspend purified nuclei in high-salt/ detergent buffer (1 M NaCl, 1% Triton X-100, 20 mM HEPES pH 8.0) to extract lamins and nuclear membrane proteins, and centrifuge (30 min, 40,000g, 4°C) to pellet insoluble material. Dilute the supernatant 10-fold with 20 mM HEPES (pH 8.0), incubate 10 min at 4°C to allow reformation of complexes that may have dissociated during lysis, and recentrifuge (30 min, 40,000g, 4°C) to yield a clarified supernatant enriched for lamins and other nucleoskeletal proteins, some of which may retain biologically relevant associations. Protein(s) of interest can be immunoprecipitated or studied by other methods. For quality control, verify and estimate the percent solubilization of nucleoskeletal marker proteins (e.g., lamins, emerin) and protein(s) of interest, by western blot analysis of small aliquots of the starting nuclear pellet, soluble versus insoluble nuclear fractions, and each clarified supernatant.

2. Efficient Solubilization of Lamins Lamin polymers resist biochemical extraction. This resistance can be exploited by treating cells or isolated nuclei with moderate salt buffers, to solubilize most nuclear proteins, and then pelleting to obtain a lamin-enriched pellet. Lamins are solubilized

112

Zhixia Zhong et al.

relatively efficiently by high salt (e.g., 1 M NaCl) or high pH (pH 8.0–9.0) or preferably both as summarized next from Aebi et al. (1986) and Herrmann and Aebi (2004). Isolate nuclear envelopes (or nuclei) and wash with 1 M KCl, then incubate 30 min in (300 mM KCl, 2% Triton X-100, 10% sucrose, 20 mM MES–KOH pH 6.0, 2 mM EDTA, 1 mM DTT) and centrifuge (6000g, 20 min) to obtain a lamin-enriched pellet. Resuspend this pellet in (500 mM KCl, 2% Triton X-100, 20 mM Tris–HCl pH 9.0, 2 mM EDTA, 1 mM DTT) for 30 min on ice, and then centrifuge (200,000g, 40 min, 4°C) to yield a supernatant of solubilized lamins, and an insoluble pellet that may include any contaminating cytoplasmic intermediate filaments (Aebi et al., 1986; Herrmann and Aebi, 2004). Interestingly, the solubility properties of A- and B-type lamins in living cells can be manipulated. For example, endogenous A- and B-type lamins are more resistant to extraction from cells that overexpress lamin A, whereas endogenous lamins B1 and B2 are more easily solubilized from cells that overexpress lamin B2 (Schirmer and Gerace, 2004). The potential effects of these manipulations on other nucleoskeletal components will be interesting to test.

3. Solubilization Conditions Suitable for Nesprins Several nesprin isoforms, from large to small (e.g., the INM-localized nesprin-1), can be immunoprecipitated from mouse C2C12 myoblasts after lysis in (100 mM NaCl, 1% Triton X-100, 20 mM Tris–HCl pH 7.5, 1 mM EDTA–Na2) (Zhang et al., 2001). If cytoskeletal contaminants are a concern, consider first isolating nuclei through a sucrose density gradient Section IIA2).

D. Micropipette Aspiration and Recoil Micropipette aspiration (MPA) is a widely used micromanipulation technique for quantifying the viscoelastic properties of cells, isolated nuclei, and other biological samples. The mechanical contribution of a target nuclear structural protein can be measured by comparing the MPA results of control versus manipulated (overexpres sing or depleted) cell nuclei. We developed a new “recoil” step to the traditional MPA assay, which reflects both the energy storage capacity and the elasticity of the nucleus. This method enables one to measure both nuclear deformation and recovery (Zhong et al., submitted). 1. Pull (or purchase) micropipettes of 8- to 10-mm diameter from 1-mm borosilicate capillaries (WPI, Sarasota, FL, USA) with commercial pipette pullers (MicroData Instrument Inc. or Sutter), also used to make pipettes for microinjection and patch clamping. Recoil assays require larger micropipette diameters (10 mm) than those normally used for nuclei MPA (3–5 mm). The larger diameter reduces damage to nuclear structure and facilitates recoil by reducing the plasticity of aspiration. Quantitative recoil assays require pipettes with relatively constant tip diameter (10 mm) along the entire length contacted by the aspirated nuclei to ensure constant cross section of aspiration.

5. The nucleoskeleton beyond lamins

113

2. Backfill the micropipette with 1% BSA in PBS using a syringe with needle to coat the tip and thereby lower static charge and prevent adhesion. New commercial devices are being developed to facilitate micropipette handling and backfilling (e.g., Warner Instruments). 3. Connect the filled micropipette to the pressure system and micromanipulators (e.g., Narishige or Eppendorf) via a waterline to a water-filled reservoir. Aspirate sample buffer into the micropipette to ensure solution compatibility. 4. Cells in suspension can be incubated with both a cell-permeable DNA dye (such as Hoechst 33342) and a cell-impermeable DNA dye (propidium iodide). Thus, nuclear deformation can be specifically tracked in live cells, since ruptured cells become PI positive. One caveat is that Hoechst 33342 intercalates into DNA and can slightly affect nuclear stiffness. However, it is still useful when comparing samples on this short time scale. 5. Apply and measure aspiration pressures. Since nuclei are typically stiffer than other parts of the cell, negative pressures of 0.1–10 kPa are required for aspiration. Aspiration pressures this high are typically generated by syringe suction. A pressure transducer or manometer in parallel (or within) the apparatus is required to monitor applied pressure. 6. Observe cells and nuclei in brightfield and via the Hoechst 33342 signal, respectively, with a high-resolution fluorescent microscope. Check the propidium iodide signal before and after aspiration, to verify cell viability. During aspiration at constant pressure, the nucleus will be gradually sucked into the micropipette forming a “projection.” The projection area increases as a function of time. 7. The best strategy is to produce and hold a set aspiration pressure, then watch deformation over time. These “creep” tests reveal viscoelastic deformation. Instantaneous removal of pressure allows nuclei to recoil. To measure recoil (reversibility of aspiration), release pressure to zero when the projection area of the aspirated nucleus is less than one-third of total nuclear area. Aspiration into too narrow pipettes (<5 mm-diameter), or release after projections become too large, may lead to incomplete recoil. 8. The creep compliance of the MPA process can be quantified by the modified equation: J ðtÞ ¼ �DA=3DPR2p , where � depends on micropipette wall thickness (Theret et al., 1988), DP is the constant pressure applied, DA is the aspiration area, and Rp is the pipette radius (Fig. 4). Creep compliance (J) is useful to compare between experimental conditions and to examine the form of the deformation over time (viscoelastic, purely elastic). However, extracting exact measures of viscoelasticity, particularly for large pipette aspiration, is not recommended. For nuclear aspiration and recoil, time-dependent creep compliance matches a power law, J ðtÞ ¼ Aðt=t0 Þ , and the data spots can be linear fit on a log–log scale. We have found that the power-law coefficient a is similar for both aspiration and recoil in wildtype HeLa cell nuclei and TC7 cell nuclei (Dahl et al., 2005). For cells with nuclear defects or deficiencies (e.g., in lamin A/C or aII-spectrin), the power-law coefficient a during aspiration versus recoil can differ dramatically. Thus, by measuring recoil

114

Zhixia Zhong et al.

(A)

Cytoplasm

ΔA

ΔP = const

Rp

Nucleus (B)

ΔP = –0.5 to ~2 MPa (aspiration) 5-min aspiration

ΔP = 0 Pa (recoil)

2-s P release

5-min P release

KD αll-sp

WT

2-s aspiration

(C)

Fig. 4

Micropipette aspiration (MPA) of nuclei and cells to show subtle nuclear structural and mechanical changes. (A) Schematic of aspirated cell showing nuclear deformation into the pipette of DA (cross-hatched) under applied pressure DP. (B) Aspiration of cells into the micropipette under constant pressure shows nuclear deformation into the pipette. (C). When pressure released to zero, the wildtype (WT) cell recoiled back gradually, while cell with reduced aII-spectrin (KD aII-sp) failed to recoil. Adapted from Zhong, Z. et al., submitted manuscript.

explicitly, one can detect new types of mechanical defects, specifically including nuclear elasticity, which were not previously detectable by MPA or AFM.

III. Discussion and Prospects Further studies of nonlamin components of the nucleoskeleton are essential to fully understand nuclear architecture and mechanics. With this information, we can better understand how the nucleoskeleton impacts the cytoskeleton, the genome, and human disease. References Aebi, U., Cohn, J., Buhle, L., and Gerace, L. (1986). The nuclear lamina is a meshwork of intermediate-type filaments. Nature 323, 560–564. Arlucea, J., Andrade, R., Alonso, R., and Arechaga, J. (1998). The nuclear basket of the nuclear pore complex is part of a higher-order filamentous network that is related to chromatin. J. Struct. Biol. 124, 51–58.

5. The nucleoskeleton beyond lamins

115

Barboro, P., D’Arrigo, C., Diaspro, A., Mormino, M., Alberti, I., Parodi, S., Patrone, E., and Balbi, C. (2001). Unraveling the organization of the internal nuclear matrix: RNA-dependent anchoring of NuMA to a lamin scaffold. Exp. Cell Res. 279, 202–218. Bennett, V., and Baines, A. J. (2001). Spectrin and ankyrin-based pathways: Metazoan inventions for integrating cells into tissues. Phys. Rev. 81, 1353–1392. Bennett, V. and Gilligan, D. M. (1993). The spectrin-based membrane skeleton and micron-scale organiza tion of the plasma membrane. Annu. Rev. Cell Biol. 9, 27–66. Bettinger, B. T., Gilbert, D. M., and Amberg, D. C. (2004). Actin up in the nucleus. Nat. Rev. Mol. Cell Biol. 5, 410–415. Broers, J. L. V., Peeters, E. A. G., Kuijpers, H. J. H., Endert, J., Bouten, C. V. C., Oomens, C. W. J., Baaijens, F. P. T., and Ramaekers, F. C. S. (2004). Decreased mechanical stiffness in LMNA–/– cells is caused by defective nucleo-cytoskeletal integrity: implications for the development of laminopathies. Hum. Mol. Genet. 13, 2567–2580. Capell, B. C., and Collins, F. S. (2006). Human laminopathies: Nuclei gone genetically awry. Nat. Rev. Genet. 7, 940–952. Cianci, C. D., Zhang, Z., Pradhan, D., and Morrow, J. S. (1999). Brain and muscle express a unique alternative transcript of alphaII spectrin. Biochemistry 38, 15721–15730. Clark, T. G., and Merriam, R. W. (1977). Diffusible and bound actin in nuclei of xenopus laevis oocytes. Cell 12, 883–891. Cola, E. D., Waigh, T. A., Trinick, J., Tskhovrebova, L., Houmeida, A., Pyckhout-Hintzen, W., and Dewhurs, C. (2005). Persistence length of titin from rabbit skeletal muscles measured with scattering and microrheology techniques. Biophys. J. 88, 4095–4106. Crisp, M., Liu, Q., Roux, K., Rattner, J. B., Shanahan, C., Burke, B., Stahl, P. D., and Hodzic, D. (2005). Coupling of the nucleus and cytoplasm: Role of the LINC complex. J. Cell Biol. 172, 41–53. Dahl, K. N., Kahn, S. M., Wilson, K. L., and Discher, D. E. (2004). The nuclear envelope lamina network has elasticity and a compressibility limit suggestive of a molecular shock absorber. J Cell Sci. 117, 4779–4786. Dahl, K. N., Engler, A. J., Pajerowski, J. D., and Discher, D. E. (2005). Power-law rheology of isolated nuclei with deformation mapping of nuclear substructures. Biophys. J. 89, 2855–2864. Dahl, K. N., Scaffidi, P., Islam, M. F., Yodh, A. G., Wilson, K. L., and Misteli, T. (2006). Distinct structural and mechanical properties of the nuclear lamina in Hutchinson-Gilford progeria syndrome. Proc. Natl. Acad. Sci. USA 103, 10271–10276. Dahl, K. N., Ribeiro, A. J. S., and Lammerding, J. (2008). Nuclear shape, mechanics, and mechanotransduc tion. Circ. Res. 102, 1307–1318. de Lanerolle, P., Johnson, T., and Hofmann, W. A. (2005). Actin and myosin I in the nucleus: what next? Nat. Struct. Mol. Biol. 12, 742–746. Dean, A. D., and Kasamatsu, H. (1994). Signal- and energy-dependent nuclear transport of SV40 Vp3 by isolated nuclei. J. Biol. Chem. 269, 4910–4916. Dechat, T., Pfleghaar, K., Sengupta, K., Shimi, T., Shumaker, D. K., Solimando, L., and Goldman, R. D. (2008). Nuclear lamins: Major factors in the structural organization and function of the nucleus and chromatin. Genes. Dev. 22, 832–853. DiDonato, D., and Brasaemle, D. L. (2003). Fixation methods for the study of lipid droplets by immuno fluorescence microscopy. J. Histochem. Cytochem. 51, 773–780. Dionne, M. A., Howard, L., and Compton, D. A. (1999). NuMA is a component of an insoluble matrix at mitotic spindle poles. Cell Motil. Cytoskel. 42, 189–203. Fabian, L., Xia, X., Venkitaramani, D. V., Johansen, K. M., Johansen, J., Andrew, D. J., and Forer, A. (2007). Titin in insect spermatocyte spindle fibers associates with microtubules, actin, myosin and the matrix proteins skeletor, megator and chromator. J. Cell Sci. 207, 2190–2204. Fiskum, G., Craig, S. W., Decker, G. L., and Lehninger, A. L. (1980). The cytoskeleton of digitonin-treated rat hepatocytes. Proc. Natl. Acad. Sci. USA 77, 3430–3434. Fomproix, N., and Percipalle, P. (2004). An actin–myosin complex on actively transcribing genes. Exp. Cell Res. 294, 140–148.

116

Zhixia Zhong et al. Gardel, M. L., Shin, J. H., MacKintosh, F. C., Mahadevan, L., Matsudaira, P., and Weitz, D. A. (2004). Elastic behavior of cross-linked and bundled actin networks. Science 304, 1301–1305. Gerull, B., Gramlich, M., Atherton, J., McNabb, M., Trombitás, K., Sasse-Klaassen, S., Seidman, J. G., Seidman, C., Granzier, H., Labeit, S., Frenneaux, M., and Thierfelder, L. (2002). Mutations of TTN, encoding the giant muscle filament titin, cause familial dilated cardiomyopathy. Nat. Genet. 30, 201–204. Gieni, R. S., and Hendzel, M. J. (2009). Actin dynamics and functions in the interphase nucleus: Moving toward an understanding of nuclear polymeric actin. Biochem. Cell Biol. 87, 283–306. Graham, J., and Rickwood, D. (1997). In “Subcellular Fractionation: A Practical Approach”. IRL Press, Oxford. Gruenbaum, Y., Margalit, A., Goldman, R. D., Shumaker, D. K., and Wilson, K. L. (2005). The nuclear lamina comes of age. Nat. Rev. Mol. Cell Biol. 6, 21–31. Guelen, L., Pagie, L., Brasset, E., Meuleman, W., Faza, M. B., Talhout, W., Eussen, B. H., de Klein, A, Wessels, L., de Laat, W., and van Steense, B. (2008). Domain organization of human chromosomes revealed by mapping of nuclear lamina interactions. Nature 453, 948–951. Hale, C. M., Shrestha, A. L., Khatau, S. B., Stewart-Hutchinson, P. J., Hernandez, L., Stewart, C. L., Hodzic, D., and Wirtz, D. (2008). Dysfunctional connections between the nucleus and the actin and microtubule networks in laminopathic models. Biophys. J. 95, 5462–5475. Hammond, A. T., and Glick, B. S. (2000). Dynamics of transitional endoplasmic reticulum sites in vertebrate cells. Mol. Biol. Cell 11, 3013–3030. Haque, F., Lloyd, D. J., Smallwood, D. T., Dent, C. L., Shanahan, C. M., Fry, A. M., Trembath, R. C., and Shackleton, S. (2006). SUN1 interacts with nuclear lamin A and cytoplasmic nesprins to provide a physical connection between the nuclear lamina and the cytoskeleton. Mol. Cell. Biol. 26, 3738–3751. Haraguchi, T., Holaska, J. M., Yamane, M., Koujin, T., Hashiguchi, N., Mori, C., Wilson, K. L., and Hiraoka, Y. (2004). Emerin binding to Btf, a death-promoting transcriptional repressor, is disrupted by a missense mutation that causes Emery–Dreifuss muscular dystrophy. Eur. J. Biochem. 271, 1035–1045. Herrmann, H., and Aebi, U. (2004). Intermediate filaments: Molecular structure, assembly mechanism, and integration into functionally distinct intracellular scaffolds. Annu. Rev. Biochem. 73, 749–789 Hofmann, W. A., Stojiljkovic, L., Fuchsova, B., Vargas, G. M., Mavrommatis, E., Philimonenko, V., Kysela, K., Goodrich, J. A., Lessard, J. L., Hope, T. J., Hozak, P., and de Lanerolle, P. (2004). Actin is part of pre-initiation complexes and is necessary for transcription by RNA polymerase II. Nat. Cell Biol. 6, 1094–1101. Holaska, J. M., and Wilson, K. L. (2007). An emerin “proteome”: Purification of distinct emerin-containing complexes from HeLa cells suggests molecular basis for diverse roles including gene regulation, mRNA splicing, signaling, mechanosensing, and nuclear architecture. Biochemistry 46, 8897–8908. Holaska, J. M., Kowalski, A. K., and Wilson, K. L. (2004). Emerin caps the pointed end of actin filaments: Evidence for an actin cortical network at the nuclear inner membrane. PLoS Biol. 2, e231. Jiang, H. P., and Serrero, G. (1992). Isolation and characterization of a full length cDNA coding for an adipose differentiation-related protein. Proc. Natl. Acad. Sci. USA 89, 7856–7860. Kiseleva, E., Rutherford, S., Cotter, L. M., Allen, T. D., and Goldberg, M. W. (2001). Steps of nuclear pore complex disassembly and reassembly during mitosis in early Drosophila embryos. J. Cell Sci. 114, 3607–3618. Kiseleva, E., Drummond, S. P., Goldberg, M. W., Rutherford, S. A., Allen, T. D., and Wilson, K. L. (2004). Actin- and protein-4.1-containing filaments link nuclear pore complexes to subnuclear organelles in Xenopus oocyte nuclei. J. Cell Sci. 117, 2481–2490. Krauss, S. W., Chen, C., Penman, S., and Heald, R. (2003). Nuclear actin and protein 4.1: Essential interactions during nuclear assembly in vitro. Proc. Natl. Acad. Sci. USA 100, 10752–10757. Lammerding, J., Schulze, P. C., Takahashi, T., Kozlov, S., Sullivan, T., Kamm, R. D., Stewart, C. L., and Lee, R. T. (2004). Lamin A/C deficiency causes defective nuclear mechanics and mechanotransduction. . J. Clin. Invest 113, 370–378. Lammerding, J., Fong, L. G., Ji, J. Y., Reue, K., Stewart, C. L., Young, S. G., and Lee, R. T. (2006). Lamins A and C but not Lamin B1 regulate nuclear mechanics. J. Biol. Chem. 281, 25768–25780. Lee, E. H., Hsin, J., Mayans, O., and Schulten, K. (2007). Secondary and tertiary structure elasticity of titin Z1Z2 and a titin chain model. Biophys. J. 93, 1719–1735

5. The nucleoskeleton beyond lamins

117

Lefferts, J. A., Wang, C., Sridharan, D., Baralt, M., and Lambert, M. W. (2009). The SH3 domain of aII spectrin is a target for the fanconi anemia protein, FANCG. Biochemistry 48, 254–263. Liu, Q., Pante, N., Misteli, T., Elsagga, M., Crisp, M., Hodzic, D., Burke, B., and Roux, K. J. (2007). Functional association of Sun1 with nuclear pore complexes. J. Cell Biol. 178, 785–798. Machado, C., and Andrew, D. J. (2000). D-Titin: A giant protein with dual roles in chromosomes and muscles. J. Cell Biol. 151, 639–652. Machado, C., Sunkel, C. E., and Andrew, D. J. (1998). Human autoantibodies reveal titin as a chromosomal protein. J. Cell Biol. 141, 321–333. McDonald, D., Carrero, G., Andrin, C., de Vries, G., and Hendzel, M. J. (2006). Nucleoplasmic -actin exists in a dynamic equilibrium between low-mobility polymeric species and rapidly diffusing popula tions. J. Cell Biol. 172, 541–552. McMahon, L. W., Walsh, C. E., and Lambert, M. W. (1999). Human alpha spectrin II and the fanconi anemia proteins FANCA and FANCC interact to form a nuclear complex. J. Biol. Chem. 274, 32904–32908. McMahon, L. W., Sangerman, J., Goodman, S. R., Kumaresan, K., and Lambert, M. W. (2001). Human spectrin II and the FANCA, FANCC, and FANCG proteins bind to DNA containing psoralen interstrand cross-links. Biochemistry 40, 7025–7034. McMahon, L. W., Zhang, P., Sridharan, D. M., Lefferts, J. A., and Lambert, M. W. (2009). Knockdown of II spectrin in normal human cells by siRNA leads to chromosomal instability and decreased DNA interstrand cross-link repair. Biochem. Biophys. Res. Commun. 381, 288–293. Metral, S., Machnicka, B., Bigot, S., Colin, Y., Dhermy, D., and Lecomte, M.-C. (2009). Alpha II-spectrin is critical for cell adhesion and cell cycle. J. Biol. Chem. 284, 2409–2418. Newport, J., Wilson, K., and Dunphy, W. (1990). A lamin-independent pathway for nuclear envelope assembly. J. Cell Biol. 111, 2247–2259. Offterdinger, M., Schofer, C., Weipoltshammer, K., and Grunt, T. W. (2002). c-erbB-3: A nuclear protein in mammary epithelial cells. J. Cell Biol. 157, 929–939. Olave, I. A., Reck-Peterson, S. L., and Crabtree, G. R. (2002). Nuclear actin and actin-related proteins in chromatin remodeling. Annu. Rev. Biochem. 71, 755–781. Parfenov, V. N., Davis, D. S., Pochukalina, G. N., Sample, C. E., Bugaeva, E. A., and Murti, K. G. (1995). Nuclear actin filaments and their topological changes in frog oocytes. Exp. Cell Res. 217, 385–394 Pederson, T., and Aebi, U. (2002). Actin in the nucleus: What form and what for? J. Struct. Biol. 140, 3–9. Pederson, T., and Aebi, U. (2005). Nuclear actin extends, with no contraction in sight. Mol. Biol. Cell 16, 5055–5060. Percipalle, P., Zhao, J., Pope, B., Weeds, A., Lindberg, U., and Daneholt, B. (2001). Actin bound to the heterogeneous nuclear ribonucleoprotein hrp36 is associated with Balbiani ring mRNA from the gene to polysomes. J. Cell Biol. 153, 229–236. Philimonenko, V. V., Zhao, J., Iben, S., Dingová, H., Kyselá, K., Kahle, M., Zentgraf, H., Hofmann, W. A., de Lanerolle, P., Hozák, P., and Grummt, I. (2004). Nuclear actin and myosin I are required for RNA polymerase I transcription. Nat. Cell Biol. 6, 1165–1172. Qi, H., Rath, U., Ding, Y., Ji, Y., Blacketer, M. J., Girton, J., Johansen, J., and Johansen, K. M. (2005). EAST interacts with megator and localizes to the putative spindle matrix during mitosis in drosophila. J. Cell. Biochem. 95, 1284–1291. Rath, U., Wang, D., Ding, Y., Xu, Y.-Z., Qi, H., Blacketer, M. J., Girton, J., Johansen, J., and Johansen, K. M. (2004). Chromator, a novel and essential chromodomain protein interacts directly with the putative spindle matrix protein Skeletor. J. Cell. Biochem. 93, 1033–1047. Razafsky, D., and Hodzic, D. (2009). Bringing KASH under the SUN: the many faces of nucleo-cytoskeletal connections. J. Cell Biol. on line, jcb.200906068. Ris, H. (1997). High-resolution field-emission scanning electron microscopy of nuclear pore complex. Scanning 19, 368–375. Roux, K. J., Crisp, M. L., Liu, Q., Kim, D., Kozlov, S., Stewart, C. L., and Burke, B. (2009). Nesprin 4 is an outer nuclear membrane protein that can induce kinesin-mediated cell polarization. Proc. Natl. Acad. Sci. USA 106, 2194–2199.

118

Zhixia Zhong et al. Rowat, A. C., Foster, L. J., Nielsen, M. M., Weiss, M., and Ipsen, J. H. (2004). Characterization of the elastic properties of the nuclear envelope. J. R. Soc. Interface. 2, 63–69. Sasseville, A. M.-J., and Langelier, Y. (1998). In vitro interaction of the carboxy-terminal domain of lamin A with actin. FEBS Lett. 425, 485–489. Schermelleh, L., Carlton, P. M., Haase, S., Shao, L., Winoto, L., Kner, P., Burke, B., Cardoso, M. C., Agard, D. A., Gustafsson, M. G. L., Leonhardt, H., and Sedat, J. W. (2008). Subdiffraction multicolor imaging of the nuclear periphery with 3D structured illumination microscopy. Science 320, 1332–1336. Schirmer, E. C., and Gerace, L. (2004). The stability of the nuclear lamina polymer changes with the composition of lamin subtypes according to their individual binding strengths. J. Biol. Chem. 279, 42811–42817. Simon, D. N., Zastrow, M. S., and Wilson, K. L. (2010). Direct actin binding to A- and B-type lamin tails, actin filament bundling by the lamin A tail and potential lamin A precursor auto-inhibition. Nucleus, 1, 1–9. Smetana, K., Steele, W., and Busch, H. (1963). A nuclear ribonucleoprotein network. Exp. Cell Res. 31, 198–201. Sridharana, D., Brown, M., Lambert, W. C., McMahon, L. W., and Lambert, M. W. (2003). Nonerythroid alpha-II spectrin is required for recruitment of FANCA and XPF to nuclear foci induced by DNA interstrand cross-links. J. Cell Sci. 116, 823–835. Sridharana, D. M., McMahona, L. W., and Lambert, M. W. (2006). II-Spectrin interacts with five groups of functionally important proteins in the nucleus. Cell Biol. Int. 30, 866–878. Stewart-Hutchinson, P. J., Hale, C. M., Wirtz, D., and Hodzic, D. (2008). Structural requirements for the assembly of LINC complexes and their function in cellular mechanical stiffness. Exp. Cell Res. 314, 1892–1905. Stricker, J., Falzone, T., and Gardel, M. L. (2010). Mechanics of the F-actin cytoskeleton. J. Biomech. 43, 9–14. Theret, D. P., Levesque, M. J., Sato, M., Nerem, R. M., and Wheeler., L. T. (1988). The application of a homogeneous half-space model in the analysis of endothelial cell micropipette measurements. J. Biomech. Eng. 110, 190–199. Tsai, M.-Y., Wang, S., Heidinger, J. M., Shumaker, D. K., Adam, S. A., Goldman, R. D., and Zheng, Y. (2006). A mitotic lamin B matrix induced by RanGTP required for spindle assembly Science 311, 1887–1893. Tzur, Y. B., Wilson, K. L., and Gruenbaum, Y. (2006). SUN-domain proteins: “Velcro” that links the nucleoskeleton to the cytoskeleton. Nat. Rev. Mol. Cell Biol. 7, 782–788. Walker, D. L., Wang, D., Jin, Y., Rath, U., Wang, Y., Johansen, J., and Johansen, K. M. (2000). Skeletor, a novel chromosomal protein that redistributes during mitosis provides evidence for the formation of a spindle matrix. J Cell Biol. 151, 1401–1411. Wang, N., Butler, J., and Ingber, D. (1993). Mechanotransduction across the cell surface and through the cytoskeleton. Science 260, 1124–1127. Wasser, M., and Chia, W. (2000). The EAST protein of Drosophila controls an expandable nuclear endoskeleton. Nat. Cell. Biol. 2, 269–275. Wen, B., Wu, H., Shinkai, Y., Irizarry, R. A., and Feinberg, A. P. (2009). Large histone H3 lysine 9 dimethylated chromatin blocks distinguish differentiated from embryonic stem cells. Nat. Genet. 41, 246–250. Wilhelmsen, K., Litjens, S. H., Kuikman, I., Tshimbalanga, N., Janssen, H., van den Bout, I., Raymond, K., and Sonnenberg, A. (2005). A. Nesprin-3, a novel outer nuclear membrane protein, associates with the cytoskeletal linker protein plectin. J. Cell Biol. 171, 799–810. Worman, H. J., and Courvalin, J.-C. (2004). How do mutations in lamins A and C cause disease? J. Clin. Invest. 113, 349–351. Young, K. G., and Kothary, R. (2005). Spectrin repeat proteins in the nucleus. BioEssays 27, 144–152. Zastrow, M. S., Vlcek, S., and Wilson, K. L. (2004). Proteins that bind A-type lamins: Integrating isolated clues. J. Cell Sci. 117, 979–987. Zastrow, M. S., Flaherty, D. B., Benian, G. M., and Wilson, K. L. (2006). Nuclear titin interacts with A- and B-type lamins in vitro and in vivo. J. Cell Sci. 119, 239–249.

5. The nucleoskeleton beyond lamins

119

Zeng, C., He, D., and Brinkley, B. R. (1994). Localization of NuMA protein isoforms in the nuclear matrix of mammalian cells. Cell Motil. Cytoskel. 29, 167–176. Zhang, Q., Skepper, J. N., Yang, F., Davies, J. D., Hegyi, L., Roberts, R. G., Weissberg, P. L., Ellis, J. A., and Shanahan, C. M. (2001). Nesprins: A novel family of spectrin-repeat-containing proteins that localize to the nuclear membrane in multiple tissues J. Cell Sci. 114, 4485–4498. Zhang, Q., Ragnauth, C., Greener, M. J., Shanahan, C. M., and Roberts, R. G. (2002). The nesprins are giant actin-binding proteins, orthologous to Drosophila melanogaster muscle protein MSP-300. Genomics 80, 473–481. Zhang, Q., Ragnauth, C. D., Skepper, J. N., Worth, N. F., Warren, D. T., Roberts, R. G., Weissberg, P. L., Ellis, J. A., and Shanahan, C. M. (2005). Nesprin-2 is a multi-isomeric protein that binds lamin and emerin at the nuclear envelope and forms a subcellular network in skeletal muscle. J. Cell Sci. 118, 673–687. Zhong, Z., Chang, S. A., Kalinowski, A., Wilson, K. L., and Dahl, K. N. (2010). The spectrin-like domains of nesprin-1a are over-stabilized by the evolutionarily conserved “adaptive” domain. Cell. Mol. Bioeng., 3, 139–150. Zhong, Z., Ribeiro, A.J.S., Simon, D.N., Wilson, K.L., Dahl, K.N. aII-spectrin is an essential ‘spring’ component of the nucleoskeleton, submitted manuscript.

CHAPTER 6

Altered Mechanical Properties of the Nucleus in Disease Maria Lucia Lombardi and Jan Lammerding Department of Medicine, Cardiovascular Division, Brigham and Women’s Hospital/Harvard Medical School, Boston, Massachusetts 02115

Abstract I. Introduction II. Rationale III. Materials A. Preparation of Isolated Nuclei B. Micropipette Aspiration C. Atomic Force Microscopy D. Cell Compression E. Substrate Strain F. Microneedle Manipulation G. Intranuclear Microrheology H. Image Acquisition I. Analysis IV. Methods A. Micropipette Aspiration Experiments B. Atomic Force Microscopy Experiments C. Cell Compression Experiments D. Substrate Strain Experiments E. Microneedle Manipulation Experiments F. Microbead Microrheology Experiments G. Alternate Methods V. Results and Discussion VI. Summary Acknowledgments References

METHODS IN CELL BIOLOGY, VOL. 98 Copyright 2010 Elsevier Inc. All rights reserved.

121

978-0-12-381009-0

DOI: 10.1016/S0091-679X(10)98006-0

122

Maria Lucia Lombardi and Jan Lammerding

Abstract In eukaryotic cells, the nucleus is the largest and most rigid organelle. Therefore, its physical properties contribute critically to the biomechanical behavior of cells, e.g., during amoeboid migration or perfusion through narrow capillaries. Further more, it has been speculated that nuclear deformations could directly allow cells to sense mechanical stress, e.g., by modulating the access of specific transcription factors to their binding sites. Defects in nuclear mechanics have also been reported in a variety of muscular dystrophies caused by mutations in nuclear envelope proteins, indicating an important role in the maintenance of cells in mechanically stressed tissue. These findings have prompted the growing field of nuclear mechanics to develop advanced experimental methods to study the physical proper ties of the nucleus as a function of nuclear structure and organization, and to understand its role in physiology and disease. These experimental techniques include micropipette aspiration, atomic force microscopy of isolated nuclei, cellular strain and compression experiments, and microneedle manipulation of intact cells. These experiments have provided important insights into the mechanical behavior of the nucleus under physiological conditions, the distinct mechanical contributions of the nuclear lamina and interior, and how mutations in nuclear envelope proteins associated with a variety of human diseases can cause distinct alterations in the physical properties of the nucleus and contribute to the disease mechanism. Here, we provide a brief overview of the most common experimental techniques and their application and discuss the implication of their results on our current understanding of nuclear mechanics.

I. Introduction With an average diameter of 10–20 µm for mammalian cells, the nucleus is typically the largest organelle in eukaryotic cells. Based on its size, and the finding that it is approximately 5- to 10-fold stiffer than the surrounding cytoskeleton (Dahl et al., 2005; Guilak et al., 1999), the nucleus plays a critical role in the overall mechanical behavior of the cell. The interior of the nucleus is separated from the cytoskeleton by the nuclear envelope, a component that significantly contributes to the mechanical properties of the nucleus (Dahl et al., 2005; Rowat et al., 2005). The nuclear envelope consists of two lipid bilayers, the inner nuclear membrane (INM) and outer nuclear membrane (ONM), and a filamentous protein network, the nuclear lamina, which underlies the nuclear membrane. The ONM and INM join at the nuclear pore complexes and enclose the perinuclear space (PNS). Importantly, the ONM, and consequently the PNS, are continuous with the endoplasmic reticulum. Therefore, membrane proteins can laterally diffuse between the INM, the ONM, and the endoplasmic reticulum, although crossing from the ONM to the INM at the nuclear pores imposes some restrictions on the size of the proteins in the INM.

6. Mechanical Properties of the Nucleus

123

The INM contains about 80 integral proteins, including lamin B receptor, MAN1, and emerin (Schirmer et al., 2003). These proteins are retained at the INM through the interactions with chromatin and/or proteins on the nuclear interior such as lamins. Underlying the INM is the nuclear lamina, composed mostly of type V intermediate filaments, known as lamins. Mammalian cells express two types of lamins, the A-type lamins and the B-type lamins. The major A-type lamins, lamins A and C (henceforth referred to as lamin A/C), are expressed in most but not all differentiated cells (Schirmer and Gerace, 2004; Stewart and Burke, 1987). The major B-type lamins, lamins B1 and B2, are expressed in all cells and throughout development (Broers et al., 1997; Goldman et al., 2002). Within the nucleus, chromatin organizes into densely packed heterochromatin and loosely packed euchromatin, often transcriptional inactive and active regions, respectively. The heterochromatin is often located under the nuclear envelope and around the nucleo lus, while the euchromatin is generally found in the nuclear interior and at nuclear pores. Outside the nucleus, nesprins physically connect the nucleus to the cytoske leton by interacting with SUN proteins (INM proteins) across the PNS and by binding to cytoskeletal proteins, i.e., actin filaments, intermediate filaments, and microtubules, with some of these interactions mediated through linker proteins such as plectin (Crisp et al., 2006). SUN proteins are retained at the INM through interactions with lamins, emerin, chromatin, and other, yet to be identified nuclear envelope proteins. Thus, the nesprin/SUN protein complex creates a physical link between the cytoskeleton and the nucleus, which is referred to as the LINC complex (Linker of Nucleoskeleton and Cytoskeleton) (Crisp et al., 2006). The mechanical behavior of the nucleus is determined by the nuclear lamina and the nuclear interior; the specific contribution of each of these components may depend on the particular mechanical conditions (e.g., tension or compression), the cell type, differentiation state, chromatin configuration, and other factors; the lipid inner and other nuclear membranes are thought to play only a minor role in the mechanical behavior of the nucleus. The nuclear lamina is an elastic, load-bearing component, essential in providing structural and mechanical support to the entire nucleus. Changes in the nuclear lamina, such as the depletion or alterations of lamins, in particular A-type lamins, have been suggested to modulate nuclear mechanics (Lammerding et al., 2006). Studies in lamin-depleted nuclei have shown fragile nuclei in Xenopus egg extracts (Newport et al., 1990) and mechanically softer and more fragile nuclei in lamin A/C-deficient mouse embryo fibroblasts (MEFs) (Lam merding et al., 2004). In the nuclear interior, nucleoplasmic lamins, other nuclear matrix proteins, and chromatin provide structure and mechanical stability to the nucleus. Importantly, these are not always independent factors. For example, loss of lamin A/C or specific lamin A mutations can cause alterations in chromatin organization, characterized by the loss of peripheral heterochromatin. Additionally, because the nucleus is physically coupled to the cytoskeleton, the transmission of external mechanical forces into the nuclear interior could be altered by changes in the nuclear lamina and/or chromatin organization, thereby affecting nuclear deformation under externally applied loads.

124

Maria Lucia Lombardi and Jan Lammerding

Several human diseases have been linked to defects or mutations in nuclear envelope proteins, particularly in the LMNA gene encoding A-type lamins (Bione et al., 1994; Bonne et al., 1999; Cao and Hegele, 2000; De Sandre-Giovannoli et al., 2002; Fatkin et al., 1999; Muchir et al., 2000; Novelli et al., 2002; Shackleton et al., 2000). These diseases, collectively referred to as laminopathies, include Emery–Dreifuss muscular dystrophy, dilated cardiomyopathy, Dunnigan type familial partial lipodystrophy, and Hutchinson–Gilford progeria syndrome. Cells harvested from patients suffering from laminopathys are often characterized by abnormally shaped nuclei with altered mechanical properties, as discussed in the following sections. In addition, nuclear mechanical measurements have detected that lamins and changes in chromatin structure are associated with specialized cell types, such as stem cells during differentiation and cancer cells (Constantinescu et al., 2006; Friedl and Wolf, 2009; Gotzmann and Foisner, 2006; Hou et al., 2009; Hudson et al., 2007; Konety and Getzenberg, 1999; Li et al., 2009; Pajerowski et al., 2007; Scaffidi and Misteli, 2008; Spencer et al., 2001; Wolf and Friedl, 2006; Zink et al., 2004). In this chapter, we describe several experimental techniques that have been developed to probe the physical properties of the nucleus and discuss the physiological implications of altered nuclear mechanical properties.

II. Rationale Studying the mechanical properties of the nucleus serves several important pur poses. Firstly, it reveals important insights into the relationship between nuclear structural organization and cellular function. For example, micropipette aspiration studies by Dahl et al. (2005) indicate that the elastic nuclear lamina can serve as a “shock-absorber” in cells, thereby protecting the nuclear interior. Additionally, a better understanding of the ultrastructural deformations that occur within the nucleus under applied mechanical load can also help unravel some of the unsolved puzzles of mechanotransduction signaling. For instance, mechanically induced changes in chro matin could directly modulate gene expression by altering access to transcription factors. Furthermore, differences in nuclear mechanical properties can serve as indi cators for changes in chromatin structure that are associated with the transition from an open to a closed state, and other important changes in nuclear and cellular structure. One striking example is the recent demonstration that differentiating stem cells develop increasingly stiffer nuclei (Pajerowski et al., 2007). Lastly, comparison of the nuclear mechanical properties of cells taken from patients or mouse models of diseases caused by mutations in nuclear envelope proteins (e.g., Emery–Dreifuss muscular dystrophy) with normal controls can reveal further details on the relevance of nuclear structure and mechanics on normal cellular function and offer clues into the pathobiology of the disease. For all of these applications, sophisticated experimental techniques and appropriate analysis methods are required to obtain important information on nuclear structure and mechanics.

6. Mechanical Properties of the Nucleus

125

III. Materials A. Preparation of Isolated Nuclei Several protocols are available for the isolation of nuclei from mammalian cells (Caille et al., 2002; Dahl et al., 2005; Deguchi et al., 2005). In general, the plasma membrane of a cell is mechanically disrupted with hypotonic swelling and subse quent mechanical shearing (e.g., Dounce homogenizer) or chemically disrupted with mild detergents, such as digitonin. Ultracentrifugation through a sucrose gradient separates the isolated nuclei from cellular debris and nondisrupted cells. The following materials are necessary for isolating nuclei by mechanical disruption (Dahl et al., 2005): 1. Hypotonic buffer: 10 mM Hepes and 1 mM dithiothreitol, pH 7.5. [Suggested modification by the authors to further improve the protocol: Addition of protease inhibitors to the hypotonic buffer, such as Protease Inhibitor Cocktail (Sigma, P8340) and phenylmethanesulfonylfluoride.] 2. Sucrose gradient: STKMC (1.25 M sucrose, 250 mM Tris, pH 7.6, 125 mM KCl, 15 mM MgCl2, 15 mM CaCl2, 10 mg/ml of each protease inhibitor, pepstatin and leupeptin) (Sigma-Aldrich) and 2.3 M sucrose in TKMC (50 mM Tris, pH 7.6, 25 mM KCl, 3 mM MgCl2, 3 mM CaCl2, 2 mg/ml of each leupeptin and pepstatin). 3. Dounce homogenizer, tight pestle (Wheaton Scientific, Millville, NJ, USA). 4. Ultracentrifuge (Optima TLX, Bechman Coulter, Palo Alto, CA, USA). 5. Isolated nuclei are resuspened in 200 ml TKMC or 200 ml Tris (10 mM Tris–HCl, pH 7.6, 2 mg/ml each of pepstatin and leupeptin and 1% bovine serum albumin).

B. Micropipette Aspiration 1. Borosilicate capillaries, 1 mm inner diameter (World Precision Instruments, Sarasota, FL, USA). 2. Pipette puller (Sutter Instruments Company, Novato, CA, USA). 3. Custom-made micropipette setup consisting of manometer system, pressure transducer, and micromanipulators (Dahl et al., 2005; Evans, 1989; Henriksen and Ipsen, 2004; Pajerowski et al., 2007).

C. Atomic Force Microscopy 1. Custom-built or commercially available atomic force microscope such as the MFP 3D-BIO AFM (Asylum Research, Santa Barbara, CA, USA) or the Bioscope AFM (Veeco, Woodbury, NY, USA). 2. Flexible cantilever with a small and very sharp pyramidal or spherical probe tip (e.g., Veeco). Cantilevers should be suitable for forces up to 10 nN.

126

Maria Lucia Lombardi and Jan Lammerding

D. Cell Compression 1. Microindentor or microplate made from regular borosilicate bars using a micropipette puller (Sutter Instrument Company, Broers et al., 2004; Peeters et al., 2003). 2. A probe attached to micromanipulators, piezoelectric system, and force transducer (Broers et al., 2004; Peeters et al., 2003). E. Substrate Strain 1. Custom-made strain devices (Caille et al., 1998; Lammerding et al., 2004, 2005) or commercially available strain devices, e.g., Stageflexer (Flexcell International, Hillsborough, NC, USA) or STREX Mechanical Cell Strain Instrument (B-Bridge International, Inc., Mountain View, CA, USA). 2. Thin, transparent silicone membranes (Silastic, Specialty Mfg, Pineville, NC, USA) that can be coated with fibronectin or other extracellular matrix proteins of interest, or precoated silicone membranes (Flexcell International). F. Microneedle Manipulation 1. Borosilicate capillaries (World Precision Instruments), pulled to tip diameters of 1–3 µm with commercial pipette pullers (Sutter Instrument). 2. Micromanipulator (e.g., InjectMan NI 2 micromanipulator, Eppendorf, Hauppauge, NY, USA). G. Intranuclear Microrheology 1. Micromanipulator and microinjector (e.g., InjectMan NI 2 micromanipulator and the FemtoJet microinjector, Eppendorf, Hauppauge, NY, USA).

1. Active Rheology 1. MyOne magnetic bead, 0.5 µm radius (Dynal Biotech Inc., Lake Sucesss, NY, USA). 2. A micropipette TIP04TW1F, 0.4 µm inner diameter (World Precision Instruments) to microinject magnetic beads into the nucleus. 3. Custom-built magnetic tweezer system, consisting of magnetic core, copper coils wrapped around the core, and power supply to control the current through the coils (de Vries et al., 2007).

2. Passive Rheology 1. Fluorescent polystyrene nanoparticles, 100 nm in diameter (Invitrogen, Carlsbad, CA, USA). 2. Borosilicate microneedles, 0.3 µm inner diameter and 0.4 µm outer diameter (World Precision Instruments) to microinject nanospheres into the nucleus.

6. Mechanical Properties of the Nucleus

127

H. Image Acquisition For each of the methods, images are acquired with (inverted) epifluorescence microscopes that have a digital charge-coupled device camera suitable for fluorescence microscopy (e.g., CoolSNAP HQ, Roper Scientific, Tucson, AZ, USA) or laser scanning confocal microscope. For the atomic force microscopy (AFM) techniques, images can also be acquired with the AFM tip by raster scanning. While confocal microscopes generally come with their own image acquisition software, several image acquisition packages that can also control the microscope and other hardware acces sories are available for epifluorescence microscopes, e.g., IPLab (Becton Dickinson, Franklin Lakes, NJ, USA), Metamorph (Molecular Devices, Sunnyvale, CA, USA), or Image Pro (Media Cybernetics, Bethesda, MD, USA). I. Analysis Image analysis software, e.g., MATLAB (Mathworks, Natick, MA, USA), ImageJ (National Institutes of Health, Bethesda, MD, USA) or any other image analysis software, is required for each of the methods to quantitatively analyze and interpret the deformations of the nuclei both in isolation and in intact, living cells.

IV. Methods A number of experimental techniques have been developed to measure and character ize the physical properties of the nucleus. Generally, intact cells or isolated nuclei are mechanically perturbed with a known force or deformation (substrate strain or cell compression), while simultaneously imaging the induced nuclear deformation. Analysis of the induced nuclear deformation relative to the cellular deformation or the applied force can then be used to gain insights into the nuclear mechanical properties and the underlying structures of the nucleus. Experimental techniques vary from cellular strain devices, which quantify gross nuclear deformations under strain in living cells, to micropipette aspiration experiments, which exclusively probe nuclear envelope mechanics in isolated nuclei. The advantages of studying nuclei within intact living cells include maintaining normal nuclear and cytoskeleton architecture and preserving the correct chemical composition of the nucleoplasm and cytoplasm. The major limita tion is that direct measurements of nuclear mechanical properties are not possible since the nucleus is surrounded by the cytoskeleton. The isolation of nuclei, on the other hand, allows for these measurements to be directly quantified and for forces to be applied directly to the nucleus rather than reallocate to the cytoskeleton. However, isolated nuclei can be damaged easily, particularly when examining more fragile nuclei carrying lamin mutations. Furthermore, chromatin and nuclear structures are strongly dependent on the appropriate buffer conditions during the experiment (Dahl et al., 2005), such that experimental results may vary with the buffer used. In the following sections, we will describe and discuss some of the most commonly used experimental methods for probing nuclear mechanics in isolated nuclei and in nuclei within intact cells.

128

Maria Lucia Lombardi and Jan Lammerding

A. Micropipette Aspiration Experiments Micropipette aspiration is a well-established technique to measure the mechanical properties of the nuclear envelope and nuclear interior (Dahl et al., 2005; Pajerowski et al., 2007; Rowat et al., 2006). In this technique, a single nucleus—either isolated or within a cell whose cytoskeleton has been disruption by depolymerizing drugs—is gently aspirated into a glass micropipette at a precisely controlled aspiration pressure, inducing large deformations of the nucleus into the pipette which are monitored and quantified by live microscopy. The micropipette setup consists of glass capillaries pulled into micropipettes with inner diameters ranging from 1 to 8 µm. The micropip ettes are connected to a custom-made manometer system, pressure transducer, and micromanipulators (Dahl et al., 2005; Evans, 1989; Henriksen and Ipsen, 2004; Pajerowski et al., 2007). Nuclear isolation buffer is aspirated into the micropipette prior to nuclear aspiration. Aspiration pressures ranging from 1 to 7 kPa are then used to draw the nucleus into the mouth of the pipette, inducing deformations of the isolated nucleus into the micropipette (Fig. 1). Brightfield microscopy captures images of the deformed nuclei. Images of the nuclear deformations are then converted into quantitative mea surements by normalizing the aspirated length of the nucleus (the tongue length within the micropipette) to the micropipette radius and the aspiration pressure. The creep compliance, J, of the nucleus can be calculated from these measurements using Eq. (1): J ðtÞ =

2 1 DLðtÞ 3 P Rp

ð1Þ

where is the shape factor, depending on micropipette wall thickness (here, = 2.1; Theret et al., 1988), P is the constant applied aspiration pressure, DL(t) is the aspirated length, and Rp is the pipette radius (Dahl et al., 2005; Theret et al., 1988). To eject

(A) Rp L

(B)

Fig. 1 Micropipette aspiration technique. (A) Schematic of a nucleus deformed by micropipette aspiration. (B) An isolated nucleus partially aspirated into a micropipette.

129

6. Mechanical Properties of the Nucleus

the nucleus from the micropipette, a positive pressure is applied. Nuclear envelope deformations can also be directly analyzed during aspiration by the fluorescence intensity of green fluorescent protein (GFP)–lamins incorporated into the nuclear envelope and using high resolution confocal microscopy (Dahl et al., 2005; Rowat et al., 2005, 2006). The combination of these techniques can be applied to isolated nuclei and nuclei in intact living cells to determine how specific nuclear proteins respond to external mechanical forces. An advantage of the micropipette aspiration technique is the capability to directly measure nuclear mechanics in isolated nuclei. The main limitation of this technique is that it requires isolated nuclei or cytoskeleton disruption if studying nuclei in intact cells. Isolating nuclei can potentially damage the nuclei, altering nuclear and chromatin struc ture and nuclear–cytoskeleton coupling. In addition, the nuclear mechanical properties of isolated nuclei can be affected by buffer conditions, and the load application may not resemble physiological conditions found, e.g., in muscle. B. Atomic Force Microscopy Experiments Nuclear mechanical properties can also be probed by atomic force microscopy (AFM), which enables imaging and quantifying local indentations of isolated nuclei at nanometer spatial resolution while providing precise control over the applied force. Briefly, a flexible cantilever with a small and very sharp pyramidal or spherical probe tip, raster scans an isolated nucleus adhered to a rigid surface. In order to immobilize the nuclei without increasing the tension in the nuclear envelope structures (Hategan et al., 2003), the rigid substrate is coated with reagents to promote adhesion, e.g., poly-L-lysine (Dahl et al., 2005). To study the mechanical properties of isolated nuclei, AFM is typically operated in contact mode. For nuclear indentation experiments, different probe geometries are available. Sharp tipped pyramidal or conical tips (tip radius 30 nm) cause local deformations and are thus subjected to local variations in nuclear composition. Tipless probes or probes with small spherical microscopic beads (diameter 1–5 µm) attached to the tip are more suitable to determine the bulk material properties of the nucleus, as they distribute the applied force over a wider area. The cantilever acts as an elastic spring, so that the applied force can be determined by the stiffness of the cantilever and the deflection of the cantilever arm from the resting position. The local indentation of the nucleus is measured by the vertical displacements of the cantilever from the point of contact and then subtracting off the cantilever deflection. For a spherical tip and assuming a semi-infinite elastic half-space, the force– indentation relationship is given by Eq. (2): F=

4 E R1=2 3=2 3 1 2

ð2Þ

where F is the applied force, E is the Young’s modulus of the nucleus, is the Poisson’s ratio (0.5 for most biological materials), R is the radius of the spherical tip, and is the indentation depth. For a conical tip, and again assuming a semi-infinite elastic half-space, the force–indentation relationship is given by Eq. (3):

130

Maria Lucia Lombardi and Jan Lammerding

F=

2 E tan 2 1 2

ð3Þ

where F is the applied force, is the angle of the conical tip (typically 18–35°), E is the Young’s modulus of the nucleus, is the Poisson’s ratio (0.5 for most biological materials), and is the indentation depth. The Young’s modulus, representing the nuclear stiffness, is then determined by fitting Eq. (2) or (3) to the experimentally obtained force–indentation curves. The main advantage of the AFM technique is that it provides extremely accurate resolution of the applied forces and the induced deformations in isolated nuclei. In addition, images and force–indentation curves can be acquired in aqueous conditions with physiological temperatures and pH conditions; therefore, fixations, staining, or labeling are not necessary. Limitations to this technique include interpretation of the results, the assumption that the nucleus is a homogeneous, isotropic, linear elastic material for the above equations, the effect of the underlying stiff substrate for large indentations and the requirement to isolate the nucleus (see comments under micro pipette aspiration). C. Cell Compression Experiments Cellular compression devices have been used to measure nuclear mechanics by studying the sensitivity and deformations of isolated nuclei and single intact cells during mechanical uniaxial compressions (Broers et al., 2004; Caille et al., 2002; Thoumine and Ott, 1997). In these experiments, isolated nuclei or cells are cultured on coverslips and a microindentor or microplate is placed above a nucleus or single cell, parallel to the coverslip. For the microindentor method, isolated nuclei or cells are compressed with a glass probe with a tip diameter of 500 µm. The probe is controlled by micromanipulators, a piezoelectric system and force transducer allowing for controlled displacement of the glass indenter probe and for directly determining the magnitude of the applied compression force. For the microplate experiments, a rigid microplate with tip dimensions of 3 30 µm2 and a stiffness greater than 107 N/µm and a flexible microplate with tip dimensions of 1 130 µm2 and stiffness of 109–108 N/µm are made from regular borosilicate bars using a micropipette puller. Briefly, isolated nuclei or cells are adhered to the rigid microplate which is coated with extracellular matrix proteins (e.g., fibronectin) or glutaraldehyde and silane to create a covalent link between the glass and the cell or nucleus generating a strong adhesion. The microplates are then turned at a 90° angle to allow for side viewing of the cell or nucleus between the two microplates. The thinner, more flexible microplate is then used to compress the cell or nucleus and the deflection of the flexible microplate is observed and recorded using an inverted microscope to determine the magnitude of the applied compression force. One advantage of the cell compression method is that it applies a compressive force uniformly to an entire isolated nuclei or cell unlike AFM with conical- or pyramid-shaped tips which applies a highly localized force. Further more, this technique is suitable to probe nuclear mechanics in intact cells in their

131

6. Mechanical Properties of the Nucleus

normal cellular environment and does not necessarily require firm cell adhesion to the substrate. Nevertheless, the nuclear deformations observed within intact cells are affected by the cytoskeleton surrounding the nucleus. D. Substrate Strain Experiments Nuclear deformations have been used to quantify nuclear stiffness in living cells relative to the stiffness of the surrounding cytoskeleton. In these experiments, uniform uniaxial or biaxial strain is applied to living cells cultured on elastic substrates while monitoring induced nuclear and cellular deformations. Nuclear stiffness can then be inferred from the induced nuclear deformation relative to the applied substrate strain (Fig. 2). In general, cells are plated at low densities on thin, transparent silicone membranes coated with extracellular matrix proteins (e.g., fibronectin) that maximize cell adhesion and spreading. The membranes are then placed on a microscope-mounted strain device and subjected to uniform biaxial or uniaxial strain. Biaxial (isotropic) strain application allows for uniform cellular and nuclear deformations, since the applied strain is the same in all directions. Therefore, results are independent of cell orientation relative to the strain direction. In uniaxial strain, the strain is applied in only

(A) No strain 25% strain (stiff nucleus) 25% strain (soft nucleus)

Full-strain

Prestrain

(B)

Lmna+/+

Lmna–/–

Fig. 2 Substrate strain experiment. (A) Schematic of cellular strain on adherent cells. A stiff nucleus deforms less under strain than a softer nucleus. (B) A pair of images of nuclei from a wild-type (Lmnaþ/þ) mouse embryo fibroblast and of a lamin A/C-deficient (Lmna–/–) mouse embryo fibroblast with fluorescently labeled chromatin before (prestrain) and during (full-strain) strain application. The applied membrane strain was 25%. While the wild-type nucleus showed only a small increase in size, the nucleus in the lamin A/C deficient cell enlarged significantly more as a result of the strain application, suggesting a softer nucleus.

132

Maria Lucia Lombardi and Jan Lammerding

one direction, either the x- or y-direction, which often more closely resembles physio logical conditions, and also allows for larger extension without damaging the cells. Both methods are suitable to measure nuclear mechanical properties, however, depend ing on the cell type, e.g., elongated cells versus isotropic cell shape, one method may be preferred over the other. Strain devices are either custom-made (Caille et al., 1998; Lammerding et al., 2004, 2005) or commercially available, e.g., Stageflexer, and mounted on an inverted epifluorescence or confocal microscope, to simultaneously acquire images during, before, and after strain application. Membrane strain is trans mitted to the cytoskeleton through focal adhesion sites primarily located at the cell periphery, so the cytoskeletal strain closely matches the applied membrane strain. Induced nuclear deformations are analyzed by tracking fluorescently labeled nuclei before, during, and after strain application and normalized to membrane strain to compensate for small variations in the applied membrane strain. This method of strain induction allows for the application and analysis of nuclear mechanics in a large number of cells at the same in normal cell environments, without having to isolate nuclei, thus maintaining their cellular and nuclear structure and architecture. Another advantage of this technique is that the specific contribution of nuclear envelope proteins to nuclear stiffness and fragility can be studied. However, nuclear deformation and stiffness is dependent on cell adhesion to the substrate and can be affected by the surrounding cytoskeleton. E. Microneedle Manipulation Experiments Microneedle manipulation assays can be used to directly probe nuclear and cytos keletal mechanics by applying highly localized strain to the cytoskeleton and the nucleus in living adherent cells (Maniotis et al., 1997). Induced nuclear and cytoske leton deformations can be quantified in cells with fluorescently labeled cytoskeletal proteins (such as mitochondria) plated on glass bottom cell culture dishes. Since highly spread cells are very stiff and less amenable to micromanipulation (Maniotis et al., 1997), these dishes are commonly coated with low concentrations of fibronectin (0.1 µg/ml). Glass capillaries are pulled into microneedles with tip diameters of 1–3 µm. The microneedle, held by a micromanipulator, is then carefully inserted into the cell at a specific distance, e.g., 5 µm from the nuclear periphery, and moved a fixed distance toward the cell periphery (Fig. 3). Likewise, the microneedle can also be inserted into the nucleus, a specific distance from the nuclear envelope and moved a fixed distance toward the cell periphery. Induced nuclear and cytoskeletal deformations are recorded at one frame every 10 s. Induced nuclear deformations can be directly computed from changes in nuclear dimensions in phase-contrast images or by fluor escently labeling the nucleus with cell-permeable DNA stains (e.g., Hoechst 33342). Similarly, cytoskeletal strain is quantified by tracking fluorescently labeled cytoskele tal proteins using previously developed algorithms (Lammerding et al., 2003). Nuclear and cytoskeletal mechanics are thus determined by measuring localized nuclear and cytoskeletal strains, which can yield insights into the connection between the nucleus and cytoskeleton and how alterations in nuclear mechanics can cause a disruption in

133

6. Mechanical Properties of the Nucleus

(A)

(C)

(E)

(B)

(D)

(F)

Fig. 3 Nuclear and cytoskeletal force application by microneedle manipulation. (A–B) A microneedle is inserted into the cytoskeleton and subsequently moved a fixed distance from the nucleus toward the cell periphery. Induced cytoskeletal and nuclear displacements are quantified by tracking fluorescently labeled mitochondria and nucleus using a custom-written normalized cross-correlation algorithm and then plotted as displacement vectors. (C–F) Studies from our laboratory data show that cytoskeletal deformations measured by tracking mitochondria closely match those for GFP–actin or GFP–vimentin.

this coupling. Although this technique is a single-cell method and can be invasive to the cell, there are several advantages in using microneedle manipulation. For instance, strain can be applied directly to the cytoskeleton and can be modified varying the distance of the microneedle. The microneedle can also be positioned in the cell with high accuracy, making the experiment very reproducible, and the force application rate, which is directly related to the microneedle speed, can be precisely controlled by the programmed micromanipulator to move between two defined positions. F. Microbead Microrheology Experiments Local mechanical (rheological) properties of intracellular structures, such as the nucleus, can be determined by active or passive microrheology. In active microrehol ogy (magnetic bead microrheology), a magnetic bead is inserted into the nuclear interior (de Vries et al., 2007) and the bead deflection, under a given force applied to the bead by magnetic tweezers, is used to determine nuclear stiffness. In passive microrheology (particle-tracking), submicrometer-sized particles or beads are injected

134

Maria Lucia Lombardi and Jan Lammerding

into the nucleus of cells and the thermal fluctuation (Brownian motion) of the beads is used to determine the stiffness of the nuclear material surrounding the bead (Tseng et al., 2004). In magnetic bead microrheology (active rheology), a magnetic bead (about 0.5 µm radius) is placed within the nucleus with a fine microinjection needle. The response of the bead to an externally applied magnetic force is tracked using an inverted micro scope and displacement amplitudes are then computed as described previously (de Vries et al., 2007). This technique allows application of precisely controlled forces and high-resolution imaging of the induced deformations. The major limitations are the disruptive insertion of the relatively large magnetic bead into the nucleus and the fact that a stiffness measurement is obtained at only a single spot, making it subject to local variations. In particle-tracking (passive) microrheology, nanosized beads (100 nm in dia meter) are microinjected into the nucleus of living cells using a microinjector. The trajectories of the beads are tracked over time by video- or laser-tracking techniques (Tseng et al., 2004). Particle-tracking software is used to analyze the movement of the nanoparticles by calculating the mean squared displacement of single particles, to determine the elasticity of the intranuclear region (Tseng et al., 2002). The main advantage of passive microrheology compared to the active microrheology is that the smaller bead sizes cause less perturbation to the cell or the nucleus. However, even more so than in active rheology, results from this technique can be affected by nanoparticle/bead size. Particles that are not sufficiently entangled in the surrounding nuclear matrix/chromatin can move more freely and result in apparently lower nuclear stiffness values that represent the viscosity of the nucleoplasm rather than that of the surrounding matrix. In addition, the act of microinjecting foreign bodies into the nucleus could perturb nuclear structure and integrity and affect experimental results. G. Alternate Methods Additional experimental methods have been developed to probe the mechanical properties of the nucleus, including time-lapse videomicroscopy and microinjection of large, fluorescently labeled molecules into the nucleus. In time-lapse videomicroscopy, live cells can be imaged with minimal invasiveness in order to visualize the localiza tion and dynamic behavior of proteins in the nucleus and cytoskeleton or to quantify dynamic changes in nuclear shape (Lammerding et al., 2005; Liu et al., 2000). To detect changes in nuclear shape, phase contrast or differential interference contrast (DIC) images acquired at low magnification (e.g., 20) are often sufficient, and the easily detectable nucleoli can serve as fiducial markers to quantify changes in nuclear shape or size over time periods spanning minutes to hours (Lammerding et al., 2005). To probe nuclear architecture and function in living cells, nuclear proteins fused with fluorescent proteins, such as enhanced green fluorescent protein can be introduced into living cells before taking time-lapse images with a confocal or an epifluorescence microscope. Variations of this technique include fluorescence recovery after photobleaching, fluorescence correlation spectroscopy, and photoactivation to visualize and

6. Mechanical Properties of the Nucleus

135

measure the mobility and dynamic interactions of nuclear envelope proteins in living cells (Houtsmuller, 2005; Schmiedeberg et al., 2004; Weidtkamp-Peters et al., 2009; Wiesmeijer et al., 2008). We recently demonstrated that microinjection into the nucleus or cytoplasm can be used to measure nuclear envelope integrity and rupture strength. In this technique, large fluorescently labeled molecules (e.g., dextran with a molecular weight of over 60 kDa) are injected into the nucleoplasm with varying injection pressures and observed over time (Lammerding et al., 2004). Due to its size, and the lack of a nuclear localization sequence, the fluorescently labeled dextran cannot cross an intact nuclear envelope. Thus, in cells with an intact nuclear envelope, the 60-kDa dextran is retained inside the nucleoplasm when it is injected into the nucleus, whereas it can diffuse into the cytoplasm if the nuclear envelope is leaky or has become disrupted due to the injection pressure. To determine whether nuclear envelope integrity is main tained in cells at baseline, i.e., without nuclear injection, fluorescently labeled dextran can be injected into the cytoplasm. In this case, the large dextran molecules will be excluded from the nucleus in cells with intact nuclear envelopes, but diffuse into the nucleus if the nuclear membranes or nuclear pores are leaky.

V. Results and Discussion Over the past 15 years, several groups have developed and applied sophisticated experimental techniques to measure the mechanical properties of the nucleus of eukaryotic cells. These experiments have provided important insights into the physical contribution of the nuclear envelope and the nuclear interior to the deformation behavior of the nucleus under mechanical stress, the dependence of nuclear mechanics on nuclear and cellular structure and organization, and how mutations in nuclear envelope proteins can alter nuclear mechanics. In the following, we will describe these results in more detail and discuss their implications on our current understanding of nuclear mechanics in physiology and disease. Even though the measurements of nuclear stiffness vary between cell types and experimental methods used, numerous studies of nuclear mechanical properties have convincingly demonstrated that the cell nucleus is a stiff structure compared to the surrounding cytoskeleton. For example, the nuclear stiffness of chondrocytes, mea sured by compression, was reported to be 3–4 times stiffer than cytoplasmic stiffness, while the nuclei in endothelial cells are 10 times stiffer than the cytoplasm (Caille et al., 2002; Guilak, 1995). The elasticity of isolated nuclei from endothelial cells was measured by parallel plate compression and found to be 8 kPa, compared to 0.5 kPa for the cytoplasm (Caille et al., 2002); in chondrocytes, the elasticity of nuclei was 1–5 kPa as measured by micropipette aspiration (Guilak et al., 2000). Using another method, microneedle manipulation, Maniotis et al. (1997) reported that the nucleus in intact endothelial cells is very rigid and is estimated to be 9 times stiffer than the cytoplasm. Similar results were obtained by active microrheology, which found that the nuclear interior is also much stiffer than the cytoplasm in HeLa cells (de Vries et al., 2007) and by passive microrheology in 3T3 fibroblasts (Tseng et al., 2004).

136

Maria Lucia Lombardi and Jan Lammerding

Interestingly, recent studies suggest that nuclear stiffness can increase in response to the mechanical environment of a cell. For example, application of fluid shear stress to endothelial cells results in increased nuclear stiffness (Deguchi et al., 2005), possibly due to the upregulation and reorientation of A-type lamins to the nuclear envelope under fluid shear stress (Philip and Dahl, 2008). The nuclear envelope and nuclear interior, both contribute to nuclear stiffness and the overall mechanical behavior of the nucleus. Confocal-imaged microdeformations during micropipette aspiration suggest that the nuclear envelope behaves as an elastic shell comprised of the elastic nuclear lamina and fluid-like membranes, while the nuclear interior, i.e., chromatin and nucleoli, resembles a highly compres sible gel-like, viscous structure (Dahl et al., 2005; Rowat et al., 2005, 2006). In the nuclear envelope, A-type and B-type lamins widely contribute to nuclear mechanics. Lamins A and C are important for the mechanical stiffness of nuclei, as observed by nuclear deformations under strain in MEFs deficient in lamin A/C (Lammerding et al., 2006). Interestingly, MEFs deficient in lamin B1 have normal nuclear mechanics, despite the presence of severe nuclear blebs, suggesting that lamins A/C and lamin B1 form separate structural networks with distinct physical functions (Lammerding et al., 2006; Shimi et al., 2008). Recent AFM studies have demon strated that isolated nuclei of the Xenopus oocyte with overexpressed lamin A have an increase in nuclear rigidity and stiffness and a less degree of nuclear deformation (Schape et al., 2009), whereas previous studies have reported that nuclei of Xenopus oocyte extracts depleted of lamin are extremely fragile (Newport et al., 1990). These studies indicate that A-type and B-type lamins have distinct roles in nuclear mechanics. Lamins also play a role in the organization of chromatin into hetero chromatin and euchromatin and can modulate gene expression (Shimi et al., 2008). Chromatin organization and lamina composition can also contribute to nuclear stiffness in differentiation. Micropipette aspiration of nuclei from human embryonic stem cells revealed highly deformable nuclei that become sixfold stiffer through differentiation (Pajerowski et al., 2007). Since the stem cells lack A-type lamins, the chromatin reorganizes while the lamina is stretched, causing the nucleus to become more deformable. Taken together, these studies strongly suggest that nuclear stiff ness is related to chromatin and lamina organization. A number of the experimental techniques have been applied to characterize the physical properties of the nucleus in disease models, namely diseases associated with mutations in A-type lamins. Over 330 mutations have been identified in the gene-encoding A-type lamins (LMNA) that can cause a variety of human diseases, including Emery–Dreifuss muscular dystrophy, limb–girdle muscular dystrophy, and Hutchinson–Gilford progeria syndrome (HGPS), all commonly referred to as lami nopathies. While the precise molecular mechanisms underlying these diseases remain incompletely defined, the broad disease spectrum suggests multifactorial pathways and overlapping functions of lamins that can be specifically affected by different mutations. At the same time, the many muscle-specific phenotypes indicate that increased nuclear fragility due to a functional loss of lamins could contribute to the progressive loss of cells in mechanically stressed tissue such as muscle. For

6. Mechanical Properties of the Nucleus

137

example, lamin A/C-deficient MEFs have a decreased mechanical stiffness and increased fragility measured by cell compression or when subjected to mechanical strain (Broers et al., 2004; Lammerding et al., 2004). In contrast, nuclei of HGPS patient fibroblasts develop increasingly stiffer nuclei, possibly caused by changes in chromatin organization or accumulation of the nonfarnesylated mutant lamin A (i.e., progerin) at the nuclear lamina (Dahl et al., 2006; Verstraeten et al., 2008). In addition to mutations in the gene encoding A-type lamins, mutations in the inner nuclear envelope protein, emerin, can cause Emery–Dreifuss muscular dystrophy. Alterations in the mechanical properties of MEFs deficient in emerin were observed in time-lapse observations and by micropipette aspiration (Lammerding et al., 2005; Rowat et al., 2006). Similar to results obtained in lamin A/C-deficient cells, emerin deficient cells showed increased rates of cell death in response to repetitive mechan ical strain and decreased activation of mechanosensitive genes, suggesting impaired mechanotransduction signaling as a further contributor to muscular phenotypes in laminopathies. Together, these findings suggest that the nuclear envelope has unique mechanical properties that might contribute to the tissue-specific effects observed in laminopathies; however, it remains unknown to what extent the alterations in nuclear stiffness are due to modifications in the nuclear lamina or in chromatin structure, requiring additional experiments. Alterations in nuclear structure and mechanics can also result in defective nuclear-cytoskeletal coupling and cytoskeletal stiffness, which are important for the transmission of extracellular forces to the nuclear interior. A defective coupling between the nucleus and the cytoskeleton can lead to defects in nuclear orientation and cellular functions as reported in lamin A/C-deficient fibroblasts and laminopathic mouse models (Hale et al., 2008; Houben et al., 2009; Lee et al., 2007). Although most of the techniques to measure nuclear mechanics have been previously applied to gain insights into the disease mechanisms of laminopathies, recent studies have begun to use similar experimental methods to elucidate the role of nuclear mechanics in neutrophil and cancer cell biology (Couzon et al., 2009; Olins et al., 2008, 2009; Wolf and Friedl, 2006; Wolf et al., 2007; Zink et al., 2004). Neutrophils and cancer cells are known to move through tight spaces; neutrophils migrate quickly to areas of infection (Olins et al., 2008, 2009; Sanchez and Wangh, 1999), while cancer cells migrate to normal tissues, spreading the cancer (Couzon et al., 2009; Friedl and Wolf, 2009; Wolf et al., 2007). A deform able nucleus is, therefore, important for the translocation of these cells through tight spaces. Nuclear mechanical properties, i.e., nuclear stiffness, could contribute to nuclear deformability during translocation of these cells. The nuclei of granulocytes, e.g., are lobulated with low expression levels of lamin proteins and associated proteins. In cancer cells, nuclei are misshapen and the organization of heterochro matin and the expression of lamins are altered (Hudson et al., 2007; Zink et al., 2004). These studies indicate that determining the mechanical properties of the nucleus could lead to a better understanding of many cellular functions in normal healthy cells and in disease.

138

Maria Lucia Lombardi and Jan Lammerding

VI. Summary In this chapter, we have presented several of the commonly used methods to probe the nuclear mechanical properties of nuclei in intact living cells and in isolation. Results obtained from these experimental techniques reveal that the nucleus is the stiffest organelle in mammalian cells, and that the nuclear stiffness and the overall mechanical behavior of the cell nucleus are largely determined by the composition of the nuclear lamina and the chromatin organization. In the nuclear lamina, lamins A and C, encoded by the LMNA gene, are the major contributor to nuclear stiffness, and mutations in the LMNA gene can cause defects in nuclear shape and structure and altered nuclear mechanics. Changes in nuclear stiffness are also associated with cellular differentiation and could play important roles in other human diseases such as cancer. Future experiments, which combine the discussed techniques with emerging technological developments, such as microfluidics and micropatterning, provide pro mise to further improve our understanding of the mechanical behavior of the nucleus under physiological conditions and its pathobiology in disease, thereby potentially providing the basis for the development of new diagnostic tools and therapeutics. Acknowledgments This work was supported by the National Institutes of Health (R01 HL082792 and R01 NS059348) and the American Heart Association (0635359N).

References Bione, S., et al. (1994). Identification of a novel X-linked gene responsible for Emery-Dreifuss muscular dystrophy. Nat. Genet. 8, 323-327 Bonne, G., et al. (1999). Mutations in the gene encoding lamin A/C cause autosomal dominant Emery– Dreifuss muscular dystrophy. Nat. Genet. 21, 285–288. Broers, J. L., et al. (1997). A- and B-type lamins are differentially expressed in normal human tissues. Histochem. Cell Biol. 107, 505–517. Broers, J. L., et al. (2004). Decreased mechanical stiffness in LMNA–/– cells is caused by defective nucleo-cytoskeletal integrity: Implications for the development of laminopathies. Hum. Mol. Genet. 13, 2567–2580. Caille, N., et al. (1998). Assessment of strain field in endothelial cells subjected to uniaxial deformation of their substrate. Ann. Biomed. Eng. 26, 409–416. Caille, N., et al. (2002). Contribution of the nucleus to the mechanical properties of endothelial cells. J. Biomech. 35, 177–187. Cao, H., and Hegele, R. A. (2000). Nuclear lamin A/C R482Q mutation in canadian kindreds with Dunnigan type familial partial lipodystrophy. Hum. Mol. Genet. 9, 109–112. Constantinescu, D., et al. (2006). Lamin A/C expression is a marker of mouse and human embryonic stem cell differentiation. Stem Cells 24, 177–185. Couzon, C., et al. (2009). Critical stresses for cancer cell detachment in microchannels. Eur. Biophys. J. 38, 1035–1347. Crisp, M., et al. (2006). Coupling of the nucleus and cytoplasm: Role of the LINC complex. J. Cell Biol. 172, 41–53. Dahl, K. N., et al. (2005). Power-law rheology of isolated nuclei with deformation mapping of nuclear substructures. Biophys. J. 89, 2855–2864.

6. Mechanical Properties of the Nucleus

139

Dahl, K. N., et al. (2006). Distinct structural and mechanical properties of the nuclear lamina in HutchinsonGilford progeria syndrome. Proc. Natl. Acad. Sci. USA 103, 10271–10276. De Sandre-Giovannoli, A., et al. (2002). Homozygous defects in LMNA, encoding lamin A/C nuclearenvelope proteins, cause autosomal recessive axonal neuropathy in human (Charcot-Marie-Tooth disorder type 2) and mouse. Am. J. Hum. Genet. 70, 726–736. de Vries, A. H., et al. (2007). Direct observation of nanomechanical properties of chromatin in living cells. Nano Lett. 7, 1424–1427. Deguchi, S., et al. (2005). Flow-induced hardening of endothelial nucleus as an intracellular stress-bearing organelle. J. Biomech. 38, 1751–1759. Evans, E. A. (1989). Structure and deformation properties of red blood cells: Concepts and quantitative methods. Methods Enzymol. 173, 3–35. Fatkin, D., et al. (1999). Missense mutations in the rod domain of the lamin A/C gene as causes of dilated cardiomyopathy and conduction-system disease. N. Engl. J. Med. 341, 1715–1724. Friedl, P., and Wolf, K. (2009). Proteolytic interstitial cell migration: A five-step process. Cancer Metastasis Rev. 28, 129–135. Goldman, R. D., et al. (2002). Nuclear lamins: Building blocks of nuclear architecture. Genes Dev. 16, 533–547. Gotzmann, J., and Foisner, R. (2006). A-type lamin complexes and regenerative potential: A step towards understanding laminopathic diseases? Histochem. Cell Biol. 125, 33–41. Guilak, F. (1995). Compression-induced changes in the shape and volume of the chondrocyte nucleus. J. Biomech. 28, 1529–1541. Guilak, F., et al. (1999). Viscoelastic properties of intervertebral disc cells. Identification of two biomecha nically distinct cell populations. Spine (Phila Pa 1976) 24, 2475–2483. Guilak, F., et al. (2000). Viscoelastic properties of the cell nucleus. Biochem. Biophys. Res. Commun. 269, 781–786. Hale, C. M., et al. (2008). Dysfunctional connections between the nucleus and the actin and microtubule networks in laminopathic models. Biophys. J. 95, 5462–5475. Hategan, A., et al. (2003). Adhesively-tensed cell membranes: Lysis kinetics and atomic force microscopy probing. Biophys. J. 85, 2746–2759. Henriksen, J. R., and Ipsen, J. H. (2004). Measurement of membrane elasticity by micro-pipette aspiration. Eur. Phys. J. E Soft Matter 14, 149–167. Hou, H. W., et al. (2009). Deformability study of breast cancer cells using microfluidics. Biomed. Microdevices 11, 557–564. Houben, F., et al. (2009). Disturbed nuclear orientation and cellular migration in A-type lamin deficient cells. Biochim. Biophys. Acta 1793, 312–324. Houtsmuller, A. B. (2005). Fluorescence recovery after photobleaching: Application to nuclear proteins. Adv. Biochem. Eng. Biotechnol. 95, 177–199. Hudson, M. E., et al. (2007). Identification of differentially expressed proteins in ovarian cancer using highdensity protein microarrays. Proc. Natl. Acad. Sci. USA 104, 17494–17499. Konety, B. R., and Getzenberg, R. H. (1999). Nuclear structural proteins as biomarkers of cancer. J. Cell Biochem. Suppl. 32–33, 183–191. Lammerding, J., et al. (2003). Quantitative measurements of active and passive mechanical properties of adult cardiac myocytes. IEEE Eng. Med. Biol. Mag. 22, 124–127. Lammerding, J., et al. (2004). Lamin A/C deficiency causes defective nuclear mechanics and mechan otransduction. J. Clin. Invest. 113, 370–378. Lammerding, J., et al. (2005). Abnormal nuclear shape and impaired mechanotransduction in emerin deficient cells. J. Cell Biol. 170, 781–791. Lammerding, J., et al. (2006). Lamins A and C but not lamin B1 regulate nuclear mechanics. J. Biol. Chem. 281, 25768–25780. Lee, J. S., et al. (2007). Nuclear lamin A/C deficiency induces defects in cell mechanics, polarization, and migration. Biophys. J. 93, 2542–2552. Li, Y., et al. (2009). Intracellular particle tracking as a tool for tumor cell characterization. J. Biomed. Opt. 14, 064005.

140

Maria Lucia Lombardi and Jan Lammerding Liu, J., et al. (2000). Essential roles for Caenorhabditis elegans lamin gene in nuclear organization, cell cycle progression, and spatial organization of nuclear pore complexes. Mol. Biol. Cell 11, 3937–3947. Maniotis, A. J., et al. (1997). Demonstration of mechanical connections between integrins, cytoskeletal filaments, and nucleoplasm that stabilize nuclear structure. Proc. Natl. Acad. Sci. USA 94, 849–854. Muchir, A., et al. (2000). Identification of mutations in the gene encoding lamins A/C in autosomal dominant limb girdle muscular dystrophy with atrioventricular conduction disturbances (LGMD1B). Hum. Mol. Genet. 9, 1453–1459. Newport, J. W., et al. (1990). A lamin-independent pathway for nuclear envelope assembly. J. Cell Biol. 111, 2247–2259. Novelli, G., et al. (2002). Mandibuloacral dysplasia is caused by a mutation in LMNA-encoding lamin A/C. Am. J. Hum. Genet. 71, 426–431. Olins, A. L., et al. (2008). The human granulocyte nucleus: Unusual nuclear envelope and heterochromatin composition. Eur. J. Cell Biol. 87, 279–290. Olins, A. L., et al. (2009). The LINC-less granulocyte nucleus. Eur. J. Cell Biol. 88, 203–214. Pajerowski, J. D., et al. (2007). Physical plasticity of the nucleus in stem cell differentiation. Proc. Natl. Acad Sci. USA 104, 15619–15624. Peeters, E. A., et al. (2003). Monitoring the biomechanical response of individual cells under compression: A new compression device. Med. Biol. Eng. Comput. 41, 498–503. Philip, J. T., and Dahl, K. N. (2008). Nuclear mechanotransduction: Response of the lamina to extracellular stress with implications in aging. J. Biomech. 41, 3164–3170. Rowat, A., et al. (2005). Characterization of the elastic properties of the nuclear envelope. J. R. Soc. Interface 2, 63–69. Rowat, A. C., et al. (2006). Mechanical properties of the cell nucleus and the effect of emerin deficiency. Biophys. J. 91, 4649–4664. Sanchez, J. A., and Wangh, L. J. (1999). New insights into the mechanisms of nuclear segmentation in human neutrophils. J. Cell Biochem. 73, 1–10. Scaffidi, P., and Misteli, T. (2008). Lamin A-dependent misregulation of adult stem cells associated with accelerated ageing. Nat. Cell Biol. 10, 452–459. Schape, J., et al. (2009). Influence of lamin A on the mechanical properties of amphibian oocyte nuclei measured by atomic force microscopy. Biophys. J. 96, 4319–4325. Schirmer, E. C., and Gerace, L. (2004). The stability of the nuclear lamina polymer changes with the composition of lamin subtypes according to their individual binding strengths. J. Biol. Chem. 279, 42811–42817. Schirmer, E. C., et al. (2003). Nuclear membrane proteins with potential disease links found by subtractive proteomics. Science 301, 1380–1382. Schmiedeberg, L., et al. (2004). High- and low-mobility populations of HP1 in heterochromatin of mamma lian cells. Mol. Biol. Cell 15, 2819–2833. Shackleton, S., et al. (2000). LMNA, encoding lamin A/C, is mutated in partial lipodystrophy. Nat. Genet. 24, 153–156. Shimi, T., et al. (2008). The A- and B-type nuclear lamin networks: Microdomains involved in chromatin organization and transcription. Genes Dev. 22, 3409–3421. Spencer, V. A., et al. (2001). Altered profiles in nuclear matrix proteins associated with DNA in situ during progression of breast cancer cells. Cancer Res. 61, 1362–1366. Stewart, C., and Burke, B. (1987). Teratocarcinoma stem cells and early mouse embryos contain only a single major lamin polypeptide closely resembling lamin B. Cell 51, 383–392. Theret, D. P., et al. (1988). The application of a homogeneous half-space model in the analysis of endothelial cell micropipette measurements. J. Biomech. Eng. 110, 190–199. Thoumine, O., and Ott, A. (1997). Time scale dependent viscoelastic and contractile regimes in fibroblasts probed by microplate manipulation. J. Cell Sci. 110(Pt 17), 2109–2116. Tseng, Y., et al. (2002). Micromechanical mapping of live cells by multiple-particle-tracking microrheology. Biophys. J. 83, 3162–3176.

6. Mechanical Properties of the Nucleus

141

Tseng, Y., et al. (2004). Micro-organization and visco-elasticity of the interphase nucleus revealed by particle nanotracking. J. Cell Sci. 117, 2159–2167. Verstraeten, V. L., et al. (2008). Increased mechanosensitivity and nuclear stiffness in Hutchinson-Gilford progeria cells: Effects of farnesyltransferase inhibitors. Aging Cell 7, 383–393. Weidtkamp-Peters, S., et al. (2009). Fluorescence correlation spectroscopy to assess the mobility of nuclear proteins. Methods Mol. Biol. 464, 321–341. Wiesmeijer, K., et al. (2008). Chromatin movement visualized with photoactivable GFP-labeled histone H4. Differentiation 76, 83–90. Wolf, K., and Friedl, P. (2006). Molecular mechanisms of cancer cell invasion and plasticity. Br. J. Dermatol. 154, 11–15. Wolf, K., et al. (2007). Multi-step pericellular proteolysis controls the transition from individual to collective cancer cell invasion. Nat. Cell Biol. 9, 893–904. Zink, D., et al. (2004). Nuclear structure in cancer cells. Nat. Rev. Cancer 4, 677–687.

CHAPTER 7

Theoretical Concepts and Models of Cellular Mechanosensing Rumi De1*, Assaf Zemel1†, and Samuel A. Safran‡ *

Indian institute of Science Education and Research, Kolkata, Mohanpur 741252, Nadia, West Bengal, India

†

Institute of Dental Sciences, Faculty of Dental Medicine, and Fritz Haber Center for Molecular Dynamics, Hebrew University-Hadassah Medical Center, Jerusalem, 91120, Israel

‡

Department of Materials and Interfaces, Weizmann Institute of Science, Rehovot, 76100, Israel

1

These authors contributed equally to this article

Abstract I. Introduction II. Macromolecular Components Involved in Cellular Mechanosensing III. Cell Adhesion A. Models of Mechanosensitivity of Focal Adhesions IV. Active and Passive Mechanics of the Cytoskeleton V. Slow Mechanical Processes in the Cytoskeleton A. Reorganization of Cytoskeletal Structure: Stress-Fiber Polarization VI. Cellular Response to External Mechanical Stress A. Active Elastic Dipole Model of Cells B. Predictions of the Model VII. Discussion

Acknowledgments

References

Abstract Recent discoveries have established that mechanical properties of the cellular environ ment such as its rigidity, geometry, and external stresses play an important role in determining the cellular function and fate. Mechanical properties have been shown to influence cell shape and orientation, regulate cell proliferation and differentiation, and METHODS IN CELL BIOLOGY, VOL. 98 Copyright 2010 Elsevier Inc. All rights reserved.

143

978-0-12-381009-0 DOI: 10.1016/S0091-679X(10)98007-2

144

Rumi De et al.

even govern the development and organization of tissues. In recent years, many theoretical and experimental investigations have been carried out to elucidate the mechanisms and consequences of the mechanosensitivity of cells. In this review, we discuss recent theore tical concepts and approaches that explain and predict cell mechanosensitivity. We focus on the interplay of active and passive processes that govern cell–cell and cell–matrix interac tions and discuss the role of this interplay in the processes of cell adhesion, regulation of cytoskeleton mechanics and the response of cells to applied mechanical stresses.

I. Introduction Cellular mechanosensitivity plays a central role in the viability and function of cells as well as in the development, maintenance, and pathology of tissues (Bershadsky et al., 2003; Discher et al., 2005; Huang and Ingber, 1999; Jaalouk and Lammerding, 2009; Nelson et al., 2005; Peyton et al., 2007; Wozniak and Chen, 2009). The cytoskeleton of virtually all adherent cell types, for example, fibroblasts, endothelial cells, and muscle cells, comprises actin–myosin stress fibers that actively pull on the surrounding matrix and generate elastic stress and strain fields inside and outside the cells (Bershadsky et al., 2003). The generic purpose of these forces is to provide cells with a means of actively sensing and responding to mechanical properties such as topography, rigidity of the environment (Schwarz and Bischofs, 2005). Indeed, experiments have shown that cells adjust their internal activity, overall shape, alignment, and adhesion to the extracellular matrix in a manner that depends on the mechanical nature of their environment (Bershadsky et al., 2003; Discher et al., 2005; Engler et al., 2004, 2006; Huang and Ingber, 1999; Vogel and Sheetz, 2006). When placed in rigid matrix, cells spread in larger areas, develop more stable adhesion complexes, exhibit more prominent stress fibers, and generate stronger forces compare to soft matrix (Discher et al., 2005). The rigidity of the surrounding matrix was also shown to determine the shape and spreading speed of cells (Discher et al., 2005; Lo et al., 2000; Pelham and Wang, 1997) and govern cellular self-assembly into organized tissue structures in vitro (Bissell, 2007; Paszek et al., 2005; Guo et al., 2006). Moreover, recent studies have established that even very fundamental processes such as the lineage specification of stem cells, the rate of cell proliferation, and the switch from growth to apoptosis (programmed cell death) are governed by the mechanical and geometrical properties of the cells and their environ ments (Chen et al., 1997; Cusachs et al., 2008; Engler et al., 2006; Pirone et al., 2004). An additional aspect of cell mechanosensitivity is the ability of cells to sense and respond to the various types of in vivo loads that exist within living organisms, such as muscle tension, blood pressure, shear flow, gravity, and also the active forces generated by adjacent cells (Chen, 2008; Ingber, 2003a; Orr et al., 2006). These mechanical signals induce an active reorganization of the cell cytoskeleton and readjustment of the contractile forces exerted by the cells (Deng et al., 2006; Engler et al., 2006; Harris et al., 1980; Stamenovic et al., 2007). The active nature of this mechanotransduction is demonstrated by the fact that it often vanishes when actin–myosin contractility is

7. Theoretical Concepts and Models of Cellular Mechanosensing

145

inhibited (Sarasa-Renedo et al., 2006; Torsoni et al., 2005; Zhao et al., 2007). More over, when cells on a substrate are subject to stress, either by stretching the substrate or by applying an aligned shear flow, the cells actively reorient and align themselves in preferred directions. Interestingly, while in some studies cells were shown to align parallel to the direction of a static or quasi-static stress field (Collinsworth et al., 2000; Eastwood et al., 1998; Vandenburgh, 1988), other experiments find that cells remain randomly oriented (Jungbauer et al., 2008). On the other hand, in response to dyna mically oscillating stress and strain fields (designed originally to study the effects of heart beat and blood pressure), cells tend to orient away from the stress direction aligning themselves nearly perpendicular to the applied field direction (Hayakawa et al., 2001; Kurpinski et al., 2006; Shirinsky et al., 1989; Smith et al., 1997; Takemasa et al., 1997; Wang and Grood, 2000; Wang et al., 2001). Since cellular mechanotransduction plays a crucial role in many biological processes, for example, wound healing, muscle growth, tissue assembly, and development (Ber shadsky et al., 2003; Discher et al., 2005; Huang and Ingber, 1999; Jakab et al., 2004; Korff and Augustin, 1999; Vogel and Sheetz, 2006), it is important to conceptualize and model the active responses of cells to mechanical forces. To date, the detailed mechan isms that govern the interaction of cells with the extracellular matrix and the related biochemical processes during cell spreading, movement, or contraction are not fully understood. However, some insight into these processes can be obtained from the response of a cell to its mechanical environment, since cell–matrix interactions can trigger biochemical signaling that, in turn, generates mechanical changes both within the cell and in the surrounding environment (Engler et al., 2004; Pelham and Wang, 1997). The complex nature of cell–matrix interactions, that involves an interplay of active forces that are exerted and regulated by the cells, and of passive forces, originating from the elasticity of the cells and their environment, makes this field of study highly challenging. Nevertheless, significant progress has been made in recent years in understanding the origins of cell mechanosensitivity as well as the coupling of active and passive mechanical forces in living cellular systems. In this review, we discuss the theoretical foundations, concepts, and models that are recently developed to under stand what is known experimentally about cellular mechanics. The review is organized as follows: in Section II, we present a brief outline of the macromolecular components involved in cell mechanosensitivity. Section III reviews theoretical approaches to cell adhesion. In Sections IV and V, we describe the passive and active aspects of cytoske leton elasticity including the long-term remodeling of stress fibers in the cytoskeleton. Finally, in Section VI, we summarize theoretical models that predict the orientational response of cells in dynamically varying stress fields.

II. Macromolecular Components Involved in Cellular Mechanosensing The traditional picture of the cell as a lipid membrane that surrounds a liquid-like cytosol containing localized organelles is now understood to be relevant only for some

146

Rumi De et al.

kinds of cells (e.g., cells that are in suspension such as blood cells) but not for most (adherent) tissue cell types (Ingber, 2003b). Adherent cells such as fibroblasts, neu rons, endothelial cells, and bone cells, contain an actively regulated gel-like elastic cytoskeleton (Trepat et al., 2008; Wirtz, 2009). The elastic nature of the cytoskeleton is important not only for the mechanical stability of the cell, but also for transmitting mechanical signals in the form of elastic stress and strain fields, from the environment, through the cytoskeleton and to the nucleus (Wang et al., 2009). The cytoskeleton is a complex dynamical network of actin, intermediate filaments, and microtubules that interact with a myriad of molecular motors and cross-linking proteins (Howard, 2001). Many of these proteins are collectively responsible for cellular mechanosensing. Actin is the primary structural component of most cells and provides the cytoskeleton with its structural stability. Actin filaments can undergo very rapid polymerization–depolymerization cycles depending on physical or chemical conditions (Gardel et al., 2004; Gov and Gopinathan, 2006; Pullarkat et al., 2007). The dynamic assembly and disassembly of actin network is central to the formation of membrane protrusions—lamellipodia, filopodia, ruffles, as well as in driving cell motility (Mogilner and Keren, 2009; Shlomovitz and Gov, 2008). Intermediate fila ments differ from other cytoskeletal filaments in terms of their long-term stability. In addition, in contrast to the polymerization of other filaments, the formation of inter mediate filaments occurs without GTP or ATP hydrolysis. Intermediate filaments play a passive reinforcement role and can withstand high elastic stresses (Mofrad, 2009). Another important player in cellular mechanics are microtubules which are highly dynamic, even more so than actin. Microtubules undergo constant polymerization and depolymerization and their half lives are typically only few minutes (Mitchison and Kirschner, 1984; Mofrad, 2009). Microtubules are the stiffest of the three cytoskeletal polymers and while the dense highly cross-linked actin network plays an essential mechanical role in cells (Colombelli et al., 2009; Yoshigi et al., 2005), microtubules are also believed to contribute significantly to cell mechanosensing. The networks of microtubules can resist large-scale compressive forces exerted by the surrounding contractile cytoskeleton and hence influence the shape and the mechanical behavior of the cell (Brangwynne et al., 2006). Microtubules also play an important role in intracellular transport, cell division, and chromosome separation (Fletcher and Mullins, 2010). The active contractility of the cytoskeleton is due to the pool of molecular motors, in particular myosin-II molecules, that consume chemical energy (via hydrolysis of ATP) and convert it to mechanical work. The activity of myosin motors in association with (cross-linked) actin filaments generally produce contractile forces that maintain the cell in a prestressed state far from thermal equilibrium (Griffin et al., 2004; Kumar et al., 2006; Wang et al., 2002b). The contractile forces of well-spread cells are produced by actin– myosin stress fibers that typically connect opposite sides of the cell and terminate at protein complexes, called focal adhesions, that anchor the cell to the extracellular matrix. The development of organized acto-myosin stress fibers in the cytoskeleton occurs on a timescale of hours after cell adhesion and the initial cell spreading; this process is subject to a feedback loop that is governed by the elastic stress that these same fibers produce in the cytoskeleton (we will discuss and elaborate this point in more detail later).

7. Theoretical Concepts and Models of Cellular Mechanosensing

147

Although the mechanisms by which mechanical forces are converted to biochemical signals are known to involve mechanoreceptors at the loci of cell–matrix interactions (focal adhesion complexes) (Geiger and Bershadsky, 2002; Geiger et al., 2009), recent experiments suggest that the physical coupling of the cytoskeleton to the nucleus may also impact gene expression. Mechanics may thus have a direct impact on cell function and fate, by affecting the packing of DNA and chromatin in the nucleus (Bhattacharya et al., 2009; Cusachs et al., 2008; Lee et al., 2007; Pajerowski et al., 2007; Wang et al., 2009). Therefore, the physical interactions between cytoskeletal and nuclear compo nents and their response to stresses and strains are important to understand cellular mechanotransduction. The cell nucleus is significantly stiffer than the surrounding cytoplasm and can be structurally and functionally divided into two regions—the nuclear envelope and the nuclear interior (Dahl et al., 2008; Lammerding et al., 2007). The nuclear envelope is composed of a double lipid bilayer, the inner and outer nuclear membranes, and the nuclear lamina. The inner and outer nuclear membranes join at the nuclear pore complexes that allow the transport of molecules and ions from the nucleus to the cytoplasm. The nuclear lamina consists of a dense network of protein primarily composed of lamins that underlies the inner nuclear membrane. Nuclear lamina provides mechanical support and structural stability to the nucleus in a manner similar to the support that the cytoskeleton provides to the cell as a whole. In addition, the lamina also plays an important role in chromatin organization and gene regulation (Dechat et al., 2009; Lee et al., 2007). In addition to lamins, chromatins are proposed to occupy distinct regions within the nuclear interior. The interior is largely aqueous and less well understood. A number of subnuclear bodies such as the nucleolus and the cajal bodies are also present as distinct structural and functional element that could be influenced by mechanical forces (Dahl et al., 2008; Schiessel, 2003). Mechanical forces—arising either from external sources such as compression, ten sion, fluid shear stress, or internal cellular forces arising due to the changes in the extracellular matrix—can promote continuous dynamic remodeling not only of the cytoplasm, but also of the nucleus itself. These forces are transmitted across the cytoskeleton to the nucleus resulting in cytoskeletal as well as nuclear deformations (Cusachs et al., 2008; Wang et al., 2009) that, in turn, may produce large-scale reorganization of chromatin structure, conformational changes of DNA which can then lead to changes in transcriptional activities (Rowat et al., 2008; Wang et al., 2009). Thus, active contractile force produced by the cell cytoskeleton may be a means of actively sensing the mechanical properties of the environment and transmitting this information to the nucleus where it directly influences cell biochemical function via the up- or down-regulation of gene expression. Outside the nucleus, the mechanical feed back between the cell and its environment through focal adhesion protein complexes dictates many cellular properties, such as the strength and pattern of cell adhesion (Chen et al., 2003; Tan et al., 2003) as well as the internal architecture, polarity, and rigidity of the cytoskeleton (Koenderink et al., 2009; Solon et al., 2007; Théry et al., 2006a,b; Wang et al., 2002a; Yeung et al., 2005). Whether gene expression is regulated via the direct elastic coupling of the cytoskeleton to the nucleus, or via

148

Rumi De et al.

indirect signaling that is triggered by forces acting on focal adhesions, is an important question since both channels may play a role in the regulation of longer-term processes such as the lineage specification of stem cells and the rate of cell proliferation (Discher et al., 2009; Tee et al., 2009). In the following section we briefly outline the relation of cell mechanosensing via focal adhesion complexes; we focus on adhesions since they directly couple the cell and its cytoskeleton to the mechanical environment. This is then followed by a description of recent findings that shed light on our understanding of active processes in the cytoskeleton that govern its mechanical stability and the internal structure.

III. Cell Adhesion Both theories (De and Safran, 2008; De et al., 2007, 2008; Hsu et al., 2009) and experiments (Deng et al., 2006) on the response of cells to external force focus on the reorganization of the cellular cytoskeleton as the means by which this response is achieved. The different response of cells to slowly varying (<0.001 Hz) versus rela tively high-frequency (1 Hz) cyclic stress is attributed to the fact that cells cannot effect this reorganization on very short timescales. This relatively slow (tens of minutes to hours) reorganization involves the liquification and rebuilding of the cytoskeletal stress fibers (Trepat et al., 2007) as well as changes in the coupling of the cell to the surrounding elastic matrix via the protein assemblies called focal adhesions. These assemblies connect the stress fibers to membrane bound, integrin proteins that are then connected, on the extracellular side, to the matrix. The stress fibers exert contractile forces that are transmitted to the surrounding matrix by the focal adhesions. Other important cellular functions related to adhesion, for example, DNA duplication, division, and differentiation only take place when cells adhere (Alberts et al., 1994). Here, we briefly review the salient features of cell adhesion, with particular emphasis on mature, focal adhesions and their dependence on the matrix elasticity. Cellular adhesions are characterized by assemblies of proteins (e.g., transmembrane integrins that connect the cell to the substrate and proteins such as vinculin, talin, and filamin—among many others—that connect the integrins to the cytoskeleton), that create a complex, intracellular gel-like complex (Zamir and Geiger, 2001). A well-formed adhesion site nucleates from a seed of transmembrane proteins (Cluzel et al., 2005; Miyamoto et al., 1995). The proteins in these assemblies coexist with their respective, relatively dilute cytoplasmic or membrane proteins with a gradient in both concentra tion and mobility in the region near the adhesion (Wolfenson et al., 2009). These domains have a relatively long lifetime (tens of minutes) that suggests that the phases might be near equilibrium. However, such micron-size domains are unexpected from a thermodynamic point of view (Lenne and Nicolas, 2009). This puzzle can be resolved by noting that, in addition to diffusion, energy consuming, active processes can transport free proteins from the dilute phase in the cytoplasm or membrane to the adhesion sites. These active processes involve molecular motor proteins and are highly regulated by the cell (Kawakami et al., 2001); this suggests that nonequilibrium effects

7. Theoretical Concepts and Models of Cellular Mechanosensing

149

are important in stabilizing these finite-size domains. Recent dynamical theories have focused on features of adhesive domains, such as their shape or growth velocity (Besser and Safran, 2006; Nicolas et al., 2004; Shemesh et al., 2005b). While one might be tempted to draw an analogy between physical adhesion [e.g., of a synthetic vesicle coated with ligands that are attracted to an appropriate surface (Bruinsma et al., 2000; deGennes et al., 2003; Komura and Andelman, 2000)] and cellular adhesion, this comparison is probably limited to the fact that both processes involve attractive ligand–surface interactions. However, physical adhesion is a passive process while cellular, focal adhesions involve molecular motors that generate internal tension on the stress fibers to which these adhesions are attached. This process consumes ATP and results in the fact that in addition to normal forces, cells exert lateral forces on the surfaces to which they are attached. Such forces are observed in experiments that measure surface deformation (Balaban et al., 2001; Riveline et al., 2001) and its sensitivity to acto-myosin contractility. Also, in contrast to equilibrium assemblies that are governed by short-ranged molecular interactions and entropy, cell adhesive domains are stabilized by cytoskele tal tension; this tension is also required for them to grow in size. They differ in this aspect from passive “glue” (self-assembled ligands) that might connect a synthetic vesicle to a surface. This tension has a directionality (generally toward the nucleus) resulting in an anisotropy in the shape of the adhesive domains (Petroll et al., 2003). The force per myosin motor is typically on the order of 1.5 pN; the cytoskeleton tension arises from the fact that the force operates on two, oppositely oriented actin chain. Since hundreds of actin filaments can terminate at a single adhesion the total force is on the order of nN which yields a shear stress on the order of kPa. The molecular origins of focal adhesion assembly and the mechanosensitivity of cell adhesion are not completely understood. Experiments show that in addition to sensi tivity to the mechanical properties of the extracellular matrix, cell adhesions respond to externally applied, intra- or extracellular stresses. Applied forces, such as pipette induced shear (Riveline et al., 2001), hydrodynamic flow (Zaidel-Bar et al., 2003), and stretching forces on the substrate (Kaunas et al., 2005) cause growth of focal adhesions in the direction of the force. Recently, local perturbations of stresses within the cytoskeleton by mechanical microdevices, laser nanosurgery, and other measure ment techniques (Allioux-Guerin et al., 2008; Choquet et al., 1997; Colombelli et al., 2009; Galbraith and Sheetz, 1997; Riveline et al., 2001; Saez et al., 2007; Sniadecki et al., 2007) have supplemented biochemical tools to elucidate the basis of cell mechanosensitivity. A. Models of Mechanosensitivity of Focal Adhesions The fact that tension is required for the stability and growth of focal adhesions can be analyzed by considering conditions of either constant stress or constant strain. The work in Chan and Odde (2008), Deshpande et al. (2007b), and Schwarz et al. (2006) focuses on the effects of a constant strain applied to the adhesions. Under this condition, a fixed number of bound adhesive molecules feel a stress that increases

150

Rumi De et al.

with time which, in turn, increases the rate of detachment according to Bell’s “law” (Bell, 1978).1 In this picture, the sensitivity of the adhesion domains to the rigidity of the extracellular matrix is due to the coupling between the force-sensitive strain imposed by the cytoskeleton, and the force-dependent molecular detachment of the adhesive linkers. This effect also depends on the response of the substrate to applied force since the strain-induced stress could be localized either in the adhesive bonds (subject to Bell’s law) or in the substrate (Chan and Odde, 2008), depending on which system has a smaller elastic modulus. The matrix rigidity also impacts the time it takes the system to build up a certain force level; on more rigid substrates, the dynamics are faster. On soft substrates, these approaches predict a cascade of rupture events of the bonds of the adhesive proteins (Chan and Odde, 2008; Schwarz et al., 2006), or a limitation on the magnitude of the cytoskeleton force applied to the adhesive bonds (Deshpande et al., 2006). On the other hand, one can model the system under the assumption that the cytoskeleton pulls on the adhesions with a constant stress (Besser and Safran, 2006; Nicolas et al., 2004; Shemesh et al., 2005b). This is consistent with experiments (Balaban et al., 2001; Tan et al., 2003) that show that the size of the adhesive domains increases linearly with the total force to which they are submitted, even on very short timescales of seconds (Balaban et al., 2001). This observation suggests that the stress may be constant in a wide range of timescales. To break the symmetry of the system and promote growth only in the direction of the applied force (as is observed), the stress is assumed to act only on part of the area of the adhesion. The stress applied by the cytoskeleton pulls the adhesion toward the nucleus. The adhesion ahead of the force (closer to the nucleus) is compressed while that behind the force (closer to the cell membrane) is expanded. In the model of Besser and Safran (2006) and Nicolas et al. (2004), adhesion growth is modeled using an analogy to the nucleation and growth of domains of molecules adsorbed to a substrate but with the important difference that for focal adhesions, the adsorption of new proteins at the front (nearer to the cell nucleus) of the adhesion is more favorable than at the back (closer to the cell membrane) because the ligands near the front are compressed by the acto-myosin tension, while those near the back are compressed. The growth of the adhesion occurs in a relatively small region (on the order of the adhesion thickness) near its front. Consistent with the observation that cellular adhesive domains are larger on stiffer substrates, the model (Nicolas et al., 2004) assumes that the adhesion growth kinetics include terms related to the work done by the cytoskeleton to maintain the constant stress (or equivalently, the dynamics of the adhesions is related to the energy the cell invests in maintaining the adhesions). This differs from inert systems such as adhering vesicles, where the self-assembly of adhesive molecules results from the minimization of their free energy alone. This approach predicts a linear relationship between the size of the adhesive domain and the rigidity of the substrate (Nicolas and Safran, 2006), consistent with experiments in (Saez et al., 2005) as long as constant stress is 1

This phenomenological relationship between the rate of detachment of proteins and the force they are submitted to has, however, some exceptions such as catch bonds, for which the bond is reinforced under tension (Marshall et al., 2003).

7. Theoretical Concepts and Models of Cellular Mechanosensing

151

assumed. For very rigid substrates, a saturation of the adhesion size is observed to occur (Ghibaudo et al., 2008). In addition, the model predicts that the characteristic timescale to reach a stationary state is proportional to the rigidity of the substrate (Nicolas et al., 2008). The authors (Shemesh et al., 2005b) present another scenario and assume that new proteins join the stretched domain in order to restore the local concentration of adhesion proteins that has been depleted by the stretching of the gel-like adhesion by the acto-myosin tension. The condensation process is driven by the minimization of the energy of the adsorbing proteins. The adhesion is expected to grow much more homogeneously, with proteins adsorbing all along its length, in contrast to the nucleation-growth model described in the previous paragraph. Each of these models has different regimes in which it compares well with the observations and it is difficult to distinguish between the various theories at this point. Of course, direct observation of the growth modes of focal adhesions: whether they grow in a narrow region at their front (and disassemble—to some extent—at their rear) or whether adhesion proteins accrue much more uniformly along the adhesion, can distinguish the nucleation and growth picture from the energetic model in which proteins adsorb to restore the equilibrium density all along the stretched adhesion.

IV. Active and Passive Mechanics of the Cytoskeleton In adherent cells, such as fibroblasts, endothelial cells, and muscle cells, the cytoskeleton extends over the entire volume of the cytosol and connects the nucleus and other organelles to the cell membrane (Ingber, 2003b; Wirtz, 2009). The internal architecture and mechanical properties of the cytoskeleton are established on a timescale of days after cell adhesion. During the first hours of cell adhesion to a twodimensional substrate, the cell spreads and adjusts its anchoring to the substrate, while its cytoskeleton and shape begin to remodel (Burridge and Chrzanowska-Wodnicka, 1996; Cai et al., 2010; Döbereiner et al., 2004; Ren et al., 1999). There is, thus, a period in which the cell spreads with no or little contractility followed by a gradual buildup of the contractility of the acto-myosin cytoskeleton. We first focus on the viscoelastic properties of the cytoskeleton that are measured on timescales of typically ~102 s (Kasza et al., 2007; Trepat et al., 2007; Wirtz, 2009) and follow this with a discussion of the long-time remodeling processes in the cytoskeleton. The cytoskeleton is a complex, dynamical network of protein filaments that interact with a large number of molecular motors and cross-linking proteins. Experiments suggest that the cytoskeleton rigidity arises mostly from the interconnected network of actin filaments (Wirtz, 2009). In general, the cytoskeleton is inhomogeneous and its rigidity varies with the local concentration of filamentous actin present in the cyto plasm. For example, the actin-rich cell periphery (the lamella) is significantly stiffer than the perinuclear region, which contains less actin (Wirtz, 2009). However, unlike conventional passive materials, the cytoskeleton includes active (“live”) mechanisms that allow the cell to modulate its mechanical properties at will by controlling the

152

Rumi De et al.

elastic stress in the cytoskeleton; this is achieved by regulation of the forces exerted by the myosin II molecular motors on the actin filaments (Ingber, 2003b; Koenderink et al., 2009; Wang et al., 2002a). The authors (Solon et al., 2007) have shown that for a certain range of substrate stiffness, fibroblasts actively adjust their stiffness to match that of the surroundings. The ability of a cell to control its stiffness is one of the most useful properties since it determines the balance of forces between the cell and the surroundings and thus dictates the elastic field that propagates through the cytoskele ton; this, in turn, may control and regulate diverse cellular processes (Tee et al., 2009). A number of complementary mechanisms have been suggested to describe how cells actively modulate the elastic rigidity of the cytoskeleton: (1) the tensegrity model, that focuses on the spatial architecture of the various filaments in the cytoskeleton, (2) the semiflexible chain model, that focuses on the elastic nonlinearity of the actin filaments, and (3) the dipole polarization model that focuses on the alignment of the myosin

Y

(A) (B)

T/2

σext

T/2 X

T/2

T/2

Z

(C) Matrix stresses

Polarizable acto myosin force dipoles

Fig. 1 Schematic illustration of three suggested mechanisms for active regulation of cytoskeleton stiffness by force. The tensegrity model (panel A) focuses on the architecture of the various filaments in the cytoskeleton, where the level of tension in the structure dictates its mechanical stability (rigidity) (Ingber, 2003b). The semiflexible chain model (panel B) highlights the nonlinear elasticity of the chains themselves and their rigidification with stress (Koenderink et al., 2009). The dipole polarization model (panel C) focuses on the ability of the motors (as well as the actin filaments) to orient in response to the applied load. This degree of freedom effectively stiffens the system with respect to stretch since the acto-myosin filaments are contractile (Zemel et al., 2010a). Reprinted with permission from the publishers.

7. Theoretical Concepts and Models of Cellular Mechanosensing

153

motors (and their associated actin filaments) in response to external stresses; these are depicted in Fig. 1: i. The tensegrity model: The first model to propose that the elastic stress in the cytoskeleton is essential for the shape stability of the cytoskeleton is the tensegrity model, pioneered by Ingber and coworkers (Ainsworth, 2009; Ingber, 1993, 2003b). This model is inspired by a specific class of reticulated mechanical structures known as the tensegrity architecture. Mechanical objects with this architecture include spider webs and pup tents; these require tension for their shape stability. In these structures, the greater the tension carried by the various elements in the network, the more stable the overall structure under load (Stamenovic and Wang, 2000; Stamenovic and Ingber, 2002). In this model, the actively generated acto myosin tension in the cytoskeleton is balanced (1) internally by cytoskeletal filaments such as microtubules that act as molecular struts in the cell and bear compressional load, and (2) externally via the anchorage of the cytoskeleton to the extracellular matrix. The mechanical stability of the network thus depends on the architecture of the various elements in the network. This model was successful in explaining several phenomena as reviewed in Ainsworth (2009) and Ingber (2003b). For example, the suggestion that microtubules may behave as intercellular struts predicts that disruption of microtubules should result in the cell exerting stronger tractions on the extracellular matrix, as observed in experiments [although other explanations have also been given, see Ingber (2003b) for a more detailed discussion]. Another significant consequence of the tensegrity structure as derived from a simple model (Stamenovic and Ingber, 2002), is that the stiffness of the object increases linearly with the applied stress, consistent with some experiments on cells (Ingber, 2003b). ii. The semiflexible chain model: A complementary approach to explain how tension allows the cell to maintain and control its mechanical stability is based on the nonlinear mechanics of the cytoskeletal filaments themselves (Gardel et al., 2004; Storm et al., 2005). Among the most widely studied reconstituted systems are networks of actin filaments, whose single-chain properties are pretty well understood (Kasza et al., 2007; MacKintosh et al., 1995). Briefly, actin, like other cytoskeleton filaments, belongs to a class of polymers that are known as semiflexible (MacKintosh et al., 1995). Polymeric filaments are characterized by a persistence length which is the length above which thermally induced bending becomes appreciable. Flexible polymers have persistence lengths that are very short compared with their contour length (their fully extended length), while rigid filaments cannot be bent—their persistence length is longer than the polymer size. Semiflexible systems are intermediate between flexible and rigid filaments: their persistence length is comparable, but smaller than the fully extended length of the polymer. Thus, these filaments are neither “random” globules nor they are rigid and straight; they exhibit moderate thermal undulations which provide them with unique, nonlinear, mechanical properties (MacKintosh et al., 1995). In general, the stretching of a fluctuating chain reduces the number of thermally accessible

154

Rumi De et al.

(microscopic) chain configurations (the fully stretched chain has only one microscopic configuration). Thus, for moderate forces, the elasticity of semiflexible chains is entropic in origin; however, pulling the chain to extensions that are on the order of its contour length results in a significant rise of the chain energy that can even lead to the rupture. A detailed, statistical–mechanics calculation shows that the force–extension relation for semiflexible chains is nonlinear and that the force diverges universally as 1/(L – Lc)2 when the chain length L approaches the fully stretched contour length of the polymer, Lc (Bustamante et al., 1994; MacKintosh et al., 1995); that is, the stronger the force acting on the chain, the stiffer the chain becomes. Interestingly, even in a highly cross-linked network of semiflexible chains, where the distance between cross-linkers drops below the persistence length of the polymers, the same nonlinear dependence of the network elasticity on applied (shearing) force has been observed (Gardel et al., 2004; Storm et al., 2005). Typically, these gels show a linear phase where the elastic modulus of the network is constant but beyond some critical stress, the rigidity increases nonlinearly with the stress. From these studies, it is concluded that the nonlinear, stress-stiffening, behavior exhibited by these networks results from the loss of configurational entropy of the chains (Gardel et al., 2004; Storm et al., 2005). More recently, actin networks composed of the (large and) flexible, actin crosslinker, filamin A, have been studied (Broedersz et al., 2008; Gardel et al., 2006; Koenderink et al., 2009); in this case, the elasticity of the network is thought to be due to the compliance of the cross-linkers rather than relatively rigid actin chains (Broedersz et al., 2008). We note that in contrast to the case of rigid cross-linkers the density of flexible cross-linkers hardly changes the rigidity of the network (in the Pa range); however, application of external stresses of 100–1000 Pa, comparable to the stresses exerted by adherent cells, causes an increase in the rigidity of the network by a few orders of magnitude, reaching the kPa regime exhibited by cells (Fernadez et al., 2006; Wang et al., 2002b). Addition of myosin II to these actin– filamin networks resulted in an overall contraction of the system. The effect of the motors is similar to that of applied (shear) load ( Koenderink et al., 2009): myosin caused an increase in the stiffness of the network by 2 orders of magnitude. The elastic field produced by an isotropic and homogeneous distribution of myosin motors can be predicted from a model that assumes that each local actin–myosin interaction forms a force dipole (namely, two forces acting in opposite directions and separated by a short distance). The local pulling force of the dipoles propagates through the cytoskeleton and acts on the surface thereby mimicking the effect of an applied load (Carlsson, 2006; Koenderink et al., 2009; Zemel et al., 2010a). iii. The dipole polarization model: In the models that focus on the actin gel, the ability of the myosin motors to orient was disregarded and their distribution was assumed to be isotropic and uniform. However, an additional contribution to the apparent stiffening of the cytoskeleton can arise from the active polarizability of the acto myosin dipoles (Zemel et al., 2010a,b); that is, from the freedom of the acto myosin dipoles to reorganize and change both their orientation as well as the

7. Theoretical Concepts and Models of Cellular Mechanosensing

155

magnitude of the forces they exert in response to the local stresses. In particular, an active orientation of (contractile) acto-myosin dipoles in the direction of a tensile stress results in an active enhancement of the contractile restoring force of the system and opposes an applied stretch; this effect then acts as an effective stiffening of the system (Zemel et al., 2006).2 These conclusions are based on a variety of recent experiments that indeed demonstrate that the acto-myosin contractility in the cell changes in response to applied mechanical forces. The magnitude and relevant timescale of this response, however, have not been studied experimentally thus far. While these studies provide important insight into the nonlinear elastic nature of the cytoskeleton and its active regulation by molecular motors, the dynamic, viscoelastic properties of the cytoskeleton are much less understood at the molecular level—especially at short timescales (Trepat et al., 2008;). Recent studies reveal that the viscoelasticity of the cytoskeleton shares several features that are common to soft glassy materials (Trepat et al., 2007, 2008). In particular, the elastic (shear) modulus of the cytoskeleton shows a weak power-law dependence on the frequency of an applied oscillating load, implying that the cytoskeleton possesses a broad spectrum of relaxation times (Trepat et al., 2007). The origin of this behavior is unclear and may either reflect a high level of structural organization (Sollich, 1998) or a multitude of dynamical processes such as the binding/unbinding kinetics of the cross-linkers, polymerization or depolymerization of filaments, motion of molecular motors, intracellular trafficking, or actin–myosin contraction. In addition, in striking similarity to soft glassy materials, cells subjected to transient, stretch-and-release loads, abruptly soften and liquefy—following the instantaneous softening of the cytoskeleton its rigidity recovers back to baseline on a timescale of minutes (Trepat et al., 2007); this is in contrast to the stress-stiffening response observed when a cell is subjected to continuous load. While this behavior considered to be generic and universal, the mechanisms involved still remain elusive. Moreover, whether cells stiffen (and exert more force) or soften in response to applied stretch is not so far well understood. The work done by Brown and Eastwood (Brown et al., 1998) showed that cells on collagen matrix subjected to slowly varying stretches (on the scale of 30 min) exert smaller forces on their surroundings. On the other hand, the authors (Gavara et al., 2008) have shown that in response to quickly varying stretch, cells stiffen and increase the forces they exert. However, these studies were done on cells on polymeric substrates which are relatively stiff compared to collagen. The authors (Gavara et al., 2008) point out that both passive and active components of cell force need to be considered and that might show different responses to applied stretch. The work of Nekouzadeh et al. (2008) on cells in three-dimensional matrices indicates more complex dynamical responses with both stiffening and softening at different timescales. In addition to the intrinsic response of the cell, one has to also 2

A similar stiffening effect has been previously calculated for an ensemble of contractile cells in a gel (Zemel et al., 2006).

156

Rumi De et al.

consider matrix remodeling effects that may then feedback on cell activity and elasticity. The problem of whether cell force and stiffness increases or decreases in response to applied stretch (on various timescales) is a field that deserves further study.

V. Slow Mechanical Processes in the Cytoskeleton Thus far, we have focused on the viscoelastic properties of the cytoskeleton that are typically measured on timescales shorter than minutes. In addition, on this timescale, many of the reconstituted model systems such as actin filaments and various motor and cross-linking proteins are isotropic and homogeneous (on the scale of the viscoelastic probe). However, when cells are moved from one mechanical environment to another, such as when a cell adheres to a substrate or is acted upon by a force (for a sufficiently long time), large-scale remodeling processes take place in the cell that can lead to a change in its shape, cytoskeleton structure, and its anchorage to the extracellular matrix; these rearrangements and adjustment typically occur on a timescale of tens of minutes to days. At the end of this phase, an adhering cell acquires its “native” shape and internal cytoskeleton architecture. Thus, the active interplay of forces between the cell and its environment during cell adhesion, and their impact on long-term mechan ical processes in the cell, are important factors that determine cell behavior and fate. In the next section, we discuss a fundamental aspect of cell mechanosensitivity: stressfiber formation and its dependence on the cell shape and rigidity of the surroundings. A. Reorganization of Cytoskeletal Structure: Stress-Fiber Polarization The stresses generated in the cytoskeleton during cell adhesion and spreading play a central role in the determination of cytoskeletal structure (Chicurel et al., 1998; Kaunas et al., 2005; Pathak et al., 2008; Ren et al., 1999; Théry et al., 2006a; Wang et al., 2002a; Zemel et al., 2010a) and cell shape ( Bischofs et al., 2008; Paul et al., 2008; Tlusty et al., 1999). Of particular interest are the appearance in the cytoskeleton of long and thick acto-myosin bundles, known as stress fibers; these are formed in tens of minutes to hours after cell spreading (Hall, 1998; Ren et al., 1999; Théry et al., 2006a; Wang et al., 2002a). These fibers generally develop throughout the cytoskeleton and their density, orientation, and spatial distribution vary depending on cell type. The formation of stress fibers is governed both by soluble factors and mechanical forces (Hall, 1998). Stress fibers play an essential role in the regulation of cell adhesion. The tension produced by acto-myosin contraction is transmitted through the fibers to the focal adhesions where they activate integrin receptors and control the strength of cell adhesion to the extracellular matrix (Bershadsky et al., 2006; Burridge and ChrzanowskaWodnicka, 1996). Activation of integrins, in return, feeds back on myosin light chain assembly, and consequently stress-fiber formation, through activation of the small GTPase Rho—known to be a potent stimulator of stress-fiber formation (Burridge and Chrzanowska-Wodnicka, 1996; Cai et al., 2010; Chrzanowska-Wodnicka and Burridge, 1996; Geiger et al., 2009; Ren et al., 1999). This mechanically regulated

7. Theoretical Concepts and Models of Cellular Mechanosensing

157

feedback loop is thought to be responsible for both the early time initiation of stressfiber assembly (minutes to hours after cell adhesion), where the isotropic acto-myosin contraction results in organized anisotropic stress fibers (Besser and Schwarz, 2007; Chrzanowska-Wodnicka and Burridge, 1996; Ren et al., 1999; Zemel et al., 2010a), as well as for the longer time reinforcement of the cytoskeleton by well-developed stress fibers. A growing body of evidence indicates that stress-fiber organization plays a key role in a variety of biological processes ranging from the determination of cell shape (Noria et al., 2004; Théry et al., 2006a), global polarity (Théry et al., 2006b), and orientation in response to force (Iba and Sumpio, 1991; Kaunas et al., 2005; Noria et al., 2004), to stem cell differentiation (Pirone et al., 2004; Rodriguez et al., 2004; Wozniak et al., 2003) and cancer development (Paszek et al., 2005). It is thus of fundamental importance to understand the principles that govern stress-fiber assembly in cells. There are numerous experiments that indicate that tension is essential for stress-fiber formation (Burridge and Chrzanowska-Wodnicka, 1996; Chrzanowska-Wodnicka and Burridge, 1996; Ren et al., 1999). When cells are detached from the surface they adhere to (and this tension is released), or when myosin is inhibited, stress fibers disassemble in the cell (Burridge and Chrzanowska-Wodnicka, 1996). Conversely, when tension is applied to a localized region of the cell surface, a bundle of actin filaments is induced immediately adjacent to the site of applied tension and oriented in a direction parallel to the applied force (Kolega, 1986); a similar conclusion was obtained by attaching a fibronectin-coated bead to the cell membrane and restraining it with an optical trap (Choquet et al., 1997). In addition, since stress fibers typically terminate at focal adhesions, their development is likely to couple to the growth and maturation of focal adhesions and via this coupling, stress-fiber properties will depend on the forces acting on the focal adhesions (Riveline et al., 2001). The authors (Kozlov and Bershadsky, 2004; Shemesh and Kozlov, 2007; Shemesh et al., 2005a) have studied the formin family of proteins, known to be potent stimu lators of actin filament nucleation and elongation. Their theoretical studies predict that formin-mediated actin polymerization could be significantly enhanced by the application of pulling forces (pN) to the formin-capped filament end. Since formins are present in the focal-adhesion plaque, they may play a role in the control of actin polymerization and hence in the formation of stress fibers that emanate from focal adhesions. However, experiments show that the long-time formation of large, mature stress fibers is, in general, a global-cell phenomenon that is governed by the overall shape of the cell and elastic properties of the cell and its environment (Discher et al., 2005; Théry et al., 2006a; Wang et al., 2002a; Yeung et al., 2005). The authors (Wang et al., 2002a) and later also others (Théry et al., 2006a,b) were able to control the patterning of stress fibers and other cytoskeleton components in the cell by systematically manipulating the spreading area and shape of cells (see also Parker et al., 2002). Typically, one finds that the density of stress fibers is higher in cells with larger spreading areas; this is consistent with a variety of other experiments that have measured the force produced in the cell as a function of the rigidity of the environment (Ghibaudo et al., 2008; Paszek et al., 2005; Saez et al., 2005). In addition, stress fibers were more densely populated along the periphery and corners of the cells. Direct

158

Rumi De et al.

measurements of the forces generated by the cells in these experiments showed that regions with higher stresses, such as those along the perimeter and corners of the cell, had a larger population of stress fibers. Similarly, in more rigid environments, such as on a rigid substrate or within a gel with fixed boundaries, where more cell tension is produced, cells show more prominent stress fibers compared with softer systems, that is, soft substrates or floating gels with free boundaries (Discher et al., 2005). All these experi ments suggest that the spatial distribution and density of stress fibers in the cell may be a reflection of the stresses in the cytoskeleton. This, in turn, implies that global mechanical and geometrical properties that influence the mechanical stress in the cell also dictate the spatial distribution and alignment of stress fibers in the cell. Based on this hypothesis, a number of models were recently developed to predict the spatial and temporal distribution of stress fibers in cells in various mechanical conditions, in both the absence and presence of applied loads (Deshpande et al., 2006; Hsu et al., 2009; Zemel et al., 2010a,b). One group of authors (Deshpande et al., 2006, 2007a, 2008; Pathak et al., 2008) proposed a detailed biochemical and mechanical model that predicts the kinetics of stress fiber and focal-adhesion assembly in the cell by taking into account the kinetics of stress-fiber formation. The kinetics depends on a biochemical activation signal, the tension-dependent fiber dissociation rate, and the rate of force generation by myosin II motors. This model was extended later to include the coupling between stress-fiber formation and the kinetics of focal-adhesion growth (Deshpande et al., 2008; Pathak et al., 2008). The theoretical predictions are also compared with the systematic experimental study by Théry et al. (2006a), who investigated the relation between cell shape and the spatial distribution of stress fibers. The model successfully captures several experimental observations including: (1) the accumulation of focal-adhesions at the cell periphery, (2) the high density of stress fibers at the attachments to the extracellular matrix as well as at locations where external forces are imposed, and (3) the scaling of the force generated by the cell with the matrix rigidity. Stress-fiber orientation in response to cyclic strain applied to the substrate has also been predicted by these authors (Wei et al., 2008). An alternative approach was proposed by (Zemel et al., 2010a,b), who focused on the early stages (first hours) of cell adhesion where myosin II motors are not yet fully assembled in anisotropic stress fibers and the actin cytoskeleton is still in an isotropic gel state. The authors proposed a self-consistent model that couples the polarization (alignment) of individual acto-myosin force dipoles in the cytoskeleton to the elastic stress induced by these dipoles during cell adhesion and spreading. The initial, isotropic and uniform distribution of myosin II motors in the cytoskeleton produces an elastic stress field whose characteristics depend on the cell shape and the ratio of the Young’s modulus of the cell and its surroundings. The authors show that a small anisotropy in the cell shape can result in spontaneous polarization of these acto-myosin dipoles in a direction parallel to the long axis of the cell, as often observed in experiments (Curtis et al., 2006). The theory assumes an active feedback effect, namely, that the acto-myosin dipoles polarize in response to anisotropy of the elastic stress. Interestingly, these studies demonstrate both experimentally and theoretically that the polarization of acto-myosin dipoles (in adult mesenchymal stem cells) depends

7. Theoretical Concepts and Models of Cellular Mechanosensing

159

nonmonotonically on the rigidity of the extracellular matrix. The anisotropy of the stress-fiber orientation (the difference between the stress fibers polarized along the long axis and those polarized along the short axis of the cell) attains a maximum for a regime of matrix stiffness, where the cell and matrix are approximately matched. This anisotropy is reminiscent of the nematic order parameter of liquid crystals; however, in cells, the active elasticity as well as the matrix rigidity plays a crucial role. Although, the molecular mechanism of stress-fiber formation is not accounted by this model, the initial polarization of the acto-myosin dipoles in response to the anisotropic stress in the cell and the dependence on the geometry and elasticity of the cell and the surroundings is shown to govern the initial breaking of symmetry of the cytoskeleton. The same model was used by Zemel et al. (2010b) to explain the hyperbolic depen dence of the cell force and spread cell area on the matrix rigidity that is observed in experiments (Discher et al., 2005; Ghibaudo et al., 2008; Paszek et al., 2005; Saez et al., 2005). According to this model, the spread size of the cell and the forces it produces are dictated by the elastic balance of forces exerted by the cell and the surroundings. In a more rigid environment, cells spread more, assume larger sizes, and produce stronger forces since the elastic stress produced in the cell (whose spreading is opposed by the active, cytoskeletal contractility) can be balanced by the larger elasticity of the surrounding matrix. Finally, recent work by Schwarz and coworkers (Bischofs et al., 2008, 2009) focuses on stress fibers that are located at the cell boundary. They show that a consistent fit to experiments can be obtained if one assumes that the resulting line tension of the boundary is not a constant, but depends on the distance between the points to which the boundary is pinned; this effect is due to stretching of the stress fibers that are located along the boundary. The models described above deal with the spontaneous formation of stress fibers during cell adhesion in the absence of applied forces. The elastic stresses that influence stress-fiber polarization can be applied externally or can be generated internally by the contractile force of the cell. In the latter case, the spontaneous polarization of the stress fibers is termed self-polarization (Zemel et al., 2010a). This response occurs at relatively long times (typically tens of minutes), after the cell has a sufficient period in which it has spread and can then reorganize its stress fibers in response to stress. This regime is thus relevant for cell response to an external static field but not to a rapidly varying field.

VI. Cellular Response to External Mechanical Stress Cells in tissues are subjected to a variety of mechanical forces that influence their behavior and alignment. These forces can arise from gravity, muscle tension, blood pressure, as well as from local active tractions of nearby cells. These forces can be static as well as time varying, for example, continuous loading occurs during development of long bone growth while cyclic loading occurs due to periodic blood pressure varia tions. Cells respond to these forces by actively adjusting their internal activities and

160

Rumi De et al.

contractile forces. Experiments on adherent cells in elastic substrates subjected to timevarying strains as well as measurements of the corresponding traction map provide a valuable tool to understand both the cytoskeleton and nuclear responses to external forces. Since the propagation of elastic forces within the tissue can be long ranged, the local stress field sensed by the cells can also reflects the global mechanical character istics of the tissue, such as its shape and its stiffness relative to its surroundings. Various theoretical studies have been carried out to understand cellular responses as well as orientation of stress fibers in presence of applied forces (Civelekoglu-Scholey et al., 2005; Hsu et al., 2009; Pirentis and Lazopoulos, 2009; Wei et al., 2008). The authors (Hsu et al., 2009) proposed a simple stochastic model to predict the dynamics of stress-fiber orientation in response to an oscillating load. This model is based on a hypothesis that the stress fibers require tension to develop; however, they depolymer ize when the tension reaches above a certain threshold value (Lu et al., 2008). The model predicts that reorganization of the stress fibers is determined by the competition between the rate of stress-fiber assembly and the load-dependent disassembly. On rigid substrates under isometric tension, stress fibers orient isotropically in the cell; however, in presence of a high-frequency oscillating axial load, the anisotropic load-dependent disassembly rate exceeds the isotropic assembly rate and as a result, the stress fibers accumulate nearly perpendicular to the load. It provides a simple mechanism that explains the orientation of stress fibers away from the stretch direction in response to an oscillating load at high frequency. Another theory by Wei et al. (2008) predicts the orientation of stress fibers in response to cyclic stretch based on a biochemo-mechanical model (discussed in the previous section) by relating the con traction and extension rate sensitivity of the stress fibers to the magnitude and frequency of the applied stress. An alternative, more generic mechanism—that coarse grains over the stress fibers, myosin activity, the adhesions, and lumps all these into a cell force dipole—is used to predict the cell orientation in response to oscillating loads by the authors (De and Safran, 2008; De et al., 2007, 2008), as we now discuss.

A. Active Elastic Dipole Model of Cells The theoretical model proposed by the authors (De and Safran, 2008; De et al., 2007, 2008), idealizes a stationary, adhering cell as an elastic force dipole that can change its contractile activity and orientation by reorganizing its focal adhesions, myosin activity, and stress fibers in response to external forces. The dipole description coarse grains over the effects of the stress fibers, myosin, and the focal adhesions to provide a generic, “lumped” description of a contractile cell (Bischofs et al., 2004; De et al., 2007; Schwarz and Safran, 2002). It has the advantage of being general with only two degrees of freedom: the dipole magnitude and orientation, but with the disadvantage that it lumps several molecular processes into these coarse-grained variables. Anchorage-dependent cells constantly assemble and disassemble focal adhe sions to probe the mechanical properties of their environment as well as the presence of external stresses. The sum of the forces exerted by focal adhesions can be modeled as a

7. Theoretical Concepts and Models of Cellular Mechanosensing

161

pair of equal and oppositely directed contraction forces. The assumption that the forces are equal in magnitude is appropriate for stationary or very slowly moving cells where the net force is approximately zero (Balaban et al., 2001). Each cell, in this coarsegrained picture, is modeled by an anisotropic force dipole tensor, Pij = li fj ; that is the product of the distance ðli Þ between the two equal but oppositely directed forces ðfj Þ due to actin–myosin contractility. This model is used to show how the combination of active cellular forces due to the reorganization of the cytoskeleton and the elastic forces exerted by the matrix determine cell orientation in the presence of time-varying stresses (De et al., 2007, 2008; Safran and De, 2009). These authors proposed a model that addressed a long-standing puzzle: why do many cells align parallel to the stress direction for slow varying (nearly static, on the scale of tens of minutes) applied stresses (Brown et al., 1998; Collinsworth et al., 2000; Eastwood et al., 1998; Vandenburgh, 1988), but align nearly perpendicular to the stress direction (in some cases in the zero strain direction) for rapidly varying stresses (on the scale of seconds) (Hayakawa et al., 2001; Jungbauer et al., 2008; Kurpinski et al., 2006; Shirinsky et al., 1989; Smith et al., 1997; Takemasa et al., 1997; Wang and Grood, 2000; Wang et al., 2001). The theoretical explanation is based on a hypothesis of “tensional homeostasis” assuming that cells reorganize their focal adhe sion and stress fibers in order to maintain an optimal set-point stress or strain in the adjacent matrix (Brown et al., 1998; Freyman et al., 2002; Saez et al., 2005). Any deviation from the set-point condition results in the development of internal forces within the cell that reestablish the optimal force or strain condition. These forces can be derived from the generalized gradients of an effective free energy, F, which includes all the internal activity within the cell that reestablishes the cellular response to its local environment. The effective free energy is function of the two degrees of freedom of the problem: the dipole magnitude and its direction. For simplicity, the theory focuses on cells that show bipolar morphologies, for example, muscle cells and fibroblasts, where the forces and the relative distance are both in the direction of the long axis of the cell, taken to be along the z direction (as shown in Fig. 2); the force dipole is thus written, Pij = Piz jz . In the presence of an external uniaxial stress, a, applied at an angle relative to the cell axis (taken to be in the z direction) the effective free energy is written as 1 Fc = P2 ½p þ pa ðtÞðf f1 Þ 12 2

ð1Þ

where ¼ cos2 and is a measure of cell activity, that is, the tendency of the cell to reorganize its focal adhesions and stress fibers if the stress in the matrix is not at its setpoint value, P . This form is a generalization of the cases (De and Safran, 2008; De et al., 2008) in which (1) the cellular dipole is controlled by the matrix stress where 1 = 0 and (2) the dipole is controlled by the matrix strain and 1 = cos2 0 = 0 , where 0 is the zero strain direction given by cos2 0 = =ð1 þ Þ; is the Poisson ratio of the matrix. We define dimensionless quantities: p = P=P , pa ðtÞ = Pa ðtÞ=ð0 P Þ which are the cell dipole and external stress relative to the set point, P , where 0 is a

162

Rumi De et al.

F P θ P R

Z

F

Static or quasi-static

F

High frequency

Fig. 2 Stem cell and coarse-grained dipole model. The fluorescence image shows the cytoskeletal actin fibers (red) that generate stress, the sites of adhesion (green) to the substrate, and the cell nucleus (blue). The model consists of a contractile force dipole P along the z-axis oriented at an angle to the direction of an external force field F (or external stress field a). R is the reaction stress in the adjacent elastic matrix due to the cells contractility. In the static and low-frequency case, the cell aligns parallel to the strain; at higher frequencies, the cell orients nearly perpendicular to the oscillating stretch (Rehfeldt and Discher, 2007). Reprinted with permission from the publishers. (See Plate no. 4 in the Color Plate Section.)

163

7. Theoretical Concepts and Models of Cellular Mechanosensing

function of . The scaled stress is written as Pa = a a3 , which has the dimensions of energy so that the external stress and dipole strength are expressed in the same units [for a detailed description of the model, see Safran and De (2009)]. In addition to forces that originate from cell activity and the tendency to return to homeostasis, there are also mechanical matrix forces that act on the cell in the presence of external stresses. An important contribution is the interaction energy of a force dipole with an external strain field, uaij ; this energy is proportional to the product of the force dipole and the external strain; thus, the matrix energy is Fm = Pij uaij . F = Fc þ Fm represents the energy that the cell contractile dipole invests in deforming the matrix in the presence of the external stress.3 The effective free energy, F = Fc þ Fm , is used to derive generalized forces that drive the dynamics of the dipole magnitude and its direction, to balance the effects of the external stress and the tendency of the cell to achieve homeostasis (De and Safran, 2008; De et al., 2007; Safran and De, 2009). This then allows the theory to predict the orientational response of cells in the presence of time-varying external stresses, Pa ð1 cos !a tÞ (where !a is the frequency of the cyclic stress).4 For simple relaxational kinetics in which the dynamics of the dipole are proportional to the generalized forces (given by the gradients of the free energy), the dynamical equations for the dipole magnitude, P, and its orientation, , are given by dP 1 ∂F = ; dt

P ∂P

d 1 ∂F = dt

∂

ð2Þ

where, P and are the relaxation times for the readjustment of the magnitude and orientation of the force dipole. Intuitively, we expect that the time for the liquification and repolymerization of the stress fibers and adhesions, P , is significantly shorter than the time required for the reorientation of the cell, . This separation of timescales is used in the theoretical analysis (De and Safran, 2008; Safran and De, 2009) and is consistent with the experimental observations of a crossover between two dynamical regimes (Jungbauer et al., 2008). The theory predicts several properties that are measured experimentally, including the change of cellular contractile forces in both the absence and presence of applied mechanical stress and more interestingly, the parallel alignment of cells in response to static or quasi-static stresses and the nearly perpendicular alignment of cells in presence of high-frequency dynamic stresses. In addition, the effects of random fluctuations of the cellular orientation due to “noise” in either the cell or its surroundings can be accounted for in the model (Safran and De, 2009). This predicts the distribution of cell orientations which has been observed

3

There is also a self energy which is the energy the cell invests in deforming the matrix even in the absence of external stress, but since this energy is independent of the cellular orientation, it is omitted here. The contribution of this term to the dynamics is discussed in De and Safran (2008). 4 It is important to emphasize that F is not a thermodynamic free energy, but is used to derive forces acting on the system since dynamics of cell orientation requires a calculation of these forces. The steady states of the dynamical system are equivalent to finding the local minima of the effective free energy where all of the forces balance and the steady state is maintained as long as there is sufficient ATP in the cell.

164

Rumi De et al.

(Jungbauer et al., 2008; Kemkemer et al., 1999; Wang et al., 2001) and whose quantitative nature is currently under study (Zedler et al., to be published).

B. Predictions of the Model

1. Parallel Versus Perpendicular Orientations The theory predicts cellular orientation in response to a wide range of amplitudes and frequencies of the applied cyclic stress, as well as in the presence of random forces which may arise due to internal cell activities and molecular interaction between cells and their environment. Random forces are modeled as a thermal type noise, with a scaled, effective temperature, Ts (Safran and De, 2009). The competitions of the cell activity with the matrix forces and with the noise in the system determine the steadystate cellular orientation. At low frequency, cells have sufficient time to readjust their contractile activities by reorganizing the cytoskeleton and thus balance the active cell forces by matching their internal forces to the optimal set point in the matrix. Since the internal forces can be balanced, the cell orientation is then determined by the matrix forces that cause cells to orient parallel to the external stress field—at least in the absence of noise. However, the presence of noise causes the orientation to be random as shown in Fig. 3. On the other hand, for high-frequency applied stress, the cell cannot follow the quickly varying stress to establish its set point, whose timescale is typically in the regime of tens of minutes; the forces due to cellular activity thus tend to orient the cell perpendicular to the stress direction so that it remains unaffected by the external stress and can reach homeostasis in the adjacent matrix. For the case of set points determined by matrix strain, the cell “escapes” pﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ the high-frequency strain by ﬃ orienting in the zero strain direction given by cos = =ð þ 1Þ (De and Safran, 2008; De et al., 2008; Safran and De, 2009). The high-frequency regime is less affected by noise than the low-frequency response as shown in Fig. 3. This is because the forces due to cellular activity are taken to be much larger than the matrix forces [otherwise, there would be no perpendicular (or zero strain direction) orientation at high frequencies]; the latter may be comparable to the noise amplitude. Since the highfrequency response is governed by the cellular activity, it is less sensitive to the noise. We note that the question of cell orientation under very slow loading (on the scale of tens of minutes) may not be a universal feature and may depend on cell type. Some cells on relatively rigid surfaces seem to show random orientation (Jungbauer et al., 2008), while studies of cells in collagen (Brown et al., 1998; Eastwood et al., 1998) have been reported to align parallel to the applied, slowly varying stress. This is an area that deserves further study (Zedler et al., to be published).

2. Cellular Relaxation Time Experiments show that cytoskeletal reorganization in response to applied forces is a complex process that involves multiple timescales. Polymerization and depolymeriza tion of proteins, generation of active contractile forces by motor proteins, energy

165

7. Theoretical Concepts and Models of Cellular Mechanosensing

<φ> 1.0 0.8 0.6 0.4 0.2

0.001

0.01

0.1

1

10

100

0.01

0.1

1

10

100

ω

< φ> 1.0 0.8 0.6 0.4 0.2

0.001

ω

Fig. 3 Theoretical prediction of cell orientation as a function of frequency of the external stress. The average of ðÞ = ðcos2 Þ (in steady state) versus the scaled frequency ! = !a p for cells controlled by stress (top plot) and strain (bottom plot). The scaled temperatures are Ts = 10, 1, 0.1, and 0.001 as one looks at the graphs from top to bottom in the top panel at high frequencies and in the bottom panel from the bottom to the top at low frequencies [for details, see Safran and De (2009)]. Reprinted with permission from the publishers.

consumption by the hydrolysis from ATP to ADP, ion concentration, compliance of the extracellular matrix, and several other factors related to the mechanical and chemical environment, all play a role in determining the timescales of different cellular func tions. For example, the lifetime of biomolecular focal adhesion bonds ranges from a fraction of a second to around 100 s (Chesla et al., 1998; Evans and Calderwood, 2007). Generally, the timescale associated with the regulation of cellular forces involves reorganization of the focal adhesions and actin polymerization–depolymer ization cycles that occur on the timescale of seconds to minutes. On the other hand, the

166

Rumi De et al.

timescale associated with the cellular orientational response is a much longer process since it involves cooperative reorganization of whole cell cytoskeleton; this time is on the order of tens of minutes to hours and is supported by several experiments (Jungbauer et al., 2008; Yoshigi et al., 2005). Interestingly, recent experimental studies show that the characteristic time required for the cell to attain its steady-state orientation strongly depends on the frequency of the external mechanical stress (Jungbauer et al., 2008). The experiments show two distinct frequency regimes: for high frequencies > 1 Hz, the characteristic time is a constant; on the other hand, at lower frequencies, the characteristic time increases as the frequency decreases (see Fig. 4A). The characteristic time itself has a very large

(A)

2.5 × 104 2 × 104 1.5 × 10

REF52wt HDF

4

τ (s)

104

5 × 103

0,1 (B)

1 Frequency f (s–1)

10

τ 28,000.

24.000.

20,000

16,000

0.2

0.5

1

10

ω

100

Fig. 4 (A) Biphasic characteristic time ( ) measured in experiment; for frequencies below 1 Hz, it decreases as the frequency increases; for frequencies above 1 Hz, it saturates at a constant value (Jungbauer et al., 2008). (B) Theoretical prediction of the characteristic time ( ) as a function of frequency (! = !a P) of the external stress; exhibits two distinct regimes as observed in experiments. At high frequencies, approaches a constant value that scales with the long timescale and at low frequencies, it increases as 1=!2 (Safran and De, 2009). Reprinted with permission from the publishers.

7. Theoretical Concepts and Models of Cellular Mechanosensing

167

values (measured to be on the order of 104 s), while the crossover time is on the order of 1 s. The theory proposed by the authors (Safran and De, 2009) predicts these two very different timescales for the problem and the scaling with frequency is in qualitative agreement with experiment as shown in Fig. 4. The theoretical predictions are derived using an “adiabatic approximation” in which there is a separation of timescales into a slow scale that is related to the orientation dynamics and a fast timescale that is related to the magnitude of the stress fiber and adhesion remodeling (Safran and De, 2009). The authors explain the crossover between the two distinct frequency regimes that are observed in the experiments and also identify the physical origin of the two timescales (the short timescale related to stress fiber/adhesion density and the long timescale to orientation).

3. Homeostasis Set Point: Stress Versus Strain Many experimental studies show that cells respond to mechanical forces by remo deling their stress fibers and traction forces to maintain a tactile set point in the adjacent matrix. However, whether cell mechanosensitivity and hence its active response and set point is controlled by stress (force) in the extracellular matrix or by strain (matrix deformation) has not yet been resolved. Measurements of the traction forces between cells and the substrate by Saez et al. (2005) suggest that cellular forces are governed by the deformation of the matrix and that the cell maintains an optimal strain in the matrix. This interpretation has been questioned (Nicolas and Safran, 2006) since cellular adhesions increase in size on more rigid matrices and this might be the reason that the force is observed to increase with rigidity. On the experimental side, the authors (Freyman et al., 2002) measured the macroscopic contraction of the substrate and concluded that the force is maintained at an optimal value. The model described above (De et al., 2007, 2008; De and Safran, 2008; Safran and De, 2009) suggests that measurements of steady-state cell orientation angle as a function of the Poisson ratio, , of the matrix can identify whether the stress or strain dominates the regulation of cellular activity. The theory predicts that in presence of a high-frequency dynamic stress, the cell is frustrated by having to adapt its relatively slowly varying internal activity to the instantaneous, dynamically varying stress at a timescale that is too fast for its homeostatic response. The cell thus chooses a direction in which there is no applied stress and where it can establish its contractile machinery in a static manner to achieve the optimal set point, with no dynamically induced frustration. If the cellular activity is controlled by the strain pﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ of the medium, the ﬃ cell chooses a direction where there is no applied strain: cos = =ð þ 1Þ. For the case of > 0, the matrix is stretched in the direction of the uniaxial stress and is compressed in the perpendicular direction; in between these two directions, there is a region of zero strain direction that strongly depends on the Poisson ratio of the medium. However, if the cellular set point is determined by the stress in the medium, then the cell orients in the perpendicular direction to avoid the external stress since perpendicular direction is the zero stress direction for an applied uniaxial tensile stress.

168

Rumi De et al.

VII. Discussion The theory discussed above predicts many features observed in experimental mea surements of cellular forces and orientation. It provides a generic platform that can be elaborated and modified into a more realistic, quantitative model in order to obtain insight into the mechanical response of living cells. The theory focuses on the global relation between the cell force regulation and the stress field in the cell and enables a generic, conceptual picture of cell dynamics; this is in contrast to more microscopic models that focus on the complex biochemical rate equations whose generic features may be difficult to discern. However, many of the nonequilibrium aspects have been lumped into a small number of model parameters; therefore, many molecular details involved in the mechanosensitivity are not addressed in this approach. There are still many interesting and important questions that remain to be resolved. Even at the “coarse-grained” level, one would like to have a comprehensive experimental picture of the response of cell orientation to slowly varying stresses. In addition, the question of whether cell force increases or decreases with applied stress (either slow or fast) and how to separate the active from passive effects requires further study. The origin and theoretical nature of the “noise” in the coarse-grained picture of cell mechanical response is not well characterized (Mizuno et al., 2007). Can the noise be modeled as an effective temperature that just smears the distributions in a Boltzmann-like manner or are the timescales involved in the noise relevant to the cellular response (Mizuno et al., 2007; Safran and De, 2009)? Is the noise related to inhomogeneities in the surface treatment or does it arise in the cell activity itself? How general are the dynamical responses observed to date? How do they vary in mature versus immature (e.g., stem) cells? In the generic, coarse-grained theories, what is the dependence of the various parameters (e.g., the set-point value of the dipole, the cell activity force, and the dipole magnitude and orientation timescales) on the rigidity of the substrate? At the molecular level there are also fundamental issues that must still be resolved. For example, it remains unclear how on the molecular level, stress-fiber assembly and disassembly are governed by force, which molecules are involved in the mechanosen sing, how the thickness of a stress-fiber bundle is determined by the force it sustains, and how these processes are coupled to focal adhesion maturation and degeneration. Some progress has been made recently on the process of stress-fiber assembly on the molecular level (Hotulainen and Lappalainen, 2006; Vavylonis et al., 2008; Zimerman et al., 2004). All these are interesting questions that have yet to be answered. New approaches are needed to develop, new experimental methods and further theoretical models will help to understand and rationalize the new findings.

Acknowledgments RD gratefully acknowledges the support from University of Southampton during the writing stages of this article. AZ acknowledges support of the Israel Science Foundation. SAS wishes to acknowledge the Israel Science Foundation for its support as well as the historic generosity of the Perlman Family Foundation. The Schmidt Minerva Center and the Clore Center for Biological Physics are also acknowledged.

7. Theoretical Concepts and Models of Cellular Mechanosensing

169

References Ainsworth, C. (2009). Stretching the imagination. Nature 456, 696–699. Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K., and Watson, J. D. (1994). “Molecular Biology of the Cell,” 3rd edn. Garland Publishing, New York. Allioux-Guerin, M., Icard-Arcizet, D., Durieux, C., Henon, S., Gallet, F., Mevel, J. C., Masse, M. J., Tramier, M., and Coppey-Moisan, M. (2008). Spatio-temporal analysis of cell response to a rigidity gradient: A quantitative study by multiple optical tweezers. Biophys. J. 96, 238–247. Balaban, N. Q., Schwarz, U. S., Riveline, D., Goichberg, P., Tzur, G., Sabanay, I., Mahalu, D., Safran, S., Bershadsky, A., Addadi, L., and Geiger, B. (2001). Force and focal adhesion assembly: A close relation ship studied using elastic micropatterned substrates. Nat. Cell Biol., 3, 466–472. Bell, G. I. (1978). Models for the specific adhesion of cells to cells. Science 200, 618–627. Bershadsky, A. D., Balaban, N. Q. and Geiger, B. (2003). Adhesion-dependent cell mechanosensitivity. Annu. Rev. Cell Dev. Biol. 19, 677–695. Bershadsky, A., Kozlov, M., and Geiger, B. (2006). Adhesion-mediated mechanosensitivity: A time to experiment, and a time to theorize. Curr. Opin. Cell Biol. 18, 472–481. Besser, A. and Safran, S. A. (2006). Force-induced adsorption and anisotropic growth of focal adhesions. Biophys. J. 90, 3469–3484. Besser, A., and Schwarz, U. S. (2007). Coupling biochemistry and mechanics in cell adhesion: A model for inhomogeneous stress fiber contraction. New J. Phys. 9, 425. Bhattacharya, D., Talwar, S., Mazumder, A. and Shivashankar, G. V. (2009). Spatio-temporal plasticity in chromatin organization in mouse cell differentiation and during drosophila embryogenesis. Biophys. J. 96, 3832–3839. Bischofs, I. B., Safran, S. A., and Schwarz, U. S. (2004). Elastic interactions of active cells with soft materials. Phys. Rev. E 69, 021911. Bischofs, I. B., Klein, F., Lehnert, D., Bastmeyer, M., and Schwarz, U. S. (2008). Filamentous network mechanics and active contractility determine cell and tissue shape. Biophys. J. 95, 3488–3496. Bischofs, I. B., Schmidt, S. S. and Schwarz, U. S. (2009). Effect of adhesion geometry and rigidity on cellular force distributions. Phys. Rev. Lett. 103, 048101. Bissell, M. J. (2007). Modelling molecular mechanisms of breast cancer and invasion: Lessons from the normal gland. Biochem. Soc. Trans. 35, 1. Brangwynne, C. P., MacKintosh, F. C., Kumar, S., Geisse, N. A., Talbot, J., Mahadevan, L., Parker, K. K., Ingber, D. E., and Weitz, D. A. (2006). Microtubules can bear enhanced compressive loads in living cells because of lateral reinforcement. J. Cell Biol. 173, 733–741. Broedersz, C. P., Storm, C., and MacKintosh, F. C. (2008). Nonlinear elasticity of composite networks of stiff biopolymers with flexible linkers. Phys. Rev. Lett. 101, 118103. Brown, R. A., Prajapati, R., McGrouther, D. A., Yannas, I. V., and Eastwood, M. (1998). Tensional home ostasis in dermal fibroblasts: Mechanical responses to mechanical loading in three-dimensional substrates. J. Cell Physiol., 175, 323–332. Bruinsma, R., Behrisch, A., and Sackmann, E. (2000). Adhesive switching of membranes: Experiment and theory. Phys. Rev. E 61, 4253–4267. Burridge, K., and Chrzanowska-Wodnicka, M. (1996). Focal adhesions, contractility, and signaling. Annu. Rev. Cell Dev. Biol. 12, 463–519. Bustamante, C., Marko, J. F., Siggia, E. D., and Smith, S. (1994). Entropic elasticity of lambda-phage DNA. Science 265, 1599–1600. Cai, Y., Rossier, O., Gauthier, N. C., Biais, N., Fardin, M. A., Zhang, X., Miller, L. W., Ladoux, B., Cornish, V. W., and Sheetz, M. P. (2010). Cytoskeletal coherence requires myosin II contractility. J. Cell Sci. 123, 413–423. Carlsson, A. E. (2006). Contractile stress generation by actomyosin gels. Phys. Rev. E 74, 051912. Chan, C. E., and Odde, D. J. (2008). Traction dynamics of filopodia on compliant substrates. Science 322, 1687–1691. Chen, C. S. (2008). Mechanotransduction—a field pulling together? J. Cell Sci. 121, 3285–3292. Chen, C. S., Mrksich, M., Huang, S., Whitesides, G. M., and Ingber, D. E. (1997). Geometric control of cell life and death. Science 276, 1425–1428.

170

Rumi De et al. Chen, C. S., Alonso, J. L., Ostuni, E., Whitesides, G. M., and Ingber, D. E. (2003). Cell shape provides global control of focal adhesion assembly. Biochem. Biophys. Res. Commun. 307, 355–361. Chesla, S. E., Selvaraj, P., and Zhu, C. (1998). Measuring two-dimensional receptor-ligand binding kinetics by micropipette. Biophys. J. 75, 1553–1572. Chicurel, M. E., Chen, C. S., and Ingber, D. E. (1998). Cellular control lies in the balance of forces. Curr. Opin. Cell Biol. 10, 232–239. Choquet, D., Felsenfeld, D. P., and Sheetz, M. P. (1997). Extracellular matrix rigidity causes strengthening of integrin-cytoskeleton linkages. Cell 88, 39–48. Chrzanowska-Wodnicka, M., and Burridge, K. (1996). Rho-stimulated contractility drives the formation of stress fibers and focal adhesions. J. Cell Biol. 133, 1403–1415. Civelekoglu-Scholey, G., Orrc, A. W., Novak, I., Meistera, J. J., Schwartz, M. A., and Mogilner, A. (2005). Model of coupled transient changes of Rac, Rho, adhesions and stress fibers alignment in endothelial cells responding to shear stress. J. Theor. Biol. 232, 569–585. Cluzel, C., Saltel, F., Lussi, J., Paulhe, F., Imhof, B. A., and Wehrle-Haller, B. (2005). The mechanisms and dynamics of (alpha)v(beta)3 integrin clustering in living cells. J. Cell Biol. 171, 383–392. Collinsworth, A. M., Torgan, C. E., Nagda, S. N., Rajalingam, R. J., Kraus, W. E., and Truskey, G. A. (2000). Orientation and length of mammalian skeletal myocytes in response to a unidirectional stretch. Cell Tissue Res. 302, 243–251. Colombelli, J., Besser, A., Kress, H., Reynaud, E. G., Girard, P., Caussinus, E., Haselmann, U., Small, J. V., Schwarz, U. S., and Stelzer, E.H.K. (2009). Mechanosensing in actin stress fibers revealed by a close correlation between force and protein localization. J. Cell Sci. 122, 1665–1679. Curtis, A., Aitchison, G., and Tsapikouni, T. (2006). Orthogonal (transverse) arrangements of actin in endothelia and fibroblasts. J. R. Soc. Interface 3, 753–756. Cusachs, P. R., Alcaraz, J., Sunyer, R., Samitier, J., Farré, R., and Navajas, D. (2008). Micropatterning of single endothelial cell shape reveals a tight coupling between nuclear volume in g1 and proliferation. Biophys. J. 94, 4984–4995. Dahl, K. N., Ribeiro, A., and Lammerding, J. (2008). Nuclear shape, mechanics, and mechanotransduction. Circ. Res. 102, 1307–1318. De, R., and Safran, S. A. (2008). Dynamical theory of active cellular response to external stress. Phys. Rev. E 78, 31923–31940. De, R., Zemel, A., and Safran, S. A. (2007). Dynamics of cell orientation. Nat. Phys. 3, 655–659. De, R., Zemel, A., and Safran, S. A. (2008). Do cells sense stress or strain? Measurement of cellular orientation can provide a clue. Biophys. J. Lett. 94, L29–L31. Dechat, T., Adam, S. A. and Goldman, R. D. (2009). Nuclear lamins and chromatin: When structure meets function. Adv. Enzyme Regul. 49, 157–166. deGennes, P. G., Puech, P. H., and Brochard-Wyart, F. (2003). Adhesion induced by mobile stickers: A list of scenarios. Langmuir 19, 7112–7119. Deng, L. H., Trepat, X., Butler, J. P., Millet, E., Morgan, K. G., Weitz, D. A., and Fredberg, J. J. (2006). Fast and slow dynamics of the cytoskeleton. Nat. Mater. 5, 636–640. Deshpande, V. S., McMeeking, R. M., and Evans, A. G. (2006). A bio-chemo-mechanical model for cell contractility.Proc. Natl. Acad. Sci. USA 103, 14015–14020. Deshpande, V. S., McMeeking, R. M., and Evans, A. G. (2007a). A model for the contractility of the cytoskeleton including the effects of stress-fiber formation and dissociation. Proc. R. Soc. A 463, 787–815. Deshpande, V. S., Mrksich, M., McMeeking, R. M., and Evans, A. G. (2007b). A bio-mechanical model for coupling cell contractility with focal adhesion formation. J. Mech. Phys. Solids 56, 1484–1510. Deshpande, V. S.,Mrksich, M., McMeeking, R. M., and Evans, A. G. (2008). A bio-mechanical model for coupling cell contractility with focal adhesion formation. J. Mech. Phys. Solids 56, 1484–1510. Discher, D. E., Janmey, P., and Wang, Y. (2005). Tissue cells feel and respond to the stiffness of their substrate. Science 310, 1139–1143. Discher, D. E., Mooney, D. J., and Zandstra, P. W. (2009). Growth factors, matrices, and forces combine and control stem cells. Science 324, 1673–1677.

7. Theoretical Concepts and Models of Cellular Mechanosensing

171

Döbereiner, H. G., Dubin-Thaler, B., Giannone, G., Xenias, H. S., and Sheetz, M. P. (2004). Dynamic phase transitions in cell spreading. Phys. Rev. Lett. 93, 108105. Eastwood, M., Mudera, C. V., McGrouther, D. A., and Brown, R. A. (1998). Effect of precise mechanical loading on fibroblast populated collagen lattices: Morphological changes. Cell Motil. Cytoskeleton 40, 13–21. Engler, A. J., Griffin, M. A., Sen, S., Bönnemann, C. G., Sweeney, H. L., and Discher, D. E. (2004). Myotubes differentiate optimally on substrates with tissue-like stiffness: Pathological implications for soft or stiff microenvironments. J. Cell Biol. 166, 877–887. Engler, A. J., Sen, S., Sweeney, H. L., and Discher, D. E. (2006). Matrix elasticity directs stem cell lineage specification. Cell 126, 677–689. Evans, E. A., and Calderwood, D. A. (2007). Forces and bond dynamics in cell adhesion. Science 316, 1148–1153. Fernadez, P., Pullarkat, P. A., and Ott, A. (2006). A master relation defines the nonlinear viscoelasticity of single fibroblasts. Biophys. J. 90, 3796–3805. Fletcher, D. A., and Mullins, R. D. (2010). Cell mechanics and the cytoskeleton. Nature 463, 485–492. Freyman, T. M., Yannas, I. V., Yokoo, R., and Gibson, L. J. (2002). Fibroblast contractile force is independent of the stiffness which resists the contraction. Exp. Cell Res. 272, 153–162. Galbraith, C. G., and Sheetz, M. P. (1997). A micromachined device provides a new bend on fibroblast traction forces. Proc. Natl. Acad. Sci. USA 94, 9114–9118. Gardel, M. L., Shin, J. H., MacKintosh, F. C., Mahadevan, L., Matsudaira, P., and Weitz, D. A. (2004). Elastic behavior of cross-linked and bundled actin networks. Science 304, 1301–1305. Gardel, M. L., Nakamura, F., Hartwig, J., Crocker, J. C., Stossel, T. P., and Weitz, D. A. (2006). Prestressed factin networks cross-linked by hinged filamins replicate mechanical properties of cells. Proc. Natl. Acad. Sci. USA 103, 1762–1767. Gavara, N., Roca-Cusachs, P., Sunyer, R., Farrè, R., and Navajas, D. (2008). Mapping cell-matrix stresses during stretch reveals inelastic reorganization of the cytoskeleton. Biophys. J. 95, 464–471. Geiger, B., and Bershadsky, A. (2002). Exploring the neighborhood adhesion-coupled cell mechanosensors. Cell 110, 139–142. Geiger, B., Spatz, J. P., and Bershadsky, A. D. (2009). Environmental sensing through focal adhesions. Nat. Rev. Mol. Cell Biol. 10, 21–33. Ghibaudo, M., Saez, A., Trichet, L., Xayaphoummine, A., Browaeys, J., Silberzan, P., Buguinb, A., and Ladoux, B. (2008). Traction forces and rigidity sensing regulate cell functions. Soft Matter 4, 1836–1843. Gov, N. S., and Gopinathany, A. (2006). Dynamics of membranes driven by actin polymerization. Biophys. J. 90, 454–469. Griffin, M. A., Engler, A. J., Barber, T. A., Healy, K. E., Sweeney, H. L., and Discher, D. E. (2004). Patterning, prestress, and peeling dynamics of myocytes. Biophys. J. 86, 1209–1222. Guo, W. H., Frey, M. T., Burnham, N. A., and Wang, Y. L. (2006). Substrate rigidity regulates the formation and maintenance of tissues. Biophys. J., 90, 2213–2220. Hall, A. (1998). Rho GTPases and the actin cytoskeleton. Science 23, 509–514. Harris, A. K., Wild, P., and Stopak, D. (1980). Silicone rubber substrata: A new wrinkle in the study of cell locomotion. Science 208, 177–179. Hayakawa, K., Sato, N., and Obinata, T. (2001). Dynamic reorientation of cultured cells and stress fibers under mechanical stress from periodic stretching. Exp. Cell Res. 268, 104–114. Howard, J. (2001). “Mechanics of Motor Proteins and the Cytoskeleton.”). Sinauer Associates, Sunderland, MA. Hsu, H. J., Lee, C. F., and Kaunas, R. (2009). A dynamic stochastic model of frequency-dependent stress fiber alignment induced by cyclic stretch. PLoS ONE 4, e4853. Huang, S., and Ingber, D. E. (1999). The structural and mechanical complexity of cell-growth control. Nat. Cell Biol. 1, E131–E138. Iba, T., and Sumpio, B. (1991). Morphological response of human endothelial cells subjected to cyclic strain in vitro. Microvasc. Res. 42, 245–254. Ingber, D. E. (1993). Cellular tensegrity: Defining new rules of biological design that govern the cytoske leton. J. Cell Sci. 104, 613–627. Ingber, D. E. (2003). Mechanobiology and diseases of mechanotransduction. Ann. Med. 35, 1–14. Ingber, D. E. (2003). Tensegrity. i. cell structure and hierarchical systems biology. J. Cell Sci. 116, 1157–1173.

172

Rumi De et al. Jaalouk, D. E., and Lammerding, J. (2009). Mechanotransduction gone awry. Nat. Rev. Mol. Cell. Biol. 10, 34–43. Jakab, K., Neagu, A., Mironov, V., Markwald, R. R., and Forgacs, G. (2004). Engineering biological structures of prescribed shape using self-assembling multicellular systems. Proc. Natl. Acad. Sci. USA 101, 2864–2869. Jungbauer, S., Gao, H., Spatz, J. P., and Kemkemer, R. (2008). Two characteristic regimes in frequencydependent dynamic reorientation of fibroblasts on cyclically stretched substrates. Biophys. J. 95, 3470–3478. Kasza, K. E., Rowat, A. C., Liu, J., Angelini, T. E., Brangwynne, C. P., Koenderink, G. H., and Weitz, D. A. (2007). The cell as a material. Curr. Opin. Cell Biol. 19, 101–107. Kaunas, R., Usami, P.N.S., and Chien, S. (2005). Cooperative effects of rho and mechanical stretch on stress fiber organization. Proc. Natl. Acad. Sci. USA 102, 15895–15900. Kawakami, K., Tatsumi, H., and Sokabe, M. (2001). Dynamics of integrin clustering at focal contacts of endothelial cells studied by multimode imaging microscopy. J. Cell Sci. 114, 3125–3135. Kemkemer, R., Neidlinger-Wilke, C., Claes, L., and Gruler, H. (1999). Cell orientation induced by extra cellular signals. Cell Biochem. Biophys. 30, 167–195. Koenderink, G. H., Dogic, Z., Nakamura, F., Bendix, P. M., MacKintosh, F. C., Hartwig, J. H., Stossele, T. P., and Weitz, D. A. (2009). An active biopolymer network controlled by molecular motors. Proc. Natl. Acad. Sci. USA 106, 15192–15197. Kolega, J. (1986). Effects of mechanical tension on protrusive activity and microfilament and intermediate filament organization in an epidermal epithelium moving in culture. J. Cell Biol. 102, 1400–1411. Komura, S., and Andelman, D. (2000). Adhesion-induced lateral phase separation in membranes. Eur. Phys. J. E 3, 259–271. Korff, T., and Augustin, H. G. (1999). Tensional forces in fibrillar extracellular matrices control directional capillary sprouting. J. Cell Sci. 112, 3249–3258. Kozlov, M. M., and Bershadsky, A. D. (2004). Processive capping by formin suggests a force-driven mechanism of actin polymerization. J. Cell Biol. 167, 1011–1017. Kumar, S., Maxwell, I. Z., Heisterkamp, A., Polte, T. R., Lele, T. P., Salanga, M., Mazur, E., and Ingber, D. E. (2006). Viscoelastic retraction of single living stress fibers and its impact on cell shape, cytoskeletal organization, and extracellular matrix mechanics. Biophys. J. 90, 3762–3773 Kurpinski, K., Chu, J., Hashi, C., and Li, S. (2006). Anisotropic mechanosensing by mesenchymal stem cells. Proc. Natl. Acad. Sci. USA 103, 16095–16100. Lammerding, J., Dahl, K. N., Discher, D. E., and Kamm, R. D. (2007). Nuclear mechanics and methods. Methods Cell Biol. 83, 269–294. Lee, J. S. H., Hale, C. M., Panorchan, P., Khatau, S. B., George, J. P., Tseng, Y., Stewart, C. L., Hodzic, D., and Wirtz, D. (2007). Nuclear lamin a/c deficiency induces defects in cell mechanics, polarization, and migration. Biophys. J. 93, 2542–2552. Lenne, P. F., and Nicolas, A. (2009). Physics puzzles on membrane domains posed by cell biology. Soft Matter 5, 2841–2848. Lo, C. M., Wang, H. B., Dembo, M., and Wang, Y. L. (2000). Cell movement is guided by the rigidity of the substrate. Biophys. J. 79, 144–152. Lu, L., Feng, Y., Hucker, W. J., Oswald, S. J., Longmore, G. D., and Yin, F. C. (2008). Actin stress fiber pre-extension in human aortic endothelial cells. Cell Motil. Cytoskeleton 65, 281–294. MacKintosh, F. C., Kas, J., and Janmey, P. A. (1995). Elasticity of semiflexible biopolymer networks. Phys. Rev. Lett. 75, 4425–4428. Marshall, B. T., Long, M., Piper, J. W., Yago, T., McEver, R. P., and Zhu, C. (2003). Direct observation of catch bonds involving cell-adhesion molecules. Nature 423, 190–193. Mitchison, T., and Kirschner, M. (1984). Dynamic instability of microtubule growth. Nature 312, 237–242. Miyamoto, S., Teramoto, H., Coso, O. A., Gutkind, J. S., Burbelo, P. D., Akiyama, S. K., and Yamada, K. M. (1995). Integrin function: Molecular hierarchies of cytoskeletal and signaling molecules. J. Cell Biol. 131, 791–805. Mizuno, D., Tardin, C., Schmidt, C. F., and MacKintosh, F. C. (2007). Nonequilibrium mechanics of active cytoskeletal networks. Science 315, 370–373. Mofrad, M. R. (2009). Rheology of the cytoskeleton. Annu. Rev. Fluid Mech. 41, 433–453.

7. Theoretical Concepts and Models of Cellular Mechanosensing

173

Mogilner, A. and Keren, K. (2009). The shape of motile cells. Curr. Biol. 19, R762–R771. Nekouzadeh, A., Pryse, K., Elson, E., and Genin, G. (2008). Stretch-activated force shedding, force recovery, and cytoskeletal remodeling in contractile fibroblasts. J. Biomech. 41, 2964–2971. Nelson, C. M., Jean, R. P., Tan, J. L., Liu, W. F., Sniadecki, N. J., Spector, A. A., and Chen, C. S. (2005) Emergent patterns of growth controlled by multicellular form and mechanics. Proc. Natl. Acad. Sci. USA 102, 11594–11599. Nicolas, A., and Safran, S. A. (2006). Limitation of cell adhesion by the elasticity of the extracellular matrix. Biophys. J. 91, 61–73. Nicolas, A., Geiger, B., and Safran, S. A. (2004). Cell mechanosensitivity controls the anisotropy of focal adhesions. Proc. Natl. Acad. Sci. USA 101, 12520–12525. Nicolas, A., Besser, A., and Safran, S. A. (2008). Dynamics of cellular focal adhesions on deformable substrates: Consequences for cell force microscopy. Biophys. J. 95, 527–539. Noria, S., Xu, F., McCue, S., Jones, M., Gotlieb, A. I., and Langille, B. L. (2004). Assembly and reorientation of stress-fibers drives morphological changes to endothelial cells exposed to shear stress. Am. J. Pathol. 164, 1211–1223. Orr, A. W., Helmke, B. P., Blackman, B. R., and Schwartz, M. A. (2006). Mechanisms of mechanotransduc tion. Dev. Cell 10, 11–20. Pajerowski, J. D., Dahl, K. N., Zhong, F. L., Sammak, P. J., and Discher, D. E. (2007). Physical plasticity of the nucleus in stem cell differentiation. Proc. Natl. Acad. Sci. USA 104, 15619–15624. Parker, K. K., Brock, A. L., Brangwyne, C., Mannix, R. J., Wang, N., Ostuni, E., Geisse, N. A., Adams, J. C., Whitesides, G. M., and Ingber, D. E. (2002). Directional control of lamellipodia extension by constraining cell shape and orienting cell tractional forces. FASEB J. 16, 1195–1204. Paszek, M. J., Zahir, N., Johnson, K. R., Lakins, J. N., Rozenberg, G. I., Gefen, A., Reinhart-King, C. A., Margulies, S. S., Dembo, M., Boettiger, D., Hammer, D. A., and Weaver, V. M. (2005). Tensional homeostasis and the malignant phenotype. Cancer Cell 8, 241–254. Pathak, A., Deshpande, V. S., McMeeking, R. M., and Evans, A. G. (2008). The simulation of stress fibre and focal adhesion development in cells on patterned substrates. J. R. Soc. Interface 5, 507–524. Paul, R., Heil, P., Spatz, J. P., and Schwarz, U. S. (2008). Propagation of mechanical stress through the actin cytoskeleton toward focal adhesions: Model and experiment. Biophys. J. 94, 1470–1482. Pelham, R. J., and Wang, Y. L. (1997). Cell locomotion and focal adhesions are regulated by substrate flexibility. Proc. Natl. Acad. Sci. USA; 94, 13661–13665. Petroll, W. M., Ma, L., and Jester, J. V. (2003). Direct correlation of collagen matrix deformation with focal adhesion dynamics in living corneal fibroblasts. J. Cell Sci. 116, 1481–1491. Peyton, S. R., Ghajar, C. M., Khatiwala, C. B., and Putnam, A. J. (2007). The emergence of ECM mechanics and cytoskeletal tension as important regulators of cell function. Cell Biochem. Biophys. 47, 300–320. Pirentis, A. P., and Lazopoulos, K. A. (2009). On stress fiber reorientation under plane substrate stretching. Arch. Appl. Mech. 79, 263–277. Pirone, D. M., Nelson, C. M., Bhadriraju, K., and Chen, C. S. (2004). Cell shape, cytoskeleton tension, and RohA regulate stem cell lineage commitment. Dev. Cell 6, 483–495. Pullarkat, P. A., Fernández, P. A., and Ott, A. (2007). Rheological properties of the eukaryotic cell cytoskeleton. Phys. Rep. 449, 29–53. Rehfeldt, F., and Discher, D. E. (2007). News and views: Biophysics: Cell dipoles feel their way. Nat. Phys. 3, 592–593. Ren, X. D., Kiosses, W. B., and Schwartz, M. A. (1999). Regulation of the small GTP-binding protein rho by cell adhesion and the cytoskeleton. EMBO J. 18, 578–585. Riveline, D., Zamir, E., Balaban, N. Q., Schwarz, U. S., Ishizaki, T., Narumiya, S., Kam, Z., Geiger, B., and Bershadsky, A. D. (2001). Focal contacts as mechanosensors: Externally applied local mechanical force induces growth of focal contacts by an mdia1-dependent and rock-independent mechanism. J. Cell Biol. 153, 1175–1185. Rodriguez, J. P., Gonzalez, M., Rios, S., and Cambiazo, V. (2004). Cytoskeletal organization of human mesenchymal stem cells (msc) changes during their osteogenic differentiation. J. Cell. Biochem. 93, 721–731.

174

Rumi De et al. Rowat, A. C., Lammerding, J., Herrmann, H., and Aebi, U. (2008). Towards an integrated understanding of the structure and mechanics of the cell nucleus. BioEssays 30, 226–236. Saez, A., Buguin, A., Silberzan, P., and Ladoux, B. (2005). Is the mechanical activity of epithelial cells controlled by deformations or forces ? Biophys. Lett. 89, L52–L54. Saez, A., Ghibaudo, M., Buguin, A., Silberzan, P., and Ladoux, B. (2007). Rigidity-driven growth and migration of epithelial cells on microstructured anisotropic substrates. Proc. Natl. Acad. Sci. USA 104, 8281–8286. Safran, S. A., and De, R. (2009). Nonlinear dynamics of cell orientation. Phys. Rev. E 80, R60901–R60904. Sarasa-Renedo, A., Tunc-Civelek, V., and Chiquet, M. (2006). Role of RhoA/ROCK dependent actin contractility in the induction of tenascin-C by cyclic tensile strain. Exp. Cell Res. 312, 1361–1370. Schiessel, H. (2003). The physics of chromatin. J. Phys. Condens. Matter 15, R699–R774. Schwarz, U. S., and Bischofs, I. B. (2005). Physical determinants of cell organization in soft media. Med. Eng. Phys. 27, 763–772. Schwarz, U. S., and Safran, S. A. (2002). Elastic interactions of cells. Phys. Rev. Lett. 88, 048102. Schwarz, U. S., Erdmann, T., and Bischofs, I. B. (2006). Focal adhesions as mechanosensors: The two-spring model. Biosystems 83, 225–232. Shemesh, T., and Kozlov, M. (2007). Actin polymerization upon processive capping by formin: A model for slowing and acceleration. Biophys. J. 92, 1512–1521. Shemesh, T., Bershadsky, A. D., and Kozlov, M. M. (2005a). Force-driven polymerization in cells: Actin filaments and focal adhesions. J. Phys. Condens. Matter 17, S3913–S3928. Shemesh, T., Geiger, B., Bershadsky, A. D., and Kozlov, M. (2005b). Focal adhesions as mechanosensors: A physical mechanism. Proc. Natl. Acad. Sci. USA 102, 12383–12388. Shirinsky, V. P., Antonov, A. S., Birukov, K. G., Sobolevsky, A. V., Romanov, Y. A., Kabaeva, N. V., Antonova, G. N., and Smirnov, V. N. (1989). Mechano-chemical control of human endothelium orienta tion and size. J. Cell Biol. 40, 331–339. Shlomovitz R., and Gov, N. S. (2008). Exciting cytoskeleton-membrane waves. Phys. Rev. E 78, 41911–41918. Smith, P. G., Garcia, R., and Kogerman, L. (1997). Strain reorganizes focal adhesions and cytoskeleton in cultured airway smooth muscle cells. Exp. Cell Res. 232, 127–136. Sniadecki, N. J., Anguelouch, A., Yang, M. T., Lamb, C. M., Liu, Z., Kirschner, S. B., Liu, Y., Reich, D. H., and Chen, C. S. (2007). Magnetic microposts as an approach to apply forces to living cells. Proc. Natl. Acad. Sci. USA v, 14553–14558. Sollich, P. (1998). Rheological constitutive equation for a model of soft glassy materials. Phys. Rev. Lett. E 58, 738–759. Solon, J., Levental, I., Sengupta, K., Georges, P. C., and Janmey, P. A. (2007). Fibroblast adaptation and stiffness matching to soft elastic substrates. Biophys. J. 93, 4453–4461. Stamenovic, D., and Ingber, D. E. (2002). Models of cytoskeletal mechanics of adherent cells. Biomech. Model Mechanobiol. 1, 95–108. Stamenovic, D., and Wang, N. (2000). Cellular responses to mechanical stress. invited review: Engineering approaches to cytoskeletal mechanics. J. Appl. Physiol. 89, 2085–2090. Stamenovic, A., Rosenblatt, N., Montoya-Zavala, M., Matthews, B. D., Hu, S., Suki, B., Wang, N., and Ingber, D. E. (2007). Rheological behavior of living cells is timescale-dependent. Biophys. J. 93, L39–L41. Storm, C., Pastore, J. J., MacKintosh, F. C., Lubensky, T. C., and Janmey, P. A. (2005). Nonlinear elasticity in biological gels. Nature 435, 191–194. Takemasa, T., Sugimoto, K., and Yamashita, K. (1997). Amplitude-dependent stress fiber reorientation in early response to cyclic strain. Exp. Cell Res. 40, 407–410. Tan, J. L., Tien, J., Pirone, D. M., Gray, D. S., Bhadriraju, K., and Chen, C. S. (2003). Cells lying on a bed of microneedles: An approach to isolate mechanical force. Proc. Natl. Acad. Sci. USA 100, 1484–1489. Tee, S., Bausch, A., and Janmey, P. (2009). The mechanical cell. Curr. Biol. 19, R745–R748. Théry, M., Pépin, A., Dressaire, E., Chen, Y., and Bornens, M. (2006a). Cell distribution of stress fibres in response to the geometry of the adhesive environment. Cell Motil. Cytoskeleton 63, 341–355. Théry, M., Racine, V., Piel, M., Pépin, A., Dimitrov, A., Chen, Y., Sibarita, J. B., and Bornens, M. (2006b). Anisotropy of cell adhesive microenvironment governs cell internal organization and orientation of polarity. Proc. Natl. Acad. Sci. USA 103, 19771–19776.

7. Theoretical Concepts and Models of Cellular Mechanosensing

175

Tlusty, R.B.Z.T., Moses, E., Safran, S. A., and Bershadsky, A. (1999). Pearling in cells: A clue to under standing cell shape. Proc. Natl. Acad. Sci. USA 96, 10140–10145. Torsoni, A. S., Marin, T. M., Velloso, L. A., and Franchini, K. G. (2005). RhoA/ROCK signaling is critical to FAK activation by cyclic stretch in cardiac myocytes. Am. J. Physiol. Heart Circ. Physiol. 289, H1488–H1496. Trepat, X., Deng, L., An, S. S., Navajas, D., Tschumperlin, D. J., Tnrthoffer, W., Butler, J. P., and Fredberg, J. J. (2007). Universal physical responses to stretch in the living cell. Nature 447, 592–595. Trepat, X., Lenormand, G., and Fredberg, J. J. (2008). Universality in cell mechanics. Soft Matter 4, 1750–1759. Vandenburgh, H. H. (1988). A computerized mechanical cell stimulator for tissue-culture—effects on skeletal-muscle organogenesis. In Vitro Cell. Dev. Biol. Cell Tissue Res. 24, 609–619. Vogel, V., and Sheetz, M. (2006). Local force and geometry sensing regulate cell functions. Nat. Rev. Mol. Cell Biol. 7, 265–275. Wang, J., and Grood, E. (2000). The strain magnitude and contact guidance determine orientation response of fibroblasts to cyclic substrate strains. Connect. Tissue Res. 41, 29–36. Wang, J.H.C., Goldschmidt-Clermont, P., Wille, J., and Yin, F. (2001). Specificity of endothelial cell reorientation in response to cyclic mechanical stretching. J. Biomech. 34, 1563–1572. Wang, N., Ostuni, E., Whitesides, G. M., and Ingber, D. E. (2002a). Micropatterning tractional forces in living cells. Cell Motil. Cytoskeleton 52, 97–106. Wang, N., Tolic-Norrelykke, I. M., Chen, J. X., Mijailovich, S. M., Butler, J. P., Fredberg, J. J., and Stamenovic, D. (2002b). Cell prestress. i. Stiffness and prestress are closely associated in adherent contractile cells. Am. J. Physiol. Cell Physiol. 282, C606–C616. Wang, N., Tytell, J. D., and Ingber, D. E. (2009). Mechanotransduction at a distance: Mechanically coupling the extracellular matrix with the nucleus. Nat. Rev. Mol. Cell Biol. 10, 75–82. Wei, Z., Deshpande, V. S., McMeeking, R. M., and Evans, A. G. (2008). Analysis and interpretation of stress-fiber organization in cells subject to cyclic stretch. J. Biomed. Eng. 130, 031009. Wirtz, D. (2009). Particle-tracking microrheology of living cells: Principles and applications. Annu. Rev. Biophys. 38, 301–326. Wolfenson, H., Lubelski, A., Regev, T., Henis, J.K.Y.I., and Geiger, B. (2009). A role for the juxtamembrane cytoplasm in the molecular dynamics of focal adhesions. PLoS ONE 4, e4304. Wozniak, M. A., and Chen, C. S. (2009). Mechanotransduction in development: A growing role for contractility. Nat. Rev. Mol. Cell. Biol. 10, 63–73. Wozniak, M. A., Desai, R., Solski, P. A., Der, C. J., and Keely, P. J. (2003). Rock-generated contractility regulates breast epithelial cell differentiation in response to the physical properties of a three-dimensional collagen matrix. J. Cell Biol. 163, 583–595. Yeung, T., Georges, P. C., Flanagan, L. A., Marg, B., Ortiz, M., Funaki, M., Zahir, N., Ming, W., Weaver, V., and Janmey, P. A. (2005). Effects of substrate stiffness on cell morphology, cytoskeletal structure, and adhesion. Cell Motil. Cytoskeleton 60, 24–34. Yoshigi, M., Hoffman, L. M., Jensen, C. C., Yost, H. J., and Beckerle, M. C. (2005). Mechanical force mobilizes zyxin from focal adhesions to actin filaments and regulates cytoskeletal reinforcement. J. Cell Biol. 171, 209–215. Zaidel-Bar, R., Ballestrem, C., Kam, Z., and Geiger, B. (2003). Early molecular events in the assembly of matrix adhesions at the leading edge of migrating cells. J. Cell Sci. 116, 4605–4613. Zamir, E., and Geiger, B. (2001). Components of cell-matrix adhesion. J. Cell Sci. 114, 3577–3579. Zedler, U., Hampe, N., Rubner, W., Kirchgeßner, N., Merkel, R., and Hoffmann, B. (to be published). Force thresholds and tensile strains induce cell reorientation upon cyclic stretch. Zemel, A., Bischofs, I. B., and Safran, S. A. (2006). Active elasticity of gels with contractile cells. Phys. Rev. Lett. 97, 128103. Zemel, A., Rehfeldt, F., Brown, A.E.X., Discher, D. E., and Safran, S. A. (2010a). Optimal matrix rigidity for stress-fiber polarization in stem cells. Nat. Phys. DOI: 10.1038/NPHYS1613. Zemel, A., Rehfeldt, F., Brown, A. E. X., Discher D. E., and Safran, S. A. (2010b). Cell shape, spreading symmetry and the polarization of stress-fibers in cells. J. Phys. Condens. Matter 22, 194–210. Zhao, X. H., Laschinger, C., Arora, P., Szaszi, K., Kapus, A., and McCulloch, C. (2007). Force activates smooth muscle alpha-actin promoter activity through the rho signaling pathway. J. Cell Sci. 120, 1801–1809.

SECTION B

Impact of Nuclear Mechanics on Function

CHAPTER 8

Mechanical Induction of Gene Expression in Connective Tissue Cells Matthew W. C. Chan, Boris Hinz, and Christopher A. McCulloch Matrix Dynamics Group, Faculty of Dentistry, University of Toronto, Fitzgerald Building, Toronto, ON, Canada M5S 3E2

Abstract I. Introduction II. Extracellular Matrix Environment III. Mechanical Signaling A. Systems B. Cell Contacts IV. Overview of Methods for Mechanical Cell Stimulation A. Shear Forces B. Compression C. Stretch D. Static Mechanical Stimuli—Substrate Stiffness Matters!! E. Subcellular Mechanical Stimulation V. Cardiac Fibrosis and Mechanical Induction of Gene Expression A. Cardiac Interstitium B. Mechanical Induction of Myofibroblasts C. Regulation of Gene Expression in Mechanically Loaded Cardiac Cells D. Regulation of the a-Smooth Muscle Actin Promoter E. Cell Culture Models for Mechanical Induction of a-smooth Muscle Actin Expression F. Coating Methods for Beads G. Cell Transfection and Promoter Methods H. Identification of Adhesion-Associated Proteins VI. Conclusions Acknowledgments References

METHODS IN CELL BIOLOGY, VOL. 98 Copyright 2010 Elsevier Inc. All rights reserved.

179

978-0-12-381009-0

DOI: 10.1016/S0091-679X(10)98008-4

180

Matthew W. C. Chan et al.

Abstract The extracellular matrices of mammals undergo coordinated synthesis and degrada tion, dynamic remodeling processes that enable tissue adaptations to a broad range of environmental factors, including applied mechanical forces. The soft and mineralized connective tissues of mammals also exhibit a wide repertoire of mechanical properties, which enable their tissue-specific functions and modulate cellular responses to forces. The expression of genes in response to applied forces are important for maintaining the support, attachment, and function of various organs including kidney, heart, liver, lung, joint, and periodontium. Several high-prevalence diseases of extracellular matrices including arthritis, heart failure, and periodontal diseases involve pathological levels of mechanical forces that impact the gene expression repertoires and function of bone, cartilage, and soft connective tissues. Recent work on the application of mechanical forces to cultured connective tissue cells and various in vivo force models have enabled study of the regulatory networks that control mechanically induced gene expression in connective tissue cells. In addition to the influence of mechanical forces on the expression of type 1 collagen, which is the most abundant protein of mammals, new work has shown that the expression of a wide range of matrix, signaling, and cytoskeletal proteins are regulated by exogenous mechanical forces and by the forces generated by cells themselves. In this chapter, we first discuss the fundamental nature of the extracellular matrix in health and the impact of mechanical forces. Next we consider the utilization of several, widely employed model systems for mechanical stimulation of cells. Finally, we consider in detail how application of tensile forces to cultured cardiac fibroblasts can be used for the characterization of the signaling systems by which mechanical forces regulate myofibroblast differentiation that is seen in cardiac pressure overload.

I. Introduction In mammals, connective tissues surround and are distributed throughout organs including liver, heart, lung, and kidney. Connective tissue cells synthesize and main tain extracellular matrices and provide mechanical support and attachment for contig uous tissues and organs. The cells embedded in soft or mineralized connective tissues live in specialized environments in which they experience tissue-specific chemical signals. They may also be subjected to various mechanical forces. In solid tissues, because of their surrounding extracellular matrices, cells likely can sense and respond to mechanical forces in ways that are quite different than the forces that impact, for example, blood cells flowing past endothelial cells (Tzima et al., 2005). In general, cells of connective tissues are mechanically adapted to the rheological properties of the extracellular matrix and their responses to mechanical stimuli are strongly affected by the matrix proteins that surround them. In this chapter, which focuses on mechanical signaling to mediate gene expression, connective tissues

8. Mechanical Induction of Gene Expression in Connective Tissue

181

provide critical physical and biological elements for transmission of gravitational and muscular forces. For example, connective tissues anatomically and functionally join muscles to bone, thereby enabling force transfer between these two tissues. Another important aspect of the extracellular matrix is its contribution to stabilization of tissues that are subjected to physical forces. This general property enables organs and tissues to preserve their shape and helps to prevent cellular damage induced by mechanical forces.

II. Extracellular Matrix Environment The extracellular matrices of connective tissues are comprised of collagen fibers and a large gruop of other fibrillar and globular proteins, which may include fibro nectins, laminins, glycosaminoglycans, tenascins, and several other glycoproteins. In mineralized connective tissues, highly ordered arrays of hydroxyapatite crystals that are distributed throughout a soft connective tissue matrix, can display a wide range of stiffness values. For example, cortical bone is much stiffer than cartilage. For all connective tissues, collagen is the principal molecular building block and indeed fibrillar collagen is the most abundant protein in mammals (Perez-Tamayo, 1978). Further, in soft connective tissues, collagen fibrils can transmit tensile forces (Pro venzano and Vanderby, 2006) to fibroblasts and many other cell types in which there is appropriate expression of collagen receptors. Depending on the structure and makeup of the proteins in connective tissues, extracellular matrices are well adapted to transmit forces and to protect cells against a wide variety of mechanical loads. These loads could include tensile, compressive, and/or shearing forces (Warden et al., 2005). When connective tissues are subjected to increased loading, there is enhanced remodeling of connective tissue matrices (Ozaki et al., 2005) and increased prolifera tion of fibroblasts and osteoblasts. These cells are the principal mesenchymal cell type of soft and mineralized connective tissues, respectively. Their sensitivity to mechanical forces facilitates force-induced remodeling of extracellular matrices. It is evident from many in vivo studies that mechanical forces do indeed regulate gene expression in connective tissue cells. Further, the mechanical properties of the extra cellular matrix itself have an important influence on the morphology and function of osteoblasts and fibroblasts (Hinz and Gabbiani, 2003). For example, when collagen lattices are subjected to mechanical forces, a “synthetic” fibroblast phenotype emerges, which is characterized by increased expression of connective tissue matrix proteins and inhibition of matrix protein degradation (Kessler et al., 2001). In experiments that employ three-dimensional collagen gels, mechanical loading influ ences matrix remodeling (Mudera et al., 2000). Accordingly, while the extracellular matrix can transmit mechanical forces, its structure and protein composition is affected by the cellular responses to the forces that are applied or that are generated by the cells themselves.

182

Matthew W. C. Chan et al.

III. Mechanical Signaling A. Systems Much is known of how soluble chemical ligands, after binding to cognate receptors, activate cellular signaling pathways. In contrast, there is less definitive knowledge on transduction of mechanical signals although it is well recognized that physical forces impact the metabolic responses of many tissues including the stimulation of bone formation (Martin, 2007; Skerry and Suva, 2003; Turner and Robling, 2004), the remodeling of the periodontium during orthodontic treatment and dental occlusal trauma (Krishnan and Davidovitch, 2009; Rygh, 1973), the induction of cardiac hypertrophy by volume or pressure overload (Catalucci et al., 2008; Tarone and Lembo, 2003), the generation of ventilator-induced injury in lung (Lionetti et al., 2005; Stenqvist et al., 2008), and the sensations of pain (Lewin and Moshourab, 2004; Tsunozaki and Bautista, 2009) and hearing. Currently, the molecular identity of specific mechanotransducers has not been defined but analysis of genetic models of mechanotransduction in Caenorhabditis elegans (Syntichaki and Tavernarakis, 2004) and a large number of mammalian cell models (Tsunozaki and Bautista, 2009) have suggested several possible mechanisms. One possible mechanism of mechanotrans duction invokes force–activation of mechanosensitive plasma membrane channels (Kiselyov and Patterson, 2009; Martinac, 2004), thereby allowing inflow of Ca2þ that can act as a second messenger to regulate gene expression. Recent evidence indicates that one family of channels, the transient receptor potential channels, play central roles as specific mechanosensitive channels in hearing and mechanosensation (Corey, 2003; Yin and Kuebler, 2010), but how mechanically gated channels are regulated by forces acting on the cell surface and how these signals are translated into biological outcomes is not defined. The potential therapeutic importance of mechanically gated ion channels is underlined by their identification as possible drug targets for a variety of diseases involving dysregulated mechanotransduction (Cortright and Szallasi, 2009;Gottlieb et al., 2004).

B. Cell Contacts In connective tissues, direct transfer of forces to cells may involve cell-to-cell and/or cell-to-matrix contacts (Chen et al., 2004). One of the cell surface receptors that bind to matrix molecules, the integrins, are of particular importance because they functionally integrate cell adhesion and cell signaling processes, and because they may be able to transfer forces from the extracellular matrix to the cytoskeleton (Chiquet et al., 2009; Katsumi et al., 2004). While specific macromolecular platforms may provide cells with the ability to respond specifically to mechanical stimuli (Helmke and Schwartz, 2004), the proteins which comprise the force sensor and effector systems are not defined. This is important since tissue and organ dysfunction is mediated by high-amplitude/high frequency applied forces to cell surface receptors, including the integrins (Ingber, 2003a; Thodeti et al., 2009). Among the cellular elements that are thought to

8. Mechanical Induction of Gene Expression in Connective Tissue

183

contribute to mechanotransduction, the cytoskeleton is of particular interest because it can transmit cellular forces, contributes to the information processing of mechanically induced signals and may protect cells against damage induced by excessive force levels (Janmey and Weitz, 2004). Cells in connective tissues adhere to extracellular matrices by a wide variety of matrix receptors. Under certain circumstances these receptors can become clustered into aggregates, which, in cultured cells, are termed focal adhesions or focal com plexes. These protein complexes are potential sites for transfer of contractile forces to the cytoskeleton in cultured cells and possibly for cells in tissues (Ingber, 2003b). Because of the ease of culturing fibroblasts and for then applying a wide variety of exogenous forces to these cells, cultured fibroblasts are now widely used models for exploring mechanosensing and force response mechanisms in solid tissues. In fibro blasts, force transmission is critically dependent on the attachment of cells to matrix molecules such as fibrillar collagen or fibronectin applied to either the culture substrate (Hinz, 2006) or beads. With this methodology, mechanical induction of gene expres sion is experimentally testable: tensile forces applied through matrix proteins like collagen (but not poly-L-lysine-coated beads) promote increased expression of the actin-binding protein filamin A (D’Addario et al., 2002). Trans-membrane proteins can activate intracellular biochemical signaling pathways either by binding an extracellular ligand (chemical signaling) or when they are unfolded or otherwise deformed by force (mechanical signaling). Thus, the adhesive functions of attachment molecules such as integrins and cadherins are key elements in mechanosensing and mechanotransduction. Integrins are enriched in the extensively studied focal adhesions described above. Focal adhesions are multimolecular com plexes consisting of more than 50 different proteins that link extracellular matrixattached integrins to the actin cytoskeleton (Geiger and Bershadsky, 2002). The assembly and maintenance of focal adhesions depend in part on local mechanical forces. These forces may be generated by myosin II-driven contraction of the actin cytoskeleton or by stretching forces originating from the extracellular matrix. Force-induced assembly of focal adhesions leads to activation of a variety of signaling pathways that control cell proliferation, differentiation, the organization of the cytoskeleton, and the expression of specific genes. While attachment of connective tissue cells to the extracellular matrix is generally reliant on the formation and remodeling of integrin-mediated adhesions, connective cells can also adhere to each other by intercellular adhesive molecules (e.g., cadherins) that may also act as force sensors (Ko et al., 2001) and possibly be able to regulate gene expression. As N-cadherin-mediated adherens junctions are influenced by integ rin biology, fibroblasts may be able to integrate mechanical signals from both adhesion systems to coordinate gene responses relevant to differentiation, organogenesis, and wound healing (Linask et al., 2005). Notably, several reports have described mechan ical signaling through both cadherin and integrins (Ko et al., 2001; Potard et al., 1997). Consequently, mechanotransduction, may not be a single, restricted process but may instead be a chain of interrelated processes that require the recruitment of a large variety of attachment, cytoskeletal, and signaling proteins. These proteins may then be

184

Matthew W. C. Chan et al.

able to form docking and signaling complexes that are appropriately oriented in time and space to optimize transmission and processing of mechanical signals.

IV. Overview of Methods for Mechanical Cell Stimulation Here, we provide an overview of different culture methods that are currently used to study mechanotransduction at the cellular and subcellular levels. More detailed tech nical descriptions are covered in excellent recent reviews (Brown, 2000; Jonas et al., 2008; Kim et al., 2009; Lele et al., 2007). As introduced above, mechanical cues determine cell fate, phenotype, and behavior (Bao and Suresh, 2003; Chen, 2008; Discher et al., 2005; Hoffman and Crocker, 2009; Ingber et al., 2006; Janmey and McCulloch, 2007; Janmey and Weitz, 2004; Orr et al., 2006; Vogel and Sheetz, 2006; Wang and Thampatty, 2008). When studying cell “mechanoperception” it is important to consider that different cells types are exposed to different qualities and quantities of mechanical load in vivo. As mentioned above, circulating blood cells and endothelial cells of the vascular wall experience fluid flow shear stress (Makino et al., 2007), which is very different than the compressive and tensile forces that are sensed by cells in solid connective tissues. Thus while cells residing in bone and cartilage are under compressive load (Adams, 2006; Turner, 2006), a large number of different cell types are subjected to stretching forces. For example, cyclic stretch and compression are characteristic mechanical stimuli for cardiomyocytes (Lammerding et al., 2004) and endothelial and smooth muscle cells in the vessel wall (Halka et al., 2008; Reinhart-King et al., 2008; White and Frangos, 2007), of the intestine (Jones and Bratten, 2008), and of the airways (Choe et al., 2006; Hasaneen et al., 2005; Pugin, 2003). Skeletal muscle cells, connective tissue fibroblasts, and epidermal keratinocytes are subject to rather more gradual stretching of various degrees and rates (Chiquet et al., 2003; Hinz, 2010; Reichelt, 2007). We will describe below several different systems to manipulate cells mechanically, consider their potential pitfalls and provide simple guidelines on how to select the appropriate instruments from this mechanical toolbox. A. Shear Forces Fluid flow is commonly applied to expose cultured cells to shear forces. The physiological relevance of fluid flow shear force is most obvious for hematopoietic cells transported in the bloodstream and for cells lining the inner surfaces of fluid-filled cavities, including endothelial cells of blood and lymphatic vessels but also epithelial cells in the respiratory and gastrointestinal tract. Two major construction principles exist for two-dimensional culture systems: (1) the cone-and-plate system where rota tion of a cone-shaped body over a flat culture plate sets the cell culture medium in turbulent motion (Chung et al., 2005; Dewey, 1984) and (2) parallel plate flow chambers in which cells are grown in flow channels of defined dimensions (Bacabac et al., 2005; Usami et al., 1993). Variations of the second device are more widely used, allowing precise control over flow and corresponding shear rates.

8. Mechanical Induction of Gene Expression in Connective Tissue

185

Flow-induced shear stress also acts on cells populating three-dimensional tissue environments, such as chondrocytes, osteoblasts, and osteocytes. High levels of shear are generated in interstitial fluid with these models by the compression of the porous tissue structure (Chen et al., 2010; Fritton and Weinbaum, 2009). This can be recapitu lated in culture by perfusing cell-containing three-dimensional porous scaffolds, which are produced from various materials infused with medium (Brown, 2000; Datta et al., 2006; Griffith and Swartz, 2006; Stiehler et al., 2009). Notably, even the very small interstitial flow rates that occur during tissue swelling, microvascular permeability, and lymphatic drainage, have profound influences on the behavior of fibroblasts, tumor cells, and inflammatory cells (Rutkowski and Swartz, 2007). It is unlikely that such low flow rates produce shear forces that are sensed by the cells. Instead, cellular responses are generally explained by directed solute transport (Fleury et al., 2006). Indeed, transport processes always have to be considered as “contaminating” parameters in flow experi ments. Further, it is important to consider that shear forces are calculated on the basis of the applied flow rates and the dimensions of the fluid-filled channels, which define the pressure differential between the inlet and the outlet. This is an accurate approximation for two-dimensional flow chambers and for cells grown in flat monolayers but requires extensive mathematical modeling for more complex three-dimensional materials (Anderson and Knothe Tate, 2008; Porter et al., 2005). B. Compression Mechanically loading of cells in three-dimensional environments not only produces fluid flow and shear stress but also produces a compressive component. Compression devices are principally used to study the mechanobiology of chondrocytes, osteocytes, and osteoblasts. Physiological challenge of cartilage and bone is estimated to generate 40 times higher hydrostatic pressure in the porous structure than values measured in the vasculature (Chen et al., 2010; Zhang et al., 1998). Positive and negative hydrostatic pressures can be generated using gas pressure incubators (Brown, 2000; Yousefian et al., 1995). Hence, no direct contact is needed between the pressure-imposing device and the cells. In addition, the cells do not need to adhere to a substrate. On the other hand, high pO2 and pCO2 conditions may alter the culture medium chemistry, which requires appropriate countermeasures (Ozawa et al., 1990). Solid specimens, such as cartilage, bone, or biomimetic scaffolds can be subjected to pressure by a direct platen abutment. Loading of the sample can be unconfined so that the lateral edges are free to move under compression (Burton-Wurster et al., 1993; Torzilli et al., 1997), or the lateral edges can be confined (Freeman et al., 1994). With recent advances in microtechnological devices and their application to biological problems, miniaturized versions of compression devices are now available for multiple sample analysis (Moraes et al., 2010b). C. Stretch Stretchable substrates are probably the most frequently used tools to study the mechanisms of cell mechanosensing, the consequences of mechanical protein

186

Matthew W. C. Chan et al.

deformation, and gene expression (Jonas et al., 2008; Little et al., 2008; Sawada and Sheetz, 2002; Wipff et al., 2007; Zhong et al., 1998). Most two-dimensional culture stretch systems are based on transparent silicone-based culture membranes that require surface activation for cell adhesion (Lateef et al., 2002; Wipff et al., 2009). Standard models have used polydimethyl siloxane to allow linear extensions of up to 25%, which is in the range of physiologically relevant strain (e.g., smooth muscle cells in the arterial wall experience strain rates of 3–10% in normal conditions). Novel biocompatible silicones have been developed to reach much larger culture surface expansions (Majd et al., 2009), which allow generation of even higher strain of > 30% (e.g., those that might occur in hypertensive vessels) (Califano and Reinhart-King, 2010). Microma chined stretch systems have recently been introduced to strain cells with a high through put (Moraes et al., 2010a). Alternatively, for growth on two-dimensional stretched membranes, cells can be embedded in three-dimensional substrates such as collagen, fibrin, hydrogels, and these mixtures are then clamped to strain devices (Brown et al., 1998; Lee et al., 2008; Raeber et al., 2008). For cells growing in three-dimensional tissue environments, these systems more closely approximate in vivo conditions but are prone to cell-mediated alterations. Cell remodeling processes can lead to matrix anisotropy and local strain distributions that are difficult to predict (Balestrini and Billiar, 2009). The process of selecting the appropriate stretch protocol should be guided by the physiologically relevant conditions for a particular cell type and biological condition. Depending on the design of the stretch device, cells can be stretched uniaxially (substrate and cells strained in one direction) or biaxially (strain in multiple directions) (Banes et al., 1985; Brown, 2000; Jungbauer et al., 2008; Lee et al., 1996). In equibiaxial stretch systems, cells experience the same strain in all directions in contrast to nonequibiaxial apparatus, where different strain magnitudes act in different directions. In addition to the direction and magnitude of strain, automated stretch devices control whether cells are gradually strained or subjected to cyclic stretches of various frequencies. Some limitations of silicone-based strain devices have to be taken into considera tion: (1) the magnitude of the imposed stretch is in most cases higher than the actual stretch experienced at the cellular level. In two-dimensional systems this can be due to friction between the silicone membrane and the stretch device, loss of cell adhesion to the membrane coating, or slipping of the coating membrane with respect to the membrane (Wipff et al., 2007). In three-dimensional cultures, the protein architecture of extracellular matrix will function as a stress buffer according to the level of organization. (2) In uniaxial systems cells are compressed perpendicular to the axis of stretch if the free edges of the clamped membrane are not confined. This effect is prevented in strip-biaxial systems where the substrate is held fixed in the nonstretched axis (Lee et al., 2008). (3) Cells will react to cyclic uniaxial strains by typically aligning along or perpendicular to the axis of stretch. Once aligned, cells will perceive different mechanical inputs compared with their random initial orientation (Mata et al., 2002; Syedain et al., 2008). Cell alignment does usually not occur upon equibiaxial stretch due to the lack of a major axis of strain. It may appear trivial but morphological study and verification of the proper force application to the cells under stretch is an essential component of these types of experiments.

8. Mechanical Induction of Gene Expression in Connective Tissue

187

D. Static Mechanical Stimuli—Substrate Stiffness Matters!! In addition to dynamic mechanical stimuli, static mechanical conditions such as the rigidity of the microenvironment profoundly influence cell behavior (Engler et al., 2007; Janmey et al., 2009; Tenney and Discher, 2009). There is an expanding list of different cell types that respond to substrate stiffness including cancer cells (Paszek et al., 2005), mesenchymal stem cells (Engler et al., 2006; Winer et al., 2009), neurons (Georges et al., 2006), epithelial cells (Pelham and Wang, 1997), myotubes (Engler et al., 2004), cardiomyocytes (Engler et al., 2008), and fibroblasts (Goffin et al., 2006; Klein et al., 2009). To replicate known levels of tissue stiffness in culture, different two-dimensional polymer coatings are frequently applied. Biopolymer substrates pro duced from purified collagen, fibrin, and complex protein mixtures like Matrigel usually bracket the lower end of the physiological stiffness spectrum of tissues (Grinnell, 2009; Velegol and Lanni, 2001). However, mechanical signals are difficult to uncouple from chemical influences in biopolymer substrates. Moreover, biopoly mers are remodeled by cells, which better matches the in vivo situation (Storm et al., 2005) but compromises the reproducibility of culture experiments due to the resulting anisotropic stiffness and stress distributions. In this context, the development of synthetic polymer substrates with tunable elastic modulus and almost ideal elastic properties represent a major advance. Two-dimensional compliant substrates are mainly produced from polyacrylamide-, polyvinyl alcohol-, or silicone-based elastomers. They provide tissue-like stiffness and excellent optical properties (Brown et al., 2005; Harris et al., 1980; Kandow et al., 2007; Pelham and Wang, 1997; Zajaczkowski et al., 2003). Nevertheless, twodimensional compliant culture substrates have limitations that should be considered: (1) they cannot reproduce three-dimensional tissue environments, which is a major challenge for biomaterials and tissue engineering (Lutolf and Hubbell, 2005; Place et al., 2009). (2) The mechanical qualities of elastic synthetic polymers and of viscoelastic biopolymers are very different at the cell “perception” level (Storm et al., 2005). (3) Cells may circumvent the direct mechanical input from the compliant substrate surface by secreting their own extracellular matrix proteins, which may become stiffer than the underlying polymer. (4) The mechanical influence of inter cellular adherens and tight junctions with neighboring cells in confluent cultures may override the significance of the substrate (Yeung et al., 2005). E. Subcellular Mechanical Stimulation There are alternatives to stimulating whole cells and cell populations. A variety of methods are now available for application of local forces at different levels of force and spatial resolution. Several currently used methods to apply stress to cells locally involve microparticles that are coated with cell adhesive proteins and couple to cytoskeleton-linked receptors in the dorsal plasma membrane (see the section below for a detailed description of one system). These particles have a typical diameter of 1–10 µm and can be actuated in two principle ways, using magnetic forces or optical forces. In magnetic twisting cytometry, ferromagnetic beads are manipulated

188

Matthew W. C. Chan et al.

by transiently applying a strong magnetic field that orients the magnetic dipoles of the beads horizontally. Subsequent application of a weak but sustained magnetic field in an orthogonal direction induces bead rotation and thereby “twists” the membrane (Gosse and Croquette, 2002; Lele et al., 2007; Wang et al., 1993). Magnetically induced forces in magnetic twisting cytometry can range from pN to several nN and are applicable to whole cell populations; however, the inconvenience of magnetic twisting cytometry is its relatively poor spatial resolution. This limita tion has been overcome with the development of magnetic pulling cytometry (or “magnetic tweezers”) using magnetic microneedles to apply forces to superpara magnetic particles (Lele et al., 2007). Even higher spatial resolution is gained with optical tweezers; however, the forces produced by these optical traps are limited to the hundreds of pN range (Bar-Ziv et al., 1998; Dai and Sheetz, 1995; Grier, 2003; Nieminen et al., 2007; Svoboda and Block, 1994). One potential danger of all microparticle-based technologies is bead internalization by phagocytosis; since small beads can become engulfed within minutes the time for experiments is limited. This is prevented by using atomic force microscopy cantilever as micro- and nanostimu lators (Charras et al., 2001, 2002; Shroff et al., 1995). An elegant variation of this theme is the combination of the magnetic bead technology and microstructured elastic substrates. By growing cells on elastic micropillars with integrated cobalt nanowires, applied magnetic field lead to pillar deflections and locally stimulate the ventral cell surface (Sniadecki et al., 2007).

V. Cardiac Fibrosis and Mechanical Induction of Gene Expression The application of supraphysiological forces in vivo can lead to tissue damage, which is frequently manifest as fibrosis, a process which involves the formation of poorly organized and dysfunctional connective tissues. Impaired collagen turnover in fibrotic lesions is thought to contribute to collagen accumulation, thereby leading to loss of appropriate connective tissue function. Fibrosis disturbs the protective features of extracellular matrices by disrupting the stress-shielding, cross-linked architecture of these tissues (Hinz et al., 2007). In heart failure, the activation of cardiac fibroblasts, which involves their differentiation into myofibroblasts and the excessive accumula tion of extracellular matrix proteins, is an example of a medically important fibrotic response and which is strongly associated with abnormal cardiac diastolic function (MacKenna et al., 2000). This process is mediated in part by mechanical signals and leads to the de novo expression of a-smooth muscle actin by cardiac fibroblasts. a-Smooth muscle actin is a hallmark gene expressed by myofibroblasts (Tomasek et al., 2002). Below we will consider in detail how inappropriate remodeling of the extracellular matrix of the cardiac interstitium by myofibroblasts can be modeled by mechanical force induction of a-smooth muscle actin. We use this system as a model to study mechanical induction of gene expression in cultured fibroblasts.

8. Mechanical Induction of Gene Expression in Connective Tissue

189

A. Cardiac Interstitium The cardiac interstitium is composed of nonmyocyte cells and a structural protein network that plays a dominant role in governing the architecture and mechanical behavior of the myocardium (Brilla et al., 1992, 1995; Weber et al., 1994). The cardiac extracellular matrix is composed predominantly of collagen fibers and a variety of other extracellular matrix proteins including fibronectin, laminin and tenascin. Cardiac muscle contains about sixfold more collagen than skeletal muscle. Accordingly, differences in the resting tension relationships in cardiac and skeletal muscle may result largely from differences in the connective tissue matrix (Covell, 1990). The fibrillar elements form a stress-tolerant network that facilitates the distribution of forces generated in the heart and provide for appropriate alignment of cardiac myocytes (Carver et al., 1991). The long-term performance of cardiac muscle is regulated through a complex but poorly understood group of feedback mechanisms in which mechanical loading controls the organization of myofibrils, the size of muscle fibers, the expression of muscle-specific genes, and the synthesis and secretion of a wide variety of extracellular matrix products and trophic factors (Yamazaki et al., 1998). This phenomenon is tightly regulated during growth or adaptive responses. Increases of muscle mass occur because of hypertrophic enlargement of terminally differentiated cardiomyocytes, increased numbers of fibroblasts, and increased volume of the extra cellular matrix (Olson and Srivastava, 1996). When after-load is increased, the adult heart adapts by hypertrophy. This compensatory response in adult hearts is associated with up to a sixfold increase of type I and III collagens (Butt et al., 1995; Sun and Weber, 1996) and an increase of the ratio of type III/I collagens (Carver et al., 1991). Contemporaneous with the increased collagen synthesis is a reduction of collagen degradation, possibly mediated by reduced collagenolytic activity (Gonzalez et al., 2009). The net increase of collagen in the interstitium is an important determinant of pathological hypertrophy since it may account for abnormal myocardial stiffness (Wilke et al., 1996). Over the long term, this adaptive response can contribute to impairment of cardiac function and heart failure (Keating and Sanguinetti, 1996). Thus the regulatory mechanisms that are related to the fibrous tissue response in various cardiovascular diseases (e.g., hypertensive heart disease, dilated cardiomyo pathy, postmyocardial infarction) are of primary clinical interest (Brilla et al., 1995).

B. Mechanical Induction of Myofibroblasts While cardiac myocytes comprise the largest volume fraction of the adult heart, they represent <25% of cell number (Grove et al., 1969). By far the most abundant nonmyocytic cell in the myocardium is the fibroblast (Eghbali, 1992) (30–50% of cell number), which, along with endothelial and smooth muscle cells, pericytes, neurons, and blood-borne cells make up the remaining 75%. The fibroblast is the principal cell involved in the synthesis and remodeling of the extracellular matrix and therefore plays a central role in the hypertrophic response (Khan and Sheppard, 2006). The conversion of the fibroblast to the myofibroblast is a critical step in

190

Matthew W. C. Chan et al.

cardiac pathology since in contrast to fibroblasts, myofibroblasts elaborate a poorly organized collagen matrix that impairs diastolic function (Kuwahara et al., 2002). Currently, the mechanisms that regulate conversion of fibroblasts to myofibroblasts and increased collagen deposition in hypertrophic and fibrotic hearts are not defined. But almost certainly the application of increased forces to cardiac cells is required (Chien, 1999; Komuro et al., 1996; Sadoshima and Izumo, 1997)). At least three effector responses involved in mechanotransduction have been identified in cardiac fibroblasts: 1. Increased mechanical load by itself stimulates fibroblast proliferation, reduces collagenolytic activity and increases collagen production (Husse et al., 2007). 2. The release of autocrine and paracrine growth factors is stimulated by mechanical loading (Kaye et al., 1996; van Wamel et al., 2000). 3. Mechanical load may upregulate vessel wall permeability and thereby increase the availability of systemic factors that can activate fibroblasts (Nicoletti and Michel, 1999). All of these responses likely involve increased mechanical loading of cardiac fibroblasts, but how are these mechanical forces applied to the cells? In brief, and as described above, force transmission can be mediated by cell-to-cell interactions, cell–matrix interactions, and perhaps by cytoskeletal transmission of force directly to the nucleus (Ingber, 1997, 2003a). As physical forces in vivo can be transmitted from extracellular matrix molecules to integrins and to cytoskeletal proteins (Juliano and Haskill, 1993), the integration of the extracellular matrix, integrins, cytoskeleton, and ion channels in a connected and dynamic network provides an attractive scheme for sensing force in cardiac fibroblasts (Kiseleva et al., 1996). C. Regulation of Gene Expression in Mechanically Loaded Cardiac Cells a-Smooth muscle actin is not expressed in fibroblasts of normal hearts but is a marker for cardiac myofibroblasts in hypertrophic and fibrotic hearts (Campbell and Katwa, 1997; Leslie et al., 1991) and in heart failure in humans (Suurmeijer et al., 2003). The abundance of myofibroblasts is also locally increased at sites of myocardial infarction (Campbell et al., 1995; Katwa et al., 1997). Further, the expression of a-smooth muscle actin is reactivated during cardiac hypertrophy (Black et al., 1991). In cultured fibro blasts, expression of a-smooth muscle actin is required for integrin-mediated collagen gel remodeling (Leslie et al., 1991; Tomasek et al., 2002). D. Regulation of the a-Smooth Muscle Actin Promoter Serum response factor is a MADS-box transcription factor that regulates genes involved in cell proliferation, migration, cytoskeletal dynamics, and myogenesis by binding a conserved DNA sequence [CC(A/T)6GG], known as a CArG box or serum response element (Kuwahara et al., 2005). The ability of serum response factor to distinguish between different target genes depends on the presence of binding sites for

8. Mechanical Induction of Gene Expression in Connective Tissue

191

other transcription factors, the number of CArG boxes, and a large number of cofactors. Myocardin and the myocardin-related transcription factors A and B are expressed in a broad range of cell types (Wang et al., 2002a) and comprise a family of coactivators that mediates the transcriptional regulating activity of serum response factor (Wang et al., 2001) (Fig. 1). The CArG box serves as a docking site for myocardin and myocardin related transcription factors. These proteins enhance the transcriptional activity of serum response factor by forming a ternary complex with serum response factor on DNA (Wang et al., 2002a). MRTF-A and -B also convey stimulatory signals from the Rho GTPase and the actin cytoskeleton to serum response factor via their regulated translocation into the nucleus (Miralles et al., 2003). The a-smooth muscle actin promoter contains several conserved cis elements (Blank et al., 1992; Tomasek et al., 2005) that both positively and negatively affect transcription. Within the proximal 400 bases of the a-smooth muscle actin promoter, in addition to CArG elements are Transforming growth factor (TGF)-b control elements (Hautmann et al., 1997) as well as E-box elements that can regulate a-smooth muscle actin promoter activity (Kumar et al., 2003). We have shown that in fibroblastic cells with low basal levels of actin filaments, stretch-induced activation of a-smooth muscle actin relies on the CArG B box in the a-smooth muscle actin promoter (Wang et al., 2002b). When we first attempted these experiments in myofibroblasts that expressed high levels of a-smooth muscle actin, we did not see additional stretch-induced increases of a-smooth muscle actin protein. Instead, force reduced a-smooth muscle actin protein (Wang et al., 2000) and inhibited skeletal a actin promoter activity (Lew et al., 1999). Subsequently, we have discovered that the basal, unstimulated level of actin assembly in cultured cells is critical in these experiments since the abundance of actin filaments are now known to regulate serum response factordependent transcription (Yoshida et al., 2003). To overcome this limitation, we used ROS 17/2.8 cells that develop low levels of actin filaments in vitro and are readily transfected (Wang et al., 2002b). These experiments underline the importance of careful control of cell culture conditions when performing experiments on a-smooth muscle actin activation by force (Wang et al., 2000, 2003). E. Cell Culture Models for Mechanical Induction of a-smooth Muscle Actin Expression To study cytoskeletal involvement in the regulation of the hypertrophic response in cardiac myocytes and fibroblasts, it is essential to use models that generate membrane strain as a result of direct manipulation of integrins. As described previously (Chan et al., 2009; Glogauer and Ferrier, 1998; Glogauer et al., 1995, 1997; Wang et al., 2000, 2002b; Zhao et al., 2007), we developed and tested a model, based on magne tically generated tensile forces applied to collagen-coated beads, which does indeed deliver measurable, physiologically meaningful forces through integrins to the actin cytoskeleton. Moreover, this model provides very useful correlates for examining the expression of a-smooth muscle actin by fibroblasts in vivo and what might be important upstream regulators and mechanosensors, such as the focal adhesion kinase (Chan et al., 2009).

192

Magnet Collagen-coated beads Plasma membrane RhoA-GDP GTP

Integrins GDP RhoA-GTP

ROCK

MRTF MRTF

p-MLC

Myosin activity

p-LIMK

MRTF

p-Cofilin

SRF MRTF

CArG-B

α-SMA promoter

Fig. 1 Model to demonstrate application of tensile forces to cells using magnetically generated forces, collagen-coated beads, and the resultant activation of the Rho–LIM kinase–cofilin–actin filament signaling system that leads to enhanced expression of a-smooth actin.

Matthew W. C. Chan et al.

Actin polymerization

193

8. Mechanical Induction of Gene Expression in Connective Tissue

We have developed magnetic devices that deliver precise tensile forces to fibrillar collagen-coated magnetite beads attached to integrins on cultured cells. Force levels are evenly distributed across the whole diameter of the culture dish, unlike other substrate stretch methods that may produce a nonlinear strain gradient (Fig. 2). The beads are attached to the ventral surfaces of cultured cells through a2b1 and a11b1 integrins. The stretch system pulls the beads vertically, delivering tensile forces of 0.6 pN/µm2 (force/cell area), which are within the range of physiological force levels that are expected to be encountered by cardiac fibroblasts in vivo (Hamrell and Dey, 1993; Wang et al., 2000). Notably, to model the forces that might be expected in cardiac overload, the force levels can be increased by twofold (to 1.2 pN/µm2 cell area), which is in the range of predicted increases of force that might be generated in hypertrophic conditions in vivo (Mann et al., 1991). Thus it is possible to model “physiological” and “pathological” force levels, respectively. Further, compressive forces can be applied by placing the magnet below cells, and twisting forces can be applied by rotating the magnet placed above cells by 90° after half the total length of exposure time (Mak et al., 2008). The forces generated by magnetic fields on 5-µm beads have been estimated from Stokes’ law and from direct measurements of bead velocities in viscous fluids (Glogauer and Ferrier, 1998). Magnetic fields producing one of two force levels at the cell surface will be used for all experiments. Since we have carefully mapped out the magnetic flux densities that are generated by the permanent and electric magnets, we have precise estimates of flux densities at various distances from the magnetic pole Cell stretching model through focal adhesions Magnet

Stretching forces α2β1 integrins in focal adhesions

Collagen-coated magnetite beads

Nucleus

Glass coverslip or culture dish

Fig. 2 Cartoon of experimental model system used for in vitro cell stretching. Collagen beads are attached to the dorsal surfaces of cells through b1 integrins. Stretching forces are applied (0.5–2 pN/mm2 surface area of cell). The stretching forces can be adjusted in terms of amplitude, direction, and duration, to suit the desired experimental condition.

194

Matthew W. C. Chan et al.

face. Another important variable in determining force application to the cells is to estimate the total cell surface area and the percent surface area covered by beads. This is determined by using an image analyzer as described (Glogauer et al., 1995). From these data, precise estimates of the force applied per cell can be obtained. F. Coating Methods for Beads Based on the abundance of type I collagen in the myocardial extracellular matrix and its impact on the cardiac hypertrophic process (Pelouch et al., 1993; Shirwany and Weber, 2006) and its apparent role in force transmission to fibroblasts (Provenzano and Vanderby, 2006), collagen seems to be a good candidate molecule for bead coatings. Further, previous data (Wang et al., 2003) indicate that when force is applied through fibronectin-coated beads in rat cardiac fibroblasts, there is no activation of the p38 MAP kinase or increased expression of a-smooth muscle actin. In contrast, force applied to collagen-coated beads strongly upregulates p38 kinase activity and a-smooth muscle actin content. This finding is in agreement with in vivo findings of pressure overload-induced p38 activation in rats. To test for nonspecific binding to the cell membrane and the specificity of the collagen effect, beads are coated with the nonintegrin-dependent adhesive poly-L lysine or with bovine serum albumin (BSA) as described (Wang et al., 2003). These different bead coatings permit comparisons of loading through integrins versus stretch ing through the cell membrane and other integral membrane proteins. They should also indicate which one of the collagen receptors is most important in the stretch response. Beads can be coated with antibodies to the specific a subunits that are expected as collagen receptors and determine which antibody is most effective in inducing altera tion of a-smooth muscle actin levels in response to stretching (Wang et al., 2000, 2003). As the antibodies do not exhibit multivalent interactions with receptors, it is possible to study clustering by confocal microscopy. Beads should be coated with equimolar concentrations of proteins. Bead size can be estimated by electronic particle counting (Glogauer et al., 1997). Verification of collagen or antibody coating on beads is assessed with antibodies to type I collagen (rabbit antibovine, antiserum; Chemicon). Beads are added to well-spread cells for 10 min and cells are washed 3 times to remove unbound beads. For longer duration experiments when force is applied (1–4 h), bead internalization and bead detachment is minimal under these conditions, but this must be carefully examined as mentioned in the section above, as phagocytosis of beads confounds experimental results. For experiments in which large numbers of cells are analyzed, phase-contrast microscopy is used to assess the equality of bead loading and the relatively even distribution of beads across the dish. Cultures in which bead loading does not conform to previously established protocols (Glogauer et al., 1995) should not be used as this would indicate that the cells were not loaded equivalently. The force applied is proportional to the volume of the beads (Glogauer and Ferrier, 1998). From the distribution of bead diameters we have calculated a cross-sectional area-weighted mean diameter of 5 µm. The latter value is appropriate for calculating the force on a per unit area basis.

8. Mechanical Induction of Gene Expression in Connective Tissue

195

G. Cell Transfection and Promoter Methods The tensile force model system described above is particularly well suited to study of force-induced gene expression. Consequently, the use of rapid readout cell systems is helpful. For promoter studies, a Rous sarcoma virus (RSV; 124 to þ34)/luciferase construct and a similar Renilla luciferase construct are used as internal controls as they are not affected by cell stretching (Lew et al., 1999). Subconfluent cardiac fibroblasts are transiently transfected with Qiagen-purified plasmid DNA using the transfection reagent Lipofectamine 2000 or Fugene (Roche) as described (Lew et al., 1999; Sayegh et al., 2005; Wang et al., 2002a). Standardized amounts of plasmid (typically 10 µg/100 mm plate of cells) are used in quadruplicate replicate cultures for each promoter construct. Based on our experience with cardiac fibroblasts, we usually obtain reproducible and meaningful results with up to threefold increased a-smooth muscle actin promoter activity in response to stretch in cells with no detectable baseline a-smooth muscle actin expression. After 16 h incubation, the medium is replaced with serum-free medium. Collagen or poly-L-lysine or BSA-coated beads are added to the transfected cells for 15 min at 37°C and force applied. After washing in phosphate-buffered saline (PBS), cells are collected and processed for b-galactosidase activity as described (Lew et al., 1999). The assessment of b-galactosidase or luciferase activity can be used to analyze the effect of force on promoter activity and of the various perturbations that are sought.

H. Identification of Adhesion-Associated Proteins The system described above can be used also to study which proteins in the collagen bead adhesion complex may be regulated by the application of force. After addition of collagen- or BSA-coated magnetite beads, and then application of force, beadassociated proteins are isolated from cells as described (Wang et al., 2006) (Fig. 3). Briefly, cells are washed with cold PBS to remove unbound beads, scraped into cold cytoskeleton extraction buffer, and sonicated for 10s. The beads are isolated from the lysate with a magnet, resuspended in cold cytoskeleton extraction buffer, homoge nized, and reisolated magnetically. Bead-associated proteins are removed by boiling in Laemmli sample buffer. In control experiments, cell cultures are treated with swinho lide A (50 nM, 25°C, 20 min) prior to bead binding to dissipate focal adhesion formation. The resulting lysates can then be analyzed for bead-associated proteins by immunoblotting (for known proteins of interest) or with the use of proteomic methods (e.g., tandem mass spectrometry and isotope-coded affinity tags; Pho et al., 2008) for a wider screen of potentially interesting proteins that are regulated by force.

VI. Conclusions The application of the methods described here should provide relatively facile approaches to identify gene responses in cultured mesenchymal cells to a variety of different mechanical perturbations. We have described in detail the utilization of one

196

Isolation of focal adhesion-associated proteins Magnet

Collagen-coated magnetite bead

Collagen-coated magnetite bead

β αβ α β αβ α

β αβ α β αβ α Cell membrane Talin Vinculin

Filamin Paxillin

Filamin Paxillin

Talin

Filamin

Vinculin

Filamin

Actin

Actin

Attachment of collagen-coated beads stimulates integrin clustering and recruitment of focal adhesion proteins and filamin A

CSKB disrupts cell membrane but preserves integrity of focal adhesions and associated proteins. CSKB

Tandem Mass Spec siRNA

FAC-bound beads

Fig. 3 Diagram to illustrate methods for purification of adhesion-associated proteins from collagen-coated beads attached to the dorsal surface of cells. Beadassociated proteins can be analyzed by immunoblotting or, for searching for novel proteins, by mass spectrometry.

Matthew W. C. Chan et al.

Magnet

8. Mechanical Induction of Gene Expression in Connective Tissue

197

tensile force model to the cardiac fibroblast situation and the specific focus on the regulation of the a-smooth muscle actin promoter, but this approach may be applicable to other genes in which short-term responses can be evaluated. Notably, there are limitations to all of these methods and we have described limitations and pitfalls, which must be considered for the appropriate interpretation and application to the in vivo situation. In particular, the use of controls and the need for reasonable force levels cannot be overstated.

Acknowledgments MC is supported by a Canadian Heart and Stroke Foundation fellowship. CAM is supported by a Canada Research Chair (Tier 1). The research described in this review was supported by a Canadian Institutes of Health Research Operating Grant to CAM MGP-37783 and by an Ontario Heart and Stroke grant to CAM (T 6022).

References Adams, M. A. (2006). The mechanical environment of chondrocytes in articular cartilage. Biorheology 43, 537–545. Anderson, E. J., and Knothe Tate, M. L. (2008). Idealization of pericellular fluid space geometry and dimension results in a profound underprediction of nano-microscale stresses imparted by fluid drag on osteocytes. J. Biomech. 41, 1736–1746. Bacabac, R. G., Smit, T. H., Cowin, S. C., Van Loon, J. J., Nieuwstadt, F. T., Heethaar, R., and Klein-Nulend, J. (2005). Dynamic shear stress in parallel-plate flow chambers. J. Biomech. 38, 159–167. Balestrini, J. L., and Billiar, K. L. (2009). Magnitude and duration of stretch modulate fibroblast remodeling. J. Biomech. Eng. 131, 051005. Banes, A. J., Gilbert, J., Taylor, D., and Monbureau. O. (1985). A new vacuum-operated stress-providing instrument that applies static or variable duration cyclic tension or compression to cells in vitro. J. Cell Sci. 75, 35–42. Bao, G., and Suresh, S. (2003). Cell and molecular mechanics of biological materials. Nat. Mater. 2, 715–725. Bar-Ziv, R., Moses, E., and Nelson, P. (1998). Dynamic excitations in membranes induced by optical tweezers. Biophys. J. 75, 294–320. Black, F. M., Packer, S. E., Parker, T. G., Michael, L. H., Roberts, R., Schwartz, R. J., and Schneider, M. D. (1991). The vascular smooth muscle alpha-actin gene is reactivated during cardiac hypertrophy provoked by load. J. Clin. Invest. 88, 1581–1588. Blank, R. S., McQuinn, T. C., Yin, K. C., Thompson, M. M., Takeyasu, K., Schwartz, R. J., and Owens, G. K. (1992). Elements of the smooth muscle alpha-actin promoter required in cis for transcriptional activation in smooth muscle. Evidence for cell type-specific regulation. J. Biol. Chem. 267, 984–989. Brilla, C. G., Maisch, B., and Weber, K. T. (1992). Myocardial collagen matrix remodelling in arterial hypertension. Eur. Heart J. 13(Suppl. D), 24–32. Brilla, C. G., Maisch, B., Rupp, H., Funck, R., Zhou, G., and Weber, K. T. (1995). Pharmacological modulation of cardiac fibroblast function. Hertz 20, 127–134. Brown, T. D. (2000). Techniques for mechanical stimulation of cells in vitro: A review. J. Biomech. 33, 3–14. Brown, R. A., Prajapati, R., McGrouther, D. A., Yannas, I. V., and Eastwood, M. (1998). Tensional home ostasis in dermal fibroblasts: Mechanical responses to mechanical loading in three-dimensional substrates. J. Cell Physiol. 175, 323–332. Brown, X. Q., Ookawa, K., and Wong, J. Y. (2005). Evaluation of polydimethylsiloxane scaffolds with physiologically-relevant elastic moduli: Interplay of substrate mechanics and surface chemistry effects on vascular smooth muscle cell response. Biomaterials 26, 3123–3129.

198

Matthew W. C. Chan et al. Burton-Wurster, N., Vernier-Singer, M., Farquhar, T., and Lust, G. (1993). Effect of compressive loading and unloading on the synthesis of total protein, proteoglycan, and fibronectin by canine cartilage explants. J. Orthop. Res. 11, 717–729. Butt, R. P., Laurent, G. J., and Bishop, J. E. (1995). Mechanical load and polypeptide growth factors stimulate cardiac fibroblast activity. Ann. N Y Acad. Sci. 752, 387–393. Califano, J. P., and Reinhart-King, C. A. (2010). Exogenous and endogenous force regulation of endothelial cell behavior. J. Biomech. 43, 79–86. Campbell, S. E., and Katwa, L. C. (1997). Angiotensin II stimulated expression of transforming growth factor-beta1 in cardiac fibroblasts and myofibroblasts. J. Mol. Cell Cardiol. 29, 1947–1958. Campbell, S. E., Janicki, J. S., and Weber, K. T. (1995). Temporal differences in fibroblast proliferation and phenotype expression in response to chronic administration of angiotensin II or aldosterone. J. Mol. Cell Cardiol. 27, 1545–1560. Carver, W., Nagpal, M. L., Nachtigal, M., Borg, T. K., and Terracio, L. (1991). Collagen expression in mechanically stimulated cardiac fibroblasts. Circ. Res. 69, 116–122. Catalucci, D., Latronico, M. V., Ellingsen, O., and Condorelli, G. (2008). Physiological myocardial hyper trophy: How and why? Front. Biosci. 13, 312–324. Chan, M. W., Arora, P. D., Bozavikov, P., and McCulloch, C. A. (2009). FAK, PIP5KIg and gelsolin cooperatively mediate force-induced expression of a-smooth muscle actin. J. Cell Sci. 122, 2769–2781. Charras, G. T., Lehenkari, P. P., and Horton, M. A. (2001). Atomic force microscopy can be used to mechanically stimulate osteoblasts and evaluate cellular strain distributions. Ultramicroscopy 86, 85–95. Charras, G., Lehenkari, P., and Horton, M. (2002). Biotechnological applications of atomic force microscopy. Methods Cell Biol. 68, 171–191. Chen, C. S. (2008). Mechanotransduction—a field pulling together? J. Cell Sci. 121, 3285–3292. Chen, C. S., Tan, J., and Tien, J. (2004). Mechanotransduction at cell-matrix and cell–cell contacts. Annu. Rev. Biomed. Eng. 6, 275–302. Chen, J. H., Liu, C., You, L., and Simmons, C. A. (2010). Boning up on Wolff’s Law: Mechanical regulation of the cells that make and maintain bone. J. Biomech. 43, 108–118. Chien, K. R. (1999). Stress pathways and heart failure. Cell 98, 555–558. Chiquet, M., Renedo, A. S., Huber, F., and Fluck, M. (2003). How do fibroblasts translate mechanical signals into changes in extracellular matrix production? Matrix Biol. 22, 73–80. Chiquet, M., Gelman, L., Lutz, R., and Maier, S. (2009). From mechanotransduction to extracellular matrix gene expression in fibroblasts. Biochim. Biophys. Acta 1793, 911–920. Choe, M. M., Tomei, A. A., and Swartz, M. A. (2006). Physiological 3D tissue model of the airway wall and mucosa. Nat. Protoc. 1, 357–362. Chung, C. A., Tzou, M. R., and Ho, R. W. (2005). Oscillatory flow in a cone-and-plate bioreactor. J. Biomech. Eng. 127, 601–610. Corey, D. P. (2003). New TRP channels in hearing and mechanosensation. Neuron 39, 585–588. Cortright, D. N., and Szallasi, A. (2009). TRP channels and pain. Curr. Pharm. Des. 15, 1736–1749. Covell, J. W. (1990). “Cardiac Myocyte Connective Tissue Interactions in Health and Disease,” Vol. 13, pp. 99–112. Karger, Basel. D’Addario, M., Arora, P. D., Ellen, R. P., and McCulloch, C. A. (2002). Interaction of p38 and Sp1 in a mechanical force-induced, beta 1 integrin-mediated transcriptional circuit that regulates the actin-binding protein filamin-A. J. Biol. Chem. 277, 47541–47550. Dai, J., and Sheetz, M. P. (1995). Mechanical properties of neuronal growth cone membranes studied by tether formation with laser optical tweezers. Biophys. J. 68, 988–996. Datta, N., Pham, Q. P., Sharma, U., Sikavitsas, V. I., Jansen, J. A., and Mikos, A. G. (2006). In vitro generated extracellular matrix and fluid shear stress synergistically enhance 3D osteoblastic differentiation. Proc. Natl. Acad. Sci. USA 103, 2488–2493. Dewey, C. F. Jr. (1984). Effects of fluid flow on living vascular cells. J. Biomech. Eng. 106, 31–35. Discher, D. E., Janmey, P., and Wang, Y. L. (2005). Tissue cells feel and respond to the stiffness of their substrate. Science 310, 1139–1143.

8. Mechanical Induction of Gene Expression in Connective Tissue

199

Eghbali, M. (1992). Cardiac fibroblasts: Function, regulation of gene expression, and phenotypic modula tion. Basic Res. Cardiol. 87(Suppl 2), 183–189. Engler, A. J., Griffin, M. A., Sen, S., Bonnemann, C. G., Sweeney, H. L., and Discher, D. E. (2004). Myotubes differentiate optimally on substrates with tissue-like stiffness: Pathological implications for soft or stiff microenvironments. J. Cell Biol. 166, 877–887. Engler, A. J., Sen, S., Sweeney, H. L., and Discher, D. E. (2006). Matrix elasticity directs stem cell lineage specification. Cell 126, 677–689. Engler, A. J., Rehfeldt, F., Sen, S., and Discher, D. E. (2007). Microtissue elasticity: Measurements by atomic force microscopy and its influence on cell differentiation. Methods Cell Biol. 83, 521–545. Engler, A. J., Carag-Krieger, C., Johnson, C. P., Raab, M., Tang, H. Y., Speicher, D. W., Sanger, J. W., Sanger, J. M., and Discher, D. E. (2008). Embryonic cardiomyocytes beat best on a matrix with heart-like elasticity: Scar-like rigidity inhibits beating. J. Cell Sci. 121, 3794–3802. Fleury, M. E., Boardman, K. C., and Swartz, M. A. (2006). Autologous morphogen gradients by subtle interstitial flow and matrix interactions. Biophys. J. 91, 113–121. Freeman, P. M., Natarajan, R. N., Kimura, J. H., and Andriacchi, T. P. (1994). Chondrocyte cells respond mechanically to compressive loads. J. Orthop. Res. 12, 311–320. Fritton, S. P., and Weinbaum, S. (2009). Fluid and solute transport in bone: Flow-induced mechanotransduc tion. Annu. Rev. Fluid Mech. 41, 347–374. Geiger, B., and Bershadsky, A. (2002). Exploring the neighborhood: Adhesion-coupled cell mechanosensors. Cell 110, 139–142. Georges, P. C., Miller, W. J., Meaney, D. F., Sawyer, E. S., and Janmey, P. A. (2006). Matrices with compliance comparable to that of brain tissue select neuronal over glial growth in mixed cortical cultures. Biophys. J. 90, 3012–3018. Glogauer, M., Ferrier, J. (1998). A new method for application of force to cells via ferric oxide beads. Pflugers Arch. 435, 320–327. Glogauer, M., Ferrier, J., and McCulloch, C. A. (1995). Magnetic fields applied to collagen-coated ferric oxide beads induce stretch-activated Ca2þ flux in fibroblasts. Am. J. Physiol. 269, C1093–C1104. Glogauer, M., Arora, P., Yao, G., Sokholov, I., Ferrier, J., and McCulloch, C. A. (1997). Calcium ions and tyrosine phosphorylation interact coordinately with actin to regulate cytoprotective responses to stretching. J. Cell Sci. 110(Pt 1), 11–21. Goffin, J. M., Pittet, P., Csucs, G., Lussi, J. W., Meister, J. J., and Hinz, B. (2006). Focal adhesion size controls tension-dependent recruitment of alpha-smooth muscle actin to stress fibers. J. Cell Biol. 172, 259–268. Gonzalez, A., Lopez, B., Ravassa, S., Beaumont, J., Arias, T., Hermida, N., Zudaire, A., and Diez, J. (2009). Biochemical markers of myocardial remodelling in hypertensive heart disease. Cardiovasc. Res. 81, 509–518. Gosse, C., and Croquette, V. (2002). Magnetic tweezers: Micromanipulation and force measurement at the molecular level. Biophys. J. 82, 3314–3329. Gottlieb, P. A., Suchyna, T. M., Ostrow, L. W., and Sachs, F. (2004). Mechanosensitive ion channels as drug targets. Curr. Drug Targets CNS Neurol. Disord. 3, 287–295. Grier, D. G. (2003). A revolution in optical manipulation. Nature 424, 810–816. Griffith, L. G., and Swartz, M. A. (2006). Capturing complex 3D tissue physiology in vitro. Nat. Rev. Mol. Cell Biol. 7, 211–224. Grinnell, F. (2010). Cell migration and signaling in three-dimensional matrix. Annu. Rev. Cell Dev. Biol. (in press). Grove, D., Zak, R., Nair, K. G., and Aschenbrenner, V. (1969). Biochemical correlates of cardiac hyper trophy. IV. Observations on the cellular organization of growth during myocardial hypertrophy in the rat. Circ. Res. 25, 473–485. Halka, A. T., Turner, N. J., Carter, A., Ghosh, J., Murphy, M. O., Kirton, J. P., Kielty, C. M., and Walker, M. G. (2008). The effects of stretch on vascular smooth muscle cell phenotype in vitro. Cardiovasc. Pathol. 17, 98–102. Hamrell, B. B., and Dey, S. K. (1993). Sarcomere shortening velocity in pressure overload hypertrophied rabbit right ventricular myocardium at physiological sarcomere lengths. J. Mol. Cell Cardiol. 25, 1483–1500.

200

Matthew W. C. Chan et al. Harris, A. K., Wild, P., and Stopak, D. (1980). Silicone rubber substrata: A new wrinkle in the study of cell locomotion. Science 208, 177–179. Hasaneen, N. A., Zucker, S., Cao, J., Chiarelli, C., Panettieri, R. A., and Foda, H. D. (2005). Cyclic mechanical strain-induced proliferation and migration of human airway smooth muscle cells: Role of EMMPRIN and MMPs. Faseb J. 19, 1507–1509. Hautmann, M. B., Madsen, C. S., and Owens, G. K. (1997). A transforming growth factor beta (TGFbeta) control element drives TGFbeta-induced stimulation of smooth muscle alpha-actin gene expression in concert with two CArG elements. J. Biol. Chem. 272, 10948–10956. Helmke, B. P., and Schwartz, M. A. (2004). Putting the squeeze on mechanotransduction. Dev. Cell 6, 745–746. Hinz, B. (2006). Masters and servants of the force: The role of matrix adhesions in myofibroblast force perception and transmission. Eur. J. Cell Biol. 85, 175–181. Hinz, B. (2010). The myofibroblast: Paradigm for a mechanically active cell. J. Biomech. 43, 146–155. Hinz, B., and Gabbiani, G. (2003). Mechanisms of force generation and transmission by myofibroblasts. Curr. Opin. Biotechnol. 14, 538–546. Hinz, B., Phan, S. H., Thannickal, V. J., Galli, A., Bochaton-Piallat, M. L., Gabbiani, G. (2007). The myofibroblast: one function, multiple origins. Am. J. Pathol. 170, 1807–1816. Hoffman, B. D., and Crocker, J. C. (2009). Cell mechanics: Dissecting the physical responses of cells to force. Annu. Rev. Biomed. Eng. 11, 259–288. Husse, B., Briest, W., Homagk, L., Isenberg, G., and Gekle, M. (2007). Cyclical mechanical stretch modulates expression of collagen I and collagen III by PKC and tyrosine kinase in cardiac fibroblasts. Am. J. Physiol. Regul. Integr. Comp. Physiol. 293, R1898–R1907. Ingber, D. E. (1997). Tensegrity: The architectural basis of cellular mechanotransduction. Annu. Rev. Physiol. 59, 575–599. Ingber, D. E. (2003a). Mechanobiology and diseases of mechanotransduction. Ann. Med. 35, 564–577. Ingber, D. E. (2003b). Tensegrity II. How structural networks influence cellular information processing networks. J. Cell Sci. 116, 1397–1408. Ingber, D. E., Mow, V. C., Butler, D., Niklason, L., Huard, J., Mao, J., Yannas, I., Kaplan, D., and VunjakNovakovic, G. (2006). Tissue engineering and developmental biology: Going biomimetic. Tissue Eng. 12, 3265–3283. Janmey, P. A., and McCulloch, C. A. (2007). Cell mechanics: Integrating cell responses to mechanical stimuli. Annu. Rev. Biomed. Eng. 9, 1–34. Janmey, P. A., and Weitz, D. A. (2004). Dealing with mechanics: Mechanisms of force transduction in cells. Trends Biochem. Sci. 29, 364–370. Janmey, P. A., Winer, J. P., Murray, M. E., and Wen, Q. (2009). The hard life of soft cells. Cell Motil. Cytoskeleton 66, 597–605. Jonas, M., So, P. T., and Huang, H. (2008). Cell mechanics at multiple scales. Methods Enzymol. 443, 177–198. Jones, M. P., and Bratten, J. R. (2008). Small intestinal motility. Curr. Opin. Gastroenterol. 24, 164–172. Juliano, R. L., and Haskill, S. (1993). Signal transduction from the extracellular matrix. J. Cell Biol. 120, 577–585. Jungbauer, S., Gao, H., Spatz, J. P., and Kemkemer, R. (2008). Two characteristic regimes in frequencydependent dynamic reorientation of fibroblasts on cyclically stretched substrates. Biophys. J. 95, 3470–3478. Kandow, C. E., Georges, P. C., Janmey, P. A., and Beningo, K. A. (2007). Polyacrylamide hydrogels for cell mechanics: Steps toward optimization and alternative uses. Methods Cell Biol. 83, 29–46. Katsumi, A., Orr, A. W., Tzima, E., and Schwartz, M. A. (2004). Integrins in mechanotransduction. J. Biol. Chem. 279, 12001–12004. Katwa, L. C., Campbell, S. E., Tyagi, S. C., Lee, S. J., Cicila, G. T., and Weber, K. T. (1997). Cultured myofibroblasts generate angiotensin peptides de novo. J. Mol. Cell Cardiol. 29, 1375–1386. Kaye, D., Pimental, D., Prasad, S., Maki, T., Berger, H. J., McNeil, P. L., Smith, T. W., and Kelly, R. A. (1996). Role of transiently altered sarcolemmal membrane permeability and basic fibroblast growth factor release in the hypertrophic response of adult rat ventricular myocytes to increased mechanical activity in vitro. J. Clin. Invest. 97, 281–291.

8. Mechanical Induction of Gene Expression in Connective Tissue

201

Keating, M. T., and Sanguinetti, M. C. (1996). Molecular genetic insights into cardiovascular disease. Science 272, 681–685. Kessler, D., Dethlefsen, S., Haase, I., Plomann, M., Hirche, F., Krieg, T., and Eckes, B. (2001). Fibroblasts in mechanically stressed collagen lattices assume a “synthetic” phenotype. J. Biol. Chem. 276, 36575–36585. Khan, R., and Sheppard, R. (2006). Fibrosis in heart disease: Understanding the role of transforming growth factor-beta in cardiomyopathy, valvular disease and arrhythmia. Immunology 118, 10–24. Kim, D. H., Wong, P. K., Park, J., Levchenko, A., and Sun, Y. (2009). Microengineered platforms for cell mechanobiology. Annu. Rev. Biomed. Eng. 11, 203–233. Kiselyov, K., and Patterson, R. L. (2009). The integrative function of TRPC channels. Front. Biosci. 14, 45–58. Kiseleva, I., Kamkin, A., Kohl, P., and Lab, M. J. (1996). Calcium and mechanically induced potentials in fibroblasts of rat atrium. Cardiovasc. Res. 32, 98–111. Klein, E. A., Yin, L., Kothapalli, D., Castagnino, P., Byfield, F. J., Xu, T., Levental, I., Hawthorne, E., Janmey, P. A., and Assoian, R. K. (2009). Cell-cycle control by physiological matrix elasticity and in vivo tissue stiffening. Curr. Biol. 19, 1511–1518. Ko, K. S., Arora, P. D., and McCulloch, C. A. (2001). Cadherins mediate intercellular mechanical signaling in fibroblasts by activation of stretch-sensitive calcium-permeable channels. J. Biol. Chem. 276, 35967–35977. Komuro, I., Kudo, S., Yamazaki, T., Zou, Y., Shiojima, I., and Yazaki, Y. (1996). Mechanical stretch activates the stress-activated protein kinases in cardiac myocytes. Faseb J. 10, 631–636. Krishnan, V., and Davidovitch, Z. (2009). On a path to unfolding the biological mechanisms of orthodontic tooth movement. J. Dent. Res. 88, 597–608. Kumar, M. S., Hendrix, J. A., Johnson, A. D., and Owens, G. K. (2003). Smooth muscle alpha-actin gene requires two E-boxes for proper expression in vivo and is a target of class I basic helix-loop-helix proteins. Circ. Res. 92, 840–847. Kuwahara, F., Kai, H., Tokuda, K., Kai, M., Takeshita, A., Egashira, K., and Imaizumi, T. (2002). Transforming growth factor-beta function blocking prevents myocardial fibrosis and diastolic dysfunction in pressure-overloaded rats. Circulation 106, 130–135. Kuwahara, K., Barrientos, T., Pipes, G. C., Li, S., and Olson, E. N. (2005). Muscle-specific signaling mechanism that links actin dynamics to serum response factor. Mol. Cell Biol. 25, 3173–3181. Lammerding, J., Kamm, R. D., and Lee, R. T. (2004). Mechanotransduction in cardiac myocytes. Ann. N Y Acad. Sci. 1015, 53–70. Lateef, S. S., Boateng, S., Hartman, T. J., Crot, C. A., Russell, B., and Hanley, L. (2002). GRGDSP peptidebound silicone membranes withstand mechanical flexing in vitro and display enhanced fibroblast adhe sion. Biomaterials 23, 3159–3168. Lee, A. A., Delhaas, T., Waldman, L. K., MacKenna, D. A., Villarreal, F. J., and McCulloch, A. D. (1996). An equibiaxial strain system for cultured cells. Am. J. Physiol. 271, C1400–C1408. Lee, E. J., Holmes, J. W., and Costa, K. D. (2008). Remodeling of engineered tissue anisotropy in response to altered loading conditions. Ann. Biomed. Eng. 36, 1322–1334. Lele, T. P., Sero, J. E., Matthews, B. D., Kumar, S., Xia, S., Montoya-Zavala, M., Polte, T., Overby, D., Wang, N., and Ingber, D. E. (2007). Tools to study cell mechanics and mechanotransduction. Methods Cell Biol. 83, 443–472. Leslie, K. O., Taatjes, D. J., Schwarz, J., vonTurkovich, M., and Low, R. B. (1991). Cardiac myofibroblasts express alpha smooth muscle actin during right ventricular pressure overload in the rabbit. Am. J. Pathol. 139, 207–216. Lew, A. M., Glogauer, M., and McUlloch, C. A. (1999). Specific inhibition of skeletal alpha-actin gene transcription by applied mechanical forces through integrins and actin. Biochem. J. 341(Pt 3), 647–653. Lewin, G. R., and Moshourab, R. (2004). Mechanosensation and pain. J. Neurobiol. 61, 30–44. Linask, K. K., Manisastry, S., and Han, M. (2005). Cross talk between cell–cell and cell–matrix adhesion signaling pathways during heart organogenesis: Implications for cardiac birth defects. Microsc. Microanal. 11, 200–208. Lionetti, V., Recchia, F. A., and Ranieri, V. M. (2005). Overview of ventilator-induced lung injury mechan isms. Curr. Opin. Crit. Care 11, 82–86.

202

Matthew W. C. Chan et al. Little, W. C., Smith, M. L., Ebneter, U., and Vogel, V. (2008). Assay to mechanically tune and optically probe fibrillar fibronectin conformations from fully relaxed to breakage. Matrix Biol. 27, 451–461. Lutolf, M. P., and Hubbell, J. A. (2005). Synthetic biomaterials as instructive extracellular microenviron ments for morphogenesis in tissue engineering. Nat. Biotechnol. 23, 47–55. MacKenna, D., Summerour, S. R., and Villarreal, F. J. (2000). Role of mechanical factors in modulating cardiac fibroblast function and extracellular matrix synthesis. Cardiovasc. Res. 46, 257–263. Majd, H., Wipff, P. J., Buscemi, L., Bueno, M., Vonwil, D., Quinn, T. M., and Hinz, B. (2009). A novel method of dynamic culture surface expansion improves mesenchymal stem cell proliferation and pheno type. Stem Cells 27, 200–209. Mak, B. C., Wang, Q., Laschinger, C., Lee, W., Ron, D., Harding, H. P., Kaufman, R. J., Scheuner, D., Austin, R. C., and McCulloch, C. A. (2008). Novel function of PERK as a mediator of force-induced apoptosis. J. Biol. Chem. 283, 23462–23472. Makino, A., Shin, H. Y., Komai, Y., Fukuda, S., Coughlin, M., Sugihara-Seki, M., and Schmid-Schonbein, G. W. (2007). Mechanotransduction in leukocyte activation: A review. Biorheology 44, 221–249. Mann, D. L., Urabe, Y., Kent, R. L., Vinciguerra, S., and Cooper, G. (1991). Cellular versus myocardial basis for the contractile dysfunction of hypertrophied myocardium. Circ Res. 68, 402–415. Martin, R. B. (2007). The importance of mechanical loading in bone biology and medicine. J. Musculoskelet. Neuronal Interact. 7, 48–53. Martinac, B. (2004). Mechanosensitive ion channels: Molecules of mechanotransduction. J. Cell Sci. 117, 2449–2460. Mata, A., Boehm, C., Fleischman, A. J., Muschler, G., and Roy, S. (2002). Growth of connective tissue progenitor cells on microtextured polydimethylsiloxane surfaces. J. Biomed. Mater. Res. 62, 499–506. Miralles, F., Posern, G., Zaromytidou, A. I., and Treisman, R. (2003). Actin dynamics control SRF activity by regulation of its coactivator MAL. Cell 113, 329–342. Moraes, C., Chen, J. H., Sun, Y., and Simmons, C. A. (2010). Microfabricated arrays for high-throughput screening of cellular response to cyclic substrate deformation. Lab Chip 10, 227–234. Moraes, C., Wang, G., Sun, Y., and Simmons, C. A. (2010). A microfabricated platform for high-throughput unconfined compression of micropatterned biomaterial arrays. Biomaterials 31, 577–584. Mudera, V. C., Pleass, R., Eastwood, M., Tarnuzzer, R., Schultz, G., Khaw, P., McGrouther, D. A., and Brown, R. A. (2000). Molecular responses of human dermal fibroblasts to dual cues: Contact guidance and mechanical load. Cell Motil. Cytoskeleton 45, 1–9. Nicoletti, A., and Michel, J. B. (1999). Cardiac fibrosis and inflammation: Interaction with hemodynamic and hormonal factors. Cardiovasc. Res. 41, 532–543. Nieminen, T. A., Knoner, G., Heckenberg, N. R., and Rubinsztein-Dunlop, H. (2007). Physics of optical tweezers. Methods Cell Biol. 82, 207–236. Olson, E. N., and Srivastava, D. (1996). Molecular pathways controlling heart development. Science 272, 671–676. Orr, A. W., Helmke, B. P., Blackman, B. R., and Schwartz, M. A. (2006). Mechanisms of mechanotransduc tion. Dev. Cell 10, 11–20. Ozaki, S., Kaneko, S., Podyma-Inoue, K. A., Yanagishita, M., and Soma, K. (2005). Modulation of extracellular matrix synthesis and alkaline phosphatase activity of periodontal ligament cells by mechan ical stress. J. Periodontal Res. 40, 110–117. Ozawa, H., Imamura, K., Abe, E., Takahashi, N., Hiraide, T., Shibasaki, Y., Fukuhara, T., and Suda, T. (1990). Effect of a continuously applied compressive pressure on mouse osteoblast-like cells (MC3T3-E1) in vitro. J. Cell Physiol. 142, 177–185. Paszek, M. J., Zahir, N., Johnson, K. R., Lakins, J. N., Rozenberg, G. I., Gefen, A., Reinhart-King, C. A., Margulies, S. S., Dembo, M., Boettiger, D., Hammer, D. A., and Weaver, V. M. (2005). Tensional homeostasis and the malignant phenotype. Cancer Cell. 8, 241–254. Pelham, R. J. Jr., and Wang, Y. (1997). Cell locomotion and focal adhesions are regulated by substrate flexibility. Proc. Natl. Acad. Sci. USA 94, 13661–13665. Pelouch, V., Dixon, I. M., Golfman, L., Beamish, R. E., and Dhalla, N. S. (1993). Role of extracellular matrix proteins in heart function. Mol. Cell Biochem. 129, 101–120.

8. Mechanical Induction of Gene Expression in Connective Tissue

203

Perez-Tamayo, R. (1978). Pathology of collagen degradation. A review. Am. J. Pathol. 92, 508–566. Pho, M., Lee, W., Watt, D. R., Laschinger, C., Simmons, C. A., and McCulloch, C. A. (2008). Cofilin is a marker of myofibroblast differentiation in cells from porcine aortic cardiac valves. Am. J. Physiol. Heart Circ. Physiol. 294, H1767–H1778. Place, E. S., Evans, N. D., and Stevens, M. M. (2009). Complexity in biomaterials for tissue engineering. Nat. Mater 8, 457–470. Porter, B., Zauel, R., Stockman, H., Guldberg, R., and Fyhrie, D. (2005). 3-D computational modeling of media flow through scaffolds in a perfusion bioreactor. J. Biomech. 38, 543–549. Potard, U. S., Butler, J. P., and Wang, N. (1997). Cytoskeletal mechanics in confluent epithelial cells probed through integrins and E-cadherins. Am. J. Physiol. 272, C1654–C1663. Provenzano, P. P., and Vanderby, R., Jr. (2006). Collagen fibril morphology and organization: Implications for force transmission in ligament and tendon. Matrix Biol. 25, 71–84. Pugin, J. (2003). Molecular mechanisms of lung cell activation induced by cyclic stretch. Crit. Care Med. 31, S200–S206. Raeber, G. P., Mayer, J., and Hubbell, J. A. (2008). Part I: A novel in vitro system for simultaneous mechanical stimulation and time-lapse microscopy in 3D. Biomech. Model Mechanobiol. 7, 203–214. Reichelt, J. (2007). Mechanotransduction of keratinocytes in culture and in the epidermis. Eur. J. Cell Biol. 86, 807–816. Reinhart-King, C. A., Fujiwara, K., and Berk, B. C. (2008). Physiologic stress-mediated signaling in the endothelium. Methods Enzymol. 443, 25–44. Rutkowski, J. M., and Swartz, M. A. (2007). A driving force for change: Interstitial flow as a morphor egulator. Trends Cell Biol. 17, 44–50. Rygh, P. (1973). Ultrastructural changes in pressure zones of human periodontium incident to orthodontic tooth movement. Acta Odontol. Scand. 31, 109–122. Sadoshima, J., and Izumo, S. (1997). The cellular and molecular response of cardiac myocytes to mechanical stress. Annu. Rev. Physiol. 59, 551–571. Sawada, Y., and Sheetz, M. P. (2002). Force transduction by Triton cytoskeletons. J. Cell Biol. 156, 609–615. Sayegh, T. Y., Arora, P. D., Fan, L., Laschinger, C. A., Greer, P. A., McCulloch, C. A., and Kapus, A. (2005). Phosphorylation of N-cadherin-associated cortactin by Fer kinase regulates N-cadherin mobility and intercellular adhesion strength. Mol. Biol. Cell. 16, 5514–5527. Shirwany, A., and Weber, K. T. (2006). Extracellular matrix remodeling in hypertensive heart disease. J. Am. Coll Cardiol. 48, 97–98. Shroff, S. G., Saner, D. R., and Lal, R. (1995). Dynamic micromechanical properties of cultured rat atrial myocytes measured by atomic force microscopy. Am. J. Physiol. 269, C286–C292. Skerry, T. M., and Suva, L. J. (2003). Investigation of the regulation of bone mass by mechanical loading: From quantitative cytochemistry to gene array. Cell Biochem. Funct. 21, 223–229. Sniadecki, N. J., Anguelouch, A., Yang, M. T., Lamb, C. M., Liu, Z., Kirschner, S. B., Liu, Y., Reich, D. H., and Chen, C. S. (2007). Magnetic microposts as an approach to apply forces to living cells. Proc. Natl. Acad. Sci. USA 104, 14553–14558. Stenqvist, O., Odenstedt, H., and Lundin, S. (2008). Dynamic respiratory mechanics in acute lung injury/ acute respiratory distress syndrome: Research or clinical tool? Curr. Opin. Crit. Care 14, 87–93. Stiehler, M., Bunger, C., Baatrup, A., Lind, M., Kassem, M., and Mygind, T. (2009). Effect of dynamic 3-D culture on proliferation, distribution, and osteogenic differentiation of human mesenchymal stem cells. J. Biomed. Mater. Res. A 89, 96–107. Storm, C., Pastore, J. J., MacKintosh, F. C., Lubensky, T. C., and Janmey, P. A. (2005). Nonlinear elasticity in biological gels. Nature 435, 191–194. Sun, Y., and Weber, K. T. (1996). Cells expressing angiotensin II receptors in fibrous tissue of rat heart. Cardiovasc. Res. 31, 518–525. Suurmeijer, A. J., Clement, S., Francesconi, A., Bocchi, L., Angelini, A., Van Veldhuisen, D. J., Spagnoli, L. G., Gabbiani, G., and Orlandi, A. (2003). Alpha-actin isoform distribution in normal and failing human heart: A morphological, morphometric, and biochemical study. J. Pathol. 199, 387–397. Svoboda, K., and Block, S. M. (1994). Biological applications of optical forces. Annu. Rev. Biophys. Biomol. Struct. 23, 247–285.

204

Matthew W. C. Chan et al. Syedain, Z. H., Weinberg, J. S., and Tranquillo, R. T. (2008). Cyclic distension of fibrin-based tissue constructs: Evidence of adaptation during growth of engineered connective tissue. Proc. Natl. Acad Sci. USA 105, 6537–6542. Syntichaki, P., and Tavernarakis, N. (2004). Genetic models of mechanotransduction: The nematode Cae norhabditis elegans. Physiol. Rev. 84, 1097–1153. Tarone, G., and Lembo, G. (2003). Molecular interplay between mechanical and humoral signalling in cardiac hypertrophy. Trends Mol. Med. 9, 376–382. Tenney, R. M., and Discher, D. E. (2009). Stem cells, microenvironment mechanics, and growth factor activation. Curr. Opin. Cell Biol. 21, 630–635. Thodeti, C. K., Matthews, B., Ravi, A., Mammoto, A., Ghosh, K., Bracha, A. L., and Ingber, D. E. (2009). TRPV4 channels mediate cyclic strain-induced endothelial cell reorientation through integrin-to-integrin signaling. Circ. Res. 104, 1123–1130. Tomasek, J. J., Gabbiani, G., Hinz, B., Chaponnier, C., and Brown, R. A. (2002). Myofibroblasts and mechano-regulation of connective tissue remodelling. Nat. Rev. Mol. Cell Biol. 3, 349–363. Tomasek, J. J., McRae, J., Owens, G. K., and Haaksma, C. J. (2005). Regulation of alpha-smooth muscle actin expression in granulation tissue myofibroblasts is dependent on the intronic CArG element and the transforming growth factor-beta1 control element. Am. J. Pathol. 166, 1343–1351. Torzilli, P. A., Grigiene, R., Huang, C., Friedman, S. M., Doty, S. B., Boskey, A. L., and Lust, G. (1997). Characterization of cartilage metabolic response to static and dynamic stress using a mechanical explant test system. J. Biomech. 30, 1–9. Tsunozaki, M., and Bautista, D. M. (2009). Mammalian somatosensory mechanotransduction. Curr. Opin. Neurobiol. 19, 362–369. Turner, C. H. (2006). Bone strength: Current concepts. Ann. N Y Acad. Sci. 1068, 429–446. Turner, C. H., and Robling, A. G. (2004). Exercise as an anabolic stimulus for bone. Curr. Pharm. Des. 10, 2629–2641. Tzima, E., Irani-Tehrani, M., Kiosses, W. B.. Dejana, E., Schultz, D. A., Engelhardt, B., Cao, G., DeLisser, H., and Schwartz, M. A. (2005). A mechanosensory complex that mediates the endothelial cell response to fluid shear stress. Nature 437, 426–431. Usami, S., Chen, H. H., Zhao, Y., Chien, S., and Skalak, R. (1993). Design and construction of a linear shear stress flow chamber. Ann. Biomed. Eng. 21, 77–83. van Wamel, J. E., Ruwhof, C., van der Valk-Kokshoorn, E. J., Schrier, P. I., and van der Laarse, A. (2000). Rapid gene transcription induced by stretch in cardiac myocytes and fibroblasts and their paracrine influence on stationary myocytes and fibroblasts. Pflugers Arch. 439, 781–788. Velegol, D., and Lanni, F. (2001). Cell traction forces on soft biomaterials. I. Microrheology of type I collagen gels. Biophys. J. 81, 1786–1792. Vogel, V., and Sheetz, M. (2006). Local force and geometry sensing regulate cell functions. Nat. Rev. Mol. Cell Biol. 7, 265–275. Wang, J. H., and Thampatty, B. P. (2008). Mechanobiology of adult and stem cells. Int. Rev. Cell. Mol. Biol. 271, 301–346. Wang, N., Butler, J. P., and Ingber, D. E. (1993). Mechanotransduction across the cell surface and through the cytoskeleton. Science 260, 1124–1127. Wang, J., Seth, A., and McCulloch, C. A. (2000). Force regulates smooth muscle actin in cardiac fibroblasts. Am. J. Physiol. Heart Circ. Physiol. 279, H2776–H2785. Wang, D., Chang, P. S., Wang, Z., Sutherland, L., Richardson, J. A., Small, E., Krieg, P. A., and Olson, E. N. (2001). Activation of cardiac gene expression by myocardin, a transcriptional cofactor for serum response factor. Cell 105, 851–862. Wang, D. Z., Li, S., Hockemeyer, D., Sutherland, L., Wang, Z., Schratt, G., Richardson, J. A., Nordheim, A., and Olson, E. N. (2002). Potentiation of serum response factor activity by a family of myocardin-related transcription factors. Proc. Natl. Acad. Sci. USA 99, 14855–14860. Wang, J., Su, M., Fan, J., Seth, A., and McCulloch, C. A. (2002). Transcriptional regulation of a contractile gene by mechanical forces applied through integrins in osteoblasts. J. Biol. Chem. 277, 22889–22895. Wang, J., Chen, H., Seth, A., and McCulloch, C. A. (2003). Mechanical force regulation of myofibroblast differentiation in cardiac fibroblasts. Am. J. Physiol. Heart Circ. Physiol. 285, H1871–H1881.

8. Mechanical Induction of Gene Expression in Connective Tissue

205

Wang, Q., Herrera Abreu, M. T., Siminovitch, K., Downey, G. P., and McCulloch, C. A. (2006). Phosphor ylation of SHP-2 regulates interactions between the endoplasmic reticulum and focal adhesions to restrict interleukin-1-induced Ca2þ signaling. J. Biol. Chem. 281, 31093–31105. Warden, S. J., Hurst, J. A., Sanders, M. S., Turner, C. H., Burr, D. B., and Li, J. (2005). Bone adaptation to a mechanical loading program significantly increases skeletal fatigue resistance. J. Bone Miner. Res. 20, 809–816. Weber, K. T., Sun, Y., Tyagi, S. C., and Cleutjens, J. P. (1994). Collagen network of the myocardium: Function, structural remodeling and regulatory mechanisms. J. Mol. Cell Cardiol. 26, 279–292. White, C. R., and Frangos, J. A. (2007). The shear stress of it all: The cell membrane and mechanochemical transduction. Philos. Trans. R. Soc. Lond. B Biol. Sci. 362, 1459–1467. Wilke, A., Funck, R., Rupp, H., and Brilla, C. G. (1996). Effect of the renin-angiotensin-aldosterone system on the cardiac interstitium in heart failure. Basic Res. Cardiol. 91(Suppl. 2), 79–84. Winer, J. P., Janmey, P. A., McCormick, M. E., and Funaki, M. (2009). Bone marrow-derived human mesenchymal stem cells become quiescent on soft substrates but remain responsive to chemical or mechanical stimuli. Tissue Eng. A 15, 147–154. Wipff, P. J., Rifkin, D. B., Meister, J. J., and Hinz, B. (2007). Myofibroblast contraction activates latent TGF beta1 from the extracellular matrix. J. Cell Biol. 179, 1311–1323. Wipff, P. J., Majd, H., Acharya, C., Buscemi, L., Meister, J. J., and Hinz, B. (2009). The covalent attachment of adhesion molecules to silicone membranes for cell stretching applications. Biomaterials 30, 1781–1789. Yamazaki, T., Komuro, I., and Yazaki, Y. (1998). Signalling pathways for cardiac hypertrophy. Cell Signal. 10, 693–698. Yeung, T., Georges, P. C., Flanagan, L. A., Marg, B., Ortiz, M., Funaki, M., Zahir, N., Ming, W., Weaver, V., and Janmey, P. A. (2005). Effects of substrate stiffness on cell morphology, cytoskeletal structure, and adhesion. Cell Motil. Cytoskeleton 60, 24–34. Yin, J., and Kuebler, W. M. Mechanotransduction by TRP channels: General concepts and specific role in the vasculature. Cell Biochem. Biophys. 56, 1–18. Yoshida, T., Sinha, S., Dandre, F., Wamhoff, B. R., Hoofnagle, M. H., Kremer, B. E., Wang, D. Z., Olson, E. N., and Owens, G. K. (2003). Myocardin is a key regulator of CArG-dependent transcription of multiple smooth muscle marker genes. Circ. Res. 92, 856–864. Yousefian, J., Firouzian, F., Shanfeld, J., Ngan, P., Lanese, R., and Davidovitch, Z. (1995). A new experi mental model for studying the response of periodontal ligament cells to hydrostatic pressure. Am. J. Orthod. Dentofacial Orthop. 108, 402–409. Zajaczkowski, M. B., Cukierman, E., Galbraith, C. G., and Yamada, K. M. (2003). Cell-matrix adhesions on poly(vinyl alcohol) hydrogels. Tissue Eng. 9, 525–533. Zhang, D., Weinbaum, S., and Cowin, S. C. (1998). Estimates of the peak pressures in bone pore water. J. Biomech. Eng. 120, 697–703. Zhao, X. H., Laschinger, C., Arora, P., Szaszi, K., Kapus, A., and McCulloch, C. A. (2007). Force activates smooth muscle alpha-actin promoter activity through the Rho signaling pathway. J. Cell Sci. 120, 1801–1809. Zhong, C., Chrzanowska-Wodnicka, M., Brown, J., Shaub, A., Belkin, A. M., and Burridge, K. (1998). Rhomediated contractility exposes a cryptic site in fibronectin and induces fibronectin matrix assembly. J. Cell Biol. 141, 539–551.

CHAPTER 9

Physical Plasticity of the Nucleus and its Manipulation Irena Ivanovska, Joe Swift, Takamasa Harada, J. David Pajerowski, and Dennis E. Discher Biophysical Engineering Laboratory, School of Engineering and Applied Science, University of Pennsylvania, Philadelphia, Pennsylvania 19104

Abstract I. Introduction II. Micropipette Aspiration A. Basic Experimental Method B. Mathematics of Physical Responses III. Molecular Mechanisms from Reengineered Nuclei A. Lamin Knockdown with RNA Interference IV. Isolation of Individual Nuclei A. Bulk Isolation Protocol V. Outlook

Acknowledgments

References

Abstract The genome is virtually identical in all cells within an organism, with epigenetic changes contributing largely to the plasticity in gene expression during both develop ment and aging. These changes include covalent modifications of chromatin compo nents and altered chromatin organization as well as changes in other nuclear components, such as nuclear envelope lamins. Given that DNA in each chromosome is centimeters long and dozens of chromosomes are compacted into a micronsdiameter nucleus through non-trivial interactions with the bounding envelope, the polymer physics of such a structure under stress can be complex but perhaps systema tic. We summarize micromanipulation methods for measuring the physical plasticity of the nucleus, with recent studies documenting the extreme flexibility of human METHODS IN CELL BIOLOGY, VOL. 98 Copyright 2010 Elsevier Inc. All rights reserved.

207

978-0-12-381009-0 DOI: 10.1016/S0091-679X(10)98009-6

208

Irena Ivanovska et al.

embryonic stem cells and the rigidification in model aging of progerin-type nuclei. Lamin-A/C is a common molecular factor, and methods are presented for its knock down and measurement.

I. Introduction The nucleus is generally the largest single organelle of a eukaryotic cell and is literally the cell’s defining feature. Within the nuclei of a given organism, the DNA is essentially identical, but the many differentiated cell types possess epigenetic differences that fix the fates of progenitors and stem cells. DNA methylation, histone isoforms, and histone modifications collectively regulate gene expression but so do nuclear envelope proteins that also change in development. For example, lamin-A/C is largely absent from human embryonic stem cells (Constantinescu et al., 2006), but spliced isoforms accumulate at the nuclear periphery in cells from aged, normal humans (Scaffidi and Misteli, 2006). We hypothesized that such epigenetic plasticity in normal development would be mechani cally measurable and meaningful. A rigid nucleus, for example, would likely imply less accessible genes and a more terminally differentiated cell fate. A methodology was therefore sought that could lend physical insight into nuclear development as well as various physiological and technological processes. For exam ple, blood capillaries at 2–3 µm in diameter are similar in size or smaller than nuclei, which means that nuclei in cells ranging from white blood cells to circulating meta static cancer cells must be able to deform and flow in order to access peripheral tissues. Micropipette aspiration is a standard method to understand the flow and deformation of cells into tubes (Fig. 1). In terms of technology motivations, somatic cell nuclear transfer (SCNT) involves micropipette manipulation of a nucleus from an adult somatic cell for transfer into a denucleated, unfertilized egg (Wakayama et al., 1998). SCNT techniques are notoriously inefficient (Wilmut and Paterson, 2003) and might benefit from a better fundamental understanding of stress effects on nuclei and their substructures.

II. Micropipette Aspiration A. Basic Experimental Method This method has several important features: (1) the length scale of deformation is similar in length scale to nuclear subdomains or “nuclear territories” (LiebermanAiden et al., 2009), (2) the change in nuclear shape under stress can be measured over a wide range of time, and (3) the deformation modes of the different nuclear subcomponents labeled fluorescently (with DNA dyes or GFP-constructs) can be visualized by fluorescence microscopy. With a nucleus that is isolated from a cell (see below), aspiration reveals nuclear responses without the affects of physical links between the nucleus and different components of the cytoskeleton. Solution

209

9. Physical Plasticity of the Nucleus and its Manipulation (A)

(B) 1 min

0.1 min

DNA condensed by histones

Human embryonic stem cell

Lnuc

Relative nuclear deformability Lnuc / LceII

ΔP = const Lcell Nuclear lamins A/C, B

Cell body

Nucleolus (B23)

1.0

E ~ 1–2 kPa

0.8

0.15 + 0.7 e–1/2.7

0.6

0

GM p17, at 200 s

0

2

4

6 days

Tissue fibroblast

AG p24, at 200 s

2.0

10

Creep prefactor, A (kPa–1)

GM p17 AG p17 AG p21 AG p24

1

1.5

1.0

0.5

0.0 1

10

100

GM p17

AG p17

AG p21

AG p24

Creep compliance, J at 300 s (kPa–1)

Compliance of HGPS nuclei

(D) Creep compliance, J(t) (kPa–1)

E ~ 5–10 kPa

0.2

Pluripotent

(C)

Differentiation

0.4

10 8 6 4 2 0 GM p17

AG p17

AG p21

AG p24

Time (s)

**P value < 0.01 in paired t-test * P value < 0.05 in paired t-test

Fig. 1 Micropipette aspiration and deformation of nuclei. (A) Schematic showing the various substructures within the nucleus, including the nuclear lamin with lamin-A/C. (B) Aspiration at constant pressure of an hESC nucleus. The nucleus and cell flow slowly or “creep” into the micropipette. The cell contributes very little resistance, and based on many measurements of tissue fibroblasts, the elasticity prefactor E is about 5–10 kPa for the differentiated nuclei, and so the hESC nucleus is estimated to have a stiffness of about 1–2 kPa. (C) Aspiration of normal (“GM”) and lamin-A/C mutant (progerin, “AG”) fibroblasts that were latrunculin-treated to fluidize the actin cytoskeleton. Wild-type GFP-lamin-A/C is expressed in some aspirated nuclei to visualize the nuclear envelope and nucleoplasmic lamin structures as well as wrinkles and creases during aspiration. (D) The mutant cells were aged in culture (p denotes passage number) and exhibit rigidity associated with an accelerated aging phenotype. Rigidity is evident in creep compliance, including prefactor A, being lower for the aged nuclei.

210

Irena Ivanovska et al.

conditions are readily changed, including both ionic strength and osmotic strength that control nuclear volume and molecular interactions (Dahl et al., 2005). However, measurements made on nuclei within intact cells can in principle provide more relevant in situ insight into the physical behavior of the nucleus provided the cytoskeleton and other cell components do not interfere. This has been achieved when necessary by depolymerizing filamentous actin in cells with latrunculin immediately prior to aspira tion (Pajerowski, 2007). Whether the nucleus is isolated or not, it is partially aspirated into a micropipette with a preset pressure and the increase in the projection length of the nucleus is measured as a function of time. The key steps are as follows: 1. Micropipettes are prepared from a 1-mm-diameter glass capillary by pulling with a micropipette puller and following methods that are standard in electrophysiology. Typically, the inner diameter is between 1 and 5 µm, and this should be constant over at least 10 µm of length in order to visualize the aspirated projection of nucleus. This is achieved by optimization of the pulling rate. 2. The micropipettes are fractured and forged to create a flat tip. This is important to ensure symmetric aspiration and flow of the nucleus into the micropipette. 3. The micropipette inner and/or outer surface can be then passivated to minimize cell or nuclear adhesion and friction. This is achieved by immersing into albumin solutions or silanizing solutions. 4. The micropipette is backfilled with physiological buffer using a syringe with a suitably small gauge needle. 5. The micropipette is mounted on a micromanipulator connected to a pressurecontrolled system in which the applied pressure (negative relative to atmospheric) is measured by a calibrated pressure transducer or manometer. 6. Using the micromanipulators, the micropipette is positioned horizontally within the focal plane of the microscope and close to the nucleus to be aspirated. After the nucleus is slightly aspirated, the micropipette should be raised above the coverslip which the cell or nucleus is settled upon in order to avoid friction between the nucleus and the supportive surface during the aspiration. 7. Negative pressures from –1 to –10 kPa are required for aspiration of the nucleus. Higher pressures should be avoided because they can rupture the nucleus. In the case of measurements of a nucleus within a cell, repetitive aspiration of the cell can be use to mechanically disrupt the cell wall and thus isolate the nucleus (Guilak, 2000). B. Mathematics of Physical Responses 1. The DNA in each of the 46 human chromosomes is a massive macromolecule (1–10 cm long) that is condensed as chromatin into a microns-diameter nucleus and would be expected—as with most polymers—to exhibit complex flow behavior (i.e., complex rheology). While some past studies of nuclear deformation have assumed fully recoverable elastic behavior (Deguchi et al., 2005), others have suggested viscoelastic behavior (Guilak, 2000) or shown more complex power law rheology (Dahl et al. 2005, Pajerowski, 2007).

211

9. Physical Plasticity of the Nucleus and its Manipulation

For small deformations most solid materials can be described by Hooke’s law of linear elasticity in which E is Young’s elastic modulus for the material. The onedimensional relationship between the applied stress and caused strain " is given by ¼ E" The inverse of the modulus E is called compliance:

J¼

1

E

A purely viscous fluid under shear obeys the following relationship between the stress and strain rate: ¼

d" dt

where is the viscosity. Real materials often deviate from pure elastic or viscous behavior and exhibit a more complex and time-dependent stress versus deformation response. If they yield or break after some period of stress and thereafter do not recover their deformation, then they are referred to as “plastic.” Creep is defined as a progressive deformation of a material held under constant stress. If a sudden stress is applied to a material (Fig. 2A), then it can be described with ¼ 0 HðtÞ where H(t) = 0 for t < 0 and H(t) = 1 for t > 0. The compliance might then prove time dependent and is called the creep compliance J ðtÞ ¼

"ðtÞ 0

Creep curves can capture material behaviors over many decades of time, but if the load is released the material might begin to recover. If the recovery never reaches zero strain, then the remaining strain is an indication of the plasticity of the material. Linear viscoelastic materials are those for which the creep compliance is independent of the stress. On the other hand, if the strain is suddenly imposed and held constant, then one can present the strain history as a step function (Fig. 2B) " ¼ "0 HðtÞ A decrease in stress in a material held under constant strain is called relaxation and can be described with relaxation modulus EðtÞ ¼

ðtÞ "0

212

Irena Ivanovska et al.

(B)

(A) Stress

σ0

Cause

Strain

Cause

ε0

t

t

Effect Strain

Effect Stress

Viscoelastic

Elastic Viscoelastic

Elastic Recovery Plastic

Viscous

t

Creep

Relaxation

Plastic

t

Recovery

Fig. 2 Mechanical responses. (A) Constant pressure effects a change in strain, and release of the stress leads to recovery except for a residual plastic strain. (B) Constant strain effects a change in stress, and release of the strain leads to recovery except for a residual plastic stress.

Creep and relaxation function can be experimentally obtained and their mathematical descriptions involve attempts to capture a solid–liquid duality. Differential constitutive models relate the stress and the strain in linear differential equations with constant coefficients by connecting the basic elements—elastic (springs) and viscous (dashpots) in different ways. A classical Maxwell model consists of a spring and dashpot connected in series with a constitutive equation that reads: d" 1 d ¼ ¼ þ dt E dt A spring and dashpot connected in parallel (Kelvin–Voigt model) has the form: d" dt For elements in series, the stresses coincide with the total strain being a sum of the elemental strains. For elements in parallel, the strains coincide and the total stress is a sum of stresses in the individual elements. This approach can be used to design more complicated models by combining different Maxwell and Voigt elements in parallel and/or in series to capture the complex rheological behavior of the biological materials. 2. Complexity in Fractional Differential Models. More complex rheological properties may be expressed in terms of fractional derivatives. An element might obey an equation of the form ¼ E" þ

213

9. Physical Plasticity of the Nucleus and its Manipulation

ðtÞ ¼ E r

d "ðtÞ dt

where E is the stiffness, r is a relaxation time constant, and is a materialdependent parameter can be used to obtain the relaxation function E(t). In such a case, the relaxation function is a power law in time that may be approximated by EðtÞ ¼ Bt n Then using the Laplace transform it can be shown that the creep function is also a power law. J ðtÞ ¼ At n For n = 0 the compliance is independent of time, which is the case for purely elastic response, and for n = 1, the response is purely viscous. For any n’s that are between these extremes, the viscoelastic complexity of the material can be captured. Power-law models describe adequately the creep and relaxation function of many materials with small number of adjustable parameters and over of many decades of time. 3. Nuclear Creep in Micropipette Aspiration. For the particular case of micropipette aspiration of a nucleus that is much larger than the micropipette diameter, the creep of a nucleus into the micropipette can be measured as J ðtÞ ¼

2F 1 DLðtÞ 3 P Rp

where is a geometry-dependent numerical prefactor (~2.1), P is the constant pressure applied by the micropipette, and the strain is given by DL, which is the dynamic aspirated length normalized to the pipette radius Rp. This equation for J(t) provides the instantaneous measurable creep; nuclei could aspirate independent of time or exhibit a more complex power-law response. Micropipette measurements performed on both stem cells and differentiated cells over time scales of about 100 s or more (Fig. 1B and C) tend to exhibit a power-law creep compliance with creep exponents of n ~ 0.2–0.6 (Pajerowski, 2007). After approximately 10 s, the deformation proves irreversible, and this provides clear evidence of the plasticity of the nucleus, that is, a permanent rearrangement of the nucleus and its chromatin. In general, the nuclear compliance of fully differentiated and aged cells (fibroblast and epithelial cells) is demonstrably lower than that in younger cells, including both human hematopoietic stem cells (HSCs) and pluripotent human embryonic stem cells (hESCs). The latter nuclei, over several days in differentiation media, exhibit a sixfold increase in stiffness (Fig. 1B). Visualization of the chromatin in differentiated cells with GFP-histones show that the chromatin is pinned at places to the nuclear envelope; the chromatin therefore extends and flows into the aspirating micropipette. Micropipette aspiration of fibroblast nuclei from patients with an accelerated aging laminopathy further show that, particularly with “aging” by passage in culture, nuclei

214

Irena Ivanovska et al.

rigidify (Fig. 1C and D) with differences in young versus old mean rheological parameters of about 20–30%. The images of GFP-lamin not only illustrate the slightly reduced extension of the aged nucleus into the micropipette after 200 sec but further show how the nuclear envelope folds and wrinkles outside the aspirating micropipette. Detailed molecular studies of nuclei from aged normal donors versus young donors suggest accumulation of a particular deletion isoform of lamin-A at the nuclear envelope in aged cells (Scaffidi and Misteli, 2006). This seems consistent with the fact that lamin-A/C is not expressed in the ESCs nor in HSCs.

III. Molecular Mechanisms from Reengineered Nuclei To clarify the relative contributions of different nuclear components to nuclear plasticity and rheology, a systematic perturbation of specific molecules is required. While B-type lamins appear expressed in all cell types, lamin-A/C varies considerably as mentioned above, suggesting that knockdown of lamin-A/C in a differentiated cell type such as an epithelial cell might test more directly the role of this lamin in nuclear mechanics. Results from micropipette aspiration confirm the hypothesized role for lamin-A/C, as elaborated elsewhere (Pajerowski, 2007). Other nuclear molecules might be approached similarly. A. Lamin Knockdown with RNA Interference RNA interference (RNAi) is a phenomenon whereby double-stranded RNA induces posttranscriptional silencing of target mRNA by catalytic cleavage in a relatively sequencespecific manner. RNAi was first discovered in Caenorhabditis elegans (Fire et al., 1998), and it was later shown that synthetic 21–23 nucleotide RNA termed short interfering RNA (siRNA) could silence in cultured mammalian cells (Elbashir et al., 2001). When siRNA is delivered to the cytoplasm, it is incorporated in a multiprotein complex called the RNAinduced silencing complex (RISC). Once in the RISC, the sense strand of siRNA is cleaved, and the remaining antisense strand directs the complex to the target mRNA. After binding of the siRNA antisense strand to its target, mRNA is catalytically cleaved by one subcomponent of the RISC. This process is call posttranscriptional gene silencing. Since its discovery in the late 1990s, RNAi has been intensively studied not only for understanding function of genes but also for clinical applications (Castanotto and Rossi, 2009). Here we outline the methods of lamin-A/C downregulation using siRNA and the commercially available transfection agent Lipofectamine 2000, with subsequent analysis of knockdown efficiency. We note here that similar methods can be used for transfections of GFP-lamins and other nuclear proteins (Pajerowski, 2007).

1. siRNA Complex for Transfection Different types of transfection reagents are commercially available. Lipofectamine 2000 (LF2k) is a cationic lipid-based nanosized particle that complexes with nucleic acids (lipoplex) through electrostatic interactions when mixed in an appropriate buffer.

9. Physical Plasticity of the Nucleus and its Manipulation

215

Excess cationic charge of the complex also promotes cell uptake by binding to the netnegative cell membrane. Cargo release into cytoplasm occurs, it is thought, via an osmotic rupture of endolysosomes. Preparation of LF2k/siRNA complex follows the protocol provided by Invitrogen (San Diego, CA, USA). Dilute LF2k and siRNA (sequence) in the same volume of Opti-MEM (Invitrogen): normally 20 µg/ml LF2k in 50 µl and siRNA 20-fold of desired concentration in 50 µl. • Incubation for 15 min at room temperature. • Mix two solutions in one and incubate for 15 min at room temperature.

2. Transfection of Adherent Epithelial Cells A549 lung epithelial cells are seeded 24 h prior to transfection. High-glucose DMEM (Invitrogen) supplemented with 10% FBS is used for cell culture and transfection. Before adding the siRNA complex, cells are washed with DPBS (Invitrogen) once and provided with fresh medium. Complex solution was added such that complex/medium = 1/10 in volume. Cells are incubated 72 h in 37°C humidified chamber with 5% CO2.

3. Immunostaining for Lamin-A/C Visualization This protocol is standard immunostaining as provided by Abcam (http://www. abcam.com/). • The A549 cells are seeded in 12-well plates (Corning) with 10,000 cells per well 24 h prior to transfection. • After transfection is done, fix cells with 3.7% formaldehyde (Fisher Scientific) in DPBS. Incubate for 15 min at room temperature. • Wash cells with ice-cold DPBS twice. • Permeabilize cells with 0.25% Triton X-100 (MP Biomedicals) in DPBS. Incubate for 10 min at room temperature. • Incubate cells with DPBS for 5 min at room temperature 3 times. • Incubate cells with 1% bovine serum albumin (BSA, Sigma-Aldrich) in phosphatebuffered saline (PBS) with 0.05% Tween-20 (Fisher Scientific). Incubation time is 30 min in 37°C humidified chamber with 5% CO2. • Dilute primary antibody against lamin-A/C (mouse monoclonal IgG, Santa Cruz Biotech) 200 times in DPBS with 1% BSA and 0.25% Tween-20 (concentration of antibody is 1 µg/ml). Add 400 µl of 200 antibody solution to each well and incubate ether 1 h in 37°C humidified chamber with 5% CO2 or overnight at 4°C. • After incubation with primary antibody, incubate cells with DPBS at room temperature for 5 min. Repeat this wash step 3 times. • Dilute Alexa Fluor 488-conjugated donkey-derived antimouse polyclonal antibody (Invitrogen) 200 times in DPBS with 1% BSA and 0.25% Tween-20 (concentration of antibody is 10 µg/ml). Incubate cells with secondary antibody (400 µl in each well) for 1 h in 37°C humidified chamber with 5% CO2. 10 min before finishing

216

Irena Ivanovska et al.

incubation, add 4 µl of 50 times-diluted Hoechst 33342 dye (Invitrogen) suspended in DPBS. • After incubation with secondary antibody and Hoechst 33342, incubate cells with DPBS at room temperature for 5 min. Repeat this step 3 times. • Cells are imaged by epifluorescence microscopy (Olympus IX71) using a 40 objective lens. • Image collection on any sensitive charge-coupled device camera under constant image settings for knockdown and control cells can be used to quantify the relatively lower intensity of lamin-A/C in the knockdown cells. DNA staining with Hoechst can be used to also determine the DNA intensity in any given nucleus as a lamin measurement, and then a scatterplot of intensities can be made: lamin versus DNA. The knockdown nuclei should be a scattering of points below the nontransfected cells. Typical knockdown levels are 50% or more, with considerable cell-to cell variation.

4. Western Blotting for Lamin-A/C Knockdown Standard protocols are also available (http://www.abcam.com/). • 106 cells were seeded in 60-mm cell culture dish (Falcon) 24 h prior to transfection. Total volume of medium and sample is 2 ml. • After 72 h, detach the cells by 1.5 ml of trypsin with 0.05% EDTA (Invitrogen). Wash cells with ice-cold DPBS twice (centrifugation is done with 5000 rpm for 5 min at 4°C). • After removing DPBS, add 250 µl of NP40 lysis buffer (1% NP40 þ 50 mM Tris Base þ 150 mM NaCl) with 1% protease inhibitor (Sigma-Aldrich). Incubate on ice for 30 min with occasional vortex. • During the incubation, sonicate sample for 15 s. • Centrifuge the sample with 12,000 rpm for 20 min at 4°C and collect supernatant. • For 200 µl lysate, add 66 µl of Laemmli buffer (Invitrogen) and 6 µl of b-mercaptoethanol. Heat the sample at 80–90°C for 5–10 min. • Prior to the electrophoresis, determine the total protein concentration of the lysate using BCA Protein Assay Kit (Pierce). • Load sample with the volume based on the aimed amount of total protein per each well. Run time is 10 min with 100 V and 65 min with 160 V. It is best to run several different protein concentrations from both knockdown and control for later analysis. • Blot protein on PVDF membrane using iBlot system (Invitrogen). • Block membrane with 10% nonfat dry milk (American Analytical) in TTBS (4.6 mM Tris Base, 15.4 mM Tris HCl, 154 mM NaCl and 0.1% Tween-20) on a rocker at room temperature for 1 h. • Wash membrane with TTBS once for 15 min and twice for 5 min on rocker. • Cut the membrane such that it includes approximately 60–80 kDa proteins. Incubate the membrane in a solution of 400-fold diluted primary antibody against lamin-A/C (mouse-derived, Santa Cruz Biotechnology) in TTBS. When using bactin as a loading control, cut out the membrane between 40 and 50 kDa and incubate it in 1000-fold diluted solution of b-actin primary antibody (mouse

9. Physical Plasticity of the Nucleus and its Manipulation

217

derived, Santa Cruz Biotechnology). Incubation can be done ether at room temperature for 1–2 h or at 4°C overnight. • Wash membrane with TTBS once for 15 min and twice for 5 min on rocker. • Dilute antimouse HRP-conjugated IgG (GE Healthcare) 2000 times in TTBS with 5% dry milk. Incubate membranes in secondary antibody solution with rocking at room temperature for 1 h. • Wash membrane with TTBS once for 15 min and twice for 5 min, and then with TBS (4.6 mM Tris Base, 15.4 mM Tris HCl, 154 mM NaCl) on rocker. • Develop with Chromosensor (GenScript) for 2–3 min at room temperature. • Scan developed membrane and analyze by densitometry using Image J. A plot of immunostaining intensity versus protein load should fit to a line, and the slope of the knockdown should be lower than that of the control cells.

IV. Isolation of Individual Nuclei Although manipulation results cited above have appeared largely consistent between in-cell nuclei and isolated nuclei (Dahl et al., 2005), whenever there is concern that the cell is contributing unduly to the apparent nuclear properties, nuclear isolation should be attempted. Spheroidal nuclei from at least some cell types can be isolated by both single-cell mechanical extraction (Guilak, 2000) and bulk methods (Caille et al., 2002; Dahl et al., 2005; Deguchi et al., 2005). Extremely fragile nuclei have not yet been isolated successfully, and since the nuclear envelope breaks down during mitosis, the methods apply only to mechanically stable interphase nuclei. Isolation also takes advantage of the nucleus’ relative rigidity and its tenuous connections to the rest of the cell. One detailed protocol for bulk isolation, described in detail previously (Dean and Kasamatsu, 1994) and shown to yield nuclei suited to micromechanical characteriza tion (Dahl et al., 2005), is briefly described below. It has the advantage that further biochemical analyses such as proteomic profiling can be applied to the nuclei (Andersen et al., 2005; Black et al., 2007), perhaps even under stressed conditions (Johnson et al., 2007). In general, the cell’s plasma membrane is disrupted mechanically with hypotonic swelling or chemically with digitonin or other surfactants that perturb the plasma mem brane but not nuclear membranes. The cell is then opened with mechanical homogeniza tion, and cellular contents are separated from the nuclei by ultracentrifugation through a sucrose gradient.

A. Bulk Isolation Protocol 1. Stock solutions • “10 TKMC”: 500 mM Tris, 250 mM KCl, 25 mM MgCl2, 30 mM CaCl2; adjust to approximately pH 8. • “2.3 STKMC”: dilute 10-fold, 2.3 M sucrose, adjust to pH 7.6 at 4°C.

218

Irena Ivanovska et al.

• “5 STKMC”: dilute 2-fold, 1.25 M sucrose, adjust to pH 7.6 at 4°C. • 10 mM HEPES, pH 7.5.

Prepare on the day of use (cool to 4°C):

• 200 µl 5 STKMC, 2.5 µl protease inhibitor cocktail (PIC). • 3 ml 2.3 STKMC, 7.5 µl PIC. • Take 900 µl of 2.3 STKMC and add BSA to make 0.2 mg/ml (“2.3 STKMC–BSA”). • 10 ml of 10 mM HEPES, pH 7.5 with 1 mM DTT (dithiothreitol). 2. Harvesting nuclei • Cells should be nearing confluency: 105 cells minimum, 108 maximum. Wash the T75 flask with 5 ml cold PBS. Repeat. • Wash once with 5 ml cold HEPES/DTT buffer. • Scrape cells and make volume up to 1 ml with HEPES/DTT buffer. • Cool on ice for 10 min. • 25 strokes in Dounce Homogenizer (avoiding air bubbles). • Take 900 µl of homogenized cells, add 180 µl 5 STKMC. • Cool on ice for 10 min. • Add 2 ml 2.3 STKMC; ensure solution is well mixed (sucrose concentration is now 1.6 M). • Pipette 200 µl 2.3 STKMC–BSA into each of four ultracentrifuge tubes

(polycarbonate 8 34 mm2).

• Layer 750 µl cell lysate into each tube (the interface between the solutions will be visible). • Spin for 1 h at 50,000 rpm in the TLS 55 rotor at 4°C. The nuclei should pellet at the bottom, other cell material should collect at the gradient interface. • Remove the supernatant and resuspend the pellet. 3. At this point nuclei can be resuspended into any media, although adding BSA helps break up clumps and facilitates manipulation later. Nuclei can be counted relatively accurately on a hemacytometer. Subsequent to isolation, it is important to assess the quality of the nuclei by comparison with nuclei in intact cells, and to make sure that the nuclear envelope remains intact while avoiding excess membranes such as the endoplasmic reticulum or cytoskeletal structures. DNA stains such as Hoechst dyes or, better, GFP-lamins allow fluorescence visualization for assessment of nuclear morphology. 4. Imaging of isolated nuclei • Resuspend nuclei in 50 µl 10 mM Tris with 10,000:1 diluted Hoescht stain. • Incubate at 37°C for 10 min. • Centrifuge to pellet the nuclei, then pipette away the supernatant. • Resuspend in 50 µl 10 mM Tris with 500:1 diluted phalloidin–rhodamine stain. • Incubate at 37°C for 30 min. • Centrifuge to pellet, pipette away supernatant, resuspend in 50 µl 10 mM Tris. • Examine nuclei by fluorescence microscopy, including immunostaining for lamin-A/C as outlined above.

9. Physical Plasticity of the Nucleus and its Manipulation

219

5. Lastly, we note that the concentrations of salts can dramatically affect nuclear mechanics since divalent and, to a lesser extent, monovalent cations affect chromatin condensation (Aaronson, 1981; Dahl et al., 2005). With changes in salt, nuclear volume changes significantly and may induce mechanical stress on the nuclear lamina. It is unclear exactly what effect these salt concentrations have on protein–protein interactions within the nucleus. Nuclei isolated in buffers with low salt tend to swell and resemble closely the contours and size of nuclei inside live cells. To accurately approximate nuclear mechanics inside the cell, it is usually desirable to mimic the intracellular ion concentrations, but the difficulty is that the ion concentrations inside the cytoplasm and within organelles have not been reproducibly determined, and values for ions such as calcium range from submillimolar to millimolar and can vary greatly as a function of disease (Dobi and Agoston, 1998). It is also difficult to predict the shift in nuclear salt homeostasis after isolation from the cell. Some studies have deliberately examined mechanics of isolated nuclei at extreme salt conditions to determine the maximum possible range of mechanical responses (Fig. 2). This strategy proves effective since the chromatin condensation and dilation appears to shift the load within the nucleus from the chromatin to the lamina (Dahl et al., 2005).

V. Outlook In outlining our methods used to understand the physical plasticity of nuclei in cell development and aging, we have focused in particular on lamin-A/C in the nuclear envelope. This is primarily because lamin-A/C is well known to undergo changes in normal stem cells thru aged cells, coupling to other epige netic changes. What our physical measurements generally show is that nuclei flow and yield, like a plastic, beyond about 10 s of stress or strain, and that lamin-A/C at the envelope contributes in part to nuclear rigidification, as evident in a decreased creep compliance. It is attractive to think that a more flexible and fluid nucleus is a more functional nucleus.

Acknowledgments HFSP (Human Frontier Science Program), NSF (including the Nano Bio Interface Center), and NIH (R01 HL062352, EB007049, R21 AR056128, and P01 DK032094) are gratefully acknowledged for support.

References Aaronson, R. P., and Woo, E. (1981). Organization in the cell-nucleus—divalent-cations modulate the distribution of condensed and diffuse chromatin. J. Cell Biol. 90(1), 181–186. Andersen, J. S., et al. (2005). Nucleolar proteome dynamics. Nature 433(7021), 77–83.

220

Irena Ivanovska et al. Black, B. E., et al. (2007). An epigenetic mark generated by the incorporation of CENP-A into centromeric nucleosomes. Proc. Natl. Acad. Sci. USA 104(12), 5008–5013. Caille, N., et al. (2002). Contribution of the nucleus to the mechanical properties of endothelial cells. J. Biomech. 35(2), 177–187. Castanotto, D., and Rossi, J. J. (2009). The promises and pitfalls of RNA-interference-based therapeutics. Nature 457(7228), 426–433. Constantinescu, D., Gray, H. L., Sammak, P. J., Schatten, G. P., and Csoka, A. B. (2006). Lamin A/C expression is a marker of mouse and human embryonic stem cell differentiation. Stem Cells 24, 177–185. Dahl, K.N., et al. (2005). Power-law rheology of isolated nuclei with deformation mapping of nuclear substructures. Biophys. J. 89(4), 2855–2864. Dean, D. A., and Kasamatsu, H. (1994). Signal- and energy-dependent nuclear transport of SV40 Vp3 by isolated nuclei. Establishment of a filtration assay for nuclear protein import. J. Biol. Chem. 269(7), 4910–4916. Deguchi, S., et al. (2005). Flow-induced hardening of endothelial nucleus as an intracellular stress-bearing organelle. J. Biomech. 38(9), 1751–1759. Dobi, A., and Agoston, D. (1998). Submillimolar levels of calcium regulates DNA structure at the dinucleo tide repeat (TG/AC)n. Proc. Natl. Acad. Sci. USA 95, 5981–5986. Elbashir, S. M., Harborth, J., et al. (2001). Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 411(6836), 494–498. Fire, A., Xu, S. Q., et al. (1998). Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391(6669), 806–811. Guilak, F. (2000). The deformation behavior and viscoelastic properties of chondrocytes in articular cartilage. Biorheology 37(1–2), 27–44. Johnson, C. et al. (2007). Forced unfolding of proteins within cells. Science 317(5838), 663–666 Lamond (2009). http://www.lamondlab.com/f7nucleolarprotocol.htm. Lieberman-Aiden, E., et al. (2009). Comprehensive mapping of long-range interactions reveals folding principles of the human genome. Science 326(5950), 289–293. Pajerowski, D. et al. Physical plasticity of the nucleus in stem cell differentiation. Proceedings of the National Academy of Science - USA (2007) 104: 15619–15624. Rowat, A. C.; et al. (2005). Characterization of the elastic properties of the nuclear envelope. J. R. Soc. Interface 2(2), 63–69. Scaffidi, P., and Misteli, T. (2006). Lamin A-dependent nuclear defects in human aging. Science 312(5776), 1059–1063. Wakayama, T., Perry, A.C., Zuccotti, M., Johnson, K. R., and Yanagimachi, R. (1998). Full-term develop ment of mice from enucleated oocytes injected with cumulus cell nuclei. Nature 394(6691), 369–374. Wilmut, I., and Paterson, L. (2003). Somatic cell nuclear transfer. Oncol. Res. 13(6–10), 303–307.

CHAPTER 10

Prestressed Nuclear Organization in Living Cells Aprotim Mazumder*, T. Roopa*, Abhishek Kumar*,†, K. Venkatesan Iyer*,†, Nisha M. Ramdas*, and G.V. Shivashankar*,† * National Centre for Biological Sciences, Tata Institute of Fundamental Research, Bellary Road, Bangalore 560065, India † Department of Biological Sciences and Research Center for Excellence in MechanoBiology, National University of Singapore, Singapore 117543

Abstract I. Introduction II. Mechanics of Isolated Nucleus A. Isolation of Nucleus from Living Cells and Study of Higher-Order Chromatin Organization B. Measurement of Entropic Force during Chromatin Decompaction C. Optical Trap: A Method to Probe Softening of Nucleus During Chromatin Decompaction D. Photo-Bleaching and Nuclear Swelling Techniques to Probe Chromatin Organization III. Nuclear Prestress in Cellular Context A. Chemical Inhibitors to Modulate Cytoskeletal Tension B. Laser Perturbation of Differentially Compacted Chromatin IV. Conclusions

Acknowledgments

References

Abstract The nucleus is maintained in a prestressed state within eukaryotic cells, stabilized mechanically by chromatin structure and other nuclear components on its inside, and cytoskeletal components on its outside. Nuclear architecture is emerging to be critical METHODS IN CELL BIOLOGY, VOL. 98 Copyright 2010 Elsevier Inc. All rights reserved.

221

978-0-12-381009-0 DOI: 10.1016/S0091-679X(10)98010-2

222

Aprotim Mazumder et al.

to the governance of chromatin assembly, regulation of genome function and cellular homeostasis. Elucidating the prestressed organization of the nucleus is thus important to understand how the nuclear architecture impinges on its function. In this chapter, various chemical and mechanical methods have been described to probe the pre stressed organization of the nucleus.

I. Introduction Chromatin, the primary constituent of nucleus, is a complex of nucleic acids and proteins packaging a meter-long DNA into a micron-scale nucleus. The interaction of histone and nonhistone proteins with DNA facilitates the condensation of DNA far beyond its radius of gyration of 220 µm defined by the entropic regime. In addition, DNA is condensed into euchromatin and heterochromatin regions enabling regulated access to genetic information. Maintenance of this state of the nucleus may require a prestressed nuclear organization with contribution from factors within the nucleus and from the cytoskeleton, but the principles of this organization are still unclear. Impor tantly, nuclear structure is emerging to be critical to the governance of chromatin assembly, regulation of genome function and cellular homeostasis. Elucidating the prestressed organization of the nucleus is thus essential to understand how the nuclear architecture impinges on its function. In eukaryotic cells, nucleus is the stiffest organelle (Dahl et al., 2008). In this context, tensegrity models have been proposed for cellular shape stability that depends on tension and compression in the cytoskeleton and physical integration of the nucleus with cytoplasm (Ingber, 1993; 2003a, b). A quantitative understanding of prestressed nucleus may also elucidate mechanisms underlying mechanotransduction of signals from extra-cellular matrix (ECM) that impinge on genome function. In this chapter, we outline some methods to probe the components involved in maintenance of nuclear shape and size.

II. Mechanics of Isolated Nucleus The prestressed organization of nucleus arises from both nuclear and cytoskeletal components. Hence it is pertinent to decouple these components and explore indivi dually their roles on nuclear prestress. Isolation of a nucleus out of its cytoplasmic milieu provides a mechanism to assess stresses on the nucleus directly. In this section, we present experimental methods for isolating nuclei from cells, and study the forces that work to maintain the prestressed state of the nucleus by enzymatic disruption of higher-order chromatin assembly. A. Isolation of Nucleus from Living Cells and Study of Higher-Order Chromatin Organization The nucleus of eukaryotic cell is 5- to 10-fold more rigid than the cytoplasm with variations depending on the cell type. Taking advantage of this property of the nucleus,

10. Prestressed Nuclear Organization in Living Cells

223

mechanical and chemical methods have been employed to isolate individual nuclei from living cells (Caille et al., 2002; Dahl et al., 2005). In general, nuclear isolation techniques require the selective mechanical or chemical perturbation of the cell mem brane and loss of cytoplasmic contents. This method retains the integrity of nuclear membrane and chromatin organization inside the nucleus. Osmotic swelling and mechanical shearing techniques have been employed in our laboratory to isolate nucleus exploiting the considerable difference in the mechanical properties of cell and nucleus. A simple method for the isolation of nucleus involves the following steps: 1. Cultured cells are harvested and washed using 1 PBS (pH 7.4). 2. Cells are re-suspended in TM2 buffer (10 mM Tris–HCl, pH 7.4, 2 mM MgCl2, and 0.5 mM PMSF (added fresh before use)). 3. Cells are incubated for 5 min on ice and 5 min at room temperature. 4. To disintegrate the cytoplasm of cells, Triton X-100 is added and mixed well, before the cells are again incubated on ice for 5 min. 5. Cells are sheared by passing them through a syringe needle (22 gauge) 10 times and centrifuged at 12,000 rpm for 5 min. 6. Triton X-100 treatment is repeated if the nuclei were found to have cellular debris adhered to them. 7. Clean isolated nuclei are used for experiments. The integrity of nuclear membrane and chromatin assembly can be assessed by staining the nucleus with DNA dyes like Hoechst 33342 and DAPI. These nuclei can then be further used for studying either their mechanical properties or prestress involved in the isolated nucleus. The nucleus devoid of any cytoplasmic milieu is still a stable structure. In the entropic regime a meter-scale genome acquires a configuration which has a radius of gyration of 220 µm. But higher-order chromatin organization involving DNA–protein interactions in isolated nucleus forces it to stabilize and restrain from going to this entropic config uration (Krajewski and Ausio, 1996; Leuba et al., 1998). Perturbation of chromatin structure provides a mechanism to investigate forces involved within the prestressed nucleus. It has been known that histone tail–tail interactions are important for compac tion of the chromatin (Bertin et al., 2004; Placek and Gloss, 2002; Schalch et al., 2005). Probing the effect of perturbing proteins involved in higher-order chromatin assembly on integrity of the nucleus is crucial to understand mechanical stresses within the nucleus. Here, we describe proteolytic-cleavage-based methods (using the enzymes clostripain and trypsin) to probe the importance of higher-order chromatin structure in maintenance of nuclear prestress. Trypsin and clostripain are a class of proteases that cause cleavage of arginine residues starting at carboxyl terminal of the proteins (Dumuis-Kervabon et al., 1986); while trypsin also acts on lysine residues. We have used trypsin and clostripain to perturb higher-order chromatin compaction by digestion of histone tails and to observe responses of nucleus to such structural perturba tion (Mazumder et al., 2008). Isolated nuclei are adhered on coverslips using poly-D lysine (PDL) following which clostripain is added in concentrations ranging from 0.5 to 4 mU/ml. The enzyme needs to be activated by 2.5 mM DTT and calcium acetate at room

224

Aprotim Mazumder et al.

(A)

0′′

407′′

815′′

1391′′

Clostripain

DNase

Nuclear area (μm2)

(B) 2000

1000

0 0

30 Time (min)

Fig. 1 Chemical perturbation of isolated nucleus to study higher-order chromatin compaction. (A) The panel of images shows the entropic swelling of an isolated H2B-EGFP nucleus digested with 0.5 mU/μl clostripain and 500 mU/μl DNase. No swelling is observed in nucleus for cleavage of nucleic acids. At high DNase concentrations (500 mU/μl), fragmented bits of chromatin are emitted from the nuclei though there is no expansion in size. The corresponding time-points are indicated. Scale bar = 10 μm. (B) Expansion kinetics for clostripain and DNase (filled circle is for clostripain and open square is for DNase).

temperature for 30 min. The action of the activated enzyme leads to an increase in isolated nuclear size which can be quantified by computing the cross-sectional area obtained from confocal images of nucleus. However, in contrast when chromatin is perturbed by nucleic acid targeted enzymes such as DNases, nucleus does not expand, instead fragmented chromatin is observed (Fig. 1). Quantification shows that the nuclear expansion increases significantly when protein interactions maintaining higher-order chromatin structure are violated resulting in an entropic swelling of chromatin. This assay can further be utilized to probe the role of nuclear scaffold proteins in the maintenance of chromatin assembly. Highly compacted heterochromatin regions of chromatin are organized along the nuclear periphery and are known to be anchored at distinct foci to the lamin matrix. The lamin scaffold via a host of proteins is involved in organizing the genome within the nucleus (Goldman et al., 2004; Gruenbaum et al., 2005; Mattout et al., 2006; Panorchan et al., 2004; Tzur et al., 2006). Perturbation of proteins that stabilize chromatin may also affect tethering of chromatin to the lamin scaffold. Cells are transiently transfected with EGFP-

10. Prestressed Nuclear Organization in Living Cells

225

LaminB1 fusion plasmid and studied following the addition of trypsin. Dynamics of expansion of lamin scaffold studied by fluorescence time-lapse imaging shows an increase in the perimeter of nucleus leading to complete nuclear disintegration marked by a loss of lamin fluorescence intensity when the nucleus ruptures. Thus, these proteolytic-cleavage-based methods allow investigation of the role of higher-order chro matin structure in the maintenance of nuclear prestress.

B. Measurement of Entropic Force during Chromatin Decompaction Various experiments have been performed to understand the mechanical properties of nucleus (Guilak et al., 2000; Ofek et al., 2009; Tseng et al., 2004). Micropipette aspiration experiments have been employed to reveal elastic modulus and viscosity of the nucleus (Pajerowski et al., 2007; Rowat et al., 2006; Vaziri and Mofrad, 2007). Here, we describe a method to measure entropic force generated by chromatin when the nuclear prestress is removed by chemical perturbation of higher-order chromatin organization. Atomic Force Microscope (AFM) has been effectively used to study forces stabilizing nanoscale structures and force fluctuations (Bao and Suresh, 2003; Milani et al., 2009). AFM cantilever sensitivity to small deflections enables them to be used for measuring minute displacements generated by the mechanics of nuclear expansion. In order to measure the entropic expansion of nucleus, we used an AFM cantilever (Veeco Instru ments Inc., NY) of stiffness kcant = 0.02 N/m which is mounted on -SNOM Microscope (WiTEC, Germany). The cantilever is employed to measure deflections on a nucleus adhered on a coverslip dish coated with PDL. A saturating concentration of trypsin can be used to hasten the process of nuclear swelling. Deflection of the cantilever is constantly monitored using a quadrant photodiode (QPD). A change in the position of the cantilever is manifested as a differential change in the voltages of four quadrants of the QPD, where voltages are precalibrated against defined displacements of the cantilever. When the enzyme starts to act, the cantilever shows an upward movement, indicat ing swelling of the nucleus. The pressure on the cantilever can be tracked as a function of time (Fig. 2A). Quantification reveals a large increase in outward pressure 3 kPa or force of 300 nN before the nuclear membrane and lamin scaffold disintegrates. This is the force that the entropic chromatin exerts on the nuclear membrane upon digestion of protein interactions. After the rupture of nuclear membrane, the cantilever recoils back to its initial position indicating a decrease in pressure owing to the absence of nuclear membrane to confine the expanding chromatin (Fig. 2A). This assay demonstrates the use of AFM to probe chromatin structure.

C. Optical Trap: A Method to Probe Softening of Nucleus During Chromatin Decompaction Perturbation to the structural components of nucleus, like deletion of Lamin A/C results in a decrease in the rigidity of nucleus as observed in micropipette aspiration experiments (Pajerowski et al., 2007). Chromatin and lamin scaffold are known to be key determinants of the structural integrity of the nucleus. In this section, we describe a method using an

226

Aprotim Mazumder et al.

Pressure (kPa)

(A) 3

2

Quadrant photodetector

Initial direction of cantilever deflection on trypain digestion of the nucleus

920 nm laser

k = 0.02 N/m

1

0 102

105

20

Change in δx (nm)

(B)

10

0 0

10

20

Time (min)

Fig. 2 Quantification of mechanical property of an isolated nuclei. (A) Pressure exerted by nucleus on the cantilever is estimated from deflection and stiffness of the cantilever. Upon trypsin digestion the pressure due to prestress is released and a large increase in the pressure on cantilever is observed followed by a dip due to rupture of the nuclear envelope. Inset shows schematic of the experimental set up. (B) Fluctuation of a bead adhered on the nuclear membrane increases with time upon trypsin digestion (open circles) suggesting softening of the nucleus. No increase in fluctuation is observed when the nucleus is fixed with paraformaldehyde (open squares).

optical trap to measure the change in mechanical properties of isolated nucleus upon perturbation of higher-order chromatin organization by enzymatic digestion. Optical traps have been used to trap dielectric beads and thereby study various biological processes (Sheetz, 1998). Fluctuations of beads trapped in an optical trap can be interpreted to elucidate the structural properties of underlying substrate (Hodges et al., 2009; Soni et al., 2003). Our laboratory has quantified the fluctuations of bead in a trap to study the stiffness of chromatin fiber and additionally the nucleus. The optical trap we used is built on an inverted fluorescence microscope (Model: IX70, Olympus, Tokyo, Japan) using a current controlled diode laser (wavelength 830 nm, GaAlAs;

227

10. Prestressed Nuclear Organization in Living Cells

SDL Inc, San Jose, CA) and controller ILX Lightwave, Bozeman, MT). In order to track position of the bead in the trap, a 5 mW, 635 nm diode laser (Coherent Inc., Santa Clara, CA) is employed to image the back-scattered light onto a photodetector. The output voltage of the photodetector is amplified using two low noise current pream plifier (Model SR570; Stanford Research Systems, Sunnyvale, CA). The difference in the two voltages provides the position of the bead from the center of the optical trap. Data acquisition and analysis are done using DAQ (PCI-MIO-16XE-10; National Instruments, Austin, TX) and LabVIEW (National Instruments, TX). In vitro experiments have been employed to study the mechanical properties of single chromatin fibers (Claudet and Bednar, 2006; Cui and Bustamante, 2000; Dame, 2008). To understand the mechanistic effect of the loss of compaction by enzymatic digestion of histone tails on stiffness of chromatin and to estimate local chromatin fluidity, we have used the optical trap assay in combination with micromanipulation methods (Roopa and Shivashankar, 2006). The quantification of chromatin stiffness rests on the assumption of treating the chromatin-trap as a system of springs in parallel. For the system used, the effective position standard deviation (PSD) of the bead can be calculated by the following equation: 1 2 trap

¼

1 2trap

þ

1 2Chr

where, Chr is PSD of the bead due to chromatin and trap is the PSD of the bead due to the trap. A PDL-coated micropipette kept at a fixed tension is used to pull a chromatin fiber from purified chromatin adhered onto a coverslip by PDL. The fluctuations of a bead which is adhered onto the chromatin fiber are monitored as a function of time, and studied in comparison to fluctuations that result after chromatin structure is modulated via the addition of trypsin. A histogram of fluctuations after enzymatic digestion of histone tails plotted as a function of time can be used to probe the rigidity of the underlying substrate quantified by the width of the histogram or the dispersion of fluctuations. Softer substrates show a larger dispersion than rigid substrates. This method can further be used to estimate the fluidity of the cell nucleus upon perturbation of the chromatin structure (Mazumder et al., 2008). A 2 µm bead is adhered on the membrane of isolated nucleus and the rigidity of the nucleus is estimated from fluctuations of the bead, i.e., standard deviation of its fluctuations (). The structural response of the nucleus to enzymatic perturbation is obtained by following of the bead on treatment with the enzyme. The bead shows 10 nm before the nucleus is digested. Upon trypsin digestion of the nucleus, increased to 30 nm which is comparable to the 50 nm of an unbound bead in the optical trap suggesting that trypsin digestion indeed leads to softening of nuclear material (Fig. 2B). D. Photo-Bleaching and Nuclear Swelling Techniques to Probe Chromatin Organization Apart from the inward force on chromatin due to histone tail–tail interactions, as described in earlier section, there exists a repulsive electrostatic force between DNA segments owing to the presence of a net negative charge (Tkachenko, 2006). DNA inside

228

Aprotim Mazumder et al.

the nucleus is in a bath of positively charged ions that screen the repulsions between the negatively charged DNA segments (Fenley et al., 1994; Taubes et al., 2005). Removal of these screening charges by altering the buffer conditions leads to an expansion of the nucleus, owing to repulsion between DNA segments. Such an expansion of the nucleus is reversible upon restoration of the buffer conditions. This phenomenon of reversibility of the nucleus can be employed as a method to probe the structural integrity of chromatin organization. Isolated nuclei are adhered onto PDL-coated coverslip dishes, stained with a DNA binding dye Hoechst or DAPI and imaged in 1 Phosphate Buffered Saline (PBS) which represents the physiological condition for the isolated nucleus. Nuclear size is then cyclically modulated by replacement of PBS buffer with water and restoration of PBS. Nuclear area increases when PBS is replaced by water and recovers completely upon restoration of buffer conditions (Fig. 3). Merging the images obtained before and after restoration of PBS buffer shows pixel-wise colocalization suggesting structural integrity of chromatin anchored within the nucleus. Further, this technique can be used to probe the effect of screening charges on the chromosome organization by titration of the buffer concentration, providing a handle on understanding the dynamics involved in the structural maintenance of chromatin assembly.

III. Nuclear Prestress in Cellular Context Within the cellular context, the eukaryotic nucleus is maintained in a prestressed condition by balance of both nuclear and cytoplasmic forces (Mazumder and Shivashankar, 2007). Apart from the forces in an isolated nucleus which stabilizes the prestressed state, cytoplasmic forces play an important role in establishing prestress in adherent cells as well as governing various functions (Janmey, 1998; Stossel, 1993). The physical links between cytoplasm and focal adhesion (FA) on one side and cytoskeleton and the nucleus on the other side are beginning to be elucidated (Ingber, 1993; 2008; Ingber et al., 1994, 1995). FAs or focal contacts are the subcellular sites where the cell contacts the ECM. FAs are points of cross-talk between trans-membrane integrin receptors and cytoplasmic filaments. Additionally, at these loci a large number of receptors are present, thus making FA a key site for various biochemical and mechan otransduction pathways (Vogel and Sheetz, 2006; Wang et al., 1993). FAs are major mechano-sensors present at the plasma membrane of the eukaryotic, mainly adherent cells (Geiger et al., 2001; Gillespie and Walker, 2001; Hamill and Martinac, 2001). These are dynamic structures and change their size and morphology in response to physical forces (Liu et al., 2010). The heterodimeric (–) trans-membrane integrin receptors interact with various anchor proteins like talin, -actinin, and tensin which either directly make connections with cytoplasmic filament actin or are mediated through other adaptor proteins like vinculin (Calderwood et al., 2003; Garcia-Alvarez et al., 2003; Geiger and Bershadsky, 2001; Jamora and Fuchs, 2002; Liu et al., 2000). By interacting with the ECM and cytoplasmic filaments, these provide a rigid structural support to the cellular structure. In cytoplasmic milieu, microtubule applies compressive load on the nucleus while actomyosin complex applies tensile force on the nuclear

229

10. Prestressed Nuclear Organization in Living Cells

Before

H2O

After

5 μm

5 μm

Fig. 3

Reversible swelling of isolated nuclei in water. Isolated H2B-EGFP HeLa nuclei in PBS buffer is washed four times in deionized water, to show large swelling. The swelling is reversed on restoring back PBS, to regain the same configuration as before. No loss of histones is apparent. Individual nuclei are imaged with photobleaching marks to discern local length scales of swelling. Every node comes back to the original configuration on restoration of physiological buffer conditions, as is apparent from the merge image (bottom).

membrane. These filaments make connections with the nuclear envelope mediated by various anchor proteins mainly SUN and KASH domain proteins (Crisp et al., 2006; Haque et al., 2006; Padmakumar et al., 2005; Tzur et al., 2006). It has now been shown that apart from diverse function in nuclear positioning and centrosome localization, SUN

230

Aprotim Mazumder et al.

(Sad1 and UNC-84 homology domain) along with its partner KASH (Klarsicht, ANC-1 and SYNE1 homology) may play an important role as mechanical couplers (Haque et al., 2006; Tzur et al., 2006). These act like bridges connecting most of the cytoplasmic filaments to nucleus (Stewart et al., 2007). The nucleus which consists mainly of chromatin and other nuclear proteins is held together by nuclear envelope (Gruenbaum et al., 2005). Further, inner and outer nuclear membranes join at nuclear pore complexes, the entry-exit site for various proteins. Also, lamins, an intermediate filament, which is connected to nuclear envelope via various anchor proteins and to chromatin mainly at heterochromatin sites via heterochromatin binding proteins like HP1a, provide a struc tural integrity to the nucleus (Georgatos and Blobel, 1987; Haithcock et al., 2005; Houben et al., 2007; Makatsori et al., 2004; Mattout et al., 2006; Nelson et al., 1986; Pajerowski et al., 2007; Panorchan et al., 2004; Shumaker et al., 2003). The size of cell nucleus reflects both total chromatin content and the balance of nuclear and cytoplasmic forces at the nuclear envelope. In addition, the size of eukaryotic nucleus is extraordi narily variable not only during different cell cycle stage but also between different cell types derived from various tissues of the same organism. In the following sections, we shall briefly describe chemical and physical methods to probe the cytoplasmic contribu tion toward a prestressed nuclear organization. A. Chemical Inhibitors to Modulate Cytoskeletal Tension In order to estimate contribution of the cytoskeleton toward prestress of nucleus, comparative measurements of size of the nucleus can be made in presence and absence of cytoskeletal filaments. Studies have explored the use of chemical reagents to perturb nucleo-skeletal and cytoskeletal balance of forces that define cellular integrity and thereby modulate nuclear prestress. A number of drugs are known to chemically perturb the cytoskeleton. Cytochalasin D is known to inhibit actin polymerization (Nair et al., 2008), while Latrunculin depolymerizes actin and Nocodozole has been shown to perturb micro tubule organization. Chemical disruption of these cytoskeletal filaments illustrate the force balance that exists within the cytoskeleton meshwork with resulting altered change to nuclear size. While different concentrations of these drugs have been utilized, a protocol for their concentrations to aid chemical perturbation of cytoskeleton is described as follows: 1. 2. 3. 4.

For For For For

de-polymerizing microtubules, use nocodazole at 1 µg/ml for 16 h. de-polymerizing actin, use cytochalasin D at 1 µM for 2.5 h. inhibiting myosin II, use blebbistatin at 5 µM for 2.5 h. inhibiting kinesin Eg5, use monastrol at 125 µM for 16 h.

The area of the nucleus is measured upon treatment with cytoskeletal inhibitors to estimate the contribution of each component to maintain the prestressed state (Fig. 4A and B). An increase in the nuclear area upon Nocodazole treatment suggests the role of microtubule as a compressive load bearer, whereas a decrease in the nuclear area upon Cytochalasin D or Blebbistatin treatment indicates the role of actin and myosin in providing outward tensile force on nucleus to maintain the prestressed state. Mild perturbation of actin cytoskeleton at low concentration of Cytochalasin D has revealed the role of a structural

231

10. Prestressed Nuclear Organization in Living Cells

(A) Control

(B)

Noc

CytoD

Monst

Blebb

140000

Area (μm2)

120000

* 300

**

** 200

**

il)

G

en

om

e

(e

nt

ro

pi

c

co

ab l et

st

eb b

H

Bl

on M

oD yt

oc C

N

C

on

t

100

Fig. 4 Chemical deploymerization of cytoplasmic filaments or inhibition of associated motor proteins cause a variation in nuclear size in PMEF cells. (A) Representative images of drug-treated and fixed nuclei with the DNAstained with Hoechst. Scale bar = 20 mm. (B) Statistics for 100 nuclei each. The error-bars are standard errors. Also shown are the size that nuclei shrink under a heterochromatin ablation (17 cells), and the estimated hydrodynamic radius of the genome in these cells. ‘*’ implies p < 0.05 and ‘**’ implies p < 0.001.

component of the actin cytoskeleton that forms a cap above the apical surface of the nucleus that modulates prestress. Disruption of this actin cap results in rounding up of nuclei (Khatau et al., 2009). The structural correlation of cell shape with nuclear shape and dependence of nuclear size on adherence is illustrated by the interplay between anchorage with cell and nuclear shape, with loss of anchorage resulting in nuclear retraction and rounding. Intermediate filaments form a continuous mesh from attachment points at the cell surface to the nuclear envelope (Herrmann et al., 2007), and the role of intermediate filaments as mechanical integrators and tensile stiffeners has been suggested more than a decade ago (Hollenbeck et al., 1989). However, their role on cellular strength and integrity requires more investigation. B. Laser Perturbation of Differentially Compacted Chromatin Though pharmacological perturbation of cytoskeletal filaments reveals their roles in maintaining cellular and nuclear prestress, this method lacks specificity in structural

232

Aprotim Mazumder et al.

perturbations. Further this does not indicate the regions in the nucleus and cytoplasm which are critical structural regions of the cell. To probe this aspect of the cellular structure, the technique of laser induced ablation is described in this section. The ablation of physical structure via laser by local heating can efficiently be employed to perturb structural components and thereby explore their contribution to cellular pres tress of the nucleus (Berns, 2007; Quinto-Su and Venugopalan, 2007). The incorporation of fluorescently tagged histone protein stably expressed in the cells of interest can be used as a method to address structures in distinct chromatin regions. The identification and comparative study of loosely packaged euchromatin and tightly com pacted heterochromatin is obtained by the differential fluorescence intensity of the core histone-tagged nucleus. Local ablation of cellular and subcellular structure within the cell is obtained by gold-nanoparticle-mediated laser perturbation. Incorporation of these particles into cells can be accomplished via methods of microinjection or hypotonic shock to cells (Mazumder and Shivashankar, 2007). Earlier studies have addressed issues such as the biocompatibility, low cytotoxicity, and method of endocytic incorporation of gold nano particles into cells (Shukla et al., 2005). Thus gold-nanoparticle incorporation can be employed as a mechanism to result in localized heating and disruption to structure. A brief protocol for incorporation of gold nanoparticle into living cells is provided below: 1. Cells pregrown on dishes are incubated in media supplemented with gold particles for 1 h to ensure the presence of the particle in the endocytosed fluid. 2. Cells are washed with PBS pH 7.4 prior to being subjected to a hypotonic shock for 3 min at 37°C to allow for incorporation of gold nanoparticles. 3. Cells are gently washed and allowed to recover for 3–4 h before experimental use. 4. Near-infrared radiation focused at specific location within the cell is employed with diffraction resolution. 5. Gold-nanoparticle-incorporated cells suffer local heating by exposure to pulsed Ti-sapphire laser mode locked at 835 nm employing laser power of 56 mW for a period of 3 s. Alternatively, since gold particles serve merely to ensure efficient local absorption of near infrared radiation, ablation without use of gold particles can achieve similar results but by employing instead higher laser powers (120–140 mW). Laser heating at powers smaller than those used results only in local photo-bleaching. Selective laser ablation of differentially compacted chromatin regions within the nucleus viz. euchromatin and heterochromatin has revealed their contributive roles toward the maintenance of nuclear prestress (Fig. 5A). Perturbation of dense heterochromatin regions results in decrease in nuclear area and volume, while perturbation of comparatively less dense euchromatin regions result only in marginal shrinkage (Fig. 5B). Analysis of shrinkage dynamics of heterochromatin ablated nucleus show mean decrease in nuclear volume by 35 ± 11% in comparison to unperturbed nuclei. This observed nuclear shrink age upon ablation reveals the role of heterochromatin as structural nodes for cytoskeleton attachment to the nucleus. The decrease in anisotropy of nuclear shape illustrates the role of outward cytoskeletal tension, as well as loss of force balance between nucleoskeletal

233

10. Prestressed Nuclear Organization in Living Cells

(A) 0′′

Before Het

18′′

71′′

232′′

419′′

5 μm

Eu

0.0

–0.2 –0.4 –0.6

Het

Eu

(C) Normalized area

Fractional Δarea

(B)

1.0 τ ~ 152 s

Het Eu

0.8 0.6 0

200 Time (s)

400

Fig. 5 Representative images of the differential response of PMEFs to heterochromatin (Het) and euchromatin (Eu) perturbations. H1e-EGFP transfected PMEF cells are ablated at the heterochromatin at indicated time-points, using a 6.8 s exposure of a 1.5 µm diameter region to a pulsed titanium sapphire laser (80 mW at a fixed spot). Experiments are done 24 h posttransfection. The ablation points are shown by the white arrows, and time-points in seconds from the ablation are indicated above the images. Nuclei showed a fall in size in response to heterochromatin ablation, while such an effect was not present upon euchromatin ablation. (B) Fractional change in area (n = 17 each) and (C) average time-trace upon heterochromatin ablation (Het) and euchromatin ablation (Eu) in PMEF cells transfected with H1e-EGFP (n = 9 each). Scale bar = 5 µm.

and cytoskeletal components. Dynamics of nuclear collapse on heterochromatin ablation reveals an initial lag phase followed by a sudden decay in nuclear volume (Fig. 5C). The reason for this lag may probably be due to the time required for the disruption of cytoskeleton filaments. Figure 6 shows the disruption of various cytoskeletal structures upon laser ablation as visualized using immunostaining methods. During mitosis majority of chromatin is highly condensed. This condensation of the nucleus is reflected at the scale of chromatin organization. Drosophila SR2þ cells have a small number of chromosomes and hence such condensation can be more tractably visualized. To test the effect of release of nuclear prestress on chromosome condensa tion, laser ablation of the nuclear envelope was performed in Drosophila SR2þ cells (Fig. 7A). These experiments resulted in a condensation of the chromatin into indivi dual chromosome territories, as envisaged by a decrease in nuclear volume and a concomitant rise in average pixel intensity (Fig. 7B). This is reminiscent of an important natural state of cell division, where there is a large condensation of nuclear volume, immediately following nuclear envelope breakdown. In this case, nuclear envelope breakdown is caused by invaginating microtubules at prophase (Beaudouin

234

Aprotim Mazumder et al.

(A) EGFP LaminB1

BA

PA

(E) Tau-EGFP

(B) H1e-mRFP

BA

PA

(F) Anti-α-tubulin

BA

PA

Normal (C) Actin-EGFP

PA

BA

Ablated

(G) Anti-vimentin

Normal

(D) H2B-EGFP

Actin

(H) Paxillin

Ablated

Merge

Fig. 6 State of the various nucleo-cytoskeletal components on heterochromatin perturbation. HeLa cells are cotransfected with (A) EGFP-Lamin B1 and (B) H1e-mRFP. Images before (BA) and after (PA) perturbation are presented for the same cell. (C) Images of HeLa cells transiently transfected with ActinEGFP before and after heterochromatin perturbation. Heterochromatin is identified by Hoechst staining of the DNA in these cases (images not shown). (D) H2B-EGFP expressing HeLa cells are perturbed at the heterochromatin, fixed and stained with an antibody against paxillin (visualized by a Cy5-labled secondary antibody), and filamentous actin bound by an Alexa-568-labeled phalloidin. Individual and merged images are presented. (E) Images of HeLa cells transiently transfected with Tau-EGFP as a microtubule-marker before and after heterochromatin perturbation. As in (C), heterochromatin is identified by Hoechst staining of the DNA. Images of H2B-EGFP expressing HeLa cells perturbed at the heterochromatin, and stained with primary antibodies against (F) -tubulin and (G) vimentin on separate plates. A Cy3-labeled secondary antibody is used. Normal and perturbed cells are imaged on the same plate. Green indicates H2B-EGFP fluorescence, while red shows the respective cytoplasmic filaments. Scale bars = 5 µm in (A), (B), (C), and (E). Scale bars = 10 µm in (D), (F), and (G). (See Plate no. 5 in the Color Plate Section.)

235

10. Prestressed Nuclear Organization in Living Cells

(A) 0′′

(B)

18′′

48′′

77′′

120

160 140 Area (μm2)

308′′

100

120 80

100 80

Intensity

Before

60

60 40

40 0

50

100 150 200 250 300 Time (s)

Fig. 7 Condensation of chromatin upon envelope ablation. (A) Representative images of condensation of individual chromatin territories in SR2þ cells. Time in seconds is indicated on top of the panel. Scale bar = 5 μm. (B) Fall in area of the confocal slice (filled squares), and corresponding rise in fluorescence intensity (open circles), both indicating condensation.

et al., 2002; Panorchan et al., 2004). The methods described in this section provide mechanical and chemical tools to probe the role of the cytoskeletal filament structure on nuclear prestress.

IV. Conclusions The prestressed state of nucleus has been illustrated by probing their structural organization across a variety of cellular systems. Employing both mechanical and chemical perturbation of cytoskeletal filament and nuclear structure, a prestressed state of the cell nucleus is studied. In Fig. 8 a schematic of this prestress is shown. On one scale, histone tail–tail interactions results in condensing DNA into a micrometer size while on the other regime entropic forces drives this assembly to several hundred microns given its polymeric nature. Balancing these two scale results in an equilibrium size of a few micron-sized nucleus. As depicted in the schematic, in interphase cell the nucleus is further prestressed by variety of nucleo-cytoplasmic skeleton. This pre stressed dynamic equilibrium can be modulated by mechanical perturbations to hetero chromatin nodes, as well as chemical alterations to nuclear histone tail–tail interactions and cytoskeletal assemblies. The nucleus is thus a part of an integrated network that spans across cytoskeletal filament structure to the plasma membrane which emerges

236

Aprotim Mazumder et al.

Entropic configuration

Isolated nucleus

Condensation due to histone tail−tail Interactions

~10 μm

~1 μm

~200 μm

Nucleus in cellular context (~20 μm)

Fig. 8 Schematic of the force balance that stabilizes chromatin assembly in an intact isolated nucleus and in cellular context. The outward entropic force due to confinement of DNA polymer to a small volume is countered by an inward condensing force due to histones and other nonhistone proteins of nucleus. These forces act even in an isolated nucleus. In a cellular context, equilibrium of forces is further modified by compressive forces due to microtubules and an outward tension due to actomyosin cytoskeleton.

through the process of cellular differentiation and development. Further, as cells are subject to various physical cues from their environment, mechanotransduction of such signals to the nucleus requires an understanding of the prestressed organization of the nucleus and its modulation (Wang et al., 2009). Cells potentially can exploit this to regulate expression level of genes responding to such mechanotransduction. Acknowledgements We thank the Nanoscience Initiative of Department of Science and Technology (DST) for funding and the NCBS Common Imaging and Flow Facility (CIFF). AK and KVI thank Council for Scientific and Industrial Research (CSIR) for their graduate research fellowships.

References Bao, G., and Suresh, S. (2003). Cell and molecular mechanics of biological materials. Nat. Mater. 2(11), 715–725. Beaudouin, J., Gerlich, D., Daigle, N., Eils, R., and Ellenberg, J. (2002). Nuclear envelope breakdown proceeds by microtubule-induced tearing of the lamina. Cell 108(1), 83–96.

10. Prestressed Nuclear Organization in Living Cells

237

Berns, M. W. (2007). A history of laser scissors (microbeams). Methods Cell Biol. 82, 1–58. Bertin, A., Leforestier, A., Durand, D., and Livolant, F. (2004). Role of histone tails in the conformation and interactions of nucleosome core particles. Biochemistry 43(16), 4773–4780. Caille, N., Thoumine, O., Tardy, Y., and Meister, J. J. (2002). Contribution of the nucleus to the mechanical properties of endothelial cells. J. Biomech. 35(2), 177–187. Calderwood, D. A., Fujioka, Y., de Pereda, J. M., Garcia-Alvarez, B., Nakamoto, T., Margolis, B., McGlade, C. J., Liddington, R. C., and Ginsberg, M. H. (2003). Integrin beta cytoplasmic domain interactions with phosphotyrosine-binding domains: A structural prototype for diversity in integrin signaling. Proc. Natl. Acad. Sci. USA 100(5), 2272–2277. Claudet, C., and Bednar, J. (2006). Pulling the chromatin. Eur. Phys. J. E Soft Matter 19(3), 331–337. Crisp, M., Liu, Q., Roux, K., Rattner, J. B., Shanahan, C., Burke, B., Stahl, P. D., and Hodzic, D. (2006). Coupling of the nucleus and cytoplasm: Role of the LINC complex. J. Cell Biol. 172(1), 41–53. Cui, Y., and Bustamante, C. (2000). Pulling a single chromatin fiber reveals the forces that maintain its higher-order structure. Proc. Natl. Acad. Sci. USA 97(1), 127–132. Dahl, K. N., Engler, A. J., Pajerowski, J. D., and Discher, D. E. (2005). Power-law rheology of isolated nuclei with deformation mapping of nuclear substructures. Biophys. J. 89(4), 2855–2864. Dahl, K. N., Ribeiro, A. J., and Lammerding, J. (2008). Nuclear shape, mechanics, and mechanotransduction. Circ. Res. 102(11), 1307–1318. Dame, R. T. (2008). Single-molecule micromanipulation studies of DNA and architectural proteins. Biochem. Soc. Trans. 36(Pt 4), 732–737. Dumuis-Kervabon, A., Encontre, I., Etienne, G., Jauregui-Adell, J., Mery, J., Mesnier, D., and Parello, J. (1986). A chromatin core particle obtained by selective cleavage of histones by clostripain. EMBO J. 5(7), 1735–1742. Fenley, M. O., Olson, W. K., Tobias, I., and Manning, G. S. (1994). Electrostatic effects in short superhelical DNA. Biophys. Chem. 50(3), 255–271. Garcia-Alvarez, B., de Pereda, J. M., Calderwood, D. A., Ulmer, T. S., Critchley, D., Campbell, I. D., Ginsberg, M. H., and Liddington, R. C. (2003). Structural determinants of integrin recognition by talin. Mol. Cell 11(1), 49–58. Geiger, B., and Bershadsky, A. (2001). Assembly and mechanosensory function of focal contacts. Curr. Opin. Cell Biol. 13(5), 584–592. Geiger, B., Bershadsky, A., Pankov, R., and Yamada, K. M. (2001). Transmembrane crosstalk between the extracellular matrix–cytoskeleton crosstalk. Nat. Rev. Mol. Cell Biol. 2(11), 793–805. Georgatos, S. D., and Blobel, G. (1987). Lamin B constitutes an intermediate filament attachment site at the nuclear envelope. J. Cell Biol. 105(1), 117–125. Gillespie, P. G., and Walker, R. G. (2001). Molecular basis of mechanosensory transduction. Nature 413(6852), 194–202. Goldman, R. D., Shumaker, D. K., Erdos, M. R., Eriksson, M., Goldman, A. E., Gordon, L. B., Gruenbaum, Y., Khuon, S., Mendez, M., Varga, R., and Collins, F. S. (2004). Accumulation of mutant lamin A causes progressive changes in nuclear architecture in Hutchinson-Gilford progeria syndrome. Proc. Natl. Acad. Sci. USA 101(24), 8963–8968. Gruenbaum, Y., Margalit, A., Goldman, R. D., Shumaker, D. K., and Wilson, K. L. (2005). The nuclear lamina comes of age. Nat. Rev. Mol. Cell Biol. 6(1), 21–31. Guilak, F., Tedrow, J. R., and Burgkart, R. (2000). Viscoelastic properties of the cell nucleus. Biochem. Biophys. Res. Commun. 269(3), 781–786. Haithcock, E., Dayani, Y., Neufeld, E., Zahand, A. J., Feinstein, N., Mattout, A., Gruenbaum, Y., and Liu, J. (2005). Age-related changes of nuclear architecture in Caenorhabditis elegans. Proc. Natl. Acad. Sci. USA 102(46), 16690–16695. Hamill, O. P., and Martinac, B. (2001). Molecular basis of mechanotransduction in living cells. Physiol. Rev. 81(2), 685–740. Haque, F., Lloyd, D. J., Smallwood, D. T., Dent, C. L., Shanahan, C. M., Fry, A. M., Trembath, R. C., and Shackleton, S. (2006). SUN1 interacts with nuclear lamin A and cytoplasmic nesprins to provide a physical connection between the nuclear lamina and the cytoskeleton. Mol. Cell Biol. 26(10), 3738–3751.

238

Aprotim Mazumder et al. Herrmann, H., Bar, H., Kreplak, L., Strelkov, S. V., and Aebi, U. (2007). Intermediate filaments: From cell architecture to nanomechanics. Nat. Rev. Mol. Cell Biol. 8(7), 562–573. Hodges, C., Bintu, L., Lubkowska, L., Kashlev, M., and Bustamante, C. (2009). Nucleosomal fluctuations govern the transcription dynamics of RNA polymerase II. Science 325(5940), 626–628. Hollenbeck, P. J., Bershadsky, A. D., Pletjushkina, O. Y., Tint, I. S., and Vasiliev, J. M. (1989). Intermediate filament collapse is an ATP-dependent and actin-dependent process. J. Cell Sci. 92(Pt 4), 621–631. Houben, F., Ramaekers, F. C., Snoeckx, L. H., and Broers, J. L. (2007). Role of nuclear lamina-cytoskeleton interactions in the maintenance of cellular strength. Biochim. Biophys. Acta. 1773(5). Ingber, D. E. (1993). Cellular tensegrity: Defining new rules of biological design that govern the cytoske leton. J. Cell Sci. 104(Pt 3), 613–627. Ingber, D. E. (2003a). Tensegrity I. Cell structure and hierarchical systems biology. J. Cell Sci. 116(Pt 7), 1157–1173. Ingber, D. E. (2003b). Tensegrity II. How structural networks influence cellular information processing networks. J. Cell Sci. 116(Pt 8), 1397–1408. Ingber, D. E. (2008). Tensegrity-based mechanosensing from macro to micro. Prog. Biophys. Mol. Biol. 97(2–3), 163–179. Ingber, D. E., Dike, L., Hansen, L., Karp, S., Liley, H., Maniotis, A., McNamee, H., Mooney, D., Plopper, G., Sims, J., et al. (1994). Cellular tensegrity: Exploring how mechanical changes in the cytoskeleton regulate cell growth, migration, and tissue pattern during morphogenesis. Int. Rev. Cytol. 150, 173–224. Ingber, D. E., Prusty, D., Sun, Z., Betensky, H., and Wang, N. (1995). Cell shape, cytoskeletal mechanics, and cell cycle control in angiogenesis. J. Biomech. 28(12), 1471–1484. Jamora, C., and Fuchs, E. (2002). Intercellular adhesion, signalling and the cytoskeleton. Nat. Cell Biol. 4(4), E101–E108. Janmey, P. A. (1998). The cytoskeleton and cell signaling: Component localization and mechanical coupling. Physiol. Rev. 78(3), 763–781. Khatau, S. B., Hale, C. M., Stewart-Hutchinson, P. J., Patel, M. S., Stewart, C. L., Searson, P. C., Hodzic, D., and Wirtz, D. (2009). A perinuclear actin cap regulates nuclear shape. Proc. Natl. Acad. Sci. USA 106(45), 19017–19022. Krajewski, W. A., and Ausio, J. (1996). Modulation of the higher-order folding of chromatin by deletion of histone H3 and H4 terminal domains. Biochem. J. 316(Pt 2), 395–400. Leuba, S. H., Bustamante, C., van Holde, K., and Zlatanova, J. (1998). Linker histone tails and N-tails of histone H3 are redundant: Scanning force microscopy studies of reconstituted fibers. Biophys. J. 74(6), 2830–2839. Liu, S., Calderwood, D. A., and Ginsberg, M. H. (2000). Integrin cytoplasmic domain-binding proteins. J. Cell Sci. 113(Pt 20), 3563–3571. Liu, Z., Tan, J. L., Cohen, D. M., Yang, M. T., Sniadecki, N. J., Ruiz, S. A., Nelson, C. M., and Chen, C. S. (2010). Mechanical tugging force regulates the size of cell-cell junctions. Proc. Natl. Acad. Sci. USA. 107, 9944–9949. Makatsori, D., Kourmouli, N., Polioudaki, H., Shultz, L. D., McLean, K., Theodoropoulos, P. A., Singh, P. B., and Georgatos, S. D. (2004). The inner nuclear membrane protein lamin B receptor forms distinct microdomains and links epigenetically marked chromatin to the nuclear envelope. J. Biol. Chem. 279(24), 25567–25573. Mattout, A., Dechat, T., Adam, S. A., Goldman, R. D., and Gruenbaum, Y. (2006). Nuclear lamins, diseases and aging. Curr. Opin. Cell Biol. 18(3), 335–341. Mazumder, A., and Shivashankar, G. V. (2007). Gold-nanoparticle-assisted laser perturbation of chromatin assembly reveals unusual aspects of nuclear architecture within living cells. Biophys. J. 93(6), 2209–2216. Mazumder, A., Roopa, T., Basu, A., Mahadevan, L., and Shivashankar, G. V. (2008). Dynamics of chromatin decondensation reveals the structural integrity of a mechanically prestressed nucleus. Biophys. J. 95(6), 3028–3035. Milani, P., Chevereau, G., Vaillant, C., Audit, B., Haftek-Terreau, Z., Marilley, M., Bouvet, P., Argoul, F., and Arneodo, A. (2009). Nucleosome positioning by genomic excluding-energy barriers. Proc. Natl. Acad. Sci. USA 106(52), 22257–22262.

10. Prestressed Nuclear Organization in Living Cells

239

Nair, U. B., Joel, P. B., Wan, Q., Lowey, S., Rould, M. A., and Trybus, K. M. (2008). Crystal structures of monomeric actin bound to cytochalasin D. J. Mol. Biol. 384(4), 848–864. Nelson, W. G., Pienta, K. J., Barrack, E. R., and Coffey, D. S. (1986). The role of the nuclear matrix in the organization and function of DNA. Annu. Rev. Biophys. Biophys. Chem. 15, 457–475. Ofek, G., Natoli, R. M., and Athanasiou, K. A. (2009). In situ mechanical properties of the chondrocyte cytoplasm and nucleus. J. Biomech. 42(7), 873–877. Padmakumar, V. C., Libotte, T., Lu, W., Zaim, H., Abraham, S., Noegel, A. A., Gotzmann, J., Foisner, R., and Karakesisoglou, I. (2005). The inner nuclear membrane protein Sun1 mediates the anchorage of Nesprin-2 to the nuclear envelope. J. Cell. Sci. 118(Pt 15), 3419–3430. Pajerowski, J. D., Dahl, K. N., Zhong, F. L., Sammak, P. J., and Discher, D. E. (2007). Physical plasticity of the nucleus in stem cell differentiation. Proc. Natl. Acad. Sci. USA 104(40), 15619–15624. Panorchan, P., Schafer, B. W., Wirtz, D., and Tseng, Y. (2004). Nuclear envelope breakdown requires overcoming the mechanical integrity of the nuclear lamina. J. Biol. Chem. 279(42), 43462–43467. Placek, B. J., and Gloss, L. M. (2002). The N-terminal tails of the H2A-H2B histones affect dimer structure and stability. Biochemistry 41(50), 14960–14968. Quinto-Su, P. A., and Venugopalan, V. (2007). Mechanisms of laser cellular microsurgery. Methods Cell Biol. 82, 113–151. Roopa, T., and Shivashankar, G. V. (2006). Direct measurement of local chromatin fluidity using optical trap modulation force spectroscopy. Biophys. J. 91(12), 4632–4637. Rowat, A. C., Lammerding, J., and Ipsen, J. H. (2006). Mechanical properties of the cell nucleus and the effect of emerin deficiency. Biophys. J. 91(12), 4649–4664. Schalch, T., Duda, S., Sargent, D. F., and Richmond, T. J. (2005). X-ray structure of a tetranucleosome and its implications for the chromatin fibre. Nature 436(7047), 138–141. Sheetz, M. P. (1998). Laser tweezers in cell biology. Introduction. Methods Cell. Biol. 55, xi–xii. Shukla, R., Bansal, V., Chaudhary, M., Basu, A., Bhonde, R. R., and Sastry, M. (2005). Biocompatibility of gold nanoparticles and their endocytotic fate inside the cellular compartment: A microscopic overview. Langmuir 21(23), 10644–10654. Shumaker, D. K., Kuczmarski, E. R., and Goldman, R. D. (2003). The nucleoskeleton: Lamins and actin are major players in essential nuclear functions. Curr. Opin. Cell Biol. 15(3), 358–366. Soni, G. V., Ali, B. M., Hatwalne, Y., and Shivashankar, G. V. (2003). Single particle tracking of correlated bacterial dynamics. Biophys. J. 84(4), 2634–2637. Stewart, C. L., Roux, K. J., and Burke, B. (2007). Blurring the boundary: The nuclear envelope extends its reach. Science 318(5855), 1408–1412. Stossel, T. P. (1993). On the crawling of animal cells. Science 260(5111), 1086–1094. Taubes, C. H., Mohanty, U., and Chu, S. (2005). Ion atmosphere around nucleic acid. J. Phys. Chem. B 109 (45), 21267–21272. Tkachenko, A. V. (2006). Electrostatic effects in DNA stretching. Phys. Rev. E Stat. Nonlinear Soft Matter Phys. 74(4 Pt 1), 041801. Tseng, Y., Lee, J. S., Kole, T. P., Jiang, I., and Wirtz, D. (2004). Micro-organization and visco-elasticity of the interphase nucleus revealed by particle nanotracking. J. Cell Sci. 117(Pt 10), 2159–2167. Tzur, Y. B., Wilson, K. L., and Gruenbaum, Y. (2006). SUN-domain proteins: “Velcro” that links the nucleoskeleton to the cytoskeleton. Nat. Rev. Mol. Cell Biol. 7(10), 782–788. Vaziri, A., and Mofrad, M. R. (2007). Mechanics and deformation of the nucleus in micropipette aspiration experiment. J. Biomech. 40(9), 2053–2062. Vogel, V., and Sheetz, M. (2006). Local force and geometry sensing regulate cell functions. Nat. Rev. Mol. Cell. Biol. 7(4), 265–275. Wang, N., Butler, J. P., and Ingber, D. E. (1993). Mechanotransduction across the cell surface and through the cytoskeleton. Science 260(5111), 1124–1127. Wang, N., Tytell, J. D., and Ingber, D. E. (2009). Mechanotransduction at a distance: Mechanically coupling the extracellular matrix with the nucleus. Nat. Rev. Mol. Cell Biol. 10(1), 75–82.

CHAPTER 11

Nanotopography/Mechanical Induction of Stem-Cell Differentiation Benjamin Kim Kiat Teo*,†, Soneela Ankam*, Lesley Y. Chan,*,§,{ and Evelyn K. F. Yim*,†,‡ *

Division of Bioengineering, National University of Singapore, Singapore Mechanobiology Institute Singapore, National University of Singapore, Singapore ‡ Department of Surgery, National University of Singapore, Singapore § NUS Graduate School of Integrative Science and Engineering, National University of Singapore, Singapore { Bioprocessing Technology Institute, A*Star, Singapore †

Abstract I. Introduction II. Types of Topological Patterns A. Physical Topography: Surface Texture B. Nanofabrication Techniques C. Chemical Topology: Protein Patterning D. Geometry III. Stem Cell Reception to Substrate Topology A. Embryonic Stem Cells B. Mesenchymal Stem Cells C. Neural Stem Cells/Neural Progenitor Cells D. Hematopoietic Stem/Progenitor Cells E. Other Progenitors IV. Cell Shape: A Regulator of Biological Processes A. Mechanotransduction: A Direct Mechanism Too Simple B. Integrins and Focal Adhesions: Inside Out and Outside In C. Cytoskeleton: A Manifestation of Cells’ Interpretation of Topography D. Nucleus: Mechanical Manipulation of Gene Regulation V. Conclusions

References

Abstract The interplay of biophysical and biochemical cues in the extracellular microenvir onment regulate and control the cell fate of stem cells. Understanding the interaction METHODS IN CELL BIOLOGY, VOL. 98 Copyright 2010 Elsevier Inc. All rights reserved.

241

978-0-12-381009-0

DOI: 10.1016/S0091-679X(10)98011-4

Benjamin Kim Kiat Teo et al.

242

between stem cells and the extracellular substrate will be crucial in controlling stem cell differentiation for regenerative medicine applications. One of the biophysical properties of the microenvironment is substrate topology, which has been demon strated to be an important mediator of stem cell lineage regulation. Biomimetic microenvironment topology can be engineered by chemical patterning or physical patterning. The rapid advancements in nanofabrication techniques have enabled ver satility in patterning types with controlled chemistries, geometries and sizes. The chapter will focus on discussing the effect on physical nanotopography on stem cell differentiation and the current theories on the topography/ mechanical force induction of stem cell differentiation possibly through integrin clustering, focal adhesion, cytos keleton organization and the nuclear mechanosensing to sense and integrate these biophysical signals from the extracellular microenvironment.

I. Introduction When stem cells are removed from the in vivo stem cell niche, they can differentiate spontaneously in vitro but this differentiation process is inefficient, is uncontrolled, and often results in highly heterogeneous cell population (Ding and Schultz, 2004). A crucial strategy of regenerative medicine is understanding how to control the microenvironment surrounding the cells to restore the niche equilibrium is crucial to control stem cell fate. A typical strategy is to enrich the biochemical environment in the in vitro culture medium with a combination of soluble growth factors, cytokines, and/or serum protein, in order to induce the stem cells to differentiate preferentially into a particular lineage. In addition to enhancing the soluble biochemical signal, gaining a fundamental understanding of the cell–substratum interaction will be required to understand and reconstruct the stem cell niche. While the importance of substrate interaction and topography may vary for different cell lineages, its relevance is increasingly unquestionable. In order to study the cell–substrate interaction, the substrate topology can be mimicked by chemical patterning and/or physical patterning. Chemical patterning refers to the patterning of adhesive and/or nonadhesive chemical to direct or restrict cell growth; meanwhile, physical patterning refers to fabrication of surface material with a depth using micro- or nanofabrication techniques. In this chapter, the emphasis will be on the physical nanotopography on stem-cell differentia tion, although studies about chemical patterning will also be discussed.

II. Types of Topological Patterns A. Physical Topography: Surface Texture During natural tissue development, cells are interacting with various nanoscaled topographical and biochemical cues in their microenvironment (Abrams et al., 2000). An example of the in vivo substratum with nanoscale topography is the

11. Nanotopography/Induced of Stem-Cell Differentiation

243

basement membrane, which is a ubiquitous component of extracellular matrix (ECM) that plays an important role in tissue development and organization. It possesses a complex mixture of pores, ridges, and fibers, which have sizes in the nanometer range. The ~200-nm-thick layer separates tissues such as epithelia, endothelia, muscle fibers, and the nervous system from connective tissue compartments. These nanometer range structures are also seen in the corneal epithelial basement membrane of the Macaque monkey as shown in Fig. 1 (Abrams et al., 2000). It consists of a porous membrane with a network of cross-linked fibers, with the pores averaging 72 nm and the fibers 77 nm in diameter, respectively. The mean elevation of features is around 160 nm. The trabecular meshwork of the human cornea is another example (Gong et al., 2002). It consists of ECM organized into a network of beams, covered by trabecular endothelial cells. Researchers have been attempting to understand the cell–substrate interaction by reproducing the substratum topography for in vitro studies. Weiss and Garber described the effect of contact guidance and the role of ECM structure in cell orientation and migration, more than 50 years ago (Weiss and Garber, 1952). Curtis also proposed the influence of microtopography on cell behavior in 1964 (Curtis and Varde, 1964), which has also been studied extensively since then. Recent findings underscore the phenomenon that mammalian cells respond to nanoscale features on synthetic surfaces, and the topic has been described in several recent reviews (Bettinger et al., 2009; Martinez et al., 2009; Seidlits et al., 2008; Yim and Leong, 2005).

Fig. 1 Scanning electron micrograph of the corneal epithelial basement membrane after incubation in 2.5 mM EDTA and epithelial removal. This shows the overall “felt-like” architecture of the surface, consisting of interwoven fibers and pores. The complex topography of intertwined fibers and pores of varying sizes is evident. Magnification = 30,000 (Abrams et al., 2000).

Benjamin Kim Kiat Teo et al.

244

B. Nanofabrication Techniques The advancements in micro- and nanofabrication technology enable the studies of cellular response to micro- and nanofeatures with a wide range of materials and topography. Most of these technologies were developed initially for microelectronic industry, but they have since been widely adapted to cellular studies. Detailed descrip tion of the micro- or nanofabrication technology has been covered in several excellent reviews (Curtis and Wilkinson, 1997, 2001; Flemming et al., 1999; Folch and Toner, 1998; Martínez et al., 2009; Patel et al., 1998; Seidlits et al., 2008); Table IA and B summarizes the nanofabrication techniques that are commonly used in studies invol ving cellular response and stem cell niche reconstruction. In general, the techniques can be divided into two categories of fabrication, mainly that of ordered features and random features.

Table IA Nanofabrication techniques for ordered features Fabrication method

Resolution

Electron beam lithography

Nanoimprint lithography

Advantages

Limitations

>3 nm (Vieu et al., 2000) • Pillars, wells • Grooves

• Precise geometry and pattern • No mask needed

>3 nm (limited by the template) (Chou et al., 1996)

• Pillars, wells, grooves • 3D (Zhang and Low, 2006) • Hierarchical structures (Zhang and Low, 2008) • Pillars, wells (2D pattern)

• Inexpensive • Fast • Easy scale up • Ability to make complex structure • Versatility in choice of material • Inexpensive • Fast • Simple

• Expensive • Time consuming • Small surface coverage • Even lower resolution with negative resist • Expensive equipment • Template needed

• Line, dots, 2D pattern

• Precise control • Maskless • Pattern and acquire image at the same time • In situ biocompatible fabrication • Precise geometry and pattern • Large surface coverage

>100 nm (commercial PDMS) >50 nm (hPDMS) (Odom et al., 2002; Schmid and Michel, 2000) Dip-pen >100 nm (Basnar and nanolithography Willner, 2009) Soft lithography

Direct-write fabrication with multiphoton Photolithography

Features

>250 nm (Kaehr et al., 2004)

• 3D complex architecture

250 nm (Norman and Desai, 2006)

• Pillars, wells, grooves

• Limitation on material • Template/mold required

• Slow • Height limitation unless used with particle deposition and/or etching • Special equipment • Limitation in choice of materials • Expensive equipment • Feature size large • Mask required

245

11. Nanotopography/Induced of Stem-Cell Differentiation

Table IB Nanofabrication techniques for random features Fabrication method

Resolution

1 nm (Desai et al., 1999; Moldovan et al., 2000) Carbon nanotubes single-walled nanotubes (SWNT) 0.4–2 nm in diameter (Malarkey and Parpura, 2007) multi-walled nanotubes (MWNT) >2 nm in diameter (Malarkey and Parpura, 2007) Etching— 2 nm chemical/ reactive ion etching Nanoporous membranes

>3 nm (Frenot and Chronakis, 2003; Matthews et al., 2002)

Features

Advantages

• Porous membrane

• • • •

• Woven mesh or vertical arrays • Patterns when combined with lithographic methods

• Craters—chemical (Fan et al., 2002) • Needles—reactive ion (Turner et al., 1997) • Pores— electrochemical etching (Low et al., 2006) • Cylindrical fibers: aligned/random

Easy fabrication • Insufficient strength Inexpensive for physiological Precise control over pore size loads Strong, flexible, and • Potential toxicity of conductive the carbon nanotubes

• Fast • Economical • No special equipment required (for chemical etching)

• Versatility in material choice • Composite with natural materials • Biological polymer • Coaxial fibers enable drug delivery Self-assembly >10 nm fibers • Hydrated 3D mesh of • Controlled fiber dimension nanofiber (Hartgerink et al., interwoven and morphology 2002; Holmes et al., nanofibers • Self-assembly 2000) Polymer demixing Vertical 13 nm • Islands, pits, ribbons • Simple (Affrossman and • Fast Stamm, 1998, 2000; • Inexpensive Affrossman et al., 1996) Colloidal 20 nm (Dalby et al., • Columns • Easy to pattern lithography 2004a) • Large surface coverage area Carbon nanofibers >40 nm in diameter • Fiber • Conductive fiber with tunable (Sen et al., 2004) mechanical and cell adhesion property Phase separation >50 nm fiber (Smith • Porous and fibrous • Simple and Ma, 2004) network • No special equipment • Porosity easily controlled Direct-write >250 nm (Kaehr • 3D complex • In situ biocompatible fabrication with et al., 2004) architecture fabrication multiphoton Electrospinning

Limitations

• No control in geometry

• Fibers only • Difficult to control pore size for nanofiber mesh

• Engineering of molecules for selfassembly required • Limitation in geometry and choice of materials

• Limitation in geometry • Fibers only • No organized pattern • Special equipment • Limitation in choice of materials

246

Benjamin Kim Kiat Teo et al.

Many fabrication techniques for ordered features are dependent on optical lithography and reactive ion etching. This can be followed by anisotropic etching or UV and glow discharge treatment (Flemming et al., 1999). Photolithography is limited by diffraction limitations. Its maximum resolution is typically of the wavelength of the light being used for the exposure (>200 nm). Electron beam lithography can be used to produce higher resolution nanoscale patterns, but it is expensive and time consuming. Some other methods include laser ablation, X-ray lithography, and dip-pen nanolithography (DPN). One such method of particular interest, nanoimprint lithography (NIL), is a versatile mechanical lithography process, in which a polymer resist will be embossed with a patterned hard mold, such as a prepatterned SiO2 mold. In conventional thermal NIL, a substrate will be spin coated with a polymer layer before embossing with a hard mold. It is of low cost and can pattern features in large areas and with a lateral resolution of <10 nm (Chou et al., 1996). In the reversal imprinting, a polymer layer is spin coated onto the mold only and then harvested by transferring to a bare substrate by imprinting under suitable temperature and pressure (Huang et al., 2002; Zhang and Low, 2006). In addition to thermal NIL, UV NIL allows imprinting of UV-curable materials at room temperature. With the different modes of imprints, NIL provides a wide choice of polymers that can be used for nanofabrication, and it would be useful for cellular engineering and the reconstruction of stem cell niche. With a template generated by optical or mechanical lithography, the pattern can be easily replicated onto elastomers such as polydimethylsiloxane (PDMS) using soft lithography. However, there are a limited number of siloxane polymers that can be used in this technique. Although most optical or mechanical lithographies are capable of generating precise and highly ordered features, the fabrication process can be both expensive and com plex. On the other hand, fabrication techniques of random structures are relatively simpler and less costly. For example, nanoislands of 13–95 nm in height can be fabricated by polymer demixing based on phase separation of polystyrene and poly (4-bromostyrene) spin coated on silicon wafers (Affrossman et al., 1996). The ability to produce different nanofeatures based on such phase separation phenomenon, how ever, would be even more restrictive. Another versatile technique of producing nanostructures for cell culture applications is electrospinning (Frenot and Chronakis, 2003; Matthews et al., 2002). In the electro spinning process, an electrode is placed into a syringe or pipette containing the polymer solution. A high voltage applied discharges an electrically charged jet of polymer solution onto a grounded collector surface. Randomly oriented fibers will be collected, forming a nonwoven mat, while aligned fibers can be collected using rotating wheel, when the rate of rotation matches the rate of fiber deposition. Many different types of polymers can be processed into fibers with diameters ranging from several nanometers to tens of microns. Although electrospinning requires relatively simple instruments, many process parameters are needed to be optimized to produce the desired fiber characteristics of size, orientation, and uniformity without bead formation. The processing parameters include viscosity, evaporation rate, surface tension, and conductivity of the polymer solution, processing variables of applied electric voltage and source-to-collector distance.

11. Nanotopography/Induced of Stem-Cell Differentiation

247

C. Chemical Topology: Protein Patterning Distribution and spatial organization of ECM can also be achieved by chemical patterning. Patterns of immobilized adhesive ECM molecules, which are usually sur rounded by nonadhesive regions, could restrict cell growth and cell spreading and also geometrically modulate the cell shape. Previous studies have demonstrated that the local changes in cell–ECM interaction could generate changes in the growth, viability, and differentiation of cells, which are critical for tissue formation (Lim and Donahue, 2007). Popular choices of adhesive protein or biochemical include fibronectin (FN), vitronectin, laminin, and collagen or the integrin-binding motif such as the arginine–glycine– aspartate (RGD) peptide. Inactive or nonadhesive molecules, such as poly(ethylene glycol) (PEG) or bovine serum albumin, can be adsorbed or immobilized to create the nonadhesive regions. Patterns of various sizes ranging from tens of nanometers to hundreds of microns can be produced by different techniques. Micropatterning can guide cell growth, cell shape, and migration. The cell shape can be controlled by growing the cells on isotropic versus anisotropic pattern or controlled by the geometry of the pattern in the case of single-cell patterning (Nelson et al., 2004). Nanopatterning regulates cell function such as cell adhesion (Arnold et al., 2004, 2008), proliferation, gene expression, and differentiation. Nevertheless, it has also been shown that changing of cell shape by micropatterning can also direct stem-cell differentiation (McBeath et al., 2004). Similar to the development of fabrication techniques for topographical pattern, advanced technology broadens the choice of fabrication techniques for micro- and nanopatterning. Examples of micropatterning fabrication included microcontact print ing (µCP), microfluidic networks, vapor deposition, and photolithography. In conven tional microcontact, siloxane-based elastomer stamp, made by soft lithography, will be used to stamp a self-assembled monolayer (SAM) of hydrophobic alkanethiol with an end functional group onto gold-coated surface (Whitesides et al., 2001). The SAMs with the functional groups have the ability to generate well-defined surfaces with a broad range of characteristics. Alternatively, various proteins can also be directly transferred from the PDMS stamp onto a flat substrate surface. The only requirement is that the material being transferred must be able to be deposited onto the surface of the stamp. Another technique using a microfluidic network involves parallel microfluidic channels to make multiprotein patterning, in which separate laminar streams of fluid flow adjacently into a common channel of a microfluidic structure from inde pendent inlets that mix only at their interface by diffusion (Whitesides et al., 2001). Examples of nanoscale protein patterning include DPN, electron beam lithography, nanoimprinting, nanoshaving, and nanoscale RGD clustering. Some of these fabrica tion techniques have already been discussed in the previous section. In DPN, relatively larger molecules such as immunoglobulin can be transferred from a cantilever probe to the substrate with nanometer resolution. The transfer mechanism is similar to microcontact printing. In nanoshaving, scanning probe microscopy is used to selectively break surface chemical bonds using mechanical or electrical tip–surface interactions (Liu et al., 2000).

248

Benjamin Kim Kiat Teo et al.

D. Geometry A diverse variety of topological features with different shapes, organization, and complexity can be fabricated with techniques described in the previous section. We have grouped the patterns according to the top-view geometry to facilitate the sum mary and discussion of the stem cell interaction with topological patterns (Table II).

III. Stem Cell Reception to Substrate Topology Stem cells are defined as cells with the ability to self-renew and differentiate into specialized cells in response to appropriate signals (Ding and Schultz, 2004). Because of its ability to differentiate into different types of functional cells, stem cells possess great value as therapeutics to regenerate and repair damaged tissue. Stem cells can be broadly classified into embryonic stem cells (ESCs) and adult stem cells. ESCs are pluripotent cells derived from the inner cell mass of blastocysts with the potential to maintain an undifferentiated state (Stojkovic et al., 2004). The ESCs are hypothetically capable of regenerating all the cell types of the three germ layers—ectoderm, meso derm, and endoderm. Recent technologies also allow the reprogramming of adult cells into pluripotent stem cells, which are referred as induced pluripotent stem cells and exhibit properties similar to ESCs (Takahashi and Yamanaka, 2006). Adult stem cells, on the other hand, are derived from adult tissues. Adult stem cells can be multipotent, hence they can generate progeny of multiple distinct cell types. Progenitor cells can also be isolated from adult tissues, and they have the capacity to differentiate into a specific type of cells; nevertheless, progenitor cells are committed to a specific lineage. With the properties of self-renewal and differentiation, stem cells have a huge potential for biotechnology and regenerative medicine. The microenvironment is composed of ECM, which is a reservoir of biochemical as well as biophysical cues. Both of these cues can be presented on substrate surface with specific spatial arrangement, while biophysical cues can be presented in the form of substrate nanotopography and/or matrix stiffness. Both the controlled self-renewal and the directed differentiation are keys to the application of stem cells in regenerative medicine, which aims to provide a therapeutic platform by creating or controlling the extracellular microenvironment in order to guide tissue growth for functional recovery and/or dictate stem-cell differentiation into the appropriate cell type. In their in vivo environment, stem cell fate is controlled by intrinsic factors and microenvironment known as the stem cell niche (Fuchs et al., 2004; Moore and Lemischka, 2006). Meanwhile, as stem cells vary in sizes and shapes in the micron range, different types of stem cells require different biochemical and biophysical cues from their niche for differentiation or maintenance. It is not surprising to see that various type of stem cells do respond to topology at the micron- and nanoscale. Even though nanotechnology is defined as the manipulation of atoms and molecules in nanometer scale range of <100 nm; nevertheless, as the focus of this chapter is not on the quantum effect of the materials but the cells’ interaction with

249

11. Nanotopography/Induced of Stem-Cell Differentiation

Table II Classification of topological patterns according to the geometry Class of geometry Isotropic Dot

Circle

Anisotropic Lines

Example of protein pattern

Examples of fabricated topography

Dimples (circles) (Doyle et al., 2009)

Pillars (Zhao et al., 2006)

Islands (Dalby et al., 2004b)

Needles (Rao and Zheng, 2009)

Hollow circle (Nelson et al., 2005)

Wells (Biggs et al., 2009)

Pits (Dalby et al., 2004b)

Honey comb

Lines (Doyle et al., 2009)

Gratings (Yim et al., Blades (He and Zhao, 2009) 2005)

Aligned fibers (Corey et al., 2007)

3D structures (Xu et al., 2006)

Fibrous mesh (Carnell et al., 2008)

Complex structures Irregular Triangular, donut, bow-tie shape shaped (Doyle et al., 2009), letters (Thery et al., 2006)

Hierarchical structure (Zhang and Low, 2006)

250

Benjamin Kim Kiat Teo et al.

substratum features, stem cells’ response to subcellular scale topography of >1000 nm will also be discussed in this chapter to give a more holistic description of the field. In the rest of this section, we will discuss how surface topology, in particular the role of topographical and chemical patterns, in the maintenance and differentiation of ESCs and adult stem cells such as mesenchymal stem cells (MSCs), hematopoietic stem cells, and neural stem cells/neural progenitor cells (NSCs/NPCs).

A. Embryonic Stem Cells ESCs are pluripotent cells that were extracted from the inner cell mass of a preimplantation blastocyst (Rao and Zandstra, 2005). Although the extraction and culture methods of ESCs may differ, all ESCs share five characteristics: long-term self-renewal capacity, multilineage developmental potential, normal euploid karyotype maintenance over extended culture, cell-specific marker expression, and telomerase activity (Hoffman and Carpenter, 2005; Ullmann et al., 2007). These shared character istics make ESCs unique and are the reason behind their enormous potential as an unlimited cell source for tissue engineering, regenerative medicine, drug discovery and testing, and studying human developmental biology (Amit et al., 2000). However, before this potential can be realized, protocols for the efficient, reproducible, and xeno free maintenance and differentiation of ESCs must be found. Traditionally, ESCs were cultured on flat surfaces coated with a layer of ECM or a layer of mitotically inactivated feeder cells. On these platforms, the ESCs were cultured in a serum-containing medium that was supplemented with additional growth factors. This continues to be the culture condition of mouse ESCs, which are con ventionally grown on a layer of gelatin in serum-containing medium that is supple mented with leukemia inhibitory factor (LIF) (Peerani et al., 2009). Human ESC culture, however, continues to evolve toward a defined and xeno-free culture system in the hope that human ESCs can be used for regenerative medicine. Human ESC culture has evolved from originally being propagated on inactivated mouse embryonic fibroblast (MEF) feeder layers (Wang et al., 2005) to being propagated on extracellular matrices such as MatrigelTM, laminin, and fibronectin with media conditioned by MEFs (Skottman and Hovatta, 2006), and recently to being propagated on extracellular matrices with defined media (Chin et al., 2010). Currently, xeno-free and fully defined culture media are being developed for human ESCs. In addition to improving the ESC expansion culture conditions, there is also a push to direct the differentiation of ESCs into a specific terminally differentiated cell type. The current approaches to achieve this end goal are to use a “differentiation medium” and/or embryoid body formation (Dang and Zandstra, 2005; Ungrin et al., 2008; Wang and Yang, 2008). The differentiation medium often contains serum and chemicals, like retinoic acid, and does not include the requisite growth factors in the maintenance medium, like fibroblast growth factor 2 (FGF2, which also known as basic FGF) or LIF. Embryoid body formation can be achieved with multiple protocols such as the hanging drop method, the spin down method, and the suspension culture.

251

11. Nanotopography/Induced of Stem-Cell Differentiation

At present, ESC proliferation and differentiation are predominantly directed using biochemical cues from growth factors and the ECM; however, the ECMs, like Matri gelTM, also provide topographical cues. In the following section, the effect of topo graphical cues on ESC proliferation and differentiation will be discussed.

1. ESCs on Grating-Like Structures Grating-like topographies cause cells to become elongated. In the case of ESCs, this elongated morphology appears to be a cue for the ESC to differentiate. This has been shown by at least two studies which have demonstrated that differentiation is enhanced when a grating-like topography is used without differentiation medium, and an even greater enhancement is observed when differentiation medium is used in conjunction. The efficacy of topographical cues to induce differentiation was recently shown by Markert et al. (2009). This group used the BioSurface Structure Arrays (BSSA) (Lovmand et al., 2009) to screen the effect of 504 distinct topographical structures on murine ESCs. This array was etched onto a silicon wafer using standard lithogra phy, before being sputter coated with tantalum oxide. This array contained a series of sharkskin structures (Fig. 2) which were formed by alternating lines of bars with lateral sizing ratios of 1:4 and 2:3. The lines and the bars were separated by 1 µm, and the lateral size unit varied from 1 to 8 µm. Even though the culture took place in expansion medium which contains LIF, the sharkskin structures promoted differentiation of the mouse ESCs as indicated by a decrease in alkaline phosphatase staining intensity, and an increase in cell elongation and colony spread. Independently, Smith et al. also demonstrated that grating-like topographies pro mote mouse ESC differentiation (Smith et al., 2009). Using a phase separation technique, this group created poly(L-lactic acid) (PLLA) matrices with nanofibrous architecture which had diameters ranging from 50 to 500 nm. They then compared the effect of the PLLA matrix on mouse ESCs to the more conventional culture surfaces

4T 3T 2T 1T 1 μm 1 μm (T) in μm

1 μm

1 2 3 4 5 6 7 8 (1) (2) (3) (4) (5) (6) (7) (8)

Fig. 2 Sharkskin structures from the BioSurface Structure Arrays (BSSA). Extracted from Lovmand et al. (2009; Fig. 1A, cropped).

Benjamin Kim Kiat Teo et al.

252 Nano

Solid

Control

Osteocalcin

TUJ1

Dapi

Fig. 3 Immunouorescence localization neuronal (TUJ1) and late bone differentiation (osteocalcin) marker expression after 26 days under osteogenic differentiation conditions on nanobrous matrix (nano), solid films (solid), and gelatin-coated tissue-culture plastic (control). Scale bar = 50 µm. Extracted from Smith et al. (2009; Fig. 6C, cropped).

gelatin-coated tissue-culture polystyrene and PLLA film. When mouse ESCs were cultured on these three platforms with differentiation media or osteogenic media, which contained osteogenic supplements, the ESCs were observed to spread more on the nanofibrous matrix and to have an increase in early ectoderm lineage and osteo genic markers (Fig. 3). More specifically, they observed a greater increase in the expression of brachyury, b1-integrin, Runx2, osteocalcin, and sialoprotein, and a decrease in nestin, as compared to flat films and the tissue-culture plastic control.

2. ESCs on Pillar-Like Structures Pillar-like topographies are supportive of mouse ESC proliferation as demonstrated by Markert et al. with the BioSurface Structure Arrays (Lovmand et al., 2009; Markert et al. 2009). With this structure array, Markert et al. surveyed the effects of pillar structures on mouse ESCs by utilizing pillars varying in shape (circle, square), place ment (aligned, staggered), feature size (1, 2, 4, 6 µm), spacing distance (1, 2, 4, 6 µm), and feature height (0.6, 1.6, 2.4 µm) (Fig. 4). This array of pillars was created in silicon and then coated with a thin layer of tantalum oxide. Using this array and expansion media, Markert et al. discovered that while pillar shape and placement have no effect on proliferation, the pillars with smaller feature sizes, bigger feature heights, and increased spacing distance increase the proliferation of these mouse ESCs. The feature sizing combination which yielded the greatest proliferation rate are pillars with 1 µm diameters and 2.4 µm heights that were spaced 2 µm apart. However, as these were the

253

11. Nanotopography/Induced of Stem-Cell Differentiation

(A)

(B)

(C)

(D)

(E)

(F)

(G)

(H)

(I)

(J)

Y X Y

1 (1,1) 5 (2,1) 9 (4,1) 13 (6,1)

(X,Y ) in μm 2 3 4 (1,2) (1,4) (1,6) 6 7 8 (2,2) (2,4) (2,6) 10 11 12 (4,2) (4,4) (4,6) 14 15 16 (6,2) (6,4) (6,6)

Fig. 4 Array of pillar structures from the BioSurface Structure Arrays. Extracted from Lovmand et al. (2009; Fig. 1A, cropped).

limits of the tested range, smaller feature sizes and taller feature heights may yield even greater proliferation rates. This was remarked upon by Markert et al. when they noticed that even without LIF in the culture media, the mouse ESC colonies exhibited rounded and more compact colonies as compared to the flat surface control. This is another indication that the cues supplied by pillar topographies are supportive of proliferation even without the support of the biochemical cues. In addition to physical pillar topographies supporting proliferation, it has been shown that circular ECM patterns can also be used to promote proliferation or induce differentia tion. Two studies conducted by Peerani et al. have shown that both mouse ESC and human ESC fates are influenced when constrained to a colony size due to the limitations of endogenous signaling (Peerani et al., 2007, 2009). The colony size was constrained using microcontact printing to pattern MatrigelTM for human ESCs or a fibronectin–gelatin mixture for mouse ESCs onto a tissue-culture-treated microscope slide before cell seeding. In the study with human ESCs, the circular patterns had a diameter varying from 200 to 800 µm and a pitch varying from 500 to 1000 µm (Peerani et al., 2007). After 48 h of culture in expansion medium, it was found that the cells in 200 µm colonies differ entiated, whereas the larger colonies of 400 and 800 µm remained undifferentiated, as determined by immunostaining and quantitative polymerase chain reaction (Fig. 5). One mechanism that determined human ESC fate by spatial colony size control was the secretion of the endogenous ligands BMP2, by human ESC-derived extraembryonic endoderm, and GDF3, by undifferentiated human ESC (Fig. 6A). In smaller colonies, the amount of GDF3 secreted is not enough to continue supporting human ESC selfrenewal; as such the BMP2 ligand will activate the differentiation pathways (Fig. 6B). Similarly, mouse ESCs grown in colonies that were restricted to diameters of <100 µm had a tendency to differentiate because endogenous Stat3 activation could

Benjamin Kim Kiat Teo et al.

254 100 0h 6h Day 2

Oct-4+ (%)

80 60 40

* 20 0

D = 200 μm D = 400 μm D = 800 μm Pattern type

T=0

Fig. 5 The percentage of pluripotent (Oct-4þ) human ESC as a function of colony size as created by microcontact printing of MatrigelTM. * indicates statistical significance of p < 0.05. Extracted from Peerani et al. (2007; Fig. 4Bi).

not be maximized (Peerani et al., 2009). Since phosphorylated Stat3 is a transcriptional factor which directly regulates LIF receptor transcription and indirectly regulates pluripotency markers transcription, it is an important factor when determining mouse ESC fate. This study utilized a range of circular patterns with diameters varying from 50 to 200 µm and pitch varying from 200 to 400 µm. The cells were then cultured on these patterns for 24 h in serum-free media without LIF. Using this range of patterns, they found that generally mouse ESCs will proliferate as the pattern size and clustering increases, and as the pitch size decreases (Fig. 7). (A) Extraembryonic endoderm (ExE) BM P2

Human embryonic stem cells (hESCs) F3 D G

BMP2

P2

BM

Pluripotency

(B)

GDF3

BMP2

GDF3

GD

F3 Single cells D = 200 μm D = 400 μm

pSmad1

pSmad1

D = 800 μm Pluripotency

Fig. 6 Two niche determinants—composition and size—regulate human ESC (hESC) self-renewal by local modulation of pSmad1 agonists and antagonists. (A) pSmad1 levels in hESCs are a balance of agonist (s) secreted by differentiated progeny including HNF3(b)+ extraembryonic endoderm and antagonists secreted by hESCs (Oct-4+). The levels of pSmad1 of a single hESC will be dependent on the localized cell density of the Oct-4+ and HNF3(b)+ neighbors around it. (B) By manipulating colony size through micropatterning, the levels of Smad1 antagonists, GDF3 and LeftyB, increase while BMP2 remains constant resulting in pSmad1 levels that decrease with colony size. This decrease in pSmad1 levels can be correlated with the increased pluripotency seen with increasing colony size. Extracted from Peerani et al. (2007; Fig. 6).

255

11. Nanotopography/Induced of Stem-Cell Differentiation

Clustering

D = 50 μm

P = 200 μm

D = 200 μm

D = 100 μm

P = 500 μm

D = 100 μm

P = 200 μm

D = 100 μm =

P = 800 μm

Pitch P = 200 μm

Diameter D = 50 μm

P 200 μm

pStat3 activation

Fig. 7 Micropatterning mouse ESC (mESC) cultures provides spatial control over endogenous Jak-Stat activation. Three parameters were explored: increasing colony diameter, decreasing colony pitch, and increasing the degree of clustering. Extracted from [Peerani et al., 2009] Fig. 5 (Peerani et al., 2009).

Restriction of the colony size by patterning can also be used to find an optimal aggregate size for differentiation, as shown by Sasaki et al. (2009). This group employed a mask-free photolithography system to develop adhesive circular microdomains with diameters ranging from 100 to 400 µm on glass substrates. After cultur ing the mouse ESCs in differentiation media on these substrates for 12 days, they found that 200 µm was the optimal diameter for the cardiac differentiation of uniformsized murine ESC aggregates (Fig. 8). These simple micropatterning techniques provide an efficient and high-throughput method to probe the signaling within the ESC colonies and to determine the constraints that will drive differentiation toward a specific lineage most efficiently. (A)

(B)

(C)

(D)

(E)

Fig. 8 Microscopic images of ESCs cultured on a cell patterning substrate with 100-mm-diameter domains (A), 200-mm-diameter domains (B), 300-mm-diameter domains (C), and 400-mm-diameter domains (D), and on a polyacrylamide (PAAm)-nongrafted substrate (E), on day 10. EGFP Fluorescence images are superimposed on phase contrast images, where EGFP fluorescence indicates that the cell is a cardiomyocyte. Scale bars = 500 mm. Extracted from Sasaki et al. (2009; Fig. 2).

Benjamin Kim Kiat Teo et al.

256

3. ESCs on Well-Like Structures Studies of the effect of well-like topographies on the proliferation of ESCs have been limited the micrometer range. Two groups have used soft lithography to show that well-like topographies made of PDMS will increase the proliferation human ESCs grown in aggregates. Mohr et al. used square microwells of PDMS coated with MatrigelTM with the following dimensional combinations (width, height): (50 µm, 50 µm), (100 µm, 100 µm), and (100 µm, 120 µm) (Fig. 9) (Mohr et al., 2006). All three of these topographical structures showed an improvement in the viability, proliferation rate, percentage of pluripotent ESCs (as measured by Oct-4 expression level), and metabolic activity of the human ESCs when compared to the traditional culture surface, tissueculture polystyrene. In these microwells, the human ESCs grew as aggregates, which is unlike the traditional culture method of monolayers. Similarly, Khademhosseini et al. also showed that human ESCs grown as aggregates in circular PDMS microwells coated with fibronectin and Mitomycin C MEFs also improved the homogeneity of the human ESC population and increased the prolifera tion rate (Khademhosseini et al., 2006). However, unlike the Mohr group,

Molded PU substrate Gold coating EG3-terminated SAM Matrigel ECM hESCs

Fig. 9 (A) Schematic of microwell fabrication depicting spatial localization of physical and chemical constraints to hESC attachment and propagation. Microwells are formed by cross-linking polyurethane (PU) prepolymer. Gold is evaporated on the surfaces surrounding the wells and the sides of the wells at an oblique angle, but the bottoms of the wells are shielded from gold. A triethylene glycol-terminated alkanethiol selfassembled monolayer (EG3-terminated SAM) is then assembled on the gold surface to repel extracellular matrix (ECM) proteins and cells. The wells are then treated with Matrigel which adsorbs to the bare PU at the bottoms of the wells. Cells are cultured in the Matrigel-coated wells. Structures are not to scale. (B) Localization of hESCs to microwells. Microwells were 120 mm deep and 100 mm per side. hESC localization within microwells after 21 days of culture on Matrigel in CMFþ visualized by phase contrast. Scale bars = 100 mm. Extracted from Mohr et al. (2006; Figs. 1 and 2B).

11. Nanotopography/Induced of Stem-Cell Differentiation

257

Khademhosseini et al. used larger dimensions with a diameter of 200 µm and a height of 120 µm. Unfortunately, neither group explored smaller dimensions to determine the effect of human ESCs when grown at the nanometer scale.

B. Mesenchymal Stem Cells MSCs are stem cells of mesodermal origin that can be derived from the adult bone marrow. MSCs are multipotent in nature, capable of differentiating into osteoblasts, myoblasts, adipocytes, chondrocytes, and skin (Caplan, 1991; Ciapetti et al., 2006; Even-Ram et al. 2006; Sanchez-Ramos et al., 2000). It is now widely accepted that MSCs stimulate host recovery and regeneration through the secretion of numerous proregenerative factors. In vitro studies have documented the secre tion of multiple anti-inflammatory, angiogenic, neurotrophic, immunomodulatory, and antifibrotic factors from MSCs (Caplan and Dennis, 2006). This therapeutic potential of MSCs could be tapped by enhancing the proliferation and directing their differentiation. The most common strategy used is a combination of growth factors/biochemical cues such as bone morphogenetic protein 6 (BMP6) and transforming growth factor b3 (TGF b3) for chondrogenic differentiation; ascorbic acid and dexamethasone for osteogenic differentiation; 5-azacytidine for myogenic induction; microRNA-21 (miR-21) and dexamethasone for adipogenic differentiation; and retinoic acid for neuronal differentiation (Drost et al., 2009; Farrell et al., 2006; Sekiya et al., 2004; Yim et al., 2007). Nevertheless, various studies have shown nanotopography alone can enhance proliferation and/or induce differentiation of MSC (Dalby et al., 2007a; Huang and Li, 2008; McBeath et al., 2004; Yim et al., 2007).

1. MSCs on Grating-Like Structures Generally, grating-like topographies have been shown to induce differentiation in MSCs. In one such study by Yim et al., the proliferation, alignment, elongation, and transdifferentiation of human MSC on the nanograting axis were studied (Yim et al., 2007). Gratings with line widths of 350 nm, 1 µm, and 10 µm on PDMS were used to study the transdifferentiation of human MSCs into neuronal lineage, with or without retinoic acid induction. They have successfully shown that the effect of nanotopogra phy on the upregulation of neuronal marker is higher compared to the retinoic acid induction alone (Fig. 10). In another interesting comparative study of different patterns by Martino et al., uniform grids and grooved nanopatterns on hydrogenated amorphous carbon film were compared for the differentiation of human bone marrow-derived MSCs (huBM-MSCs) into osteogenic and adipogenic lineage (Fig. 11) (Martino et al., 2009). These grooved nanopatterns exerted a higher effect on differentiation of huBM-MSC into osteogenic lineage due to the alignment and elongation cues presented by the grooves.

Benjamin Kim Kiat Teo et al.

258 Nanopatterned PDMS (A)

Tuj1

GFAP

DAPI

(B) Normalized MAP2 expression

axis

Grating

Day 7

50 μm

axis

Grating

Day 14

50 μm

3 2.5 2 1.5 1 0.5 0 PDMS

10μm

1μm

300nm

Normalized MAP2 expression

(C) 3.5

Nanopattern

PDMS

3 2.5 2 1.5 1 0.5 0 MSC medium

1uM RA 7 days

30uM RA

MSC medium

1uM RA

30uM RA

14 days

Fig. 10

The Transdifferentiation of hMSC on nanopatterned PDMS. (A) Immunofluorescent staining of beta III tubulin (Tuj1), a neuronal marker, and GFAP, an astrocyte marker. (B) Quantitative analysis of microtubule-associated protein 2 (MAP2) mature neuronal marker expression in hMSCs cultured on gratings with widths of 300 nm, 1 µm, and 10 µm. (C) Quantitative analysis of MAP2 expression in hMSCs cultured on nanopatterned PDMS in proliferation medium or in the presence of RA and on unpatterned PDMS on day 7 and day 14 (Yim et al., 2007).

2. MSCs on Pillar-Like Structures Similar to ESCs, pillar-like patterns favor the proliferation of MSCs. However, the optimal size of the pillar-like feature for MSC proliferation appears to be dependent on the species. In the case of human MSCs, at least two studies have shown that the optimal feature size for adhesion and proliferation is 200 nm. Khor et al. showed that human

259

11. Nanotopography/Induced of Stem-Cell Differentiation

(A)

(B)

100 μm (C)

100 μm (D)

100 μm

100 μm

Fig. 11 A and B show the bright field images of grid and grooved patterns. C and D show the fluorescent images of human bone marrow-derived MSCs seeded on grid and grooved pattern, respectively. They are stained for a-tubulin (red), F-actin (green) and DAPI (blue) (Martino et al., 2009).

mesenchymal progenitor cells cultured on poly(styrene)-block-poly(4-vinylpyrindine) diblock copolymer with pillars of average height 6 nm and width 200 nm resulted in an increased number of cell aggregates as compared to the smooth surface control on the first day post seeding (Khor et al., 2007). Interestingly, by day 7, the human MSCs cultured on these pillar structures had a flattened and cuboidal morphology with prominent and large adhesion sites and a diffused actin filament structure, which was quite different from the flattened and bat-like structure of the MSCs on the unpatterned mica surface, as seen in Fig. 12. Similarly, Dulgar-Tulloch et al. demonstrated that 200 nm is the optimal feature size for human MSC adhesion and proliferation (Dulgar-Tulloch et al., 2009). They cultured the human MSC on alumina and titania, whose surface was patterned with grain-like structures. This group also showed that the 200- nm grain size was most favorable for human MSC proliferation independent of the surface chemistry, the surface roughness, and the crystal phase. When the grain size was increased to 1500 nm, there was a decrease in the proliferation rate; however, the 1500 nm proliferation rate was still better than the proliferation rate when the grain size decreased to 24 or 50 nm as seen in Fig. 13. The optimal topographical features seem to be different for proliferation of rat MSCs cultured on titanium oxide (TiO) nanotubes. Park et al. have demonstrated that rat MSCs have increased adhesion, spreading, motility, and proliferation when cultured on these TiO nanotubes with a diameter of 15 nm (Park et al., 2007). These 15- nm TiO nanotubes also increased the osteogenic differentiation of the rat MSCs when cultured in induction media. However, when the tube diameters increased to 50 nm and larger, the conditions became unfavorable to rat MSC culture and resulted in low cell attachment and apoptosis (Fig. 14). The TiO nanotubes with 15 nm diameter also showed optimal osteogenic differentiation of rat MSCs when cultured in osteogenic differentiation media.

Benjamin Kim Kiat Teo et al.

260 10.0

10.0

(A)

0

(B)

PM

2.5

5.0

7.5

10.0

(C)

Pillar-like

Worm-like

7.5

7.5

7.5

5.0

5.0

5.0

2.5

2.5

2.5

0 10.0 μm

0

2.5

7 Day (D)

7 Day (E)

50.0 μm

50.0 μm

5.0

7.5

0 10.0 μm

0

2.5

5.0

7.5

0 10.0 μm

7 Day (F)

Fig. 12 A–C show the AFM images of unpatterned mica substrate, pillar-like pattern, and worm-like pattern on diblock copolymers of poly(styrene)-block-poly(2-vinylpyridine). D–F show confocal images of human mesenchymal progenitor cells seeded on unpatterned, pillar-like and worm-like substrates, respectively. They are stained with phalloidin (actin filaments, green) and propidium iodide (nucleus, red) (Khor et al., 2007).

Engel et al. studied the human MSCs lineage commitment by varying the cell shape using topography. These human MSCs were cultured on microstructures in the range of 50–100 µm, imprinted on poly(methyl methacrylate) abbreviated as PMMA (Engel et al., 2009; Martínez et al., 2009). When human MSCs are allowed to adhere, flatten, and spread, they undergo osteogenesis, whereas compact, round cells become adipo cytes (McBeath et al., 2004). It was observed that round pillars, 50 µm width and 1 µm height, aided adipogenic lineage differentiation while square pillars, 2 mm width and 1 µm height, promoted osteogenic transition by changing the morphology of the cells to slightly elongated or star shape, respectively, as seen in Fig. 15. Cell shape is said to regulate the switch in lineage commitment of human MSCs, by modulating endogen ous RhoA activity (Fig. 15F). The mechanism of topography-induced differentiation will be discussed in greater detail in Section 4. It was also demonstrated that the heights of topographical features could affect stem-cell differentiation. In a study conducted by Terje et al., titanium nanopillar structures of near hexagonal order with

261

11. Nanotopography/Induced of Stem-Cell Differentiation Grain size

(D)

(A)

1500 nm

200 nm

50 nm

24 nm

3 Days 250

10,000 Magnification

5 μm

(B)

Cell number (cells/mm2)

200

## **

## ** **

150

**

100

50

0 Alumina

Titania

7 Days 500 100,000 Magnification

500 nm

## **

450

(C) Cell number (cells/mm2)

400 350 300

++ ## **

** **

250 200 150 100 50 0 Alumina

100,000 Magnification

Titania

500 nm

Fig. 13 Field emission scanning electron micrographs illustrating the respective grain sizes of (A) 1500 nm alumina, (B) 24 nm alumina, and (C) 50 nm titania. (D) Human MSC proliferation on alumina and titania of various grain sizes at 3 and 7 days showing maximum proliferation for 200 nm and least for 24 nm grain size (Dulgor-Tulloch et al., 2009).

Benjamin Kim Kiat Teo et al.

262 (A) 15 nm

20 nm

30 nm

50 nm

70 nm

100 nm

200 nm

(B)

Cell proliferation, 6 days

Cell migration (C)

0.3

36 h 60 h

Migration (mm)

OD420–480 nm

3

0.2

0.1

1

0

0.0 Smooth

15

20

30

50

70

Smooth

100 nm

Anexin V(+) cells (%)

0.12 0.08 0.04

15

50

70

100 nm

20

30

50

70

100 nm

Cell apoptosis 2 days 4 days

80

60

40

20

0

0.00 Smooth

15

(E)

Mineralization

(D)

OD405 nm

2

Smooth

15

50

100 nm

Fig. 14 (A) SEM images showing highly ordered nanotubes of six different pore sizes between 15 and

100 nm created by controlling potentials ranging from 1 to 20 V. (B) Cell proliferation rates were measured

by colorimetric WST-1 assay conducted 6 days after plating. (C) For measuring cell migration in a wounding

assay, cells were plated at a density of 50,000 cells/cm2, and 3 h later a track of 3.4 mm width was created in

the confluent cell layer. Cell motility was evaluated by measuring the remaining width after 36 and 60 h.

(D) Osteogenic differentiation of stem cells after 2 weeks of cell culture in osteogenic differentiation medium was assessed by staining for calcium phosphate mineral deposition with alizarin red and colorimetric analysis of the dye. (E) Apoptosis was analyzed by staining cells–cells were counted using a fluorescence-activated cell sorter (Park et al., 2007).

263

11. Nanotopography/Induced of Stem-Cell Differentiation

(A)

(B)

124.4 nm

124.4 nm

(D)

(C)

(E)

(F) Cell shape

Osteocalcin

25

Soluble factors

PS PS+0M

20

Rho

mg/ml*106 cells

PMMA flat PMMA flat+OM

15

hMSC

Tension

PMMA square

10

Rock

PMMA square+OM PMMA round

5

PMMA round+OM

Adipocyte

Osteoblast

0 Substrate

Fig. 15 Interferometry images of the microstructured PMMA. (A) The rounded structure of 50 µm width and 1 µm height and (B) the square structure of 2 µm width and 1 µm height. (C) Immunohistochemical staining of actin (red) and cell nuclei (blue) in rat MSC on round structures. Three cells are located inside the structures and are connected to one another among the structures. (D) Rat MSC on square structures showing two types of morphologies: one more elongated, with no spreading at all, and another that is star shaped. Rat MSC differentiation using osteogenic medium (OM). (E) Cells cultured for 11 days in the presence or absence of OM: osteocalcin amount. (F) Model of a mechanically mediated switch in hMSC commitment to adipogenic or osteogenic fate. RhoA signaling appears necessary and sufficient to replace soluble growth factor signaling while ROCK activity acts downstream of cell shape (A–E, Engel et al., 2009; F, Mc. Beath et al., 2004).

Benjamin Kim Kiat Teo et al.

264

varying heights of 15, 55, or 100 nm on titanium surfaces were synthesized by anodization through a porous alumina mask (Terje et al., 2009). Osteogenic differ entiation of human MSCs grown on these titanium structures was observed. The 15 nm high topographical features resulted in the most significant bone matrix nodule forma tion after 21 days of culture in comparison to the pillars of sizes 55 and 100 nm, which can be seen from Fig. 16. In this study, the authors demonstrated that the stem cell behavior is a function of pillar heights and not of the size and spacing. The parameter will serve as another important guiding parameter to consider when designing a scaffold for MSC differentiation.

3. MSCs on Complex Structures There have emerged three different approaches toward the formation of nanofibrous materials: self-assembly, electrospinning, and phase separation. A substantial three-dimensional (3D) network of nanofiber scaffolds formed by the self-assembly of peptide-amphiphile (PA) molecules was fabricated by Hosseinkhani et al. (2006). The chemical structure of PA contains RGD, a glutamic acid (Glu) residue, four alanine (Ala), and three glycine (Gly) residues (A4G3), followed by an alkyl tail of 16 carbons. This PA is capable of self-assembly into sheets, spheres, rods, disks, or channels depending on the shape, charge, and environment. The fate of rat MSCs seeded on these nanofibers was then investigated. Improved cell attachment, significantly higher alkaline phosphatase activity, and higher osteocalcin were observed for the rat MSCs seeded on PA with RGD compared to those grown on tissue-culture plastic (Fig. 17). Commendable studies have also been carried out with differentiation of MSCs on electrospun fibers. Dang et al. fabricated a thermosensitive electrospun fiber scaffold that could be coupled with the drug encapsulation capacity of electrospinning to serve as a multifunctional platform to encapsulate growth factors as well as providing suitable topographical cues (Dang et al., 2007). Thermosensitive polymers have a lower critical solution temperature (LCST) below which they are hydrophilic and water soluble. When the temperature is increased above the LCST, they become hydrophobic and water insoluble. In this study, hydroxybutyl chitosan, a thermosensitive material, was electrospun into nanofibers, with an average diameter of 437 nm, and myogenic induction of human MSCs was observed in response to topographical features without differentiating medium. This technique is beneficial in enabling us to differentiate MSCs as well as harvest the differentiated cells as a cell sheet without disrupting the cell–ECM interaction (Fig. 18). The significance of a combinatorial approach of using both topographical cues and chemical cues in order to upregulate the neuronal markers has also been demonstrated by Prabhakaran et al. In their study, electrospun poly(L-lactic acid)-co-poly-(3-capro lactone)/collagen (PLCL/Coll) nanofibrous scaffolds were used to study the transdif ferentiation of bone marrow-derived MSCs. The fiber diameter was in the range of 230 ± 31 nm and neural induction factors such as BDNF, EGF, and b-mercaptoethanol were used (Prabhakaran et al., 2009). The mechanism behind transdifferentiation in response to nanotopographical cues has yet to be unraveled. Nonetheless, these studies

265

11. Nanotopography/Induced of Stem-Cell Differentiation

Tubulin

Osteopontin

Osteocalcin

Control

(B) Control

(A)

50 μm

15

15

55

55

100 μm

100

100

Fig. 16 Cytoskeleton and focal adhesion staining in hMSCs cultured on planar control and nanostructured Ti surfaces. (A) Actin microfilaments are stained red, tubulin microtubules are stained green, and nucleus is stained blue. Cells cultured on 15-nm-high structured surfaces display a well-defined cytoskeleton and large focal adhesion sites, whereas on the higher structured surfaces cells have a less organized cytoskeleton and fewer, smaller focal adhesions. (B) Osteopontin and osteocalcin expression in hMSCs cultured on control and titania nanostructures. Actin microfilaments are stained red, osteopontin or osteocalcin is stained green, and nucleus is stained blue. The arrows show sites of bone matrix nodule formation. Matrix nodule formation is most advanced on the 15-nm-high titania structures (Terje et al., 2009).

contribute to materializing the enormous potential for the generation of functional neurons from patient-derived MSCs. Having learnt about the topographies and how geometry could affect proliferation and differentiation of MSCs, it is interesting to understand how symmetry of a topographical

Benjamin Kim Kiat Teo et al.

266

(A)

(B)

(C)

(D)

25

40

20

Osteocalcin content (ng/cells)

ALP activity (mM/cells)

* *

15 * 10 5 0

*

* 30

20

10 * 0

1

2

3

Incubation time (week)

4

1

2

3

Incubation time (week)

Fig. 17 Histological cross sections of MSC attached to the self-assembled PA nanofibers (A) without RGD and (B) with RGD incorporation, 4 weeks after the culture in the osteogenic medium. (C) and (D) show the time course of alkaline phosphatase activity and osteocalcin contents of MSC cultured in the bone differentiation medium in self-assembled PA nanofibers with RGD incorporation (▲), self-assembled PA nanofibers without RGD incorporation (■), and tissue-culture plate (●). As controls, the cells were cultured in the normal medium in the same self-assembled PA nanofibers with RGD incorporation (Δ), self-assembled PA nanofibers without RGD incorporation (□), and tissue-culture plate (○). Asterisk in (A) and (B) refer to the newly formed bone. * p < 0:05; significant against the ALP activity (C) or osteocalcin content (D) of cells grown in the bone differentiation medium in self-assembled PA nanofibers without RGD incorporation at the corresponding week. Extracted from Hosseinkhani et al (2006; Figures 7, 8, and 9).

pattern can affect stem cell fate. Dalby and coworkers have shown that random topo graphies were better than ordered ones in inducing the MSCs to differentiate toward an osteogenic lineage (Fig. 19) (Dalby et al., 2007a). In yet another study by the same group, a controlled disorder nanopit topography was fabricated by electron beam lithography (Kantawong et al., 2009). It was termed NSQ50 with pits in a near-square arrangement, 300-nm center–center spacing and was used as a scaffold to direct osteoblast differentia tion of human bone marrow osteoprogenitor cells. Their results showed that stem cells were sensitive to nanostructure regularity, and that the symmetry of structures had significant influence on stem-cell differentiation. In addition to the mesenchymal lineage, MSCs have also been shown to acquire atypical neuronal-like phenotypes. There has been one study by Khor et al. who showed that a complex topography pattern also described as “worm-like” is able to support the proliferation of human

267

11. Nanotopography/Induced of Stem-Cell Differentiation

(A)

(B)

5 μm

(C)

5 μm

Nanofibers H

H/C

H-Film

TCPS

B-actin Pax7 Myogenin

Fig. 18 Scanning electron microscopy images of (A) HBC nanofibers and (B) HBC/collagen blend nanofiber. (C) Gene expression analysis of human MSC after 2 week culture period on thermosensitive nanofibers. H refers to HBC nanofibers, H/C refers to HBC/collagen blend nanofibers, and H-film refers to HBC films. Human MSC cultured on HBC films and TCPS act as control surfaces. Extracted from Dang et al. (2007; Figures 1 and 3).

MSCs. These patterns were fabricated by the assembly of poly(styrene)-block-poly (2-vinylpyrindine) diblock copolymer material on mica. This study showed that these “worm-like” patterns of average height 3.2 nm and average width 160 nm yielded even higher numbers of cell aggregates as compared to a pillar-like topography on a similar material. The morphology of these human MSCs, however, was quite different from that of the pillar-like surface (Fig. 12) and smooth surface, exhibiting a more elongated morphology with parallel actin bundles and smaller adhesion sites. It was postulated that these worm-like patterns forced the cells into high tension, which allowed the microtubules to distribute homogenously throughout the cell, thus better supporting smaller adhesion sites for enhanced motility (Khor et al., 2007). In this section, we have tried to recapitulate the importance of topography in MSC proliferation and fate regulation by various topographies and showed that topography was able to supersede the effect of biochemical cues but the overall outcome seems to be dependent on the species, cell type, and topography. The next section will discuss about neural stem cells and neural progenitor cells giving a concise overview of some interesting studies to date in order to understand their interaction with topographies. C. Neural Stem Cells/Neural Progenitor Cells Neural stem cells are multipotent stem cells and are present in both developing and adult central nervous systems. The NSCs are capable of self-renewal and differentia tion into astrocytes, oligodendrocytes, and neurons while the NPCs are lineage restricted and only capable of fewer divisions. The three major types of progenitors found in the adult human brain are ventricular zone neural progenitors, hippocampal neuronal progenitor cells, and white matter glial progenitors. Isolation, expansion, and

Benjamin Kim Kiat Teo et al.

268

1 μm

(A)

(B)

200 μm

1 μm

1 μm

1 μm

(C)

(D)

(E)

(H)

(I)

(J)

OPN

(F)

(G)

OCN

Fig. 19 OPN and OCN staining of osteoprogenitors after 21 days of culture. The top row shows images of nanotopographies fabricated by electron beam lithography. All have 120 nm diameter pits with hexagonal, square, displaced square 50 (±50 nm from true center), and random placements. Osteoprogenitors cultured on (A, F) control. Note the lack of positive OPN and OCN stains. (B, G) Hexagonally arranged nanopattern. Note the loss of cell adhesion. (C, H) Nanopattern with square arrangement. Note the reduced cell numbers compared with the control, but some OPN and OCN positive cells. (D, I) Nanopattern with near-square displacement named DSQ50. The bone nodule formation is shown by white arrows. (E, J) on randomly arranged nanopattern. It shows cells expressing OPN and OCN. Actin microfilaments are stained red while osteopontin/osteocalcin is stained green. Adapted from Dalby et al. (2007a; Figure 1).

in vitro differentiation of these stem cells and progenitors have been studied exten sively in the last two decades (Goldman, 2006; Johansson et al., 1999a, b; Morshead et al., 1994; Roy et al., 1999; Scolding et al., 1998; Steven et al., 2002; Svendsen et al., 1999). They can be grown as either neurospheres or adherent culture when grown in media containing basic fibroblast growth factor and epidermal growth factor (Kantawong et al., 2009). NSCs and NPCs possess the potential for differentiation into transplantable neurons and so their identification and isolation gives the hope for treating debilitating neurodegenerative disorders. Many studies have been done to direct differentiation of NSCs/NPCs to neural cells using coculture systems or biochemical cues like fetal bovine serum for NSC differentiation to astrocytes; nerve growth factor, retinoic acid, and laminin for neuronal differentiation; and a combination of Sonic

11. Nanotopography/Induced of Stem-Cell Differentiation

269

hedgehog, fibroblast growth factor 8, and brain-derived neurotrophic factor for dopa minergic neuron differentiation (Boote Jones and Mallapragada, 2007; Obayashi, et al., 2009; Park et al., 2009). Our focus in this section will be on the role of topography in influencing the proliferation and directing differentiation of NSCs and NPCs.

1. Neuronal Differentiation on Grating-Like Structures Topography often works synergistically with the appropriate biochemical cues as seen in the earlier section in a study conducted by Yim et al. Similarly, Recknor et al. have studied the effect of topography on neural differentiation of adult rat hippocampal progenitor cells (AHPCs) on micropatterned polystyrene substrates, chemically mod ified with laminin (Recknor et al., 2006). When these AHPCs were cocultured along with the astrocytes on the micropatterns, the percentage of neuronal differentiation and the neurite outgrowth showed a clear increase, suggesting a synergistic effect of nanotopography and the biochemical factors secreted by the astrocytes. The orientation of the AHPCs on the micropatterned substrate seemed to depend on the alignment of the astrocytes with the micropattern (Fig. 20).

2. NPCs on Pillar-Like Structures In an interesting study to understand the developmental fate of neural precursor cells and to investigate the ECM components and signaling factors involved, Soen et al. created a protein microenvironment using a noncontact arrayer (BCA arrayer, Perkin– Elmer) on a glass surface (Fig. 21). Primary human neural precursor cells isolated from whole cortex of a 22-week human fetus were used for the study. Each spot of the protein mixture was 400 µm in size capturing around 100 cells creating eight replicates of each combination to derive a statistically strong data. From immunostaining experi ments, it was analyzed that a combination of Notch ligand, jagged-1, rhDLL-4, TGF-b, CNTF, and BMP4 encourages glial morphology while a combination of Wnt3a and laminin has a neurogenic effect (Soen et al., 2006).

3. NSCs on Well-Like Structures Well-like topographies are able to support the proliferation of NSCs but when placed in differentiation culture media, the well-like topographies can promote differentiation. Leipzig et al. created a well-like topography from photopolymerizable methacrylamide chitosan hydrogels coated with laminin. The polymers of the hydrogel were tightly cross-linked to form a porous structure. They focused more on the Young’s elastic modulus of the hydrogel rather than the size of the pores, but it was noted that as the stiffness increased, the pores became smaller. This group showed that proliferation of rat NSCs could occur when Young’s elastic modulus was less than 10 kPa and was maximal at 3.59 ± 0.51 kPa. They also showed that the stiffness of the hydrogel could change the type of differentiation being induced when coupled with suitable differ entiating media. When the hydrogel stiffness was <1 kPa, neuronal differentiation was

Benjamin Kim Kiat Teo et al.

270

(A)

(B)

(C)

Fig. 20 SEM images of adult rat hippocampal progenitor cells (AHPCs) cultured on a poly-L-lysine (100 mg/ml)- and laminin (10 mg/ml)-coated polystyrene substrates. (A) On the micropatterned substrate, AHPC’s elaborate processes were aligned in the direction of the grooves. (B) On the nonpatterned (smooth) side of the substrate, AHPC processes were oriented randomly. Images were taken from cultures at 7 days after plating. (C) SEM images of astrocytes and AHPCs cocultured on polystyrene substrates. AHPCs are in contact with and extending along astrocyte cytoskeleton filaments with processes aligning in the groove direction. AHPCs are also in contact with the grooves. Adapted from Recknor et al. (2006; Figures 2 and 4).

enhanced while oligodendrocyte differentiation was enhanced when the hydrogel stiffness was >7 kPa; however, to promote maturation and myelination of the oligo dendrocytes, the substrate stiffness needed to be decreased to <1 kPa. Astrocyte differentiation appears to occur when the stiffness was <3.5 kPa; however, even at this stiffness the astrocytes constitutes a very small percentage of the cell population. This study suggests that the spatial presentation of the ECM-binding ligands to the integrin receptors of the cell is altered correspondingly with the compliance of the substrate, consequently changing the binding affinities between the ligand and the receptors, which would affect the transcriptional regulation of the NSC fate to pro liferate or to differentiate (Leipzig and Shoichet, 2009).

271

11. Nanotopography/Induced of Stem-Cell Differentiation

Glia (GFAP+) Neuron (TUJ1+)

Human neural

progenitor cell

3 Days of differentiation

Notch Notch ligand Frizzled Wnt

Laminin + Notch ligand

Laminin + Wnt

EC

M/

GF

ar

ra

y

~400 μm

650 μm 1.45 mm

8 replicates/combination

Fig. 21

Schematics of a molecular microenvironment array experiment. Arrays of premixed combinations of signaling molecules were printed using a noncontact, piezoelectric arrayer. Each spot typically included a combination of laminin and one or more recombinant proteins that were previously implicated in cell fate decision processes. Each combination was printed in two separate groups of four replicates. Bipotent human neural precursors were captured onto the printed spots by adherence to the laminin. Adapted from Soen et al. (2006; Figure 1A).

4. NSCs on Complex Structures The effect of topography as contact guidance has been studied extensively with a focus on axonal guidance and neuronal regeneration. It was observed that in addition to serving as contact guidance, topography can also affect the proliferation and differ entiation of NSCs. One study done by Christopherson et al. found that electrospun polyethersulfone fiber mesh coated with poly-L-ornithine and laminin will allow for the maintenance of rat hippocampus-derived adult NSC (Christopherson et al., 2009). This study showed that as the diameter decreased from 1452 to 283 nm, the proliferation and spreading of the NSCs increased and aggregated less. However, the proliferation rate of the NSCs on the fiber mesh was still lower than that of the tissue-culture polystyrene flat surface control. The fibers instead were better at promoting differen tiation when the NSCs were cultured under differentiation conditions, where they were

Benjamin Kim Kiat Teo et al.

272

cultured in the presence of fetal bovine serum and retinoic acid. They analyzed the oligodendrocyte and neural differentiation of rat hippocampus-derived neural stem cells seeded on nanofibers of various sizes. It was observed that in the presence of 1 µM retinoic acid and 1% fetal bovine serum, rat NSCs showed a 40% increase in oligodendrocyte differentiation on 283 nm fibers and 20% increase in neuronal differ entiation on 749 nm fibers, in comparison to tissue-culture polystyrene surface. This study showed the significance of fiber topography and fiber dimension in regulating rat NSC differentiation as well as proliferation (Fig. 22). D. Hematopoietic Stem/Progenitor Cells HSPCs are multipotent stem cells predominantly found in the bone marrow of adults; however, other sources of HSPCs include the fetal liver, the fetal spleen, and (B)

(A)

(C)

LEI

(D) 0.7

TCPS 283-nm fibers 749-nm fibers

Fraction Positive Staining

0.6 0.5 0.4 0.3 0.2 0.1 0.0 Nestin

Tuj-1

RIP

GFAP

Fig. 22 SEM characterization of electrospun fiber meshes.(A) 283 nm, (B) 749 nm, and (C) 1452 nm. All images were obtained at 11,000; scale bars = 1 µm. (D) Immunofluorescence analysis of rat NSCs cultured on various substrates in the presence of 1 µm retinoic acid and 1% fetal bovine serum for 5 days. Cell nuclei (blue) are stained using DAPI as a counter staining. Quantification of staining results is shown. Adapted from Christopherson et al. (2009; Figures 1 and 2).

11. Nanotopography/Induced of Stem-Cell Differentiation

273

the umbilical cord blood (Peerani and Zandstra, 2010). HSPCs are capable of selfrenewal and differentiation into all blood cells of the myeloid lineage (e.g., macro phage, erythrocytes, dendritic cells) and of the lymphoid lineage (e.g., T-cells, B-cells) (Peerani and Zandstra, 2010). HSPCs are currently used for transplantations to treat hematological disorders and to support the treatment of malignant diseases (Peerani and Zandstra, 2010). Unfortunately, only a less number of HSPCs are able to be extracted and collected from their sources, as such ex vivo expansion of HSPCs is necessary to use HSPCs for widespread treatments. The expansion of HSPCs is notoriously difficult to achieve in vitro as HSPCs will differentiate when removed from their niche. In the last several decades, however, scientists have persevered and developed two methods of culturing HSPCs ex vivo with some success. The first is to separate the HSPCs into Linþ and Lin– cells and then take the Lin– cells and culture them in suspension within a cell culture bag with serum-free media and various growth factors (Kirouac et al., 2009). As the culture progresses, the Linþ cells are removed from culture by magnetic separation, thus keeping the Lin cell population pure and preventing the Linþ cells from inducing the Lin– cells to differ entiate. The second method is to culture the HSPCs on tissue-culture plates, which produces a population with a far greater number of differentiated cells (Kirouac et al., 2009). Both methods require the use of an orbital shaker.

1. HSPCs on Complex Structures As stated above, the culture of HSPCs proves to be challenging. Chua et al. has explored the use of nanofibers to culture HSPCs while maintaining the HSPC pheno type (CD34þCD45þ) (Chua et al., 2006). They have shown that human umbilical cord HSPCs could be cultured on aminated electrospun polyethersulfone nanofiber scaf folds (Fig. 23). Chua et al. have shown that aminated nanofiber mesh and films with average fiber diameters of 529 ± 114 nm can be used to expand the CD34þCD45þ cells by 195-fold and 178-fold, respectively, which is an improvement on the tissue-culture polystyrene culture system that only expanded the CD34þCD45þ cells by 50-fold. Chua et al. then continued their studies to show that they can improve the expansion of the CD34þCD45þ cell population by functionalizing these scaffolds with poly (acrylic acid) grafting and by performing the conjugation of the amino groups with different spacers—ethylene, butylenes, and hexylene (Chua et al., 2007). When the HSPCs are cultured on scaffolds with aminoethyl- or aminobutyl-conjugated nanofi bers, they are able to obtain a total cell expansion of 773- and 805-fold, respectively, where the percentage of CD34þCD45þ cells were 25.9 and 29.2%, respectively. Culture on scaffolds of aminohexyl-conjugated nanofibers only produced a total cell expansion of 86-fold, but the CD34þCD45þ cell population constituted 41.1% of the total population. All three of these substrates were an improvement over the unmodi fied nanofiber mesh scaffold, which had a total cell expansion of 85-fold and a cell population that was 13.3% CD34þCD45þ; and the tissue-culture polystyrene control, which had a total cell expansion of 895-fold but only a cell population that was 5.9% CD34þCD45þ (Fig. 24). This group then concluded that the substrates, which

Benjamin Kim Kiat Teo et al.

274 (A)

(B)

(C)

(D)

(E)

(F)

Fig. 23 SEM images of human cord blood HSPCs after a 10-day expansion culture on aminated PES nanofiber mesh (A–C) and on aminated PES film (D–F) at various magnifications. Abundant distinct, circular cell colonies are evident on the aminated nanofiber scaffold (black arrows). Filopodia extend from the cells and interact with the aminated nanofibers (white arrows). On aminated film, fewer cells are adherent without colony formation; cells appear to attach along cracks. Extracted from Chua et al. (2006; Fig. 5).

promoted cell adhesion, also enhanced the preservation of the CD34þCD45þ pheno type and the primitive characteristics of cord blood CD34þ HSPCs.

E. Other Progenitors The culture of progenitors is quite varied and is dependent on the type of progenitor that is used. Progenitors are often used to create tissue grafts; as such, many attempts have been made to seed progenitors onto 3D scaffolds with properties that mimic the natural environment. The following studies will detail progress that has been made to determine which topographies will yield the best results for adhesion, proliferation, and differentiation of progenitors for future regenerative medicine work.

1. Osteogenic Progenitors on Grating-Like Structures Grating-like topographies provide cues to the progenitors for differentiation. Lovmand et al. demonstrated that murine osteoblastic cells underwent differentiation when cultured on a grating-like topography on silicon (Lovmand et al., 2009). They used the sharkskin structure from the BioSurface Structure Arrays, which consist of alternating lines of bars with lateral sizing ratios of 1:4 and 2:3, where each bar and line

275

11. Nanotopography/Induced of Stem-Cell Differentiation

Total cell number CD34+CD45+ cell number

Fold expansion of total nucleated cells or CD34+CD45+ cells

1000 800 600 350 300 250 200 150 100 50 0

TCPS Unmodified PAAc

EtDA

BuDA

HeDA

Fig. 24 Fold expansion of total nucleated cells and CD34þCD45þ cells following a 10-day culture of human cord blood hematopoietic stem/progenitor cells on tissue-culture polystyrene surface and various polyethersulfone nanofiber scaffolds. Bars represent mean ± SD of three to eight independent experiments, each conducted in triplicates. BuDA, 1,4-butanediamine; EtDA, 1,2-ethanediamine; HeDA, 1,6-hexanediamine; PAAc, poly(acrylic acid); TCPS, tissue-culture polystyrene. Extracted from Chua et al. (2007; Figure 2).

was spaced by 1 µm (Fig. 2). Lovmand et al. varied the lateral size unit from 1 to 8 µm and found the general trend that with decreasing lateral size unit there was an increase in mineralization and the appearance of differentiation markers (Fig. 25).

2. Osteogenic Progenitors on Pillar-Like Structures Studies done with mouse progenitor cells on pillar-like topographies do not provide a clear indication that pillar-like topographies support proliferation; however, they do indicate that when grown on pillar structures, proliferation decreases with increasing feature height. Lovmand et al. demonstrated this using an array of pillar-like topographies on silicon wafers from the BioSurface Structure Arrays (Lovmand et al., 2009). These pillars varied in shape (circle, square), placement (aligned, staggered), feature size (1, 2, 4, 6 µm), spacing distance (1, 2, 4, 6 µm), and feature height (0.6, 1.6, 2.4 µm) (Fig. 4). In general, they showed that proliferation decreased with increasing feature height and decreasing feature size. They also showed the converse in differentiation, where differentiation increased with increasing height, decreasing feature size and decreasing spacing distance. Unlike Lovmand et al., Beihl et al. worked with larger feature sizes and mouse ESC progeny. Beihl et al. showed that mouse ESC who had previously undergone 4 days of

Benjamin Kim Kiat Teo et al.

276 Two weeks of induction

Osteopontin Osteocalcin

DAPI

K7, K8 Control

K5, K6 Control

K3, K4 Control

K1, K2 Control

Alizarin red

Fig. 25 Mineralization, OPN, OC, and DAPI after 2 weeks of induction. MC3T3 cells were seeded upon a BioSurface Structure Array (BSSA) library of vertical dimension of Z = 2.4 mm and induced to mineralize for 2 weeks. The cells were subsequently stained for OPN (green), OC (red), and DAPI (blue). Following fuorescence microscopy for OPN, OC, and DAPI, the wafer was stained with alizarin red. Images showing all K-structures together with unstructured control surface starting with K1, K2 above unstructured surface at the edge of the BSSA library ending with K7, K8 above unstructured surface (last row). Adapted from Lovmand et al. (2009; Figure 3, cropped).

differentiation proliferated at half the rate when on PDMS pillars that were 15 µm high as compared to flat areas of PDMS (Biehl et al., 2009). However, when the pillar height was decreased to 5 µm, the proliferation rates were equivalent (Fig. 26). Unfor tunately, the study did not continue to explore the effect of shorter pillars on the proliferation rate of mouse ESC progeny; however, when considered with the work done by Lovmand et al., there does appear to be an indication that shorter feature heights may improve the proliferation of these mouse ESC progeny.

3. Osteogenic Differentiation of Stromal Cells on Well-Like Structures Well-like topographies also support the proliferation of mouse bone marrow stromal cells on nanoporous aluminum oxide (alumina) created by anodization (Popat et al., 2007). Popat et al. found that nanoporous alumina with an average pore diameter of 72 nm would increase the proliferation of the mouse bone marrow stromal cells by 45%, as compared to amorphous alumina. In addition, the well-like topography also

277

11. Nanotopography/Induced of Stem-Cell Differentiation

*

Proliferation relative to flat

1.2 1 0.8

*

Near

0.6

Far

0.4 0.2 0 5 μm

15 μm Height of projections

Fig. 26 Microprojection height of 5 µm does not attenuate mouse ESC-derived progeny proliferation. Histogram showing infuence microprojection height has on proliferation compared with cells grown on fat membranes. Data are mean ± SE. *P < 0.01. Extracted from Biehl et al. (2009; Figure 3c).

increased viability and adhesion by 45%, alkaline phosphatase activity by 30%, and matrix production by 50%. Unfortunately, the study did not expand the investigation to include nanoporous alumina with different pore diameters, but this study does indicate that the well-like topography may be a good support for stromal cell proliferation.

4. Osteogenic Progenitors on Complex Structures Using chemical deposition techniques, Elias et al. compared the effect of large and small multiwalled carbon nanofibers on human osteoblasts (Fig. 27) (Elias et al., 2002). The nanofibers with 100 nm diameters or less were able to increase the

PR-24 PS (nanophase) (60 nm diameter without a pyrolytic outer layer)

PR-23 PS (conventional) (125 nm diameter without a pyrolytic outer layer)

Fig. 27 High-magnification scanning electron micrographs of carbon fiber compacts. Representative scanning electron micrographs of PR-24 PS (nanophase) and PR-23 PS (conventional) carbon fiber compacts. Original magnification = 5000; 5 kV; scale bar = 1 µm. Extracted from Elias et al. (2002; Figure 1c,d).

278

Benjamin Kim Kiat Teo et al.

proliferation rate, the synthesis of alkaline phosphatase, and the deposition of extra cellular calcium. Unfortunately, Elias et al. did not find an optimal nanofiber diameter size for the maximal proliferation rate, but they did show that larger nanofibers had decreased proliferation rates.

IV. Cell Shape: A Regulator of Biological Processes For anchorage-dependent stem cells, the topographical approaches to control stem cell processes like proliferation and differentiation involve the precise control of cell spreading. In a pioneering study by Folkman and Moscona in 1978, cell shape was found to be linked intimately to DNA synthesis and growth in cells (Folkman and/or Moscona, 1978). Changes in cell shape were also deemed to be a possible mechanism for myocardial development (Manasek et al., 1972) and also the growth and differ entiation of capillary endothelial cells (Ingber, 1991). With regard to stem cells, the previous sections have shown numerous studies that stem cell lineage can be dictated through the use of engineered topographical features (both geometric patterning and/or topographical cues) to regulate their cell shape. The manipulation of the physical stem cell shape using engineered extracellular topographical features can result from subsets of effects including altered adhesive interactions although the exact mechanism remains unknown. It will be important to recognize that the effects of cell shape on intracellular signaling pathways appear to be of higher complexity than adhesion signaling alone (Meyers et al., 2006). In the later sections of this chapter, however, we will discuss the components of the adhesion signaling pathway with a focus on the cytoskeleton and its postulated mechanism in translating these topography-derived signals in the ECM into patterns of gene regulation in the nucleus for stem-cell differentiation. The discussion on the hypothesized mechanism is meant for a more complete understanding of the chapter while interested readers can obtain more detailed information on cell nucleus and the mechanism in mechanosensing and gene regulation in the chapters 1–6 and 9 of the book. Stem-cell differentiation can result from biochemical cues, structural cues, or likely a combination of both. The structural cues from topography change the orientation of the cells and reorganize its body, shape, and functional state (Bissell et al., 1999; Brunette and Chehroudi, 1999; Dunn and Brown, 1986). Studies employing simplistic two-dimensional (2D) topography models to mimic native ECM demonstrated that the mechanical cues do play an important role in regulating stem-cell differentiation. Intriguingly, the cells must have an intrinsic fundamental mechanism to sense the underlying substrate topography, generating mechanical signals that are translated by intracellular signaling pathways before ultimately regulating the genomic expression and cell function (Lutolf and Hubbell, 2005). This is also known as mechanotransduc tion. The postulated theories to model how modifications in the mechanical environ ment can elicit a cellular response are extensively discussed in the respective papers (Bissell et al., 1982; Ingber, 2003a, b; Ingber, 2008; Nelson and Bissell, 2006); however, one important aspect of the theories is the existence of a continuous physical

11. Nanotopography/Induced of Stem-Cell Differentiation

279

connection from the ECM to the nucleus. The presence of such a physical continuity will allow cellular structural rearrangements to result in gene regulation, such as topography-induced stem-cell differentiation. A. Mechanotransduction: A Direct Mechanism Too Simple As simple as it may sound, mechanotransduction involves a complex interplay of different cellular organelles and components (e.g., focal adhesion (FA) and cytoskele ton) that by themselves are highly dynamic in vivo, making the process more con voluted (Geiger et al., 2009). However, advancement of experimental techniques has helped to provide increasing evidence to the mechanism and it appears that cellular components such as the integrins, FAs, and cytoskeleton organization collectively play important roles in topography-induced cellular behavior (Dalby et al., 2003; Geiger et al., 2009; Riveline et al., 2001). The topographical extracellular signals have to be transduced from the ECM to the nucleus, where the genome can be regulated as a response to the extracellular signals. Although various other indirect and chemical mechanisms have also been postulated (e.g., G-proteins, ion channels), we have limited our discussion to a direct transmission of forces (via cellular organelles) to the nucleus as a mechanism for gene regulation (Maniotis et al., 1997). Under the effect of nanotopography, the physical substrates do exert differential mechanical stress onto the cells, as indicated by the cytoskeletal reorganization. In the previous section, we have presented an overview of the use of patterned substrates as a form of physical cue to regulate various stem-cell differentiation lineages. In this section, we will briefly introduce the important components of the mechanotransduction machinery, namely the integrins, FAs with a particular focus on the cytoskeleton that translates these topographical cues into secondary pathways, ultimately affecting stem-cell differentiation. B. Integrins and Focal Adhesions: Inside Out and Outside In Anchorage-dependent cells are able to anchor onto the underlying substrate. Various different types of adhesions exist between the cells and the ECM, differing in size, shape, and biochemical composition. It is therefore no wonder that with a large diversity of adhesions, they perform different and specific functions in cells (Zamir and Geiger, 2001). These include cell–cell adhesions (e.g., cadherins) and cell–matrix interactions. The cell–matrix interactions are probably the most fundamental adhesions involved in stem cells’ response to nanotopography. Cells anchor to the ECM through adhesions that are mediated by integrins. Integrins are heterodimeric transmembrane cell adhesion proteins that bind to specific motifs present on the ECM (Lutolf and Hubbell, 2005; Ruoslahti and Obrink, 1996). In relation to nanotopography-induced stem-cell differentiation, the change in the physi cal structure of the underlying substrate can influence this clustering of integrins and other adhesion molecules. Arnold et al. were able to use precise nanoscale adhesive islands to establish a relationship between integrin clustering, FA formation, and actin stress fibers which influenced the adhesion and spreading of cells (Arnold et al., 2004).

Benjamin Kim Kiat Teo et al.

280

In fact, the maximal distance where two individual integrin molecules can bind is in the range of 50–70 nm, showing the importance of integrin clustering in the regulation of integrin-mediated signal transduction (Arnold et al., 2004). Using a novel nanoscale ligand spacing gradient, the same group suggested that cells expressed delicate sensitivity to interparticle spacing of about 1 nm, demonstrating the sensitivity of the cellular sensing mechanism (Arnold et al., 2008). It was also suggested that the sensitivity to minute variations has physiological implications where in vivo ECM collagen fiber has a 67 nm banding periodicity (Jiang et al., 2004) and fibronectin fiber presents nanoscale epitopes (Little et al., 2008; Smith et al., 2007). Upon binding to the ECM ligands, integrins cluster and activate specific signaling pathways that are important for various cellular functions such as migration, proliferation, and differentiation. It is therefore likely that the nanotopographical cues modify the extent and activation of integrin clustering as the initial step in subsequent signal transduction in stem cell genomic regulation (Fig. 28). In fact, stem cells in the central nervous systems seem to have high levels of b1-integrins (Campos, 2005),

ECM fibers

Plasma membrane

Integrin receptors

FAK

FAK

Src

Src Paxillin

Ras

Crk

ERK

JNK

FAK

PIX

FAK

PI3K

Src

ILK

p190

Rac Cdc42 Rho

Akt

Fig. 28 Examples of biochemical signaling pathways from ECM to the nucleus. A large number of pathways induced by integrin activation implicate cytoskeletal alterations (e.g., ERK, Rac, Rho) and involve mechanosensitive proteins like focal adhesion kinase (FAK), ultimately leading to differential gene expression. Figure adapted from Gjorevski (2009).

11. Nanotopography/Induced of Stem-Cell Differentiation

281

suggesting that the modulation of integrin expression through topographical cues can regulate stem-cell differentiation. However, the role of integrins in gene regulation is complex because these receptors participate in both the sensory and operational functions of the cellular machinery, also commonly known as the outside-in (sensory) and inside-out (operational) signaling activities. The observed dynamics of the integ rins on topography can be due to both the response of the cell to the underlying ECM and a secondary effect of the actin cytoskeleton–FA feedback machinery. The complexity in the feedback network connecting the sensory and operational functions is also reflected in the highly intertwined integrin adhesome network (Zaidel-Bar et al., 2007). One of the most important integrin-mediated adhesions involved in mechanotrans duction is the FA (Burridge et al., 1988; Geiger et al., 2009). FAs play an important role in linking actin cytoskeleton to the transmembrane integrins (Geiger et al., 2001; Zaidel-Bar et al., 2007). The exact molecular nature of FAs is unclear; however, FAs are composed of a large complex network of adhesion molecules (Burridge et al., 1988; Zamir and Geiger, 2001). Some of the important structural proteins include talin, vinculin, and FA kinase (FAK). The formation and maturation of the FA are driven by feedback from the actin cytoskeleton and integrin (Geiger et al., 2009). Briefly, talins connect integrin dimmers with the actin filaments before the recruitment of additional components in the complex (Gingras et al., 2008). The subsequent maturation of the complex requires a contractile pulling force generated by the actin–myosin machinery (Choi et al., 2008; Even-Ram et al., 2007), which will be discussed in more detail in the later section. For a more detailed description of the steps and mechanism of the FA assembly, readers can refer to an excellent review by Geiger et al. (2009). Topography-imparted mechanical force thus plays an important role in the promo tion of FAs. An important component of FA, vinculin, has been shown to trigger the clustering of activated integrins (Humphries et al., 2007). The binding of vinculin to talin during the initial stages of FA assembly is shown to be force mediated and the binding site for vinculin requires unfolding that is achieved by mechanical forces (Gingras et al., 2006). Similarly for p130Cas and fibronectin, such mechanical forces can expose cryptic sites present in the fibronectin protein sequence that becomes available for interaction with cell surface receptors (Garcia et al., 1999; Sawada et al., 2006; Smith et al., 2007). It seems that the actomyosin contractile stress that actin exerts on the adhesions is essential for the formation of FA. The global forces that are experienced by the cells under nanotopographical cues can alter such forces that the FAs are experiencing, ultimately changing their differentiation lineage. All these studies indicate that FAs play an important role in mechanotransduction, possibly also in regulating nanotopography-induced stem-cell differentiation.

C. Cytoskeleton: A Manifestation of Cells’ Interpretation of Topography

1. Cell Exerting Forces on the Underlying Substrate Force generation in the cytoskeleton is required for cell adhesion and migration on ECM. The cytoskeleton consists of actin filaments (microfilaments), microtubules, and

282

Benjamin Kim Kiat Teo et al.

intermediate filaments. They form a network of filamentous protein that extends throughout the cell cytoplasm in eukaryotic cells. The cytoskeleton has been well studied where it is involved in cellular metabolism and movement ranging from mitosis to migration (Berrier and Yamada, 2007; Delon and Brown, 2007) and an increasing amount of evidence has demonstrated the importance of the cytoskeleton in stem-cell differentiation (Engler et al., 2006; McBeath et al., 2004). Different groups have tried to characterize the cellular forces that cells exert on the underlying substrate. An early method of cellular force measurement technique uses a silicon rubber membrane that deforms and wrinkles upon the exertion of force by adherent cells (Harris et al., 1980). Another more recent study employs the use of an array of vertical elastomeric microcantilevers that bend under the exertion of contractile force by cells (Tan et al., 2003). The contractile forces present in the actin stress fibers of the cytoskeleton appear to be integral in modulating cellular functions such as adhesion, migration, and stem-cell differentiation. The generation of contractile forces in nonmuscle cells is mediated by a class of motor proteins, the nonmuscle myosin II (NM II). Briefly, these NM II molecules assemble into filaments which then bind to actin through their head domains. A conformational change elicited from ATPase activity enables the contraction of the actin filaments. These thick bundles of actin–myosin cross-linked together are also known as stress fibers (Fig. 29B). A recent concept suggested that cells use actomyosin contractility for a two-way interaction with the ECM. This cell contraction through the stress fibers will be resisted by the matrix at the sites of integrin sites with the subsequent recruitment of additional molecules for FA formation. Cells’ response to topography is thus not passive but rather they often are able to “tune” their mechanical properties through the dynamic remodeling of the actin cytoskeleton. The balance of tension forces at these interfacial sites allows the cell to sense the ECM (Fig. 30). This tension-mediated signaling is manifested in the reorganization of actin microfila ments or stress fibers with the observed alignment of these stress fibers to nanogratings and differential organization to the other surface features (Dalby et al., 2003; Fujita et al., 2009; Yim et al., 2007). This cellular force sensing in turn leads to altered levels of Rho guanosine triphosphatase (GTPase) and mitogen-activated protein (MAP) kinase activity as down stream biochemical signals for stem cell gene regulation (Fig. 28). In a study by Engel et al., the use of matrices with different elasticities regulates the differentiation of MSCs into different lineages (Engler et al., 2006). The use of specific NM II inhibitor blebbistatin blocks all elasticity-directed lineage specification without strongly affecting cell function and shape significantly, providing evidence of cytoskeletal force generation in ECM sensing. Human ESCs also appeared to be aligned and elongated when they were cultured on nanometer-scale gratings (Gerecht et al., 2007). It seems that the cytoskeletal-mediated nanotopographical sensing mechanism is also present in human ESCs where mouse ESCs were sensitive to local cyclic stress that were applied to FAs. This applied stress led to the downregulation of Oct3/4 gene expression in mouse ESCs, and myosin II contractility is shown to be essential in mouse ESCs’ stress sensitivity (Chowdhury et al., 2010). It is likely that the stem cells interpret such changes in force signals to regulate stem cell fate, although topography is probably only one of many contributing factors among others like mechan ical and osmotic stress.

283

11. Nanotopography/Induced of Stem-Cell Differentiation

(A) 10S assembly-incompetent NM II

6S assembly-competent NM II Globular head domain (actin binding and Mg2+-ATPase motor domains) ELC RLC

RLC phosphorylation

Heavy chain

Coiled-coil rod domain Nonhelical tail S-1 HMM

MHC phosphorylation

(B) Bipolar NM II filaments

Actin filament

Fig. 29 (A) Structural domains of nonmuscle myosin II (NM II). Interested readers can obtain an in-depth description of the NM II in Vicente-Manzanares et al. (2009). (B) Schematic diagram of an actin–myosin motor protein. NM II molecules assemble to form bipolar filaments and bind to actin through the head domains. ATPase activity induces a conformational change in the head domain, allowing movement of the actin filaments in an antiparallel direction to exert contractility in the cytoskeleton. This is also named as stress fibers. Schematic diagram obtained from Vicente-Manzanares et al. (2009).

The ability of human MSCs to generate corresponding contractile forces in response to the underlying substrate stiffness suggests that the cell-generated forces are dis sipated within the cell on a rigid matrix (Engler et al., 2008). This has led to the postulation that these cellular forces change protein conformation, exposing binding sites that enable plasma membrane proteins to become functional (Fig. 31). It seems that optimal matrix elasticity is crucial for a functional mechanosensitive protein (Reilly and Engler, 2010). Similar force-sensitive regulation of plasma proteins have been observed in mature cells although its role in stem cell regulation is largely speculative. However, a growing amount of evidence has demonstrated the importance of external force loading from the ECM (Tan et al., 2003) or actin–myosin contractility (Galbraith et al., 2002; Riveline et al., 2001) in the formation of FAs and ECM

Benjamin Kim Kiat Teo et al.

284 Actin

NM II Mechano responsive molecules

Integrin Vinculin Talin

ROCK

FAK p130CaS

RHOA

Myosin phosphatase

AM

MYPT1 PP1

Arp2/3 Vinculin

P β-

Src illin Pax

ix GIT

Rac

FA K

RHOGEF

CRK DOCK180

xin

Zy Talin

p130CAS

α-Actinin

Plasma membrane β-Integrin subunit

α-Integrin subunit

ECM

Fig. 30 A schematic overview of the physical continuity at the plasma membrane that links the actomyosin-mediated stress fibers to the ECM. Key components of the adhesion complex including adaptor molecules and downstream kinases are shown. Downstream signaling pathways of integrin activation are spatially and temporally regulated with a convoluted feedback to the mechanosensitive molecules through the NM II in order to ”tune” the cytoskeleton mechanical properties. Diagram and signaling pathways are reviewed in Vicente-Manzanares et al. (2009).

285

11. Nanotopography/Induced of Stem-Cell Differentiation

Cryptic binding site; nonfunctional

Exposed, functional binding site

Unfolded binding site; nonfunctional

Myosin

F-Actin

Focal

contacts

Force

Mechano

sensor

Soft matrix

Intermediate matrix

Stiff matrix

Fig. 31 Cryptic binding sites otherwise hidden within mechanosensitive proteins are exposed upon the exertion of an appropriate cellular force (regulated by the matrix) mediated through the actomyosin motor. The accessibility of such sites makes these proteins functional, activating downstream signaling pathways. It is likely that stem cells on topography probe the geometrical cues using such mechanism to regulate gene expression. Schematic diagram as shown in Reilly (2010).

sensing. The cellular force contractility induces a downstream of events including the recruitment of adhesions molecules and kinases like the mechanosensitive FAK, subsequently triggering the activation of Rho GTPases (reviewed in VicenteManzanares et al., 2009) (see also Fig. 30). For example, increasing ECM stiffness activates Rho (Paszek et al., 2005), which promotes actomyosin stress fiber assembly (Chrzanowska-Wodnicka and Burridge, 1996), significantly changing the mechanical properties of the cell (Hall, 1998). These stress fibers are constantly tuning its mechanical properties with feedback from these downstream molecules. It remains difficult to unravel the entire mechanism but an increasing amount of evidence directs the importance of cytoskeleton in mechanotransduction, with possibly a similar mechanism in topography-induced stem-cell differentiation.

D. Nucleus: Mechanical Manipulation of Gene Regulation The intricate physical network described previously sets the framework for the physical continuity spanning from the ECM to the nucleus (Fig. 32). Mechanical

Benjamin Kim Kiat Teo et al.

286 ECM fibers

Integrin

receptors

Plasma membrane

Cytoskeleton

Functional proteins

KASH SUN

Chromatin

Fig. 32 Mechanical coupling of the ECM to the nucleus through the actin cytoskeleton. Physical perturbations to the network complex induce a rearrangement of proteins/chromatins in the nucleus, thus triggering a differential gene expression. Diagram adapted from Gjorevski (2009).

signals such as topographical perturbations from the ECM can be transduced by structural alterations in the network to elicit differential gene expression in stem cells. Although an in-depth discussion of laminar proteins is out of the scope of this chapter, this section briefly describes key components in nuclear mechanotransduction that might play key roles in topography sensing in stem cells.

11. Nanotopography/Induced of Stem-Cell Differentiation

287

While the actin microfilaments are anchored to the ECM through the integrins, they are also physically connected to the nuclear membrane on the other end. Nesprins are a class of large outer nuclear membrane proteins on which the actin microfilaments bind onto the Klarsicht-Anc1-syne1 homology (KASH) domains present in these proteins (Crisp et al., 2006; Zhang et al., 2001). The KASH domains are then physically connected to the Sad1-UNC-84 homology (SUN) domain of the inner nuclear mem brane forming a KASH/SUN complex to mechanically bridge the actin stress fibers to the nuclear membrane. The other main structure of the nucleus that appears to be important in topographymediated mechanotransduction is the nuclear lamina. Similar to the cytoskeleton, the nuclear lamina consists of a meshwork of intermediate filaments and lamin proteins that is associated with the KASH/SUN complex (Alberts et al., 1994; Dechat et al., 2008), which in turn is connected to the chromatin (reviewed in Gieni and Hendzel, 2008; Wang et al., 2009). Recently, the LINC, a specialized structure that links both the nucleoskeleton and the cytoskeleton, has been identified, providing more evidence that mechanical forces that arise due to nanotopography can physically affect the structural organization of the nucleus (Crisp et al., 2006; Fey et al., 1984). Forces that are transmitted to the nuclear scaffolds via LINC complex may regulate critical DNA enzymes or factors. Furthermore, in an earlier study, the disruption of intermediate filaments leads to the mechanical decoupling of the integrins and nuclei (Maniotis et al., 1997), demonstrating that a direct physical connection exists between the two. It has also been shown that local forces applied to apical integrins are indeed transmitted to the basal FAs and nucleus, suggesting that a physical continuity does exist between the ECM and the cell nucleus (Hu et al., 2003, 2005). Other research groups and our group have similarly observed nuclei shape changes and altered gene expression in response to topography (Dalby et al., 2002, 2007b; Yim et al., 2007). While Dalby et al. observed spatial alteration of chromosomes in fibroblasts under topographical influence, our work involving human MSC on nanogratings suggests that topography may exert an effect on the structural organization of the nucleus as indicated by the alignment and elongation of the MSC nuclei (Yim et al., 2007). This study suggests the direct mechanical coupling of chromatin to the ECM through the intricate mechanotransduction network in stem cells. This physical coupling allows chro matin to be tightly regulated through cellular forces to unravel DNA regulatory motifs for transcription factors to interact, analogous to the exposure of cryptic binding sites in mechanosensitive plasma proteins. Although there are clearly evidences of nuclear mechanotransduction, the molecular and biophysical bases for such a mechanism are still left to be understood. Nuclear mechanotransduction is currently an area of active research and a more elaborate discussion can be obtained from Chapters 1-6 and 9-10 of this book.

V. Conclusions This chapter reviews how topology: topography of culturing substrate or cell-interacting scaffolds and protein patterning affect stem cell maintenance and differentiation. The vast

288

Benjamin Kim Kiat Teo et al.

amount of evidences has substantiated the importance of cell–substrate topology interac tion for stem cell fate control. Even though the details of the underlying mechanism remains unknown to date, effort invested by various groups in the past few years began to point out the mechanotransduction link between substrate topology and stem cell regula tion. The cell shape control, FA, cytoskeletal force, and nucleus regulation will likely be the key players in the regulation. Further effort is needed to identify the mechanism of cell–substrate topolopy for the reconstruction of the stem cell niche.

References Abrams, G. A., et al. (2000). Nanoscale topography of the basement membrane underlying the corneal epithelium of the rhesus macaque. Cell Tissue Res. 299(1), 39–46. Affrossman, S., and Stamm, M. (1998). Topography and surface composition of thin films of blends of polystyrene with brominated polystyrenes: Effects of varying the degree of bromination and annealing. Macromolecules 31(18), 6280–6288. Affrossman, S., and Stamm, M. (2000). The effect of molecular weight on the topography of thin films of blends of poly (4-bromostyrene) and polystyrene. Colloid Polym. Sci. 278(9), 888–893. Affrossman, S. et al. (1996). Surface topography and composition of deuterated polystyrene–poly(bromos tyrene) blends. Macromolecules 29(14), 5010–5016. Alberts, B. B. D., Lewis, J., Raff, M., Watson, J. (1994). “Molecular Biology of the Cell.” Garland Publishing Inc., New York. Amit, M., et al. (2000). Clonally derived human embryonic stem cell lines maintain pluripotency and proliferative potential for prolonged periods of culture. Dev. Biol. 227(2), 271–278. Arnold, M. et al., (2004). Activation of integrin function by nanopatterned adhesive interfaces. Chem physchem. 5(3), 383–388. Arnold, M. et al., (2008). Induction of cell polarization and migration by a gradient of nanoscale variations in adhesive ligand spacing. Nano Lett. 8(7), 2063–2069. Basnar, B., and Willner, I. (2009). Dip-pen-nanolithographic patterning of metallic, semiconductor, and metal oxide nanostructures on surfaces. Small 5(1), 28–44. Berrier, A. L., and Yamada, K. M. (2007). Cell-matrix adhesion. J. Cell Physiol. 213 (3), 565–573. Bettinger, C. J., Langer, R., and Borenstein, J. T. (2009). Engineering substrate topography at the micro- and nanoscale to control cell function. Angew. Chem. Int. Ed. 48(30), 5406–5415. Biehl, J. K., et al., (2009). Proliferation of mouse embryonic stem cell progeny and the spontaneous contractile activity of cardiomyocytes are affected by microtopography. Dev. Dyn. 238(8), 1964–1973. Biggs M. J. P., et al. (2009). The use of nanoscale topography to modulate the dynamics of adhesion formation in primary osteoblasts and ERK/MAPK signaling in STRO-1þ enriched skeletal stem cells. Biomaterials 30, 5094–5103. Bissell, M. J., Hall, H. G., and Parry, G. (1982). How does the extracellular matrix direct gene expression? J. Theor. Biol. 99(1), 31–68. Bissell, M. J., et al., (1999). Tissue structure, nuclear organization, and gene expression in normal and malignant breast. Cancer Res. 59(7 Suppl): 1757–1763s; discussion 1763s–1764s. Boote Jones, E. N., and Mallapragada, S. K. (2007). Directed growth and differentiation of stem cells towards neural cell fates using soluble and surface-mediated cues. J. Biomater. Sci. Polym. Ed. 18(8), 999–1015. Brunette, D. M., and Chehroudi, B. (1999). The effects of the surface topography of micromachined titanium substrata on cell behavior in vitro and in vivo. J. Biomech. Eng. 121(1), 49–57. Burridge, K. et al. (1988). Focal adhesions: Transmembrane junctions between the extracellular matrix and the cytoskeleton. Annu. Rev. Cell Biol. 4, 487–525. Campos, L. S. (2005). Beta1 integrins and neural stem cells: Making sense of the extracellular environment. Bioessays 27(7), 698–707.

11. Nanotopography/Induced of Stem-Cell Differentiation

289

Caplan, A. I. (1991). Mesenchymal stem cells. J. Orthop. Res. 9(5), 641–650. Caplan, A. I., and Dennis, J. E. (2006). Mesenchymal stem cells as trophic mediators. J. Cell. Biochem. 98(5), 1076–1084. Carnell L. S., Siochi E. J., et al. (2008). Aligned Mats from Electrospun Single Fibers. Macromolecules 41(14), 5345–5349. Chin, A. C., et al. (2010). Defined and serum-free media support undifferentiated human embryonic stem cell growth. Stem Cells Dev. 19(6), 753–761. Choi, C. K., et al. (2008). Actin and alpha-actinin orchestrate the assembly and maturation of nascent adhesions in a myosin II motor-independent manner. Nat. Cell Biol. 10(9), 1039–1050. Chou, S. Y., Krauss, P. R., and Renstrom, P. J. (1996). Imprint lithography with 25-nanometer resolution. Science 272(5258), 85–87. Chowdhury, F., et al. (2010). Material properties of the cell dictate stress-induced spreading and differentia tion in embryonic stem cells. Nat. Mater. 9(1), 82–88. Christopherson, G. T., Song, H., and Mao, H.-Q. (2009). The influence of fiber diameter of electrospun substrates on neural stem cell differentiation and proliferation. Biomaterials 30(4), 556–564. Chrzanowska-Wodnicka, M., and Burridge, K. (1996). Rho-stimulated contractility drives the formation of stress fibers and focal adhesions. J. Cell Biol. 133(6), 1403–1415. Chua, K. N., et al. (2006). Surface-aminated electrospun nanofibers enhance adhesion and expansion of human umbilical cord blood hematopoietic stem/progenitor cells. Biomaterials 27(36), 6043–6051. Chua, K. N., et al. (2007). Functional nanofiber scaffolds with different spacers modulate adhesion and expansion of cryopreserved umbilical cord blood hematopoietic stem/progenitor cells. Exp. Hematol. 35(5), 771–781. Ciapetti, G., et al. (2006). Human bone marrow stromal cells: In vitro expansion and differentiation for bone engineering. Biomaterials 27(36), 6150–6160. Corey, J. M., et al. (2007). Aligned electrospun nanofibers specify the direction of dorsal root ganglia neurite growth. J. Biomed. Mater. Res. A83(3), 636–645. Crisp, M., et al. (2006). Coupling of the nucleus and cytoplasm: Role of the LINC complex. J. Cell Biol. 172(1), 41–53. Curtis, A. S., and Varde, M. (1964). Control of cell behavior: Topological factors. J. Natl. Cancer Inst. 33, 15–26. Curtis, A., and Wilkinson, C. (1997). Topographical control of cells. Biomaterials 18(24), 1573–1583. Curtis, A., and Wilkinson, C. (2001). Nantotechniques and approaches in biotechnology. Trends Biotechnol. 19(3), 97–101. Dalby, M. J., et al. (2002). Increasing fibroblast response to materials using nanotopography: Morphological and genetic measurements of cell response to 13-nm-high polymer demixed islands. Exp. Cell Res. 276(1), 1–9. Dalby, M. J., et al. (2003). Fibroblast reaction to island topography: Changes in cytoskeleton and morphol ogy with time. Biomaterials 24(6), 927–935. Dalby, M. J., et al. (2004a). Investigating filopodia sensing using arrays of defined nano-pits down to 35 nm diameter in size. Int. J. Biochem. Cell Biol. 36(10), 2005–2015. Dalby, M. J., et al. (2004b). Investigating the limits of filopodial sensing: A brief report using SEM to image the interaction between 10 nm high nano-topography and fibroblast filopodia. Cell Biol. Int. 28(3), 229–236. Dalby, M. J., et al. (2007a). The control of human mesenchymal cell differentiation using nanoscale symmetry and disorder. Nat. Mater. 6(12), 997–1003. Dalby, M. J., et al. (2007b). Nanomechanotransduction and interphase nuclear organization influence on genomic control. J. Cell Biochem. 102(5), 1234–1244. Dang, J. M., and Leong, K. W. (2007). Myogenic induction of aligned mesenchymal stem cell sheets by culture on thermally responsive electrospun nanofibers. Adv. Mater. 19(19), 2775–2779. Dang, S. M., and Zandstra, P. W. (2005). Scalable production of embryonic stem cell-derived cells. Meth. Mol. Biol. 290, 353–364. Dechat, T., et al. (2008). Nuclear lamins: Major factors in the structural organization and function of the nucleus and chromatin. Genes Dev. 22(7), 832–853. Delon, I., and Brown, N. H. (2007). Integrins and the actin cytoskeleton. Curr. Opin. Cell Biol. 19(1), 43–50. Desai, T. A., et al. (1999). Nanopore technology for biomedical applications. Biomed. Microdevices 2(1), 11–40.

290

Benjamin Kim Kiat Teo et al. Ding, S., and Schultz, P. G., (2004). A role for chemistry in stem cell biology. Nat. Biotechnol. 22(7), 833–840. Doyle, A. D., et al. (2009). One-dimensional topography underlies three-dimensional fibrillar cell migration. J. Cell Biol. 184(4), 481–490. Drost, A. C., et al. (2009). In vitro myogenic differentiation of human bone marrowderived mesenchymal stem cells as a potential treatment for urethral sphincter muscle repair. Ann. N Y Acad. Sci. 1176, 135–143. Dulgar-Tulloch, A. J., Bizios, R., and Siegel, R. W. (2009). Human mesenchymal stem cell adhesion and prolifera tion in response to ceramic chemistry and nanoscale topography. J. Biomed. Mater. Res. A 90(2), 586–594. Dunn, G. A., and Brown, A. F. (1986). Alignment of fibroblasts on grooved surfaces described by a simple geometric transformation. J. Cell Sci. 83, 313–340. Elias, K. L., Price, R. L., and Webster, T. J. (2002). Enhanced functions of osteoblasts on nanometer diameter carbon fibers. Biomaterials 23(15), 3279–3287. Engler, A. J., et al. (2006). Matrix elasticity directs stem cell lineage specification. Cell 126(4), 677–689. Engler, A. J., et al. (2008). Embryonic cardiomyocytes beat best on a matrix with heart-like elasticity: Scarlike rigidity inhibits beating. J. Cell Sci. 121(Pt 22), 3794–3802. Engel, E., et al. (2009). Mesenchymal stem cell differentiation on microstructured poly (methyl methacry late) substrates. Ann. Anat. 191(1), 136–144 Even-Ram, S., Artym, V., and Yamada, K. M. (2006). Matrix control of stem cell fate. Cell 126(4), 645–647. Even-Ram, S., et al. (2007). Myosin IIA regulates cell motility and actomyosin-microtubule crosstalk. Nat. Cell Biol. 9(3), 299–309. Fan, Y. W., et al. (2002). Culture of neural cells on silicon wafers with nano-scale surface topograph.J. Neurosci. Methods 120(1), 17–23. Farrell, E., et al. (2006). A collagen-glycosaminoglycan scaffold supports adult rat mesenchymal stem cell differentiation along osteogenic and chondrogenic routes. Tissue Eng. 12(3), 459–468. Fey, E. G., Wan, K. M., and Penman, S. (1984). Epithelial cytoskeletal framework and nuclear matrixintermediate filament scaffold: Three-dimensional organization and protein composition. J. Cell Biol. 98(6), 1973–1984. Flemming, R. G., et al. (1999). Effects of synthetic micro- and nano-structured surfaces on cell behavior. Biomaterials 20(6), 573–588 Folch, A., and Toner, M. (1998) Cellular micropatterns on biocompatible materials. Biotechnol. Prog. 14(3), 388–392. Folkman, J., and Moscona, A. (1978). Role of cell shape in growth control. Nature 273(5661), 345–349. Frenot, A., and Chronakis, I. S. (2003). Polymer nanofibers assembled by electrospinning. Curr. Opin. Colloid Interface Sci. 8, 64–75. Fuchs, E., Tumbar, T., and Guasch, G. (2004). Socializing with the neighbors: Stem cells and their niche. Cell 116(6), 769–778. Fujita, S., Ohshima, M., and Iwata, H. (2009). Time-lapse observation of cell alignment on nanogrooved patterns. J. R. Soc. Interface 6(Suppl 3), S269–S277. Galbraith, C. G., Yamada, K. M., and Sheetz, M. P. (2002). The relationship between force and focal complex development. J. Cell Biol. 159(4), 695–705. Garcia, A. J., Vega, M. D., and Boettiger, D. (1999). Modulation of cell proliferation and differentiation through substrate-dependent changes in fibronectin conformation. Mol. Biol. Cell 10(3), 785–798. Geiger, B., et al. (2001). Transmembrane crosstalk between the extracellular matrix—cytoskeleton crosstalk. Nat. Rev. Mol. Cell Biol. 2(11), 793–805. Geiger, B., Spatz, J. P., and Bershadsky, A. D. (2009). Environmental sensing through focal adhesions. Nat. Rev. Mol. Cell Biol. 10(1), 21–33. Gerecht, S., et al. (2007). The effect of actin disrupting agents on contact guidance of human embryonic stem cells. Biomaterials 28(28), 4068–4077. Gieni, R. S., and Hendzel, M. J. (2008). Mechanotransduction from the ECM to the genome: Are the pieces now in place? J. Cell Biochem. 104(6), 1964–1987. Gingras, A. R., et al. (2006). Structural and dynamic characterization of a vinculin binding site in the talin rod. Biochemistry 45(6), 1805–1817.

11. Nanotopography/Induced of Stem-Cell Differentiation

291

Gingras, A. R., et al. (2008). The structure of the C-terminal actin-binding domain of talin. EMBO J. 27(2), 458–469. Gjorevski, N., and Nelson, C. M. (2009). Bidirectional extracellular matrix signaling during tissue morpho genesis. Cytokine Growth Factor Rev. 20(5–6), 459–465. Goldman, S. A. (2006). Neural progenitor cells of the adult human brain, In “Neural Development and Stem Cells,” (pp. 267–297). 2nd edition, Mahendra S. Rao ed. Humana Press, New Jersey. Gong, H., et al. (2002). A new view of the human trabecular meshwork using quick-freeze, deep-etch electron microscopy. Exp. Eye Res. 75(3), 347–358. Hall, A., (1998). Rho GTPases and the actin cytoskeleton. Science 279(5350), 509–514. Harris, A. K., Wild, P., and Stopak, D. (1980). Silicone rubber substrata: A new wrinkle in the study of cell locomotion. Science 208(4440), 177–179. Hartgerink, J. D., Beniash, E., and Stupp, S. I. (2002). Peptide-amphiphile nanofibers: A versatile scaffold for the preparation of self-assembling materials. Proc. Natl. Acad. Sci. USA 99(8), 5133–5138. He, Y., and Zhao, Y. (2009). Hydrogen storage and cycling properties of a vanadium decorated Mg nanoblade array on a Ti coated Si substrate. Nanotechnology 20(20), 204008. Hoffman, L. M., and Carpenter, M. K. (2005). Characterization and culture of human embryonic stem cells. Nat. Biotechnol. 23(6), 699–708. Holmes, T. C., et al. (2000). Extensive neurite outgrowth and active synapse formation on self-assembling peptide scaffolds. Proc. Natl. Acad. Sci. USA 97(12), 6728–6733. Hosseinkhani, H., et al. (2006). Osteogenic differentiation of mesenchymal stem cells in self-assembled peptide-amphiphile nanofibers. Biomaterials 27(22), 4079–4086. Hu, S., et al. (2003). Intracellular stress tomography reveals stress focusing and structural anisotropy in cytoskeleton of living cells. Am. J. Physiol. Cell Physiol. 285(5), C1082–C1090. Hu, S., et al. (2005). Prestress mediates force propagation into the nucleus. Biochem. Biophys. Res. Commun. 329(2), 423–428. Huang, N. F., and Li, S. (2008). Mesenchymal stem cells for vascular regeneration. Regener. Med. 3(6), 877–892. Huang, X. D., et al. (2002). Reversal imprinting by transferring polymer from mold to substrate. J. Vac. Sci. Technol. B 20(6), 2872–2876. Humphries, J. D., et al. (2007). Vinculin controls focal adhesion formation by direct interactions with talin and actin. J. Cell Biol. 179(5) 1043–1057. Ingber, D. E. (1991). Control of capillary growth and differentiation by extracellular matrix. Use of a tensegrity (tensional integrity) mechanism for signal processing. Chest 99(3 Suppl), 34S–40S. Ingber, D. E. (2003a). Tensegrity II. How structural networks influence cellular information processing networks. J. Cell Sci. 116(Pt 8), 1397–1408. Ingber, D. E. (2003b). Tensegrity I. Cell structure and hierarchical systems biology. J. Cell Sci. 116(Pt 7), 1157–1173. Ingber, D. E. (2008). Tensegrity-based mechanosensing from macro to micro. Prog. Biophys. Mol. Biol. 97(2–3), 163–179. Jiang, F., et al. (2004). Assembly of collagen into microribbons: Effects of pH and electrolytes. J. Struct. Biol. 148(3), 268–278. Johansson, C. B., et al. (1999a). Identification of a neural stem cell in the adult mammalian central nervous system. Cell 96(1), 25–34. Johansson, C. B. et al. (1999b). Neural stem cells in the adult human brain. Exp. Cell Res. 253(2), 733–736. Kaehr, B., et al. (2004). Guiding neuronal development with in situ microfabrication. Proc. Natl. Acad. Sci. USA 101(46), 16104–16108. Kantawong, F., et al. (2009). Whole proteome analysis of osteoprogenitor differentiation induced by disordered nanotopography and mediated by ERK signalling. Biomaterials 30(27), 4723–4731. Khademhosseini, A., et al. (2006). Co-culture of human embryonic stem cells with murine embryonic fibroblasts on microwell-patterned substrates. Biomaterials 27(36), 5968–5977. Khor, H. L., et al. (2007). Response of cells on surface-induced nanopatterns: Fibroblasts and mesenchymal progenitor cells. Biomacromolecules 8(5), 1530–1540. Kirouac, D. C., et al. (2009) Cell–cell interaction networks regulate blood stem and progenitor cell fate. Mol. Syst. Biol. 5, 293.

292

Benjamin Kim Kiat Teo et al. Leipzig, N. D., and Shoichet, M. S. (2009). The effect of substrate stiffness on adult neural stem cell behavior. Biomaterials 30(36), 6867–6878. Lim, J. Y., and Donahue, H. J. (2007). Cell sensing and response to micro and nanostructured surfaces produced by chemical and topographic patterning. Tissue Eng. 13(8), 1879–1891. Little, W. C., et al. (2008). Assay to mechanically tune and optically probe fibrillar fibronectin conformations from fully relaxed to breakage. Matrix Biol. 27(5), 451–461. Liu, G. Y., Xu, S., and Qian, Y., (2000). Nanofabrication of self-assembled monolayers using scanning probe lithography. Acc. Chem. Res. 33(7), 457–466. Lovmand, J., et al. (2009). The use of combinatorial topographical libraries for the screening of enhanced osteogenic expression and mineralization. Biomaterials 30(11), 2015–2022. Low, S. P., et al. (2006). Evaluation of mammalian cell adhesion on surface-modified porous silicon. Biomaterials 27(26), 4538–4546. Lutolf, M. P., and Hubbell, J. A. (2005). Synthetic biomaterials as instructive extracellular microenviron ments for morphogenesis in tissue engineering. Nat. Biotechnol.23(1), 47–55. Malarkey, E. B., and Parpura, V. (2007). Applications of carbon nanotubes in neurobiology. Neurodegen erative Dis. 4(4), 292–299. Manasek, F. J., Burnside, M. B., and Waterman, R. E. (1972). Myocardial cell shape change as a mechanism of embryonic heart looping. Dev. Biol. 29(4), 349–371. Maniotis, A. J., Chen, C. S., and Ingber, D. E. (1997). Demonstration of mechanical connections between integrins, cytoskeletal filaments, and nucleoplasm that stabilize nuclear structure. Proc. Natl. Acad. Sci. USA 94(3), 849–854. Markert, L. D., et al. (2009). Identification of distinct topographical surface microstructures favoring either undifferentiated expansion or differentiation of murine embryonic stem cells. Stem Cells Dev. 18(9), 1331–1342. Martinez, E., et al. (2009). Effects of artificial micro- and nano-structured surfaces on cell behaviour. Ann. Anat. 191(1), 126–135. Martino, S., et al. (2009). Hydrogenated amorphous carbon nanopatterned film designs drive human bone marrow mesenchymal stem cell cytoskeleton architecture. Tissue Eng. A 15(10), 3139–3149. Matthews, J. A., et al. (2002). Electrospinning of Collagen Nanofibers. Biomacromolecules 3(2), 232–238. McBeath, R., et al. (2004). Cell shape, cytoskeletal tension, and RhoA regulate stem cell lineage commit ment. Dev. Cell 6(4), 483–495. Meyers, J., Craig, J., and Odde, D. J. (2006). Potential for control of signaling pathways via cell size and shape. Curr. Biol. 16(17), 1685–1693. Mohr, J. C., de Pablo, J. J., and Palecek, S. P. (2006). 3-D microwell culture of human embryonic stem cells. Biomaterials 27(36), 6032–6042. Moldovan, N. I., et al. (2000). Contribution of monocytes/macrophages to compensatory neovascularization: The drilling of metalloelastase-positive tunnels in ischemic myocardium. Circ. Res. 87(5) 378–384. Moore, K. A., and Lemischka, I. R. (2006). Stem cells and their niches. Science 311(5769), 1880–1885. Morshead, C. M., et al. (1994). Neural stem cells in the adult mammalian forebrain: A relatively quiescent subpopulation of subependymal cells. Neuron 13(5), 1071–1082. Nelson, C. M., and Bissell, M. J. (2006). Of extracellular matrix, scaffolds, and signaling: Tissue architecture regulates development, homeostasis, and cancer. Annu. Rev. Cell Dev. Biol. 22, 287–309. Nelson, C. M., et al. (2004). Vascular endothelial-cadherin regulates cytoskeletal tension, cell spreading, and focal adhesions by stimulating RhoA. Mol. Biol. Cell 15(6), 2943–2953. Nelson, C. M., et al. (2005). Emergent patterns of growth controlled by multicellular form and mechanics. Proc. Natl. Acad. Sci. USA 102(33), 11594–11599. Norman, J. J., and Desai, T. A. (2006). Methods for fabrication of nanoscale topography for tissue engineering scaffolds. Ann. Biomed. Eng. 34(1), 89–101. Odom, T. W., et al. (2002). Improved pattern transfer in soft lithography using composite stamps. Langmuir 18(13), 5314–5320. Obayashi, S., et al. (2009). Gene expression profiling of human neural progenitor cells following the seruminduced astrocyte differentiation. Cell Mol. Neurobiol. 29(3), 423–438.

11. Nanotopography/Induced of Stem-Cell Differentiation

293

Park, J., et al. (2007). Nanosize and vitality: TiO2 nanotube diameter directs cell fate. Nano Lett. 7(6), 1686–1691. Park, J. Y., et al. (2009). Differentiation of neural progenitor cells in a microfluidic chip-generated cytokine gradient. Stem Cells 27(11), 2646–2654. Paszek, M. J., et al. (2005). Tensional homeostasis and the malignant phenotype. Cancer Cell 8(3), 241–254. Patel, N., et al. (1998). Spatially controlled cell engineering on biodegradable polymer surfaces. FASEB J. 12(14), 1447–1454. Peerani, R., and Zandstra, P. W. (2010) Enabling stem cell therapies through synthetic stem cell-niche engineering. J. Clin. Invest. 120(1), 60–70. Peerani, R., et al. (2007). Niche-mediated control of human embryonic stem cell self-renewal and differ entiation. EMBO J. 26(22), 4744–4755. Peerani, R., et al. (2009). Manipulation of signaling thresholds in “Engineered stem cell Niches” identifies design criteria for pluripotent stem cell screens. PLoS ONE 4(7), e6438. Popat, K. C., et al. (2007). Osteogenic differentiation of marrow stromal cells cultured on nanoporous alumina surfaces. J. Biomed. Mater. Res. A 80(4), 955–964. Prabhakaran, M. P., Venugopal, J. R., and Ramakrishna, S. (2009). Mesenchymal stem cell differentiation to neuronal cells on electrospun nanofibrous substrates for nerve tissue engineering. Biomaterials 30(28), 4996–5003. Rao, B. M., and Zandstra, P. W. (2005). Culture development for human embryonic stem cell propagation: Molecular aspects and challenges. Curr. Opin. Biotechnol. 16(5), 568–576. Rao, P. M., and Zheng, X. (2009). Rapid catalyst-free flame synthesis of dense, aligned alpha-Fe2O3 nanoflake and CuO nanoneedle arrays. Nano Lett. 9(8), 3001–3006. Recknor, J. B., Sakaguchi, D. S., and Mallapragada, S. K. (2006). Directed growth and selective differentia tion of neural progenitor cells on micropatterned polymer substrates. Biomaterials 27(22), 4098–4108. Reilly, G. C., and Engler, A. J. (2010). Intrinsic extracellular matrix properties regulate stem cell differentia tion. J. Biomech. 43(1), 55–62. Riveline, D., et al. (2001). Focal contacts as mechanosensors: Externally applied local mechanical force induces growth of focal contacts by an mDia1-dependent and ROCK-independent mechanism. J. Cell Biol. 153(6), 1175–1186. Roy, N. S., et al. (1999). Identification, isolation, and promoter-defined separation of mitotic oligodendrocyte progenitor cells from the adult human subcortical white matter. J. Neurosci. 19(22), 9986–9995. Ruoslahti, E., and Obrink, B. (1996). Common principles in cell adhesion. Exp. Cell Res. 227(1), 1–11. Sanchez-Ramos, J., et al. (2000). Adult bone marrow stromal cells differentiate into neural cells in vitro. Exp. Neurol. 164(2), 247–256. Sasaki, D., et al. (2009). Mass preparation of size-controlled mouse embryonic stem cell aggregates and induction of cardiac differentiation by cell patterning method. Biomaterials 30(26), 4384–4389. Sawada, Y., et al. (2006). Force sensing by mechanical extension of the Src family kinase substrate p130Cas. Cell 127(5), 1015–1026. Schmid, H., and Michel, B. (2000). Siloxane polymers for high-resolution, high-accuracy soft lithography. Macromolecules 33(8), 3042–3049. Scolding, N., et al. (1998). Oligodendrocyte progenitors are present in the normal adult human CNS and in the lesions of multiple sclerosis. Brain 121(12), 2221–2228. Seidlits, S. K., Lee, J. Y., and Schmidt, C. E. (2008). Nanostructured scaffolds for neural applications. Nanomedicine 3(2), 183–199. Sekiya, I., et al. (2004). Adipogenic differentiation of human adult stem cells from bone marrow stroma (MSCs). J. Bone Miner. Res. 19(2), 256–264. Sen, R., et al. (2004). Preparation of single-walled carbon nanotube reinforced polystyrene and polyurethane nanofibers and membranes by electrospinning. Nano Lett. 4(3), 459–464. Sjostrom, T., et al. (2009). Fabrication of pillar-like titania nanostructures on titanium and their interactions with human skeletal stem cells. Acta Biomater. 5(5), 1433–1441. Skottman, H., and Hovatta, O. (2006). Culture conditions for human embryonic stem cells. Reproduction 132(5), 691–698. Smith, L. A., and Ma, P. X. (2004). Nano-fibrous scaffolds for tissue engineering. Colloids Surf., B Biointerfaces 39(3), 125–131.

294

Benjamin Kim Kiat Teo et al. Smith, M. L., et al. (2007). Force-induced unfolding of fibronectin in the extracellular matrix of living cells. PLoS Biol. 5(10), e268. Smith, L. A., et al. (2009). Enhancing osteogenic differentiation of mouse embryonic stem cells by nanofibers. Tissue Eng. A 15(7), 1855–1864. Soen, Y., et al. (2006). Exploring the regulation of human neural precursor cell differentiation using arrays of signaling microenvironments. Mol. Syst. Biol. 2, 37. Steven, A. G., et al. (2002). Isolation and induction of adult neural progenitor cells. 2(1), 70–79. Stojkovic, M., et al. (2004). Derivation, growth and applications of human embryonic stem cells. Reproduc tion 128(3), 259–267. Svendsen, C. N., Caldwell, M. A., and Ostenfeld, T. (1999). Human neural stem cells: Isolation, expansion and transplantation. Brain Pathol. 9(3), 499–513. Takahashi, K., and Yamanaka, S. (2006) .Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell. 126(4), 663–676. Tan, J. L., et al. (2003). Cells lying on a bed of microneedles: An approach to isolate mechanical force. Proc. Natl. Acad. Sci. USA 100(4), 1484–1489. Thery, M., et al. (2006). Cell distribution of stress fibres in response to the geometry of the adhesive environment. Cell Motil. Cytoskeleton 63(6), 341–355. Turner, S., et al. (1997). Cell attachment on silicon nanostructures. J. Vac. Sci. Technol. B 15(6), (November), 2848–2854. Ullmann, U., et al. (2007). Epithelial-mesenchymal transition process in human embryonic stem cells cultured in feeder-free conditions. Mol. Hum. Reprod. 13(1), 21–32. Ungrin, M. D., et al. (2008). Reproducible, ultra high-throughput formation of multicellular organization from single cell suspension-derived human embryonic stem cell aggregates. PLoS One 3(2), e1565. Vicente-Manzanares, M., et al. (2009). Non-muscle myosin II takes centre stage in cell adhesion and migration. Nat. Rev. Mol. Cell Biol. 10(11), 778–790. Vieu, C., et al. (2000). Electron beam lithography: Resolution limits and applications. Appl. Surf. Sci. 164(1–4), 111–117. Wang, X., and Yang, P. (2008). In vitro differentiation of mouse embryonic stem (mES) cells using the hanging drop method. J. Vis. Exp. (17). 10.3791/825. Wang, G., (2005). Noggin and bFGF cooperate to maintain the pluripotency of human embryonic stem cells in the absence of feeder layers. Biochem. Biophys. Res. Commun. 330(3), 934–942. Wang, N., Tytell, J. D., and Ingber, D. E. (2009). Mechanotransduction at a distance: Mechanically coupling the extracellular matrix with the nucleus. Nat. Rev. Mol. Cell Biol. 10(1), 75–82. Weiss, P., and Garber, B. (1952). Shape and movement of mesenchyme cells as functions of the physical structure of the medium. Proc. Natl. Acad. Sci. USA 38(3), 264–280. Whitesides, G. M., et al. (2001). Soft lithography in biology and biochemistry. Annu. Rev. Biomed. Eng. 3, 335–373. Xu Y., Zhao W., Low H. Y. (2006). Sacrificial film-assisted nanoimprint lithography. Microelectronic Engineering 83, 542–546. Yim, E. K., and Leong, K. W. (2005). Significance of synthetic nanostructures in dictating cellular response. Nanomedicine 1(1), 10–21. Yim, E. K., Pang, S. W., and Leong, K. W. (2007). Synthetic nanostructures inducing differentiation of human mesenchymal stem cells into neuronal lineage. Exp. Cell Res. 313(9), 1820–1829. Zaidel-Bar, R., et al. (2007). Functional atlas of the integrin adhesome. Nat. Cell Biol. 9(8), 858–867. Zamir, E., and Geiger, B. (2001). Molecular complexity and dynamics of cell-matrix adhesions. J. Cell Sci. 114(Pt 20), 3583–3590. Zhang, F. X., and Low, H. Y. (2006). Ordered three-dimensional hierarchical nanostructures by nanoimprint lithography. Nanotechnology 17(8), 1884–1890. Zhang, F. X., and Low, H. Y. (2008). Transfer printing of 3D hierarchical gold structures using a sequentially imprinted polymer stamp. Nanotechnology 19(41), 415305-1–415305-5 Zhang, Q., et al. (2001). Nesprins: A novel family of spectrin-repeat-containing proteins that localize to the nuclear membrane in multiple tissues. J. Cell Sci. 114(Pt 24), 4485–4498. Zhao, W., Low, H. Y., and Suresh, P. S. (2006). Cross-linked and chemically functionalized polymer supports by reactive reversal nanoimprint lithography. Langmuir 22(12), 5520–5524.

CHAPTER 12

Mechanical Induction in Embryonic Development and Tumor Growth: Integrative Cues Through Molecular to Multicellular Interplay and Evolutionary Perspectives Maria-Elena Fernandez-Sanchez, Fanny Serman, Padra Ahmadi, and Emmanuel Farge Mechanics and Genetics of Embryonic and Tumoral Development group, UMR168 CNRS, Institut Curie, 11 rue Pierre et Marie Curie, F-75005, Paris, France

Abstract I. Introduction II. Genetic Control of Morphogenetic Movements: Underlying Molecular Mechanisms and Evaluation of Forces A. Genetics of Morphogenesis B. From Genes to Shape: Molecular and Cell Biology of In Vivo Force Generation C. Measuring the Active Forces Developed in Morphogenetic Movements In Vivo III. Mechanical Control of Gene Expression: Testing Mechanical Induction by Application of Endogenous Forces from the Inside of the Embryo A. Twist: A Master Mechanosensitive Gene in Early Drosophila Embryos B. Mechanical Induction of Twist in the Future Anterior Gut Cells, in Response to Myo-II-Dependent Germ-Band Extension Compression IV. Mechanotransduction in the Control of Posttranslational Morphogenetic Events: Mechanical Activation of the Myosin-II Apico-Basal Polarity Triggering Mesoderm Invagination A. Mechanical Activation of Myo-II Apical Redistribution Hypothesis Emerging from the Theoretical Analysis of the Genetics of Drosophila Mesoderm Invagination B. In Silico Physical Tools to Test Theoretically the Viability of the Mechanotransduc tion Hypothesis

METHODS IN CELL BIOLOGY, VOL. 98 Copyright 2010 Elsevier Inc. All rights reserved.

295

978-0-12-381009-0 DOI: 10.1016/S0091-679X(10)98012-6

296

Maria-Elena Fernandez-Sanchez et al.

C. Coupling of Mechanical and Genetic Tools to Experimentally Test Mechanotrans duction in Myo-II Apical Redistribution D. Mechanical Modulation of Fog Endocytosis: A Potential Underlying Molecular Mechanism of Mechanotransduction E. Mechanical Induction of Twist Expression into the Mesoderm F. Incidences of the Mechanical Induction of Apical Redistribution of Myosin-II in Developmental Biology V. Incidences of Mechanical Induction in Tumor Development VI. Mechano-Genetics Network in Perspective of Evolution: Reactivation of a Primitive Feeding Response of Ancient Embryos Recapitulated in Embryonic Morphogenetic Inva gination? Mechanical Induction in First Multicellular Organism Emergence? Acknowledgments References

Abstract Embryonic development is a coordination of multicellular biochemical patterning and morphogenetic movements. Last decades revealed the close control of myosin-II dependent biomechanical morphogenesis by patterning gene expression, with constant progress in the understanding of the underlying molecular mechanisms. Reversed control of developmental gene expression and of myosin-II patterning by the mechan ical strains developed by morphogenetic movements was recently revealed at Droso phila gastrulation, through mechanotransduction processes involving the Armadillo/b catenin and the downstream of Fog Rho pathways. Here, we present the theoretical (simulations integrating the accumulated knowledge in the genetics of early embryonic development and morphogenesis) and the experimental (genetic and biophysical con trol of morphogenetic movements) tools having allowed the uncoupling of pure genetic inputs from pure mechanical inputs in the regulation of gene expression and myosin-II patterning. Specifically, we describe the innovative magnetic tweezers tools we have set up to measure and apply physiological strains and forces in vivo, from the inside of the tissue, to modulate and mimic morphogenetic movements in living embryos. We discuss mechanical induction incidence in tumor development and perspective in evolution.

I. Introduction The most ancient known scientific report on embryonic morphogenesis is by Aristotle, four centuries BCE (Aristotle). In his report, Aristotle emphasizes that the different parts of the body of the chicken embryo form in a sequential process rather than all at the same time. This seminal observation led to the so-called “epigenetic” conception of embryogenesis, through which the existence of the structure of the embryo of a given stage conditions the emergence of the structure of the next stage.

12. Mechanical Induction in Embryonic Development and Tumor Growth

297

This conception does not require a preexistent body plan, in opposition to the earlier Platonic preformationist conception of morphogenesis. Much later, the progresses of optical microscopy from the 17th to beginning of the 20th century allowed increasingly accurate observation of embryonic development. If preformationists initially thought that they could detect the existence of small preformed human shapes in the head of the male spermatozoid (called the “homunculus” by Leeuwenhoek (1683), the epigenetic view of development was rapidly confirmed by the observation of morphogenetic movements of tissues that correlate to growth, which progressively shape the embryo from ovoids to complex body shapes in a step-by-step sequential process. At that time, the privileged observable feature of embryogenesis was thus the morphogenetic move ments, which appeared as hydrodynamic fluid movements. After the Newtonian evolution of physics, these observations were naturally interpreted by many of the embryologists of that period, like His and Leduc (His, 1875; Leduc, 1912), as passive flows exclusively governed by the Newtonians laws of hydrodynamics. One of these embryologists, however, D’Arcy Thompson, suspected the existence of still unknown hidden underlying additive physiological factors driving the morphogenetic processes (Fox-Keller, 2003; Thomson, 1917). The discovery of the genome and the emergence of molecular biology and the genetics of developmental biology in the middle of 20th century revealed the nature of these factors, both genetic and biochemical in nature. From this evolution of biology, developmental biology focused most of its efforts on the study of the genetic control of the elaboration of the biochemical differentiation of the tissues designing the body plan of the future organism (Garber et al.,1983; Lewis, 1978; NussleinVolhard and Wieschaus, 1980; St Johnston and Nusslein-Volhard, 1992). However, the end of 20th century (at the onset of the 1990s) was marked by a return to geometrical morphogenetic considerations, with the discovery of master develop mental genes regulating the generation of morphogenetic movements in embryogen esis (Sweeton et al., 1991). Today, our understanding of the molecular mechanism linking patterning gene expression to the production of mechanical forces that shape the embryo increasingly progresses (Martin, 2009). In the beginning of the 21st century, reverse signals were discovered showing the mechanical control by the morphogenetic movements of the expression of patterning and developmental genes, based on biochemical mechanotransduction processes (Brouzes et al., 2004). Here, we will describe the up-to-date state of the art in this emerging field, reciprocally coupling genetics to mechanical physics. The study of such coupling necessarily requires the establishment of methods allowing the uncoupling of pure genetic inputs from pure mechanical inputs in the regulation of patterning gene expression. After a first part describing our knowledge on the genetic control of morphogenetic movements in embryogenesis, we will review today’s knowledge of the mechanical control of patterning and developmental gene expression, and the distinct genetic and biophysical methods that have been set up to uncouple mechanical inputs from biochemical inputs in the control of developmental gene expression in vivo. Specifically, we will describe the innovative tools we have set up to measure and apply physiological strains and forces in vivo, from the inside of the tissue, to

298

Maria-Elena Fernandez-Sanchez et al.

inhibit or quantitatively mimic morphogenetic movements. We will report our under standing of the underlying molecular mechanisms that translate mechanical strains applied to cells and tissues in vivo into the activation of transduction pathways connected to major developmental biochemical events during embryogenesis. Next, we will describe the influence of such mechanotransduction processes in medicine, and more specifically in carcinogenesis. Finally, we will then rapidly speculate on evolu tionary perspectives potentially related to the emergence of such mechanotransduction processes.

II. Genetic Control of Morphogenetic Movements: Underlying Molecular Mechanisms and Evaluation of Forces A. Genetics of Morphogenesis Embryogenesis is composed of two major morphogenetic processes: the biochemical patterning of the embryo and the mechanical morphogenetic movements that geometrically shape the embryo. Since 20 years ago, experiments initiated in early Drosophila embryos have shown that the morphogenetic movement sequence is tightly controlled by patterning gene expression (Sweeton et al., 1991). For instance, embryo nic mesoderm invagination requires the expression of the Fog (expressed under the control of Twist) and Snail zygotic proteins in the mesoderm (Seher et al., 2007), whereas the germ-band extension movement (i.e., the anterior–posterior elongation of the invaginated embryo) requires the expression of the Bicoid, Nanos, and Torso-like maternal proteins, which control the anterior–posterior polarity of the embryo (Irvine and Wieschaus, 1994) (Fig. 1). However, the elucidation of the relationship between gene expression and generation of strains leading to the tissue shape changes remained until recently largely unknown. The key role of cell polarities, in terms of the cortical (subplasma membrane) concentration of the molecular motor myosin-II protein (MyoII), in generating multicellular morphogenetic movements was first found in the generation of invaginations during Drosophila embryo gastrulation. In this process, the apical concentration of myosin-II leads to the constriction of cell apexes and generates the trapezoidal cell shape changes leading to posterior pole cell invaginations (Young et al., 1991). More recently, the role of the dorsoventral and anterior–posterior patterning genes was established, which induce the embryo polarities at the multicellular embryonic scale, in the generation of apicobasal and planar polarities in Myo-II concentration leading to morphogenetic movements (Bertet et al., 2004; Dawes-Hoang et al., 2005). B. From Genes to Shape: Molecular and Cell Biology of In Vivo Force Generation

1. Patterning Planar and Apicobasal Polarities in Contractile Actomyosin Concentration Regarding germ-band extension, the Bicoid, Nanos, and Torso-like genes establishing the anterior–posterior polarity of the embryo regulate the concentration of Par-3 in plasma

12. Mechanical Induction in Embryonic Development and Tumor Growth

299

membrane surfaces parallel, but not perpendicular to the axis, through a still poorly understood mechanism. As Myo-II interaction with the subcellular cortex is impaired by Par-3, this establishes a high cortical concentration of Myo-II on membranes perpendi cular to the axis (Zallen and Wieschaus, 2004). As a consequence, the surface tension increases and relaxes through a decrease of the cell–cell surface contact perpendicular to the axis, which leads to cell intercalation and extension of the germ band (Fig. 1) (Bertet et al., 2004). Understanding of this process was reinforced by successful in silico simulations mimicking germ-band extension with only these ingredients (Rauzi et al., 2008). Interestingly, planar polarity genes also appear to be critical in the generation of convergent extension morphogenetic movements in the embryos of other species, including zebrafish and Xenopus (Heisenberg and Tada, 2002; Keller, 2002; Roszko et al., 2009). Regarding mesoderm invagination, the Fog signaling pathway has been demon strated to involve the apical attraction of Myo-II through the activation of a Rho signaling pathway. Fog is a secreted signaling molecule that is expressed under the control of Twist in the mesoderm and in the posterior pole, activating apical redistribution of Myo-II (Dawes-Hoang et al., 2005). T48, another gene acting downstream of Twist, cooperates with Fog in triggering the apical attraction of RhoGEF2, a protein required for apical redistribution of Myo-II and for mesoderm invagination. In addition to the Twist-dependent activation of the RhoGEF2 apical redistribution process, Snail is also necessary for stable apical redistribution of Myo-II and mesoderm invagination, through a still unknown molecular mechan ism (Fig. 1). Different simulations were developed to test whether the apical surface tension increase induced by redistribution of Myo-II would be the only genetically con trolled active perturbation necessary for mesoderm invagination, or the invagination would require additive active movements, such as cell shortening. Whereas simula tions describing cells as a continuous viscoelastic medium suggest the necessity of an active shortening of mesodermal cells to accomplish invagination (Conte et al., 2008), hydrodynamical simulations describing the tissue as composed of individual cells with individual plasma membranes characterized by an actomyosin cortical tension and contractile apical rings connected by intercellular junctions suggest that the Myo-II-dependent increase of apical surface tension of mesodermal cells is sufficient to trigger the movements observed during invagination (Pouille and Farge, 2008). Simulations of mesoderm invagination with apical constriction only were also performed in sea urchin embryos, which have an extracellular matrix that should be specifically compliant to allow gastrulation (Davidson et al., 1995). However, the Myo-II activity was proposed to not be the only player of early embryonic morphogenesis. Effectively, recent simulations proposed the involve ment of the microtubule network within Caenorhabditis elegans epithelial cells in redistributing the stress originally produced by actomyosin-oriented actin filaments, thus leading to the elongation morphogenetic movement of the embryo (Ciarletta et al., 2009).

300

Maria-Elena Fernandez-Sanchez et al.

2. Cell Genetic Identity and Cell Sorting: Actomyosin Versus Adherence Surface Cortical Tensions In addition to the genetic control of myosin-II cell polarities, the origin of multicellular morphogenetic movements was proposed to be driven by the difference of adhesive surface tensions between cells of different differentiation states. In analogy with the physics of liquid mixing, cells characterized by a higher adhesive surface tension with their homologous cell types than with other cell types will have a tendency to aggregate, with the higher surface tension cells inside and the lower outside. Such proposal was first (A)

Torso-like Nanos

Torso-like Bicoid

Twist, Snail

(B)

E-Cad Myo-II

(C)

Germ-Band Extention

Mesoderm Invagination

(D)

Gastrulation

Fig. 1

(Continued)

12. Mechanical Induction in Embryonic Development and Tumor Growth

301

experimentally tested ex vivo by mixing two types of cells with differential adhesive properties (Foty and Steinberg, 2005). Interestingly, the adhesive surface tension of different cell types was measured by the application of a uniaxial deformation on monodifferentiated aggregates of cells. The pressure applied to the aggregate being controlled led to the evaluation of the tension by measuring the deformation of the aggregate “droplet” as a function of the applied force. Interestingly, the cells characterized by a higher surface tension adhesive interaction were effectively found in the core of the aggregate, when the two types of cells were mixed (Foty and Steinberg, 2005). The dynamics of associated demixing behavior could be simulated in the case of the hydra demixing process preceding the onset of regeneration morphogenesis (Graner and Glazier, 1992). Such a demixing mechanism was proposed ex vivo to be at the origin of the sorting between mesendoderm and ectoderm cells in the process of blastopore involution in zebrafish (Schotz et al., 2008). More recently, the involvement of the cytoplasmic cortical tension of cells associated to the cortical actomyosin tension was taken into consideration in addition to the adhesive surface tension and was proposed to drive demixing of endoderm, mesoderm, and ectoderm cells and the movement of the blastopore in vivo (Krieg et al., 2008). To ensure the stability and integrity of the tissue, the adhesive surface tension and cortical tension should be on the same order of magnitude, with slight differences in between the two governing the geometry of demixing and cell rearrangement (Lecuit and Lenne, 2007). For instance, simulations combining the effect of the cortical tension to the adhesive surface tension generated the shape of the Drosophila retina cells observed experimentally, in contrast to simulations privileging the adhesive surface tension (Kafer et al., 2007). C. Measuring the Active Forces Developed in Morphogenetic Movements In Vivo In the preceding case, the origin of the multicellular morphogenetic movements of the embryonic tissue is the anisotropy of Myo-II concentration within individual cells, Fig. 1 Control of multicellular morphogenetic movements in gastrulation of Drosophila embryos, via genetically controlled intracellular polarities in Myo-II concentration. Germ-band extension (A) Before gastrulation, the pattern of expression of developmental genes determining the anteroposterior polarity of the embryo is controlled by the expression of the maternal gene products Bicoid in the anterior and Nanos in the posterior (in red). (B) This combines to the expression of the terminal patterning genes controlled by the maternal gene Torso-like product, to establish the planar polarity of Myo-II submembranar concentration (in red, left). The origin of the underlying molecular mechanism linking anteroposterior patterning gene expression to planar polarity remains to be fully understood. (C) The consequence of the polarity is an increase of tension in membranes perpendicular to the anteroposterior axis, leading to a decrease of these surface areas, then to the dorsoventral cell intercalation (adapted from Bertet et al., 2004) extending the anteroposterior length of the tissue at gastrulation (D, green arrows). Mesoderm invagination. (A) Before gastrulation, the pattern of expression of developmental genes determining the dorsoventral polarity of the embryo is controlled by the expression of the maternally induced nuclear translocation of the transcription factor Dorsal, which activates the expression of the ventral mesodermal genes twist and snail (in green). (B) These genes are necessarily together to induce the submembrane apical accumulation of Myo-II (in red, right) that increases the apical surface tension. (C, D) This leads to the decrease of apical surface area compared to basal surface areas, triggering the inward curvature and invagination of the mesoderm at gastrulation. The understanding of the underlying molecular mechanisms linking the expression of the patterning genes twist and snail to apical attraction of Myo-II are better and better understood (see Fig. 4). (See Plate no. 6 in the Color Plate Section.)

302

Maria-Elena Fernandez-Sanchez et al.

leading to both cell migration intercalation movements in response to polar planarity anisotropies and cell shape changes in response to apicobasal polarities. Even though the link between the patterning of gene expression and the generation of a threedimensional embryonic morphology begins to be understood, the evaluation of the forces developed by these morphogenetic movements has until today been very rarely measured in vivo. For instance, looking at Xenopus embryonic explants, the measure ment of the deflection of the beam emerging from an optical fiber, of which the bending elastic constant has been calibrated and applied to the tissue submitted to convergent extension, leads to a maximum force of 1 µN (Moore, 1994). Such an apparatus is able to measure forces in the range of 50 nN to 10 µN, but necessitates working ex vivo, on tissue explants (Davidson and Keller, 2007). Another extensively studied morphogenetic movement is the Drosophila embryo dorsal closure, which combines the action of a closing purse string surrounding the amniosera tissue and constricting it, with contractile amniosera cell oscillations, extru sion of cells inside the embryo, and cell apoptosis in the amniosera (Jacinto et al., 2000; Solon et al., 2009; Tokoyama et al., 2008). In this case, fine mathematical analysis of the geometry of the tissue elements of dorsal closure, combined with the quantitative analysis of tissue relaxation photoablations of specific domains of the dorsal closure, lead to the evaluation of the ratio between the purse string and the amniosera tissue tensions collaborating in the driving of dorsal closure (Hutson et al., 2003). Relative forces can be evaluated directly by studying tissues dynamics but only a local mechan ical deformation allows access to the constraints field and the absolute forces. Measuring the forces associated to morphogenetic movements in vivo, within the developing embryo, requires the use of magnetic nanotechnologies to mechanically manipulate the living multicellular mechanical medium of the embryo from the inside. Injection of ferrofluids composed of 7-nm magnetic particles into the cytoplasm of Drosophila embryos at the end of cellularization allows the magnetization of a con densed pack of 50 µm cells, on which a force of 60 nN was applied by using a calibrated magnetic tweezer to quantitatively mimic the rate of deformation of anterior pole stomodeal cells of the embryo normally due to the convergent extension of the mesoderm at gastrulation (see protocol details in Section III-B-2-ii) (Desprat et al., 2008). The difference of two orders of magnitude of these forces is coherent with the fact the Drosophila embryo is typically 10 times smaller than the Xenopus embryo, which develops 1 µN forces, as the forces developed are generally proportional to the section of the tissue involved, which is square the size.

III. Mechanical Control of Gene Expression: Testing Mechanical Induction by Application of Endogenous Forces from the Inside of the Embryo Developmental genes control both the biochemical patterning and the generation of morphogenetic movements that geometrically shape the embryo. How does the

12. Mechanical Induction in Embryonic Development and Tumor Growth

303

genome control the state of elaboration of the patterns and shapes it is charged to develop? Regarding biochemical patterning, the genome is constantly testing the pattern of expression of developmental proteins through classical biochemical induc tion: the pattern of expression of the RNAs of a given stage of development is triggered in response to the pattern of expression of the proteins of the previous stage. Because multicellular morphology is not biochemical in nature, the existence of such feedback cannot be based on classical biochemical inductive cues. We proposed that it was rather due to mechanical cues associated to tissue deformation. A. Twist: A Master Mechanosensitive Gene in Early Drosophila Embryos The existence of mechanosensitive patterning genes making the expression of the genome sensitive to the biomechanical shape of its tissues was first postulated and demonstrated in early Drosophila embryos, just before gastrulation (Farge, 2003). A simple device composed of a tensed semipermeable membrane (BioFOLIE 25 Her aeus) clipped in between two homemade plexiglass concentric rings and a coverslip controlled by x,y,z micromanipulators coupled to a piezoelectric device (Physics Instruments M-UMR5,16 micromanipulators and P-762-1L Polytec PI piezoelectric, respectively) is used to softly deform embryos laterally by a uniaxial pressure applied to the entire embryo. In order to do so, embryos are first deposited on the membrane after dechorionation. They are then immersed in a clearing oil permeable to oxygen (halocarbon oil 27, Sigma) (Wieschaus and Nusslein-Volhard, 1998). The embryos are stably orientated laterally, and deformed by the micromanipulated coverslip in order to increase the dorsoventral size of the embryo on the order of 10% and of 10 min, which are the orders of magnitude of the length and time scales of Drosophila embryo morphogenetic movements at gastrulation. The semipermeable nature of the membrane allows the diffusion of the oxygen even in the presence of the coverslip, preventing hypoxia. The expression of the first zygotically expressed master genes of dorsoventral and anterior–posterior patterning can thus be screened during late stage 5 of develop ment, just before gastrulation (stage 6 designates the initiation of gastrulation, namely of morphogenetic movements). These expressions were first checked mostly by using LacZ reporter genes, as the b-galactosidase product of LacZ is particularly stable compared to endogenous proteins within embryonic contexts (Bradshaw et al., 1996). While the anterior–posterior patterns were found to be unchanged (the hunch back gradient and the even-skipped stripes remained unperturbed), the dorsoventral patterning was profoundly modified, with a repression of expression of the dorsal genes zen and dpp, and an ectopic expression of the ventral genes snail and twist all around the embryo, including the dorsal tissues (Farge, 2003). Strikingly, looking at endogenous morphogenesis, the expression of the Twist protein is found to be strongly amplified in anterior pole stomodeal cells, after 10–20 min of compression of these cells by the morphogenetic movement of germband extension, suggesting that Twist expression might be mechanically induced in the stomodeum at the onset of gastrulation (Fig. 2) (Desprat et al., 2008; Farge, 2003). Before beginning experiments of control of the deformation of anterior pole stomodeal

304

Maria-Elena Fernandez-Sanchez et al.

(B)

(A) 6

+0.05

6 I = 17,2% +− 13% (n = 37)

+0.03 +0.01

−0.01

7

7 −0.03

I = 75% +− 19% (n = 48)

−0.05

nls-GFP

PIV

Twi Twist pole

Fig. 2 Increase in Twist expression correlates with compression by GBE in stomodeal cells at the onset of gastrulation. (A) Nls-GFP nuclear labeling shows the compression of stomodeal cells (between red arrows) in between early stage 6 and late stage 7 due to GBE movements (orange arrows). Quantitative Particle imaging velocity analysis shows a 2% mn1 dynamics of compression during the first 10 min of GBE (encoded in red). (B) Twist is overexpressed at late stage 7 in stomodeal compressed cells (quantitative mean values combine Farge (2003) and Desprat et al. (2008) data, and figures adapted from the same references). (See Plate no. 7 in the Color Plate Section.)

cells, the expression of the functional protein Twist had to be checked in response to the uniaxial global deformation of the embryo using specific antibodies, instead of the LacZ constructs, which only signal the initiation of transcription in mechanical response to deformation. The classical 30 min at 20°C protocol of fixation of embryos (during which the exact time of fixation is not defined) being longer than the physiological 10 min characteristic time scale of stomodeal compression, the embryos were system atically fixed at 4°C in the 4% PFA fixative for 1 h, after 1 min–15 min of deformation, in order to immediately stop any active biochemical process at the onset of the fixation time. Within these time scale-controlled physiological conditions, the Twist protein was found to be expressed ectopically all around the embryo in response to mechanical deformation after typically 12 min, which is the dynamics of the stomodeal compression correlated to the overexpression of Twist at the onset of gastrulation (Farge, 2003). Interestingly, we subsequently found that embryos classically fixed at 20°C lose most of the ectopic dorsal expression of Twist (but not in anterior stomodeal pole cells, see below), in contrast to twi-lacZ embryos stably expressing b-galactosidase within the same conditions, possibly suggesting the existence of a highly dynamical negative feedback repressing any accidental mechanical induction of Twist in tissues not prepatterned to stabilize Twist mechanical induction (unpublished data). Consistent with this, the use of highly sensitive DAB peroxydase substrate amplification tools (DAB Vectastain ABC Elite kit, Vector) and good contrast camera with related imaging treatment program (Hamamatsu C4-742 95 and Hi-Pic) were necessary to detect ectopic expression of the Twist protein in whole-mount embryos, in contrast to anterior pole cell experiments in which immunofluorescence was sufficient (see below).

12. Mechanical Induction in Embryonic Development and Tumor Growth

305

B. Mechanical Induction of Twist in the Future Anterior Gut Cells, in Response to Myo-II-Dependent Germ-Band Extension Compression Testing the existence of mechanical cues leading to Twist mechanical induction in stomodeal cells in response to the endogenous morphogenetic movements of germband extension requires the elaboration of tools allowing the inhibition and rescue of the Myo-II-dependent germ-band extension morphogenetic movement in vivo.

1. Genetic Tools to Control Morphogenetic Movements: Mechanical Rescue of Twist Expression After Inhibition of Germ-Band Extension Taking advantage of the well-characterized genetics of early Drosophila embryo morphogenesis, one can use the anterior–posterior apolar triple mutant bicoid nanos torso-like to block germ-band extension (Irvine and Wieschaus, 1994). In these mutants, anterior stomodeal cells are found to not be compressed, with no amplification of Twist expression in these cells (Farge, 2003). A 50 µm micromanipulated needle was then used to compress stomodeal anterior pole cells in these mutants, with a physiological order of magnitude of deformation. In response to deformation, the amplification of the expression of Twist was rescued in these mutants, suggesting a mechanical induction of Twist by stomodeal cell compression in response to their compression by germ-band extension in wild-type embryos (Farge, 2003). The rescue of the strong expression of Twist in stomodeal cells was also triggered in response to the rescue of germ-band extension by using the bicoid torso-like double mutant, in which the expression of the posterior gene nanos allows the first 20 min of germ-band extension (Irvine and Wieschaus, 1994), without affecting the genetic background of anterior pole cells (Farge, 2003). Altogether, these results suggest that the endogenous morphogenetic activity of Myo-II, leading to germ-band extension (Bertet et al., 2004), mechanically induces the activation of Twist gene expression at the onset of Drosophila gastrulation.

2. Biophysical Tools in Morphogenetic Movement Control: Magnetic Tweezers Mechanical Rescue of Twist Expression After Mechanical Inhibition of Germ-Band Extension by Two-Photon Local Ablation The ability to use genetics and simple micromanipulated needles to control stomo deal cell deformation suggests endogenous mechanical induction of Twist in stomodeal cells. However, working with non-wild-type embryos using nonphysiological forces of tissue deformation prevents the establishment of a definitive conclusion, which would require experiments performed within fully physiologic genetic and biomechanical conditions. One has thus to develop new biophysical tools allowing control of the deformation of stomodeal cells, with finely controlled physiological forces and within the wild-type genetic context. i. Inhibiting Stomodeal Cell Compression by Photoablation Because the germ band largely extends posteriorly, leading to a strong dorsal wave of compression from

306

Maria-Elena Fernandez-Sanchez et al.

the posterior to the anterior pole of the embryo (Fig. 1), the ablation of the most dorsal tissue of the embryo was inhibited to prevent anterior pole stomodeal cell compression in wild-type embryos (Fig. 3A). In fact, the ablation of the most dorsal part of the embryo blocks the germ-band extension on a time scale largely higher than the characteristic time of stomodeal cell compression during normal development, inhibit ing both the posterior extension and the anterior extension which efficiently compress stomodeal cells (Farge, 2003; Supatto et al., 2005). The ablation uses high-power two-photon femtosecond microscopy. The twophoton technology allows the generation of a powerful irradiation only at the point of focus of the infrared incident beam. At this point only, the probability of condensing two photons at the same place and time becomes high, leading to a destructive beam whose energy is square the energy of the nondestructive original infrared beam. As a consequence, the incident beam can cross tissues without destroying them and will be destructive only at the point of convergence. In addition, because destructive ablation effects are proportional to the power of the energy of the laser (namely the energy deposited by unit of time), the fact that the femtosecond laser impulsions are very short (1015 s) means that the energy deposited in the impulses can remain very small in order to trigger a power high enough to generate destruction. Indeed, the heating of the embryos after ablation was evaluated to be on the order of 0.1°C only (Supatto et al., 2005). As a result of the laser treatment, wild-type ablated embryos showed an inhibition of stomodeal cell compression Fig. 3B, with an inhibition of the amplification of Twist expression in these cells Fig. 3C (Farge, 2003; Supatto et al., 2005). ii. Quantitative Rescue of Physiological Stomodeal Cell Compression: Ferrofluid Injection and Magnetic Manipulation To rescue the compression of stomodeal cells with physiological dynamics, a magnetic ferrofluid is injected into the anterior dorsal cells neighboring the stomodeal cells at the end of cellularization, after the photoablation of middle dorsal cells Fig. 3A (Desprat et al., 2008). The ferrofluid (composed of a 5 mol l1 volumic concentration of superparamagnetic monodisperse 5 nm maghemite -Fe2O3 nanoparticles stabilized with sodium citrate (Mayer et al., 1999)) is injected in front of the basal side of the cells still open to the yolk. As simple uncalibrated electromagnet (composed of a cylindrical pole piece of diameter 5 mm ended with a sharp tip and surrounded with 700 turns of coiled copper wire of diameter 0.7 mm) is applied perpendicularly to the tissue surface, to attract the ferrofluid into the cytoplasm of the cells, after successive 1 s pulses of 1 A. Once cells are magnetized by ferrofluid insertion, the electromagnet is removed, and the cells are subsequently attracted by a calibrated magnetic tweezers made of two conical small pieces of AFK1 pole pieces (length 6 mm; Imphy Alloys, France) at the tip of the two magnets (diameter 5 mm, length 10 mm; Binder Magnetic, gift of V. Croquette) so as to set a 1 mm interspace between the two magnetic poles (Fig. 3A). The position of the magnetic tweezer is systematically explored in order to quantitatively tune the dynamics of compression rescue of the ablated embryos to

307

12. Mechanical Induction in Embryonic Development and Tumor Growth

(A)

2-electro-magnet attraction

S

N

N

S

4-magnetic tweezers

3-photoablation

1-injection

(B)

nls-GFP

PIV

t = 0s

Ferrofulid

t = 600s (c)

F = 60nN Ablated

Ablated Rescued

Twi I = 18% +− 10% (n = 42)

I = 65% +− 14% (n = 25)

Fig. 3 Testing Twist mechanical induction in controlling stomodeal cells compression photoablation and magnetic manipulations. (A) The dorsal domain of the embryo is photoablated to block GBE, after injection and concentration of a calibrated ferrofluid of magnetic nanoparticles into anterodorsal cells. Calibrated magnetic tweezers are positioned in order to attract magnetized dorsal cells to compress anterior pole stomodeal cells with a deformation rate that mimics GBE endogenous compression dynamics. (B) Dynamics of stomodeal cells compression (in between red arrows) induced by magnetic manipulation mimicking the first 10 min of compression due to GBE during normal development, observed in nls-GFP and in PIV. A force of 60 nN is applied to stomodeal cells to quantitatively phenocopy the endogenous compression measured by PIV in Fig. 2A. (C) Inhibition of Twist overexpression in noncompressed stomodeal cells into the ablated embryo (in between red arrows), and recovery in the stomodeal cells of the ablated embryos in which the physiological compression is rescued by magnetic manipulation. Adapted from Desprat et al. (2008). (See Plate no. 8 in the Color Plate Section.)

308

Maria-Elena Fernandez-Sanchez et al.

the value of the endogenous compression of the nonablated wild-type embryos. Such comparison can be performed by using particle imaging velocimetry, a pro gram adapted from hydrodynamics to embryogenesis, in order to evaluate the local field of velocity as well as deformation of multicellular tissues (Desprat et al., 2008). After injection of 30 fl of ferrofluid, the endogenous dynamics of stomodeal compression of 2% min1 (namely 2 µm min1) is rescued for a distance between the tweezers and the magnetized cells leading to a gradient of magnetic field of 120 T m1 (measured by means of a small Hall probe KS14, Siemens) (Fig. 3B). Applied to the ferrofluid characterized by an magnetization of 2.1 105 A m1 (determined by a custom-made vibrating sample magnetometer according to Foner, LI2C, University Paris 6), this corresponds to the application of a force of 60 nN by magnetized cells onto stomodeal cells (Desprat et al., 2008). This com pression strain was applied for 10 min in order to mimic endogenous stomodeal cell compression, with immediate subsequent fixation of the embryos and classical labeling with anti-Twist antibodies. As a result, the strong expression of Twist was rescued in stomodeal cells in response to compression. Quantitative analysis revealed a level of expression rescue of Twist in stomodeal cells normalized to mesoderm cells of 65 ± 14%, which is amplified compared to the 18 ± 10% characterizing the ablated embryos (ablated and injected without magnetic field application) and comparable to the value of expression after compression by the endogenous movement of germ-band extension of 71 ± 19% (Fig. 3C) (Desprat et al., 2008). Student test confidence values of these analyses were p < 0.001. Thus, rescuing stomodeal cell compression from ablated noncompressed embryos by using the physiological biomechanical deformation of 2%. mn1 (representing slow movement of 2 µm mn1), within the physiological wild-type genetic background, quantitatively rescues the high level of Twist expression in the stomodeal cells compressed by germ-band extension, which is lost after inhibition of compression due to dorsal cell photoablations. This demonstrates that Twist overexpression is mechanically induced by stomodeal cell compression due to germ-band extension during endogenous development.

C. Underlying Molecular Mechanism of Mechanotransduction and Physiological Function in Development The underlying molecular mechanism of Twist mechanical induction is the mechani cally induced release of Armadillo/b-catenin from the junctions into the nuclei. Armadillo/b-catenin is the cotranscription factor of TCF. Dominant negative mutations of TCF and overexpression of Axin (which traps Armadillo/b-catenin in the cyto plasm, preventing any nuclear translocation) are both characterized by a lack of Twist mechanical induction, showing the necessity of the transcriptional activity of the b-catenin in Twist mechanical induction (Farge, 2003). Such release belongs to a

12. Mechanical Induction in Embryonic Development and Tumor Growth

309

mechanotransduction process that is dependent on Src42A, a protein triggering the inhibition of the interaction of b-catenin with E-cadherin in Drosophila embryos. Because Src42A is already activated (phosphorylated) before compression, and not overactivated in response to stomodeal cell mechanical strains, Src42A appears to be permissive in Armadillo/b-catenin released from the junctions, and is not directly involved in the mechanotransduction pathway (Desprat et al., 2008). One possibility, among others, is that mechanically induced conformational changes of junctional Armadillo/b-catenin would open sites of phosphorylation with activated Src42A, leading to the inhibition of its interaction with E-cadherin and to its release into the cytoplasm. Such activated Src-dependent mechanical activation of p130Cas was demonstrated in vitro, in response to mechanical induction of p130Cas changes of conformation (Sawada et al., 2006). Interestingly, mechanical activation of Armadillo/b-catenin was also found to be involved in mouse bone development, through muscle contractions inducing mechan ical shocks in between the bone synovial joints. The pluripotency of synovial bone cells, which control both bone growth and synovial joint differentiation, is indeed maintained by such mechanically induced nuclear translocation of b-catenin (Kahn et al., 2009). In addition, other pathways were suggested to be mechanically activated during development, like the MAL-D pathway during border-cell migration in oogen esis. The deformation due to the movement of these migrating cells in between static cells was suggested to activate the nuclear translocation of MAL-D, an event necessary to build a robust actin cytoskeleton in migrating cells (Somogyi and Rorth, 2004). Furthermore, one of the major problems to be solved in embryonic and animal development is what tells an achieved organ or tissue to stop growing. This issue is especially crucial to prevent tumor development where growing is often unrestrained. Whereas a consensus exists about the fact that morphogen gradients control such processes, it has been recently proposed in the case of the Drosophila imaginal disc that a mechanical stress, arising in the tissue from the nonuniformity of morphogen, could induce the arrest of cell proliferation (Hufnagel et al., 2007). This alternative scenario comes from a numerical simulation but could be tested precisely by the novel experimental approaches described here. Regarding the physiological function of Twist mechanical induction, the high expression of Twist at stage 7, which is mechanically induced by germ-band extension, was found to be necessary for midgut cell functional differentiation at late embryonic stages 14–16 (11–15 h of development) through the control of Dve expression, as well as for survival of 4- to 5-day larvae. Effectively, genetically controlled defects in Twist overexpression to the level of expression of noncompressed stomodeal cells (due to patterned and staged expression of Twi-RNAi interfering with Twist expression in stomodeal cells only and during compression only) lead to the loss of Dve expression in anterior midgut cells, as well as to lethality in typically 80% of cases (Desprat et al., 2008). Therefore, in Drosophila embryos, mechanical induction of Twist expression in stomodeal cells during their compression appears to control vital functional differen tiation of the anterior midgut cells of the embryo.

310

Maria-Elena Fernandez-Sanchez et al.

IV. Mechanotransduction in the Control of Posttranslational Morphogenetic Events: Mechanical Activation of the Myosin-II Apico-Basal Polarity Triggering Mesoderm Invagination In addition to controlling the sate of expression of the genome, mechanotransduction processes in development can also control posttranslational events, involving, for instance, myosin-II intracellular behaviors at the onset of gastrulation. In Drosophila embryos, gastrulation begins by a Snail- and Twist-dependent apical redistribution of Myo-II that leads to a constriction of apical cell surfaces. Such constriction generates a trapezoidal shape change of individual cells, leading to the decrease of the apical surface area of the mesoderm compared to the basal surface area, which induces the inward bending constraints of mesoderm invagination (Sweeton et al., 1991).

A. Mechanical Activation of Myo-II Apical Redistribution Hypothesis Emerging from the Theoretical Analysis of the Genetics of Drosophila Mesoderm Invagination Interestingly, there exist two phases of apical constriction. The first 4 min one is stochastic, randomly involves the uncorrelated reversible pulses of constriction and relaxation of individual cells, and is unable to trigger mesoderm invagination (Sweeton et al., 1991). These pulses are associated with reversible pulses of apical spots of Myo-II (Martin et al., 2009). The second one is collective and involves the constriction of all mesodermal cells (Sweeton et al., 1991), through a process of pulsatile constrictions, including a ratchet process progressively stabilizing cell apexes into more and more constricted states. This is associated with the progressive stabilization of the apical spots of Myo-II, leading to apical Myo-II coalescence and redistribution (Martin et al., 2009). Because mutants of twist only show the stochas tic phase, the collective phase is Twist dependent. In fact, the Fog secreted factor, which is expressed under the control of Twist, is the key signaling protein triggering the collective phase (Costa et al., 1994). However, the snail mutants are defective in both the stochastic and collective phases, indicating that the stochastic phase is indeed Snail dependent, but also that the two phases interact (Fig. 4A). Strikingly, a purely biochemical interaction between the Snail and Twist/Fog underlying genetic and biochemical networks can be excluded by the following observations. In mutants of Snail, in which Fog is still expressed in the mesoderm (Morize et al., 1998), no apical redistribution of Myo-II can be observed in the mesoderm (Martin et al., 2009; Pouille et al., 2009). In addition, the ectopic expression of Fog all around the embryonic tissue does not rescue apical constriction in snail mutants mesoderm, whereas it does in twist mutants (Morize et al., 1998). Thus, Fog alone is not sufficient to trigger apical redistribution of Myo-II and mesoderm invagination, but Fog and Snail together are necessary for Myo-II apical redistribution (Seher et al., 2007). On the other hand, the ectopic expression of Fog all around wild-type embryo shows an apical redistribution of myosin-II in all tissues of the embryo, including in

(B)

dl

Snail

(E)

Twist Fog+

Fog+

Stage 6

Early Stage 6

Fog+

WT

Late Stage 6

Late Stage 6

(G)

a

sna-/sna

Wild Type

Wild Type

b

sna-/sna-

Frog, T48

time First Stochastic Second Collective Constriction Wave Constriction Wave

(C)

position

Inv GBE Mesoderm

Twi

Mesoderm

Sna

(F)

dl

sna-/sna- Stage6

Snail

c Twist Fog

sna-/sna- indented invaginated

d sna-/sna- indented non-invaginated

AND

(D) WT

sna

sna indented

Myo-ll apical concentration

First Stochastic Second Collective Constriction Wave Constriction Wave Myo

311

Fig. 4 Mechanical induction of Myo-II apical redistribution leading to mesoderm invagination. (A) The genetic network controlling mesoderm invagination. The mesoderm Snail and Twist/Fog,T48 pathways cooperate in triggering the apical constriction leading to mesoderm invagination (green arrow). (B) Simulating an embryo in which Fog is expressed ectopically, and responding to mechanical strain by mechanical activation of the Fog signaling pathway, phenocopies the experimentally observed propagation of Myo-II apical redistribution and flattening from mesoderm to lateral and dorsal tissues (see text). (C) Indenting sna mutants to rescue the lack of mechanical strain of snail mutants and to test the mechanical reactivation of the Fog signaling pathway controlling both apical redistribution of Myo-II and mesoderm invagination. (D) Rescuing apical accumulation of Myo-II and mesoderm invagination, lost in the sna mutant, after soft indent of the mesoderm of sna mutants, in a Fog-dependent process. (E) Kymographs of the mesoderm constriction movements at earliest stage 6 in the wild type, which is not observed in earliest stage 6 nonconstricting sna mutant embryos, as a phenotype criteria to select sna mutant embryos a priori, before indent (see text). The sna mutants are doubled checked through the delay in anterior midgut formation of snail homozygous compared to wild-type and heterozygous mutants (yellow arrow). (F) The mechano-genetic network controlling mesoderm invagination. Snail initiates apical constriction fluctuations, which activates the Fog signaling pathway (through the mechanical blocking of Fog endocytosis, see text), leading to the coordinated Myo-II apical stabilization and coordinated apical constriction of mesoderm cells necessary for mesoderm invagination. Adapted from Pouille et al. (2009). (G) Strong expression of Twist at late stage 8 1 h after mesoderm invagination into wild types (A) is lost in homozygous mutants of snail ((B), 84%, n = 6) and rescued after indentation of the mesoderm in the complete pool of both the embryos having invaginated (C, n = 7) or having not invaginated (D, n = 9) after indent. Green is Twist labeling (Alexa 488 emission following (Desprat et al., 2008) labeling procedure). Snail homozygous were determined following the phenotype criteria described in Pouille et al. (2009), or using Snail labeling in red (anti-snail made in guinea pig was a generous gift of Yutaka Nibu and was used at 1:200 with Alexa 546 anti-guinea pig from Molecular Probes at 1:100). Note that the nuclear expression of Snail in the wild-type mesoderm is low at stage 8, as described in Alberga et al. (1991) and inexistent in snail mutants. (See Plate no. 9 in the Color Plate Section.)

12. Mechanical Induction in Embryonic Development and Tumor Growth

(A)

312

Maria-Elena Fernandez-Sanchez et al.

ectoderm cells in which Snail is not expressed (Dawes-Hoang et al., 2005). So, contrary to the previous observations, this would suggest that Fog alone is able to trigger the apical redistribution of Myo-II. This apparent contradiction can be solved by considering that the interaction between the two phases is not mediated biochemically, but by cell–cell mechanical interactions via a Fog-dependent mechanotransduction process leading to apical redis tribution of Myo-II (Pouille et al., 2009). Effectively, considering that the mechanical strains (mean surface tension or mean cell pressure) developed by the stochastic phase of constriction could trigger a Fog-dependent mechanical activation of apical redis tribution of Myo-II in the mesoderm, this would mean that stable apical redistribution of Myo-II would require mechanical strains plus Fog expression. Such mechanical strains being absent in the mesoderm in a snail mutant, no mesodermal apical redis tribution of Myo-II is observed in the snail mutants, even in the presence of Fog. On the other hand, if Snail is expressed in the mesoderm, the mechanical strains would activate the Fog-dependent mechanotransduction pathway and lead to the apical redistribution of Myo-II, then to mesoderm invagination. In turn, mesoderm invagina tion mechanically strains ventrolateral cells. In the wild type, in the absence of Fog in these cells, the apical redistribution of Myo-II is not induced in nonmesoderm cells and is thus restricted to the Fog expressing mesoderm. On the other hand, in embryos in which Fog is expressed ectopically all around the embryo, the apical redistribution of Myo-II will be first activated in ventrolateral cells in response to stretching by mesoderm cells, in which the apical stress will increase. In turn, these cells will stress more lateral cells and activate apical redistribution of Myo-II, a process that will propagate from ventral cells to dorsal cells after the initiation of mesoderm invagina tion of stage 6.

B. In Silico Physical Tools to Test Theoretically the Viability of the Mechanotransduction Hypothesis Such a scenario can be predicted by simulations of mesoderm invagination in which the Snail-dependent stochastic phase is introduced with a Fog-dependent mechanical activation of stable apical redistribution of Myo-II. The Fog-dependent mechanical activation is first tuned in the mesoderm to mimic mesoderm invagination in the wild type. It is then added all around the embryo to mimic the genetic background associated with the ectopic expression of Fog. The output of the simulation is the production of an embryo characterized by apical flattening in lateral ectoderm and dorsal cells, leading to a lateral tension preventing the formation of a complete mesoderm invagination (Pouille et al., 2009) (Fig. 4B), a phenotype characteristic of embryos overexpressing Fog ectopically (Morize et al., 1998). This suggests, but yet does not prove, the possibility of an underlying mechanotransduction mechanism of interaction between the Snail (initiating deformations) and Fog (actively responding to the strain by apical stabilization of Myo-II) networks, necessary for early Drosophila embryo mesoderm invagination.

12. Mechanical Induction in Embryonic Development and Tumor Growth

313

C. Coupling of Mechanical and Genetic Tools to Experimentally Test Mechanotransduction in Myo-II Apical Redistribution The core of the model is that Snail expression in itself is not necessary with Fog to trigger the apical redistribution of Myo-II, but the mechanical strains developed by Snail expres sion with Fog are. Thus, if the model is correct, rescuing the existence of a mechanical strain in the mesoderm of snail mutants should rescue both the apical redistribution of Myo-II and the mesoderm invagination, both of which are missing in these mutants. The snail mutant embryos are thus indented 5 µm (namely 30% of the thickness of the ectoderm and 2% of the thickness of the embryo) locally in the middle of the mesoderm, with a micromanipulated needle, precisely 2–3 min after the end of cellularization, which signals the initiation of the stochastic phase in wild-type embryos (Fig. 4C). As a result, both apical redistribution of Myo-II and invagination were rescued in 67% of the indented sna homozygous mutant embryos (Fig. 4D), not only in the indented tissue, but throughout the complete mesoderm, suggesting a propagation of the contractile wave prepatterned by Twist expression (Pouille et al., 2009). This percentage decreases to 38% when the indent is realized 10 min after at the onset of Germ-Band Extension (GBE), probably because of a competition with the GBE morphogenetic movement (Pouille et al., 2009). Interestingly, twi sna double mutants do not show any response of the mesoderm to mechanical indent, showing that the mechanotransduction pathway is dependent on the expression of Twist in the mesoderm. On the other hand, indenting a twi sna double mutant, in which Fog has been additionally expressed only in the mesoderm, rescues the apical redistribution of Myo-II and mesoderm invagination. In contrast, in the absence of indent, Fog expression alone does exhibit any rescue within the sna twi context (Pouille et al., 2009). Thus, Fog expression alone, without Snail, does not induce the apical redistribution of Myo-II, but rescues the apical redistribution of Myo-II in response to mechanical strains. Such genetic manipulation can be realized by crossing a sna twi double mutant with a twi PE-Fog transgenic mutant (Seher et al., 2007), in which Fog is expressed under the control of the proximal element of the promoter of Twist (PE) which is known to control the expression of Twist in the mesoderm only. Thus, Fog is expressed in the mesoderm only, within a sna twi context (because the single mutation of snail adds to the fact that Snail expression is highly deficient at stage 6 within the twi mutant context (Leptin, 1991), as confirmed by no effect of mesoderm invagination of the twi sna/twi; PE-Fog nonindented embryos (Pouille et al., 2009)). The determination of the snail homozygous mutant phenotype is realized a posteriori just before the indent, based on the fact that the Snail-dependent stochastic phase of constriction is signaled by a contraction of the mesoderm initiating precisely at the end of mesoderm cellularization, 10 min before the initiation of germ-band extension (precise timing is allowed by sagittal observations of the embryos), which can be quantified by using kymographs (Fig. 4E) (Pouille et al., 2009). Thus, waiting for typically 2–3 min after the end of cellularization without a constriction ensures that the embryo is sna homozygous, allowing the indent of the embryo at 3 min, which is sufficient to rescue the apical redistribution of myosin-II in all cases and the mesoderm invagination in nearly 90% of cases (and in nearly 40% of cases for an indent at 10 min, probably because late mesoderm invagination enters into competition with germ-band extension) (Pouille et al., 2009). The snail homozygous are

314

Maria-Elena Fernandez-Sanchez et al.

double checked a posteriori by the fact the timing of the anterior midgut formation compared to GBE initiation is delayed of 10 min in sna homozygous mutants compared to wild-type and heterozygous mutants (Pouille et al., 2009). D. Mechanical Modulation of Fog Endocytosis: A Potential Underlying Molecular Mechanism of Mechanotransduction Fog is a secreted signaling protein that activates the Rho pathway through its interaction with its putative receptor Cta (Costa et al., 1994). The activation of the pathway leads to the apical attraction of Myo-II, possibly through its release from microtubules, and its attraction by Cta, helped by T48 (Kolsch et al., 2007). One of the underlying mechan otransduction mechanisms involving secreted signaling proteins is the mechanical mod ulation of endocytosis. Cell culture experiments have already shown the possibility of enhancing or triggering the activation of transduction pathways due to the increase of membrane mechanical tension, leading to the flattening of the membrane, and thus to the inhibition of endocytosis of signaling proteins (Rauch et al., 2002). Membrane tension can be activated by the increase of the volume pressure in the cells (for instance, due to a mechanical deformation by pressure applied to the cells). Generally, the endocytosis of signaling proteins involves the degradation of the interaction with its specific receptor inside the endosomal compartments and the arrest of downstream signaling pathway inhibition. Therefore, the mechanical inhibition of signaling protein endocytosis leads to the enhancement of the activation of the downstream transduction pathway. Under sub threshold concentrations, the signaling protein is unable to activate the pathway within normal conditions. Within these conditions, the mechanical blocking of endocytosis is also able to rescue the activation of the pathway (Rauch et al., 2002). Fog being a secreted signaling protein, the role of the pressure developed in the mesoderm by the Snail-dependent stochastic phase of constriction, triggering the inhibi tion of Fog endocytosis, leading to the activation of the downstream Rho transduction, can also be investigated by coupling mechanical with genetic tools. Building a double shi sna mutant allows blockage of endocytosis, thanks to the temperature-sensitive shi mutation of dynamin which inhibits endocytosis within 2 min. Labeling Fog with a specific anti body confirmed the accumulation of Fog at the plasma membrane under conditions of inhibited endocytosis, as compared to the permissive temperature. Such plasma mem brane accumulation of Fog is specifically observed in indented mutants of sna, showing that the indent of the mesoderm mechanically induces the blockage of Fog endocytosis. This plasma membrane accumulation is also observed during the first 4 min of Sna-dependent stochastic constrictions in wild-type embryos, showing that Snail induces the inhibition of Fog endocytosis. Finally, the blocking of endocytosis in the sna shi double mutant rescues the apical redistribution of Myo-II and mesoderm invagination, both of which lack in the sna shi mutants in endocytosis permissive temperature conditions. Together these observations suggest that the mechanical strains developed by Sna dependent stochastic constrictions in the mesoderm lead to the inhibition of Fog endocytosis, in turn leading to the activation of the downstream Rho pathway (Fig. 4F) (Pouille et al., 2009).

12. Mechanical Induction in Embryonic Development and Tumor Growth

315

E. Mechanical Induction of Twist Expression into the Mesoderm Interestingly, Twist expression into the mesoderm is known to dramatically decrease at late stage 8, 1 h after mesoderm invagination (Leptin, 1991) in sna mutants lacking the morphogenetic movement of mesoderm invagination (Fig. 4G), suggesting a possible participation of mechanical strains associated with mesoderm invagination at stage 6 into Twist expression at stage 8 (Brouzes et al., 2004). Consistent with this, after labeling indented sna homozygous mutants at late stage 8 by using classical procedures (Desprat et al., 2008), we found a rescue of Twist expression into the mesoderm, both in the pool of embryos having responded by mesoderm invagination rescue and in the pool having not responded that show a typical snail mutant noninvagination phenotype at late stage 8 (Fig. 4G, see Section IV-C regarding the conditions to generate the two pools). F. Incidences of the Mechanical Induction of Apical Redistribution of Myosin-II in Developmental Biology Interestingly, Myo-II dynamics was also found to be regulated by tension in actin cables, proposed to be maintained by another positive feedback mechanism to generate efficient germ-band extension tissue elongation in Drosophila embryos (FernandezGonzalez et al., 2009). In Xenopus, the correct spatiotemporal assembly of the fibronectin matrix, a key process in the morphogenesis of the embryo, was suggested to be regulated by a tension integrin-dependent process (Dzamba et al., 2009). A role of mechanical strains in the regulation of microtubules orientation during meristem development was also suggested (Hamant et al., 2008). Such positive mechanical feedback from strains to tensile or structural molecule redistribution could also be at work in processes of tissue reactive contraction resistance to stress having been proposed to be involved in Xenopus embryogenesis (Beloussov et al., 2006). Here, the finding of Fog signaling as a mechanotransduction pathway specifically possesses two potential distinct implications in developmental biology.

1. Long-range and Rapid Cell–Cell Interactions Through Mechanical Cues The first implication, very directly addressing the understanding of the respective roles of Snail and Fog in the apical redistribution of Myo-II, indicates the existence of mechanical cues allowing rapid and long-range interactions between nonadjacent cells mediated by mechanical cues and mechanotransduction. Effectively, the fact that Myo-II is redistributed apically within the mesoderm in the presence of Fog and Snail, but all around the embryo when expressing Fog ectopically despite the expres sion of Snail remaining restricted to mesodermal cells, shows that the expression of both Snail and Fog is necessary, but not necessarily within the same cells, to trigger the apical redistribution of Myo-II. In other words, Snail and Fog interact across very distant cells (mesodermal to dorsal, (Fig. 4B), through the lateral propagation of mechanical strains initiated by the stochastic pulses of the apical constriction generated by Snail in the mesoderm (Dawes-Hoang et al., 2005).

316

Maria-Elena Fernandez-Sanchez et al.

2. Coordinated Constrictions Integrated by Mechanical Cues and Mechanotransduction Leading to Mesoderm Invagination? The second implication addresses the question of the intimate mechanisms leading to the coordination of apical constriction which is necessary for mesoderm invagina tion. It indicates the existence of processes triggering collective ordered cell behaviors directly by the increase of the stochastic fluctuations of behavior of individual cells. Effectively, these experiments suggest that the Snail-dependent fluctuations feedback to individual cell states through Fog-dependent mechanotransduction, leading to the activation of a strong constriction of individual cells which is coordinated via mechan ical interactions between cells that propagate very rapidly through the mesoderm, via cell surface deformations or possibly via internal cells hydrostatic pressure (Pouille et al., 2009). In contrast to physical system behaviors, in which fluctuations at a given scale fight against ordered collective behaviors at the same scale, here fluctuations would trigger ordering and coordination restricted to mesoderm, because of mechanotransduction patterned by Fog expression. Such collective coordination determines the very efficient multicellular morphogenetic movement of mesoderm invagination.

V. Incidences of Mechanical Induction in Tumor Development The activation of tumor genetic programs has long been proposed to possibly be associated with the anomalous reactivation of embryonic programs in adult tissues (Brabletz et al., 2005). Strikingly, the nuclear translocation of b-catenin is a signature of tumor initiation and progression in many tissues, and especially in human and mouse colon tumors (Kirchner and Brabletz, 2000; Morin et al., 1997). Because the nuclear translocation of b-catenin from the junctions mediates mechanical activation of Twist in early Drosophila embryos (Desprat et al., 2008; Farge, 2003), we asked the question of a putative mechanical activation of b-catenin nuclear translocation in response to the strains developed by the mechanical pressure associated with tumor growth in the tissue surrounding the tumor (i.e., the stroma) (Brouzes et al., 2004; Whitehead et al., 2008). Following our Drosophila embryo protocols, we began to deform mouse tissues with uniaxial pressures applied to colon tissue explants. These experiments were designed to test the potentiality of an oncogenic biochemical response of the tissue in response to artificial mechanical strains. However, in this specific case, the pressure can be controlled at the level of the intestinal transit pressure of the mice, the colon being submitted to such natural pressure daily. Wild-type tissues did not exhibit any response at all to the b-catenin, or to the two target genes Twist-1 (involved in invasivity) and Myc (involved in cell division and tumor growth). Inter estingly, regarding the b-catenin pathway, a major difference between the wild-type mouse tissue and the early Drosophila embryo tissue is the state of expression of adenomatous polyposis coli (APC) collaborates with the GSK-3 system to send the

12. Mechanical Induction in Embryonic Development and Tumor Growth

317

cytoplasmic b-catenin into a degradation pathway, preventing b-catenin translocation into the nucleus. APC is not expressed in the early Drosophila embryo (Hayashi et al., 1997), but is expressed in mouse colon tissues. To test that the expression of APC in wild-type colon tissues could prevent the nuclear translocation of b-catenin after mechanical strain, we strained heterozygous mutants of APC. In these tissues, some of the b-catenin is translocated into the nulcei, with the observation of the expression of Twist and Myc target genes and protein products (Whitehead et al., 2008). Thus, loss of 50% of APC expression leads to a defect of the degradation of the b-catenin released from the junctions to the cytoplasm in response to mechanical strain and to the translocation of a certain pool of b-catenin into the nuclei, where it is able to trigger the activation of oncogene transcription. Eighty percent of human colon cancer tumors carry APC mutations, of which 10% are hereditary mutations. In these cases, the question of the sensitivity of such pretumoral colon tissues to intestinal transit is potentially addressed by our observa tions. Should in vivo studies confirm such behavior, adopting an alimentary regime regulating the stiffness of the food might decrease the probability of developing tumors in the APCþ/ context. The other 90% of cases first develop a sporadic mutation in one cell, leading to the natural growth of a clonal APCþ/ domain. Inside this domain, a second sporadic event involving loss of the second allele of APC in one cell is thought to trigger the transition to cancer (a cell with complete loss of APC is no longer able to prevent the nuclear translocation of the b-catenin which is constantly produced by the cells). Thus, an APC/ clonal domain grows within the APCþ/ pretumoral domain. Our observations thus ask the question of a potential activation of b-catenin nuclear translocation and target oncogene expression in the APCþ/ pretumoral tissue domain in response to the pressure developed by APC/ tumor growth, which might amplify tumor progression. Mechanical activation of cytoskeletal elements in response to the variation of the stiffness of cell substrates (related to tumor rigidity) was already systematically studied in breast cancer (Bissell et al., 2005). A correlation has also been established between prostate tumorigenicity in vivo and disorganized growth in a laminin-rich matrix gel determined by the dysregulation of vimentin and b1-integrin (Zhang et al., 2009). Furthermore, a novel both chemical and mechanosensitive signaling pathway that controls angiogenesis has been found, whose deregulation contributes to development of many diseases, including cancer, involving a direct regulation of p190RhoGAP by growth factors, integrin-dependent Extra Cellular Matrix (ECM) binding, and mechanical distortion of the cytoskeleton, which in turn controls VEGFR2 (vascular endothelial growth factor receptor 2) expression by modulating the balance between two mutually antagonistic transcription factors, TFII-I and GATA2 (Mammoto et al., 2009). Here, we ask the question of the activation of signaling pathways connected to oncogene expression directly, with future investigations designed to probe the involve ment of the mechanical induction process in response to tumor growth during tumor progression (Alexander et al., 2008).

318

Maria-Elena Fernandez-Sanchez et al.

VI. Mechano-Genetics Network in Perspective of Evolution: Reactivation of a Primitive Feeding Response of Ancient Embryos Recapitulated in Embryonic Morphogenetic Invagination? Mechanical Induction in First Multicellular Organism Emergence? Coming back to Myo-II apical redistribution mechanosensitivity in the early Drosophila embryo, a last implication belongs to evolutionary speculation addres sing the question of the emergence of the feeding reflex of ancient embryos in response to touch, and its evolution by integration of the underlying mechanisms regulating embryo morphogenesis. The idea that ancient embryos must have devel oped primitive motor sensorial responses of multicellular phagocytosis in response to touch has long been suggested and developed (Jaegerstem, 1956; Wolpert, 1992). Such phagocytosis was thought to be a response of tissues to touch due to contact of embryos with the ground after gravity-driven migration. The contact was proposed to activate a primitive motor sensorial response of invagination to touch, leading to the phagocytosis of sediments. Strikingly, sna mutant embryos react by an active gen eration of invagination in response to touch in Fog expressing domains (Fig. 4C and D). We thus proposed that we might have reactivated in early Drosophila embryos an ancient feeding reflex response to touch (Farge, 2003; Pouille et al., 2009). In other words, we suggest that the emergence of the Fog/Myo-II mechanosensitive pathway, or a primitive equivalent, might have been at the origin of the generation of a transient primitive gastric organ in response to external stimuli of touch. Strikingly, this would mean that the emergence of such mechanotransduction pathway would have been the key event, leading to the emergence of the first organisms (by definition, a multicellular system with an organ) from the earliest embryos defined as an aggregation of cells without collective functional cell behavior. Following this view, the “cell aggregate” to “first organism” transition could be thought of as the consequence of the emergence of a mechanosensitive Myo-II apical redistribution in response to external strains. With regard to the polarized mechanical rescue of Twist expression in sna mutants (see paragraph IV-E and Fig. 4), it would also be tempting to speculate that the local response of earliest embryos tissues to mechan ical contact with the ground after gravity sedimentation could have participated in the determination of the primary axis formation of earliest embryos through mechano transduction. Then, we speculated that the generation of mechanical strains due to Snaildependent stochastic oscillations developed from the inside of the mesoderm tissue has replaced the external stimuli, in such a way that a permanent primitive gastric organ evolved to develop independently of the external stimuli. This might have initiated the process of morphogenesis by co-opting a favorable response of the embryo to external mechanical stimuli for use in response to the internal mechanical stimuli, leading to the generation of a gastric organ (Pouille et al., 2009).

12. Mechanical Induction in Embryonic Development and Tumor Growth

319

Acknowledgments We thank Joanne Whittehead for her reading of the manuscript and Anne-Christine Brunet for her contribution to embryos labelings. Padra Ahmadi realized the indent experiments. The author’s lab is funded by the ANR (PCV and PiriBio ANR-09-Piri-0013-02), the ARC, Microsoft, NanoIdF, the Foundation Pierre Gilles de Gennes and the HFSP RGP001-14/2006 grants.

References Alberga, A., Boulay, J. L., Kempe, E., Dennefeld, C., and Haenlin, M. (1991). The snail gene required for mesoderm formation in Drosophila is expressed dynamically in derivatives of all three germ layers. Development 111, 983–992. Alexander, N. R., Branch, K. M., Parekh, A., Clark, E. S., Iwueke, I. C., Guelcher, S. A., and Weaver, A. M. (2008). Extracellular matrix rigidity promotes invadopodia activity. Curr. Biol. 18, 1295–1299. Aristotle II, 22, 523 a 15. Beloussov, L. V., Luchinskaya, N. N., Ermakov, A. S., and Glagoleva, N. S. (2006). Gastrulation in amphibian embryos, regarded as a succession of biomechanical feedback events. Int. J. Dev. Biol. 50, 113–122. Bertet, C., Sulak, L., and Lecuit, T. (2004). Myosin-dependent junction remodelling controls planar cell intercalation and axis elongation. Nature 429, 667–671. Brabletz, T., Jung, A., Spaderna, S., Hlubek, F., and Kirchner, T. (2005). Opinion: Migrating cancer stem cells—an integrated concept of malignant tumour progression. Nat. Rev. Cancer 5, 744–749. Bradshaw, M. S., Shashikant, C. S., Belting, H. G., Bollekens, J. A., and Ruddle, F. H. (1996). A long-range regulatory element of Hoxc8 identified by using the pClasper vector. Proc. Natl. Acad. Sci. USA 93, 2426–2430. Brouzes, E., and Farge, E. (2004). Interplay of mechanical deformation and patterned gene expression in developing embryos. Curr. Opin. Genet. Dev. 14, 367–374. Brouzes, E., Supatto, W., and Farge, E. (2004). Is mechanosensitive expression of twist involved in mesoderm formation? Biol. Cell 96, 471–477. Ciarletta, P., Ben, Amar M., and Labouesse, M. (2009). Continuum model of epithelial morphogenesis during

Caenorhabditis elegans embryonic elongation. Philos. Trans. A Math. Phys. Eng. Sci. 367, 3379–3400.

Conte, V., Munoz, J. J., and Miodownik, M. (2008). 3D finite element model of ventral furrow invagination

in the Drosophila melanogaster embryo. J. Mech. Behav. Biomed. Mater. I, 188–198. Costa, M., Wilson, E. T., and Wieschaus, E. (1994). A putative cell signal encoded by the folded gastrulation gene coordinates cell shape changes during Drosophila gastrulation. Cell 76, 1075–1089. Davidson, L., and Keller, R. (2007). Measuring mechanical properties of embryos and embryonic tissues. Methods Cell Biol. 83, 425–439. Davidson, L. A., Koehl, M. A., Keller, R., and Oster, G. F. (1995). How do sea urchins invaginate? Using biomechanics to distinguish between mechanisms of primary invagination. Development 121, 2005–2018. Dawes-Hoang, R. E., Parmar, K. M., Christiansen, A. E., Phelps, C. B., Brand, A. H., and Wieschaus, E. F. (2005). Folded gastrulation, cell shape change and the control of myosin localization. Development 132, 4165–4178. Desprat, N., Supatto, W., Pouille, P. A., Beaurepaire, E., and Farge, E. (2008). Tissue deformation modulates Twist expression to determine anterior midgut differentiation in Drosophila embryos. Dev. Cell 15, 470–477. Dzamba, B. J., Jakab, K. R., Marsden, M., Schwartz, M. A., and DeSimone, D. W. (2009). Dev. Cell 16, 421–432. Farge, E. (2003). Mechanical induction of twist in the Drosophila foregut/stomodeal primordium. Curr. Biol. 13, 1365–1377. Fernandez-Gonzalez, R., Simoes Sde, M., Roper, J. C., Eaton, S., and Zallen, J. A. (2009). Myosin II dynamics are regulated by tension in intercalating cells. Dev. Cell 17, 736–43. Foty, R. A., and Steinberg, M. S. (2005). The differential adhesion hypothesis: A direct evaluation. Dev. Biol. 278, 255–263. Fox-Keller, E. (2003). “Making Sense of Life.” Harvard University Press, Cambridge.

320

Maria-Elena Fernandez-Sanchez et al. Garber, R. L., Kuroiwa, A., and Gehring, W. J. (1983). Genomic and cDNA clones of the homeotic locus Antennapedia in Drosophila. EMBO J. 2(11), 2027–2036. Graner, F., Glazier, A., and Gehring, J. A (1992). Simulation of biological cell sorting using a 2-dimensional extended potts-model. Phys. Rev. Lett. 69, 2013–2016. O. Hamant, M. G., Heisler, H., Jšnsson, P., Krupinski, M., Uyttewaal, P., Bokov, F., Corson, P., Sahlin, A., Boudaoud, E. M., Meyerowitz, Y., Couder, J., Traas, Developmental patterning by mechanical signals in Arabidopsis. Science 322, 1650–1655 (2008). Hayashi, S., Rubinfeld, B., Souza, B., Polakis, P., Wieschaus, E., and Levine, A. J. (1997). A Drosophila homolog of the tumor suppressor gene adenomatous polyposis coli down-regulates �-catenin but its zygotic expression is not essential for the regulation of Armadillo. Proc. Natl. Acad. Sci. USA 94, 242–247. Heisenberg, C. P., and Tada, M. (2002). Zebrafish gastrulation movements: Bridging cell and developmental biology. Semin. Cell Dev. Biol. 13, 471–479. Hufnagel, L., Teleman, A. A., Rouault, H., Cohen, S. M., and Shraiman, B. I. (2007). On the mechanism of wing size determination in fly development. Proc. Natl. Acad. Sci. USA 104, 3835–3840. Hutson, M. S., Tokutake, Y., Chang, M. S., Bloor, J. W., Venakides, S., Kiehart, D. P., and Edwards, G. S. (2003). Forces for morphogenesis investigated with laser microsurgery and quantitative modeling. Science 300, 145–149. Irvine, K. D., and Wieschaus, E. (1994). Cell intercalation during Drosophila germband extension and its regulation by pair-rule segmentation genes. Development 120, 827–841. Jacinto, A., Wood, W., Balayo, T., Turmaine, M., Martinez-Arias, A., and Martin, P. (2000). Dynamic actinbased epithelial adhesion and cell matching during. Drosophila dorsal closure. Curr. Biol. 10, 1420–1426. Jaegerstem, G. (1956). The early phylogeny of the metazoan. The bilaterogastrea theory. Zoot. Bidrag. (Uppsala) 30, 321–354. Kafer, J., Hayashi, T., Maree, A. F., Carthew, R. W., and Graner, F. (2007). Cell adhesion and cortex contractility determine cell patterning in the Drosophila. Proc. Natl. Acad. Sci. USA 104, 18549–18554. Kahn, J., Shwartz, Y., Blitz, E., Krief, S., Sharir, A., Breitel, D. A., Rattenbach, R., Relaix, F., Maire, P., Rountree, R. B., Kingsley, D. M., and Zelzer, E. (2009). Muscle contraction is necessary to maintain joint progenitor cell fate. Dev. Cell 16, 734–743. Keller, R. (2002). Shaping the vertebrate body plan by polarized embryonic cell movements. Science 298, 1950–1954. Kirchner, T., and Brabletz, T. (2000). Patterning and nuclear beta-catenin expression in the colonic adenomacarcinoma sequence. Analogies with embryonic gastrulatio. Am. J. Pathol. 157, 1113–1121. Kolsch, V., Seher, T., Fernandez-Ballester, G. J., Serrano, L., and Leptin, M. (2007). Control of Drosophila gastrulation by apical localization of adherens junctions and RhoGEF2. Science 315, 384–386. Krieg, M., Arboleda-Estudillo, Y., Puech, P. H., Kafer, J., Graner, F., Muller, D. J., and Heisenberg, C. P. (2008). Tensile forces govern germ-layer organization in zebrafish. Nat. Cell Biol. 10, 429–436. Lecuit, T., and Lenne, P. F. (2007). Cell surface mechanics and the control of cell shape, tissue patterns and morphogenesis. Nat. Rev. Mol. Cell Biol. 8, 633–644. Leduc, S. (1912). “La Biologie Synythétique. Etude de Biophysique.” Poinat Paris. Leeuwenhoek (1683). Philos. Trans. 14, 568–574. Leptin, M. (1991). Twist and snail as positive and negative regulators during Drosophila mesoderm devel opment. Genes. Dev. 5, 1568–1576. Lewis, E. B. (1978). A gene complex controlling segmentation in Drosophila. Nature 276, 565–570. Mammoto, A., Connor, K. M., Mammoto, T., Yung, C. W., Huh, D., Aderman, C. M., Mostoslavsky, G., Smith, L. E., and Ingber, D. E. (2009). A mechanosensitive transcriptional mechanism that controls angiogenesis. 2009. Nature 457, 1103–1108. Martin, A.C. (2009). Dev. Biol. Martin, A. C., Kaschube, M., and Wieschaus, E. F. (2009). Pulsed contractions of an actin-myosin network drive apical constriction. Nature 457, 495–9. Mayer, C. R., Cabuil, V. V., Lalot, T., and Thouvenot, R. (1999). Incorporation of magnetic nanoparticles in new hybrid networks based on heteropolyanions and polyacrylamide. Angew. Chem. Int. Ed Engl. 38, 3672–3675.

12. Mechanical Induction in Embryonic Development and Tumor Growth

321

Moore, S. W. (1994). A fiber optic system for measuring dynamic mechanical properties of embryonic tissues. IEEE Trans. Biomed. Eng. 41, 45–50. Morin, P. J., Sparks, A. B., Korinek, V., Barker, N., Clevers, H., Vogelstein, B., and Kinzler, K. W. (1997). Activation of -catenin-Tcf signaling in colon cancer by mutations in b-catenin or APC. Science 275, 1787–1790. Morize, P., Christiansen, A. E., Costa, M., Parks, S., and Wieschaus, E. (1998). Hyperactivation of the folded gastrulation pathway induces specific cell shape changes. Development 125, 589–597. Nusslein-Volhard, C., and Wieschaus, E. (1980). Mutations affecting segment number and polarity in Drosophila. Nature 287, 795–801. Pouille, P. A., and Farge, E. (2008). Hydrodynamic simulation of multicellular embryo invagination. Phys. Biol. 5, 15005. Pouille, P. A., Ahmadi, P., Brunet, A. C., and Farge, E. (2009). Mechanical signals trigger Myosin II redistribution and mesoderm invagination in Drosophila embryos. Sci. Signal 2, ra16. Rauch, C., Brunet, A. C., Deleule, J., and Farge, E. (2002). C2C12 myoblast/osteoblast transdifferentiation steps enhanced by epigenetic inhibition of BMP2 endocytosis. Am. J. Physiol. Cell Physiol. 283, C235–C243. Rauzi, M., Verant, P., Lecuit, T., and Lenne, P. F. (2008). Nature and anisotropy of cortical forces orienting Drosophila tissue morphogenesis. Nat. Cell Biol. 10, 1401–1410. Roszko, I., Sawada, A., and Solnica-Krezel, L. (2009). Regulation of convergence and extension movements during vertebrate gastrulation by the Wnt/PCP pathway. Semin. Cell Dev. Biol. 20, 986–997. Sawada, Y., Tamada, M., Dubin-Thaler, B. J., Cherniavskaya, O., Sakai, R., Tanaka, S., and Sheetz, M. P. (2006).

Force sensing by mechanical extension of the Src family kinase substrate p130Cas. Cell 127, 1015–1026.

Schotz, E. M., Burdine, R. D., Julicher, F., Steinberg, M. S., Heisenberg, C. P., and Foty, R. A. (2008).

Quantitative differences in tissue surface tension influence zebrafish germ layer positioning. HFSP J. 2, 42–56. Seher, T. C., Narasimha, M., Vogelsang, E., and Leptin, M. (2007). Analysis and reconstitution of the genetic cascade controlling early mesoderm morphogenesis in the Drosophila embryo. Mech. Dev. 124, 167–179. Solon, J., Kaya-Copur, A., Colombelli, J., and Brunner, D. (2009). Pulsed forces timed by a ratchet-like mechanism drive directed tissue movement during dorsal closure. Cell 137, 1331–1342. Somogyi, K., and Rorth, P. (2004). Evidence for tension-based regulation of drosophila mal and srf during invasive cell migration. Dev. Cell 7, 85–93. St Johnston, D., and Nusslein-Volhard, C. (1992). The origin of pattern and polarity in the Drosophila embryo. Cell 68, 201–219. Supatto, W., Debarre, D., Moulia, B., Brouzes, E., Martin, J. L., Farge, E., and Beaurepaire, E. (2005). In vivo modulation of morphogenetic movements in Drosophila embryos with femtosecond laser pulses Proc. Natl. Acad. Sci. USA 102, 1047–1052. Sweeton, D., Parks, S., Costa, M., and Wieschaus, E. (1991). Gastrulation in Drosophila: The formation of the ventral furrow and posterior midgut invaginations. Development 112, 775–789. Thomson, D.A.W. (1917). “On Growth and Form.” Cambridge University Press, Cambridge. Tokoyama, Y., Peralta, X. G., Wells, A. R., Kiehart, D. P., and Edwards, G. S. (2008). Apoptotic force and tissue dynamics during Drosophila embryogenesis. Science 321, 1683–6. von His, W. (1875). “Unsere Korperform und Das Physiologische problem ibrer entstebung.” F. C. W. Vogel, Leipzig. Whitehead, J., Vignjevic, D., Futterer, C., Beaurepaire, E., Robine, S., and Farge, E. (2008). Mechanical factors activate beta-catenin-dependent oncogene expression in APC mouse colon. HFSP J. 2, 286–294. Wieschaus, E., and Nusslein-Volhard, C. (1998). in “ Drosophila a Practical Approach” (D. B. Roberts, ed.), pp. 179–214. Oxford University Press, Oxford. Wolpert, L. (1992). Gastrulation and the evolution of development. Dev. Suppl. 7–13. Young, P. E., Pesacreta, T. C., and Kiehart, D. P. (1991). Dynamic changes in the distribution of cytoplasmic myosin during Drosophila embryogenesis. Development 111, 1–14. Zallen, J. A., and Wieschaus, E. (2004). Patterned gene expression directs bipolar planar polarity in Drosophila. Dev. Cell 6, 343–355. Zhang, X., Fournier, M. V., Ware, J. L., Bissell, M. J., Yacoub, A., and Zehner, Z. E. (2009). Inhibition of vimentin or beta1 integrin reverts morphology of prostate tumor cells grown in laminin-rich extracellular matrix gels and reduces tumor growth in vivo. Mol. Cancer Ther. 8, 499–508.

CHAPTER 13

Informatics-Based Analysis of Mechanosignaling in the Laminopathies Frank P. L. Lai, Radfidah A. Mutalif, Siew Cheng Phua, and Colin L. Stewart Institute of Medical Biology, Immunos, 8A Biomedical Grove, Biopolis, Singapore 138648

Abstract I. Introduction II. Methods A. Derivation of Primary Mouse Embryonic Fibroblasts III. Reagents and Equipment for Subjecting Cells to Uniaxial Strain A. Reagents B. Equipment IV. Discussion V. Summary

References

Abstract The A- and B-type lamins are the primary building blocks of the lamina—a proteinaceous meshwork underlying the nuclear envelope (NE). In the last decade, some 25 diseases have been linked to mutations in genes encoding proteins of the NE and lamina, with about half being caused by mutations in the Lamin A gene. Cells, either from patients or from mice carrying lamin mutations, frequently exhibit deformed nuclei accompanied by compromised mechanical properties in both the nucleus and the cytoplasm, implying that defects in the mechanical integrity of the nuclei and in mechanosignaling contribute to the pathology of these diseases. We describe a procedure to study total gene expression of mutant cells subjected to uniaxial mechanical strain by culturing them on a deformable surface. Using our procedure, enough high-quality RNA can be collected from these samples for microarray and informatics analysis. Such analysis may provide valuable information METHODS IN CELL BIOLOGY, VOL. 98 Copyright � 2010 Elsevier Inc. All rights reserved.

323

978-0-12-381009-0 DOI: 10.1016/S0091-679X(10)98013-8

324

Frank P. L. Lai et al.

regarding the changes in gene expression and signaling pathways that may underlie the pathologies of the various diseases, which in turn may arise as a consequence of defective responses to mechanical strain

I. Introduction Till recently, the nucleus was largely regarded as a porous bag enclosing the cell’s genetic material. However, we now know that the nucleus is an integral structural component of the cell, with the nuclear envelope (NE) and lamina physically linking the nucleus to the cytoskeleton (Stewart et al., 2007). The nuclear lamina is a network of intermediate filaments forming a scaffold that underlies the inner nuclear membrane (INM) of the NE. The lamina primarily consists of A- and B-type nuclear lamins, with lamins A and C being encoded by a single LMNA gene and lamins B1 and B2 being encoded by LMNB1 and LMNB2 genes, respectively. Alternate splicing of the LMNB2 gene generates a minor variant of lamin B2, lamin B3. The A-type lamins are widely expressed in almost all tissues, although they are absent from early embryos, embryo nic stem cells, and many hematopoetic cells, whereas lamins B1 and B2 are ubiqui tously expressed throughout development (Burke and Stewart, 2006). The nuclear lamins are intermediate filament proteins with a conserved �-helical central rod domain. Unlike cytoplasmic intermediate filament proteins, the nuclear lamins contain an additional 42 amino acids in their rod domain and nuclear localiza tion signals in their C-terminal tail domains. Little is known about the mechanism(s) of lamin filament assembly but studies using live cell imaging and photobleaching established that the nuclear lamins form stable filamentous structures underlying the INM at the nuclear periphery throughout interphase (Broers et al., 1999). The nuclear lamina is completely disassembled during mitosis, following the phosphorylation of specific amino acids in the lamins, and then reassembled after mitosis (Ellenberg et al., 1997; Gerace and Blobel, 1980). During the last decade, interest in the lamins has increased, due to the dozen diseases that are caused by mutations in the LMNA gene. More than 250 mutations have been mapped to the gene resulting in a collection of diseases called the laminopathies (Worman and Bonne, 2007). These diseases range from muscular dystrophies, a lipodystrophy, cardiomyopathy and conditions affecting skeletal homeostasis, to the premature aging disease Hutchinson–Gilford progeria. Although it is not fully under stood how mutations in a single gene cause such a wide range of diseases, the diversity and tissue specificity of the diseases indicate that A-type lamins may be involved in a wide array of cellular and physiological functions. The molecular pathology of the laminopathies remains elusive. Little is known about the molecular function(s) of nuclear lamins other than their role in providing shape and support to the interphase nucleus and as protein scaffolds, important for the selective localization of proteins to the NE and nuclear periphery (Soullam and Worman, 1995; Sullivan et al., 1999). Additional roles have also been suggested for the lamina, in regulating chromatin organization, gene transcription, and mitotic organization (Shimi et al., 2008).

13. Informatics-Based Analysis of Mechanosignaling

325

In addition to these gene regulatory functions, the nuclear lamina and the LINC complex (Link between the Nucleoskeleton and Cytoskeleton) physically tether the nucleus to all components of the cytoskeleton. The LINC complex is composed of giant Klarsicht-Anc1-syne1 homology (KASH) domain proteins, the synaptic nuclear envelope proteins (SYNEs) or Nesprins, which are anchored in the outer nuclear membrane (ONM) and interact with the C-termini of the Sad1-UNC-84 homology (SUN) domain proteins in the perinuclear space between the ONM and the INM. The SUN proteins are anchored in the INM, and their N-termini protrude into the nucleus and interact with the interphase perinuclear chromatin. Such a complex results in the physical connection of the cytoskeleton with chromatin (Crisp et al., 2006; Starr and Han, 2002). This has led to speculation that nuclear lamins may both regulate cytos keletal organization/function and, through their linkage with the SUN proteins, play an important role in mechanotransduction and signaling, perhaps by interacting with chromatin and/or transcription factors (Wang et al., 2009). It has been suggested that the pathologies observed in some of laminopathies, such as the muscular dystrophies and cardiac conduction diseases, could be due, at least in part, to compromised mechanical properties and mechanosignaling in the affected cells (Lammerding et al., 2004; Philip and Dahl, 2008). The A-type lamins are required for maintaining the mechanical integrity of the nucleus, with loss of the A-type lamins resulting in muscular dystrophies and cardiac conduction defects (Sullivan et al., 1999). In vitro, the nuclei are weaker and the cells are more prone to strain-induced apoptosis and necrosis (Lammerding et al., 2004). Intriguingly, loss of the A-type lamins not only affects the mechanical properties of the nucleus, but also results in a reduction in cytoplasmic viscosity, revealing a functional link between the lamina and the cytoskeletal organization (Lammerding et al., 2004; Lee et al., 2007). We are interested in determining what effects disrupting the LINC complex has on gene expression and signaling pathways. Over the years, a variety of organisms have been utilized to facilitate investigation into the molecular functions of the lamina and basis of the laminopathies. Till now, the mouse has been the most widely used due to its amenability to genetic manipulation and as a model for human diseases. A number of mouse models of the laminopathies, including progeria, Emery–Driefuss muscular dystrophy, and lipodystrophy, have been established (Cohen and Stewart, 2008) and some are being used for testing potential therapeutic procedures (Capell et al., 2005; Muchir et al., 2009). To understand how different mutations in the human LMNA gene result in the diverse laminopathies, we generated a number of mouse lines with mutations in the Lmna gene and components of the LINC complex. We are particularly interested in how different mutations in the A-type lamins and loss of specific LINC complex members affect signaling pathways and gene expression patterns when the mutant cells are subjected to mechanical strain. Here, we describe a method to subject primary fibroblast cells (embryonic and adult, as well as myoblasts) isolated from mice to uniaxial mechanical strain and harvest RNA from microarray analysis (Melcon et al., 2006). This facilitates the analysis of changes in gene expression as a consequence of the different mutations, with the data being utilized to identify the signaling pathways and the expression of specific genes in the mutant cells. We also suggest looking at the

326

Frank P. L. Lai et al.

effects of mechanical strain on fibroblasts that form either embryonic or adult stages as it is becoming increasingly clear that fibroblasts prepared from different tissues and stages of development differ in their patterns of gene expression (Chang et al., 2002).

II. Methods A. Derivation of Primary Mouse Embryonic Fibroblasts

1. Dissecting Mouse Uteri &

&

&

&

&

Essential equipment A tissue culture (TC) microscope equipped with phase contrast optics (10–40) for viewing cells. Standard TC facilities A sterile/filtered air culture hood and a 37°C, 5% CO2-gassed incubator. Fill 60 mm TC dishes each with 4 ml PBS (Ca2þ/Mg2þ-free), at room temperature (RT). Dissection tools Medium-sized forceps and dissection scissors. Required reagents

To disaggregate embryos for the derivation of primary mouse embryonic fibroblasts (PMEFs), MEF enzyme solution is required. MEF enzyme solution consists of 100 µg/ml DNase1 (Sigma-Aldrich, cat. D-4527) and 500 µg/ml collagenase IV (Sigma-Aldrich, C-9407) in serum-free, high glucose 4.5 g/l DMEM (GIBCO® Invitrogen, cat. 10829-018). For derivation and culture of PMEFs and primary mouse adult fibroblasts (PMAFs), D10 media is used. D10 media consists of high glucose DMEM—4.5 g/l (GIBCO® Invitrogen, cat. 10829-018), 10% fetal bovine serum (GIBCO® Invitrogen, cat. 10437 028), L-glutamine (GIBCO® Invitrogen, cat. 25030-081)—2 mM final concentration, and penicillin/streptomycin (50 IU/ml) (GIBCO® Invitrogen, cat. 15140-122).

2. Mouse Preparation Day 0: Set up mating, at 1700–1800 h. Day 1: Check for copulation plug (equivalent to day 1 of pregnancy). Remove plugged females to a separate cage. Day 13: Sacrifice the pregnant females by CO2 asphyxiation.

3. Primary Mouse Embryonic Fibroblast Derivation 1. Spray the euthanized mouse with a 70% ethanol (EtOH) solution over the abdomen to disinfect the skin. 2. Cut the abdominal skin with sharp scissors. Grasp the skin at the site of incision and tear longitudinally to expose the underlying peritoneum. 3. Open the peritoneal cavity and remove the uteri with the embryos.

327

13. Informatics-Based Analysis of Mechanosignaling

4. Dissect the embryos from the uteri in Ca2þ/Mg2þ-free PBS and remove yolk sac, amnion, and placenta. Wash embryos twice in fresh Ca2þ/Mg2þ-free PBS to remove traces of blood. 5. Using forceps, pinch off the head and retain, if necessary, for DNA extraction and subsequent genotyping. Remove and discard the liver. 6. To digest the remaining embryonic tissues, 2 ml of MEF enzyme solution is required per embryo. See above for preparation of the enzyme solution. 7. Triturate and break up the embryo in the enzyme solution using a sterile 3 cc syringe (BD FalconTM) with an 18G needle (BD FalconTM) by repeatedly syringing the embryos through the needle about 5 times in a sterile 10 ml capped cell culture centrifuge tube. 8. Leave the disaggregated embryos to incubate in a 37°C water bath for 30 min, mixing occasionally. 9. Centrifuge at 1200 rpm for 5 min and remove the MEF enzyme solution from the cell pellet. 10. Resuspend the cell pellet in 3 ml D10 media and add the cell suspension to an individual well of six-well plate (Nunc, cat. 140675) and incubate at 37°C, 5% CO2. The cells will attach almost immediately and produce PMEFs over the next 2–3 days. The culture should reach confluency by the third day (Fig. 1A). (A)

(B)

(C)

(D)

Fig. 1 (A) A subconfluent culture of PMEFs. (B) A confluent culture of PMAFs. (C) Primary myoblasts. (D) Myoblasts differentiated into myotubes.

328

Frank P. L. Lai et al.

4. Passaging and Expansion of the PMEF Cultures 1. Wash the culture dish twice in Ca2þ/Mg2þ-free PBS, and add 1 ml of 0.05% trypsin þ 1 mM EDTA in PBS (GIBCO® Invitrogen, cat. 25300-120). Trypsinize cells for about 5 min, spin down at 1200 rpm for 5 min, and transfer the resuspended single cells to a 10 cm TC dish. 2. After further culture, the PMEFs should be split at no more than a 1:3 ratio. PMEFs are only good to be used for —five to six passages (P5–P6) under standard culture conditions of 5% CO2 in air, since they cease to proliferate around P5 due to PMEFs being sensitive to ambient O2 tension that induces the stress response and proliferative arrest. Long-term cultures that proliferate continuously can, however, be maintained under hypobaric conditions using a hypobaric incubator. 3. Once sufficient numbers are established, PMEFs can be either used directly for experiments or frozen for future use. Cells are typically frozen at a density of 5 million cells (per ml) per vial and the recommended freezing media: 10% DMSO (Sigma-Aldrich, cat. D2650) þ 90% fetal bovine serum (GIBCO® Invitrogen, cat. 10437-028) at –80°C.

5. Preparation of Primary Mouse Adult Fibroblasts and Primary Mouse Adult Myoblasts &

&

& &

&

Essential equipment TC microscope equipped with phase contrast optics (10–40) for viewing cells. Standard TC facilities: A sterile/filtered air culture hood and a 37°C, 5% CO2-gassed incubator. Fill 60 mm TC dishes each with 4 ml PBS (Ca2þ/Mg2þ-free), at RT. Dissection tools: Medium-sized forceps and dissection scissors.

Required reagents:

Antibiotic solution made up of PBS (Ca2þ/Mg2þ-free) containing 2% penicillin/ streptomycin (50 IU/ml) (GIBCO® Invitrogen, cat. 15140-122) and Fungizone (GIBCO® Invitrogen, cat. 15290-018) at a final concentration of 1 µg/ml. For the washing of tissues and to reconstitute some of the reagents for this procedure, 1 HBSS (GIBCO® Invitrogen, cat. 14170-112) and HEPES (GIBCO® Invitrogen, cat. 15630-080) are required. To digest the adult tissues (muscle) for PMAF and myoblast derivation, PMAF enzyme solution is required. MAF enzyme solution consists of equal volumes of dispase II (Roche, cat. 04 942 078 001) at a concentration of 2.4 U/ml and 1% collagenase II (GIBCO® Invitrogen, cat. 17101-015). Collagenase II is prepared by dissolving 1 ml of collagenase II in 100 ml of 1 HBSS/25 mM HEPES. Similarly, dispase II is prepared as a stock solution of 2.4 U/ml in 1 HBSS/25 mM HEPES. The PMAF enzyme solution should be filter sterilized prior to use and should be kept at –20°C for long-term storage. For culturing and maintaining undifferentiated adult myoblasts, Myoblast (MB) growth media is used. MB growth media consists of Ham’s F-10 (GIBCO®

13. Informatics-Based Analysis of Mechanosignaling

329

Invitrogen, cat. 11550-043), 20% horse serum (GIBCO® Invitrogen, cat. 16050-122), 1M CaCl2 (Sigma-Aldrich, cat. 21115) at 1.26 mM final concentration, gentamycin (GIBCO® Invitrogen, cat. 15750-060) at 50 µg/ml final concentration, penicillin/streptomycin (50 IU/ml) (GIBCO® Invitrogen, cat. 15140-122), Fungizone (GIBCO® Invitrogen, cat. 15140-122)—0.8 µg/ml final concentration, and basic fibroblast growth factor (bFGF) (R&D Systems, cat. 223-FB) at a final concentration of 10 ng/ml. bFGF is dissolved in 1 ml of sterile 0.1% bovine serum albumin solution. Aliquots of bFGF should be prepared and stored in –20°C. Due to the lability of bFGF in culture media, and to maintain undifferentiated myoblasts, fresh bFGF should be added to the culture medium everyday. Gentamycin and Fungizone are only required in the media during the first few days after the derivation of the primary cells. For myoblast establishment, a 0.1% gelatin solution is required to precoat TC surfaces prior to seeding with the cells. This enhances myoblast attachment to the TC surface. To prepare a 0.1% gelatin solution, dissolve 2% (w/v) gelatin solution (Sigma-Aldrich, cat. G1393) in sterile water by autoclaving followed by dilution in PBS. To prepare TC surfaces before seeding of myoblasts, coat the required surface with minimal amounts of gelatin and leave in 37°C incubator for 15 min. Remove gelatin coating prior to seeding of myoblasts.

6. Primary Mouse Adult Fibroblast and Myoblast Derivation 1. Sacrifice the adult mice by CO2 asphyxiation. 2. Spray the euthanized mice with a 70% EtOH solution and wipe off any excess liquid. Tear the skin from the body and limbs. Cut off the four limbs, cut off the feet, and place the limbs in the antibiotic solution in a 100 mm dish. 3. Rinse off all contaminating hair and place limbs in a 60 mm dish with 1 HBSS. Remove as much fat as possible. 4. Remove all muscles from the bones and place muscle pieces into a fresh 60 mm dish with 1 HBSS on ice. Repeat with all subsequent mice, keeping the muscle at 4°C or on ice until all mice have been completed. 5. Aspirate HBSS from the dish with a pipette, being careful not to aspirate the tissue. Transfer the muscle to a 50 ml tube and weigh. Add 4 ml sterile PMAF enzyme solution to every 1 g of muscle. 6. Tissue digestion should be performed as follows: a. Incubate muscle tissue in enzyme solution in a 37°C water bath for 30 min, pipetting 10 times with a 5 ml pipette at 15 min intervals. b. Add equal volume of D10 and filter through 70 µM sterile filter (BD FalconTM, cat 352350). c. Filter again through a 40 µM sterile filter (BD FalconTM, 352340). d. Centrifuge suspension at 1200 rpm for 5 min to pellet. e. Remove supernatant and resuspend cell pellet in D10 media. f. Repeat steps d and e.

330

Frank P. L. Lai et al.

7. Preplating to enrich for myoblasts as follows: a. Add 5 ml D10 media to resuspend each cell pellet and transfer resuspended cells to one 100 mm TC dish. b. Incubate cells for 60–120 min at 37°C, in a 5% CO2-gassed incubator. This allows the fibroblasts to attach, but not the myoblasts. After incubation, the following steps should be performed: i. To obtain myoblasts, carefully collect the supernatant after the 60–120 min incubation with all the unattached cells, centrifuge at 1200 rpm for 5 min, resuspend pellet in 10 ml MB growth media, and seed onto a gelatin-coated 100 mm TC dish with MB growth media. Myoblast cultures, in which the cells remain rounded and more refractile (Fig. 1C), compared to fibroblasts, should be expanded to 70–80% confluency. Overconfluency will increase the rate of myoblast differentiation. Media for myoblast culture should be changed every day to replenish the bFGF. ii. To maintain the PMAFs, fresh D10 media is added to the initial 100 mm TC dish containing the attached fibroblasts. Incubate the plate until the cells are confluent and passage them accordingly (Fig. 1B). c. If the myoblast cultures have many PMAFs still growing in the preparations, repeat preplating process. 8. To passage both PMAFs and myoblasts, wash the culture dish twice in Ca2þ/Mg2þ-free PBS, and add sufficient volumes of 0.05% trypsin þ 1 mM EDTA in PBS (GIBCO® Invitrogen, cat. 25300-120) (e.g., 5 ml for a 100 mm TC dish). Trypsinize cells for about 5 min in 37°C. When cells are detached, add an equal volume of D10 media to neutralize the trypsin, collect the suspension, and spin down at 1200 rpm for 5 min. Subsequently, remove the supernatant and resuspend the pellet in appropriate volumes of D10 media and MB growth media for fibroblasts and myoblasts, respectively. Transfer the resuspended single cells to an appropriate TC dish. For myoblasts, the TC surface should be coated with 0.1% gelatin prior to seeding of myoblasts. 9. Primary cell lines, both the MAFs and the myoblasts, should be split at no more than a 1:3 ratio. Primary cell lines are only good to be used for —five to six passages (P5–P6) under standard culture conditions of 5% CO2 in air, since they cease to proliferate around P5 due to primary cells being sensitive to ambient O2 tension that induces the stress response and proliferative arrest (Parrinello et al., 2003). Long-term cultures that proliferate continuously can, however, be maintained under hypobaric conditions using a hypobaric incubator. 10. Once sufficient numbers are established, PMAFs and myoblasts can be either used directly for experiments or frozen for future use. Cells are typically frozen at a density of 5 106 cells (per ml) per vial and the recommended freezing media: 10% DMSO (Sigma-Aldrich, cat. D2650) þ 90% fetal bovine serum (GIBCO® Invitrogen, cat. 10437-028) at –80°C. 11. To differentiate the myoblasts, 1 105 cells are plated onto 60 mm gelatin-coated TC dishes in MB media. After 24 h, the media is changed to differentiation media

13. Informatics-Based Analysis of Mechanosignaling

331

consisting of D10 þ ß2% horse serum (without bFGF). In the ensuing 4 days, the myoblasts fuse to form multinucleated myotubes (Fig. 1D).

III. Reagents and Equipment for Subjecting Cells to Uniaxial Strain A. Reagents 1. 2. 3. 4. 5. 6. 7.

Qiagen RNA isolation kit (Qiagen Inc., Germany). TRIzol reagent (Invitrogen Inc., USA). Human plasma fibronectin, 1 mg/ml (Millipore Inc., USA). Insulin–transferrin–selenium medium (ITS) 100 (Gibco Inc., Germany). Bioflex flexible bottom culture plate 6 wells (Flexcell Inc., USA). Ambion Target Amp kit (Ambion Inc., USA). Sentrix Mouse-6 Expression BeadChip v2 (Illumina Inc., USA).

B. Equipment To subject cells to uniaxial strain, we use the Tension Plus FX-4000T system (Flexcell Inc.). The system consists of a TRIVAC© B D8B rotary vane vacuum pump (Leybold Inc., Germany), a Flexcell pressure reservoir (Flexcell Inc.), FX-4000T controller (Flexcell Inc.), and 25 mm diameter loading stations mounted on baseplates (Flexcell Inc.). The stretching is controlled by FlexsoftTM V5.0 software (Flexcell Inc.) using a preprogrammed regimen of 10% uniaxial sinusoidal stretch at 1 Hz (to model the cardiac cycle). 1. Coat each well of the Bioflex flexible-bottom six-well TC plate (Flexcell Inc., USA) with 750 µl (50 µg/ml) fibronectin for 2 h at RT. Other attachment factors (laminin, collagens, thrombospondin) may also be used. However, with fibroblasts we found fibronectin to be the most effective, and for myoblasts either fibronectin or laminin can be used. 2. Trypsinize cells and seed 7 104 cells per well in 2 ml of normal culture medium. 3. Take care to disperse the cells well and gently shake the plate and inspect under an inverted microscope to avoid cells aggregating at the center of the well. 4. Culture the cells overnight at the normal growth conditions. 5. The next day, place the cells (between 50% and 70% confluence, (Fig. 2A) under proliferative arrest (G0) by replacing the culture medium with growth arrest medium (DMEM high glucose with penicillin and streptomycin, L-glutamine, and 1 ITS medium or 0.5% fetal calf serum (FCS)) and culture for 3 days (Fig. 2B). By placing the cells under proliferative arrest, this removes from the microarray analysis of gene expression “noise” that would otherwise be generated by the cells undergoing cell cycle progression under normal culture conditions.

332

Frank P. L. Lai et al.

(A)

(B)

Fig. 2 (A) Culture of PMEFs seeded at the appropriate density on Bioflex plates. (B) The same culture after 3 days under proliferative arrest.

A. Mechanical Stretching of Growth Arrested Fibroblast Cells 6. After 3 days under proliferative arrest (Fig. 2B), the Bioflex plates are mounted onto the loading station (25 mm) and vacuum is applied by adjusting parameters in FX-4000T controller software, such that the silicone membrane is subjected to stretching at a frequency of 1 Hz and 10% strain. 7. After subjecting the cells to periods of repetitive stretching (we generally subject the cells to periods of 1, 2, and 4 h), the plates are removed from the loading station, medium is removed, and 0.5 ml of TRIzol is immediately added to each well. 8. The cells are scraped off the silicone surface with a sterile plastic scraper, and the cell extract from each well is transferred into individual sterile 1.5 ml eppendorf tubes and left at RT for 5 min. 9. About 100 µl of chloroform is added to each tube and the tube vortexed vigorously for 1 min. Spin at 13,000g for 15 min at 4 °C. 10. Remove supernatant and pipette into a new sterile eppendorf tube with equal volume of 70% EtOH. Mix and extract RNA using Qiagen RNA extract column according to the manufacturer’s protocol. 11. Elute RNA in with 20–30 µl of nuclease-free ddH2O to obtain a yield of 30 ng/µl of total RNA. 12. The concentration and integrity of the RNA is analyzed with an ND-1000 Spectrophotometer (NanoDrop Technologies, USA) and Agilent 2100 Bioanalyzer (Agilent Inc., USA). The concentration of RNA should be at least 25–30 ng/µl with absorbance ratio 260:280 1.9 and an RNA integrity number 8. 13. Before proceeding to microarray analysis, the induction of mechanical response genes in the uniaxially stretched cells is confirmed by real-time polymerase chain reaction (RT-PCR). We routinely measure two prominent mechanosensitive genes, immediate early gene X1 (Iex1) and early growth response gene 1 (Egr1) to ensure that the cells have responded to repetitive mechanical strain (Fig. 3). Primer sequences: glyceraldehyde 3-phosphate dehydrogenase (GAPDH), fw-50 -cgagaatgggaagcttgtcatc-30 , rv-50 -cggcctcaccccatttg-30 ;

333

13. Informatics-Based Analysis of Mechanosignaling

IEX1

7

(B)

20

6 5 4 3 2 1

EGR1

18 Normalized fold change

Normalized fold change

(A) 8

16 14 12 10 8 6 4 2 0

0 Untreated

1h

4 h

Untreated

1h

4h

Fig. 3 Typical real-time qPCR results of Iex-1 and Egr-1 induction in PMEFs before and after 1 and 4 h of uniaxial strain at 1 Hz and 10% strain.

IEX1, fw-50 -gccatcatcttctgccagattt-30 , rv-50 -tcacggcgctggtagcat-30 ; EGR1, fw-50 -ccatgaacgcccatatgctt-30 , rv-50 -gctcatccgagcgagaaaag-30 . 14. Amplification of RNA (~300–500 ng) to cRNA is performed with an Ambion Target Amp kit (Ambion Inc., USA) according to the manufacturer’s protocol. 15. About 1.5 µg of cRNA from each sample is then hybridized, in our situation to Sentrix Mouse-6 Expression BeadChip v2 chips (Illumina Inc., USA). Other microarray platforms are available, such as those from Affymetrix. The microarray data can then be analyzed by a variety of software programs, such as Partek analysis (Partek Inc., USA), Ingenuity Pathway Analysis (Ingenuity Inc., USA), DAVID (http://david.abcc.ncifcrf.gov), and Gene Set Enrichment Analysis (GSEA) (Mootha et al., 2003; Subramanian et al., 2005).

IV. Discussion Our objectives are focused on determining how different mutations in the LMNA gene result in a dozen different diseases, ranging from muscular dystrophy, cardiomyo pathy, lipodystrophy, and premature aging, arise from a mutated protein that is almost ubiquitously expressed in adult tissues. In pursuing these objectives, we have derived about a dozen different mouse lines, each containing different mutations in the mouse Lmna gene and genes encoding other NE proteins that when mutated result in disease. Fibroblasts isolated from these mouse lines act as a resource for our attempts to understand the cellular and molecular patho physiology of these diseases. Traditionally, nearly all analyses on fibroblasts have been performed on fibroblasts from embryos (PMEFs), as these are easy to obtain and maintain. There are, however, factors complicating this embryonic fibroblast-based analysis. One of them is the stage in the mouse’s life cycle from which the fibroblasts

334

Frank P. L. Lai et al.

are isolated (Chang et al., 2002). We are concerned that studies restricted to embryonic stages maybe misleading, particularly with regard to laminopathies, which tend to manifest post-nataly (C.L. Stewart, unpublished observations). Also, a majority of the analyses performed to date are on cells in a physically unstressed environment. A more physiologically relevant approach would be to mimic the conditions nuclei are subjected to in the muscle and cardiac tissues (both mechanically stressed tissues) and the tissues affected in patients carrying some lamin mutations by applying mechanical strain. The method we described here allows us to apply mechanical strain to fibroblasts isolated from different sources, for example, pre- or postnatal mouse, and it can easily be optimized for other tissues such as muscle myotubes generated from differentiated myoblasts. Global changes in gene expression levels detected by microarray analysis are then used to identify changes in signaling pathways between the wild-type and mutant cells. However, one of the limitations in the analysis of microarray data is its sensitivity. Global gene expression changes induced by mechanical stretching are rela tively small, compared to other treatments, such as ionizing or UV radiation (Boerma et al., 2005). It is therefore advisable that a larger sample size be used to detect more subtle changes in gene expression levels. We routinely use up to three independent cell lines in triplicate for each experimental condition. Also, microarray analysis is both expensive and time consuming; therefore, it is necessary to confirm the validity of the applied mechanical strain by determining the induction of mechanical response genes by real-time quantitative PCR (RT-PCR) analysis for each sample before proceeding to hybridization to the microarray chips. Another important point to consider is to look at gene expression changes as a group or cluster of genes (e.g., using the GSEA open source analytical tools) that have been assigned certain known biological functions rather than individual genes. This could reduce potential false positive discoveries in the microarray data. Most importantly, changes in gene expression levels detected by microarray analysis must be confirmed by real-time qPCR, followed by functional studies based on the affected proteins and/or signaling pathways.

V. Summary Future research is being directed to identifying the signaling pathways that are differentially affected in different lamin and LINC component mutations utilizing a wide variety of tissue samples subjected to physiologically relevant mechanical stres ses. This may help provide a basis for understanding the molecular pathophysiology of different lamin-associated diseases, as well as the role of the lamina and LINC complexes in mechanosignaling. References Boerma, M., et al. (2005). Microarray analysis of gene expression profiles of cardiac myocytes and fibroblasts after mechanical stress, ionising or ultraviolet radiation. BMC Genomics 6 6. Broers, J. L., et al. (1999). Dynamics of the nuclear lamina as monitored by GFP-tagged A-type lamins. J. Cell Sci. 112(Pt 20), 3463–3475.

13. Informatics-Based Analysis of Mechanosignaling

335

Burke, B., and Stewart, C. L. (2006). The laminopathies: The functional architecture of the nucleus and its contribution to disease (*). Annu. Rev. Genomics Hum. Genet. 7, 369–405. Capell, B. C., et al. (2005). Inhibiting farnesylation of progerin prevents the characteristic nuclear blebbing of Hutchinson-Gilford progeria syndrome. Proc. Natl. Acad. Sci. USA 102, 12879–12884. Chang, H. Y., et al. (2002). Diversity, topographic differentiation, and positional memory in human fibroblasts. Proc. Natl. Acad. Sci. USA 99, 12877–12882. Cohen, T. V., and Stewart, C. L. (2008). Fraying at the edge mouse models of diseases resulting from defects at the nuclear periphery. Curr. Top Dev. Biol. 84, 351–384. Crisp, M., et al. (2006). Coupling of the nucleus and cytoplasm: Role of the LINC complex. J. Cell Biol. 172, 41–53. Ellenberg, J., et al. (1997). Nuclear membrane dynamics and reassembly in living cells: Targeting of an inner nuclear membrane protein in interphase and mitosis. J. Cell Biol. 138, 1193–1206. Gerace, L., and Blobel, G. (1980). The nuclear envelope lamina is reversibly depolymerized during mitosis. Cell 19, 277–287. Lammerding, J., et al. (2004). Lamin A/C deficiency causes defective nuclear mechanics and mechano transduction. J. Clin. Invest. 113, 370–378. Lee, J. S., et al. (2007). Nuclear lamin A/C deficiency induces defects in cell mechanics, polarization, and migration. Biophys. J. 93, 2542–2552. Melcon, G., et al. (2006). Loss of emerin at the nuclear envelope disrupts the Rb1/E2F and MyoD pathways during muscle regeneration. Hum. Mol. Genet. 15, 637–651. Mootha, V. K., et al. (2003). Identification of a gene causing human cytochrome c oxidase deficiency by integrative genomics. Proc. Natl. Acad. Sci. USA 100, 605–610. Muchir, A., et al. (2009). Inhibition of extracellular signal-regulated kinase signaling to prevent cardiomyo pathy caused by mutation in the gene encoding A-type lamins. Hum. Mol. Genet. 18, 241–-247. Philip, J. T., and Dahl, K. N. (2008). Nuclear mechanotransduction: Response of the lamina to extracellular stress with implications in aging. J. Biomech. 41, 3164–3170. Parrinello, S., et al. (2003). Oxygen sensitivity severely limits the replicative lifespan of murine fibroblasts. Nat. Cell. Biol. 5, 741–747. Shimi, T., et al. (2008). The A- and B-type nuclear lamin networks: Microdomains involved in chromatin organization and transcription. Genes Dev. 22, 3409–3421. Soullam, B., and Worman, H. J. (1995). Signals and structural features involved in integral membrane protein targeting to the inner nuclear membrane. J. Cell Biol. 130, 15–27. Starr, D. A., and Han, M. (2002). Role of ANC-1 in tethering nuclei to the actin cytoskeleton. Science 298, 406–409. Stewart, C. L., et al. (2007). Blurring the boundary: The nuclear envelope extends its reach. Science 318, 1408–1412. Subramanian, A., et al. (2005). Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl. Acad. Sci. USA 102, 15545–15550. Sullivan, T., et al. (1999). Loss of A-type lamin expression compromises nuclear envelope integrity leading to muscular dystrophy. J. Cell Biol. 147, 913–920. Wang, N., et al. (2009). Mechanotransduction at a distance: Mechanically coupling the extracellular matrix with the nucleus. Nat. Rev. Mol. Cell Biol. 10, 75–82. Worman, H. J., and Bonne, G. (2007). Laminopathies: A wide spectrum of human diseases. Exp. Cell Res. 313, 2121–2133.

CHAPTER 14

Autosomal Dominant Leukodystrophy Caused by Lamin B1 Duplications: A Clinical and Molecular Case Study of Altered Nuclear Function and Disease Quasar Saleem Padiath*,† and Ying-Hui Fu* *

Department of Neurology, University of California, San Francisco, San Francisco, California 94158 2324

† Department of Human Genetics, University of Pittsburgh, Graduate School of Public Health, Pittsburgh, PA - 15261

Abstract I. Introduction II. Leukodystrophies III. Adult-Onset Autosomal Dominant Leukodystrophy A. Clinical Features B. Radiological Features C. Histopathological Features IV. Molecular Genetics of ADLD V. Molecular Mechanisms of ADLD Duplication Events VI. Molecular Mechanisms Underlying ADLD Disease Pathology A. Cell Types Affected in ADLD B. Alterations in Nuclear Structure and Organization as a Disease Mechanism in ADLD C. Alterations in the Regulation of Gene Expression as a Disease Mechanism in ADLD VII. Animal Models of Lamin B1 Mutations VIII. Aging and ADLD IX. ADLD and MS X. Summary

Acknowledgments

References

METHODS IN CELL BIOLOGY, VOL. 98 Copyright 2010 Elsevier Inc. All rights reserved.

337

978-0-12-381009-0

DOI: 10.1016/S0091-679X(10)98014-X

338

Quasar Saleem Padiath and Ying-Hui Fu

Abstract Autosomal dominant leukodystrophy (ADLD) is an adult-onset demyelinating disor der that has recently shown to be caused by duplications of the nuclear lamina gene, lamin B1. This chapter attempts to collate and summarize the current knowledge about the disease and the clinical, pathological, and radiological presentations of the different ADLD families described till date. It also provides an overview of the molecular genetics underlying the disease and the mechanisms that may cause the duplication mutation event. ADLD is the first disease that has ever been linked to lamin B1 mutations and it expands the pathological role of the nuclear lamia to include disorders of the brain. The chapter also speculates on the different mechanisms that may link an important and ubiquitous structure like the nuclear lamina with the complex and cell-specific functions of myelin formation and maintenance. Understanding these mechanisms may not only prove helpful in understanding ADLD pathology but can also help in identifying new pathways that may be involved in myelin biology that can have implications for common demyelinating diseases like multiple sclerosis.

I. Introduction The nuclear lamina is a fibrous meshwork of intermediate filaments (IFs) that underlie the inner nuclear membrane (Aebi et al., 1986) (Fig. 1A). Originally thought of as a passive structural component providing architectural integrity to the nucleus, a wealth of recent data has suggested multiple and dynamic functional roles for this structure (Goldman et al., 2002; Parnaik, 2008; Stuurman et al., 1998; Worman et al., 2009). The nuclear lamina in mammalian cells is made up of two major types of lamins: the A and B types. The A-type lamins include lamins A and C which are alternate splice forms of the same gene (Gerace and Burke, 1988; Lin and Worman, 1993; Peter et al., 1989). These lamins are expressed in cells which have undergone terminal differentia tion and are not seen in some undifferentiated cells (Lin and Worman, 1997; Stuurman et al., 1998). The B-type lamins are made up of B1 and B2 proteins that are encoded for by separate genes, lamin B1 and lamin B2 (Hoger et al., 1990; Lin and Worman, 1995). A less-abundant lamin protein, B3, is thought to be restricted to spermatogenic cells and is a splice variant of lamin B2 (Furukawa and Hotta, 1993). One or other form of the B-type lamins are thought to be present in every mammalian cell (Goldman et al., 2002). Like other IF proteins, lamins have conserved a-helical central rod domains and variable head and tail domains (Fig. 1B). Two a-helices can wrap around each other in a parallel, unstaggered fashion to form a lamin–lamin dimer which is the basic filament building block (Fig. 1B) (Stuurman et al., 1998). Higher order polymers are generated from these units (Fig. 1C), but the precise mechanisms underlying polymer formation are not well understood. Lamins can interact with themselves or other lamins although

339

14. ADLD Caused by Lamin B1 Duplications (A)

(B)

N

1A

1B

2A

2B

C

(C)

Fig. 1 Nuclear lamina structure. (A) Transmission electron micrograph of nuclear envelope spread from Xenopus oocytes showing the nuclear lamina meshwork. Scale bar = 1 µm. Reprinted with permission from Aebi et al. (1986). Copyright Nature Publishing group. (B) Schematic representation of a nuclear lamin dimer. The nuclear lamin dimer has been aligned with its primary sequence. The highly a-helical central ‘‘rod’’ domain is flanked by nonhelical ‘‘head’’ and ‘‘tail’’ domains. The helical segments (1A, 1B, 2A, and 2B), which have the ability to form two-stranded �-helical coiled coils, wrapped around each other to create dimers that are arranged in a parallel/unstaggered fashion. (C) The nuclear lamin dimers associate in a head-to-tail manner to form higher order oligomers. (B) and (C) Reprinted with permission from Stuurman et al. (1998). Copyright Elsevier.

there is evidence that lamins may preferentially polymerize in distinct homopolymers (Delbarre et al., 2006; Schirmer and Gerace, 2004; Worman et al., 2009). Lamins also interact with other proteins of the inner nuclear membrane, transcription factors, DNA, and chromatin (Stuurman et al., 1998). The A- and B-type lamins are also thought to form separate, but interacting, stable meshworks in the lamina (Shimi et al., 2008). Additional evidence for the importance of the nuclear lamina in normal cellular and organismal functioning comes from the identification of a large number of diseases caused by mutations in genes encoding components of the nuclear lamina. The majority of these diseases are caused by mutations in lamin A/C. At least 12 distinct disease phenotypes have been shown to be caused by mutations in lamin A/C and these are collectively termed as the “laminopathies.” Lamin A/C associated diseases pathologies encompass a wide range of phenotypes and include the involvement of the musculos keletal system, fat metabolism, peripheral nervous system (PNS), and premature aging. Diseases caused by mutations in genes encoding proteins associated with the nuclear lamina have also been described. As diseases caused by lamin A/C mutations and nuclear lamina-associated genes have been identified for some years, various aspects of these diseases have been described in great detail and also summarized in a number of reviews (Burke and Stewart, 2006; Parnaik, 2008; Worman et al., 2009).

340

Quasar Saleem Padiath and Ying-Hui Fu

It is only in the last few years that mutations involving lamin B1 and B2 have been shown to be associated with disease pathologies. Recently, a duplication involving the gene lamin B1 was shown to be the cause of an adult-onset demyelinating diseases, adult-onset autosomal dominant leukodystrophy (ADLD) (Padiath et al., 2006). Mutations in the lamin B2 gene were shown to be associated with acquired partial lipodystrophy (APL) also called “Barraquer–Simons syndrome” (Hegele et al., 2006). However, some of these mutations were also found at a lower frequency in controls of different ethnicities, suggesting that they might contribute to higher risk for this disease rather than behaving like a monogenic disease locus. APL thus appears to behave like a complex trait with a component of genetic susceptibility, which, in some patients, is mediated by LMNB2, with the further requirement for any one of several secondary associated illnesses or conditions (Hegele et al., 2006; Worman et al., 2009). Because of the recent discovery of the genetic mutation underlying ADLD, relatively little is known about the underlying disease mechanisms. This chapter will attempt to collate and summarize clinical and genetic features that underlie this disease and speculate on the potential pathological mechanisms that may underlie the disease. This chapter will focus exclusively on the ADLD phenotype localized to Chr.5q23–q31 or caused by duplications of the lamin B1 gene.

II. Leukodystrophies ADLD is a member of a group of disorders known as leukodystrophies. Leukody strophies are hereditary disorders affecting the white matter or myelin tracts of the central nervous system (CNS) although the PNS may also be involved (Baumann and Turpin, 2000; Lyon et al., 2006). Myelin abnormalities in leukodystrophies are the primary pathology and they are not secondary to underlying neuronal disease (Coffeen et al., 2000; Lyon et al., 2006). Myelin is composed of multiple layers of lipids and proteins and is formed by the extension of oligodendrocyte cell membranes which wrap around the axons in the CNS in a spiral manner. Myelin is composed of 20–30% protein and 70–80% lipid. The two major myelin proteins in the CNS are myelin basic protein (MBP) and proteolipid protein (PLP1) which together constitute 80–90% of total myelin protein. MBP accounts for �35% and PLP1 accounts for �50% of myelin protein weight. Other myelin proteins include myelin-associated glycoprotein (MAG) and myelin oligoden drocyte glycoprotein (MOG) (Costello et al., 2009; Lyon et al., 2006). Once considered rare, the advent and widespread use of sophisticated magnetic resonance imaging (MRI) suggests that the leukodystrophies may be more common than previously thought (Costello et al., 2009). Disease pathology includes delayed myelination, dysmyelination (defective formation of the myelin sheath), demyelination (destructive removal of the myelin sheath), or a combination of these (Lyon et al., 2006). Most hereditary leukodystrophies have distinct biochemical or metabolic abnormalities and usually have a childhood onset (Costello et al., 2009; Lyon et al., 2006). The more common childhood-onset leukodystrophies such as X-linked adrenoleukodystrophy

14. ADLD Caused by Lamin B1 Duplications

341

(X-ALD), metachromatic leukodystrophy (MLD), Canavan disease, and Krabbe’s dis ease (globoid cell leukodystrophy) are caused by genetic defects in enzymes and are characterized by accumulation of toxic metabolites (Baumann and Turpin, 2000; Costello et al., 2009; Lyon et al., 2006). Pelizaeus–Merzbacher disease (PMD), another childhoodonset leukodystrophy, is caused by mutations including duplication involving the gene encoding the myelin protein PLP1 (Woodward, 2008). Vanishing white matter disease is also a common childhood leukodystrophy and is caused by homozygous mutations in the genes encoding the five subunits of the ubiquitously expressed transcription elongation factor, eIF2B (Schiffmann and van der Knaap, 2004). Alexander’s disease is a sporadic leukodystrophy with primarily a childhood onset caused by mutations in the gene encoding for the astrocyte intermediate protein, glial fibrillary acidic protein (GFAP) (Lyon et al., 2006; Messing et al., 2001; Mignot et al., 2004). Except for Alexander’s disease, all the diseases mentioned above are caused by either X-linked recessive or autosomal recessive mutations. Heterozygous mutations in GFAP may suggest a domi nant gain of function underlying Alexander’s disease (Lyon et al., 2006; Messing et al., 2001). As most of the patients are sporadic, the mode of inheritance is difficult to ascertain. Although these diseases have a mostly childhood age of onset, juvenile and adult variants have also been described (Lyon et al., 2006; Messing et al., 2001).

III. Adult-Onset Autosomal Dominant Leukodystrophy ADLD was first described in 1984 in a large Irish American family (Eldridge et al., 1984). It is unique among the leukodystrophies in that it has an adult age of onset and an autosomal dominant mode of inheritance with early autonomic symptoms (Coffeen et al., 2000; Eldridge et al., 1984; Schwankhaus et al., 1994). Since the initial description, similar disease phenotypes have been described in Japanese, Swedish, French, Italian, and French ethnic groups (Table I) (Asahara et al., 1996; Brussino et al., 2009b, 2010; Marklund et al., 2006; Meijer et al., 2008; Melberg et al., 2006). Coffeen et al. (2000) first localized the gene for ADLD to chromosome 5q31 in the large Irish American kindred described by Eldridge et al. using family-based linkage analysis (Coffeen et al., 2000). Padiath et al. (2006) were able to identify the under lying genetic defect as a duplication of a genomic segment containing the gene lamin B1 in multiple families of Caucasian and Japanese origin. Subsequent to this, lamin B1 duplications were identified in Italian and French Canadian kindreds (Table I) (Asahara et al., 1996; Brussino et al., 2009b; Meijer et al., 2008). A. Clinical Features All reports of ADLD localizing to Chr. 5q23–q31 have very similar clinical features with a few exceptions (Table II). The common features are as follows: patients usually present in the fourth or fifth decade of life and with autonomic dysfunction followed by loss of fine motor control or gait disturbances. In most of the patients, symptoms of autonomic dysfunction are usually the first symptoms of the disorder and may precede

342

Quasar Saleem Padiath and Ying-Hui Fu

Table I Summary of ADLD families with lamin B1 duplications or localizing to Chr. 5q23–q31 Reference

Ethnic origin

Coffeen et al. (2000), Eldridge et al. (1984), Padiath et al. (2006), Schwankhaus et al. (1994) Padiath et al. (2006) Asahara et al. (1996), Padiath et al. (2006) Marklund et al. (2006), Melberg et al. (2006) Meijer et al. (2008) Brussino et al. (2009b) Brussino et al. (2010)

Irish American (Caucasian) 2a

Lamin B1 duplication

American (Caucasian) Japanese

1 1

Lamin B1 duplication Lamin B1 duplication

Swedish

2b

Localized to Chr. 5q23

French Canadian Italian Italian

1 1 1

Lamin B1 duplication Lamin B1 duplication Localized to Chr. 5q23 but no lamin B1 duplication

a b

No. of families

Mutation status

Both families had the same duplication size, suggesting that they arose from a common founder.

Of the two families, only one was localized to Chr. 5q23. The genetic details of the other family were not described.

other symptoms by many years (Coffeen et al., 2000; Schwankhaus et al., 1994). In the original Eldridge kindred, four of the five patients examined had a lifelong inability to sweat and three of them had other symptoms of autonomic dysfunction such as postural hypotension, urinary dysfunction, impotence in males, and constipation (Eldridge et al., 1984). Other ADLD families with published clinical descriptions also exhibited autonomic disturbances with various forms of bowel and bladder dysfunction often as the first presenting symptoms. A notable exception to this was the kindred described by Brussino et al. (2010) which did not show any initial signs of autonomic disturbances. However, affected individuals subsequently developed urin ary incontinence. Although the genetic defect in this family also localized to Chr. 5q23, no lamin B1 duplication was identified. Increased lamin B1 levels were however detected in patient lymphoblastoid cell lines (Brussino et al., 2010). Apart from autonomic symptoms, the other major findings in these patients were those involving the cerebellar and pyramidal systems (Table II). Cerebellar dysfunction was manifested by ataxia, nystagmus, dysmetria, and action tremors. Ataxia was a common feature in most of the families, although in the Italian family with no lamin B1 duplication, ataxia was absent (Brussino et al., 2010). In the French Canadian kindred, cerebellar signs were less overt with only a slightly affected dysdiadochokin esis (Meijer et al., 2008). Pyramidal signs include spasticity and weakness of both upper and lower extremities with a greater involvement of the lower extremities (Table II). In both Italian families described by Brussino et al. (2009a, b), pseudobulbar signs such as dysarthria and dysphagia were also described. Sensory involvement was usually absent although mild visual problems and hearing loss were described by Schwankhaus et al. (1994) in the

Table II Summary of clinical details of ADLD families Reference

Age of onset

Initial autonomic symptoms

Autonomic symptoms Cerebellar signs

Pyramidal signs

Sensory

Nerve conduction tests

Cognitive impairment

Oligoclonal bands

Coffeen et al. (2000), Fourth decade Eldridge et al. (1984), Schwankhaus et al. (1994)

Yes

Ataxia, dysmetria, nystagmus, dysarthria

None

Yes

Mild symmetrical distal sensory loss, some patients showed visual and hearing problems None

Mild

Fifth to sixth decade

Hyperreflexia, weakness of upper and especially lower limbs, spasticity, upward plantar response Pyramidal tract involvement

Normal

Marklund et al. (2006), Melberg et al. (2006), Sundblom et al. (2009)

Normal

Not described Not described

Meijer et al. (2008)

Early fifth decade

Yes

Less overt, slightly affected dysdiadochokinesis

Hyperreflexia, limb weakness, extensor plantar response

Intact, no visual or hearing loss

Normal

None

Brussino et al. (2009b)

Fourth decade

Yes

Bowel/bladder dysfunction, orthostatic hypotension, impotence, decreased sweating Micturition urgency, impotence, constipation, decreased sweating, postural hypotension Bowel/bladder dysfunction, orthostatic hypotension, impotence Bladder disturbances, fecal incontinence, impotence

Ataxia

Brussino et al. (2010)

Fourth or fifth decade

No

None

No ataxia, action tremor of head and hands

Lower limb weakness, Loss of vibratory hyperreflexia, sense extensor plantar response None Hyperreflexia, weakness of upper and lower limbs, extensor plantar response

Ataxia

None

Not described Mild

Not described

Not described None

Not described

344

Quasar Saleem Padiath and Ying-Hui Fu

Irish American kindred. Nerve conduction velocities and peripheral nerve biopsies were normal. Cognitive defects were either mild or absent. Cerebrospinal fluid (CSF) analysis did not reveal the presence of oligoclonal bands in patients from the Irish American kindred (Schwankhaus et al., 1994). Patients usually showed a progressive deterioration in symptoms and showed a survival rate of �20 years after the onset of symptoms (Coffeen et al., 2000). The French Canadian family, however, did sometimes exhibit periods of remission (Meijer et al., 2008). B. Radiological Features Before the widespread use of computed tomography (CT) and MRI scans, many of the patients with ADLD were diagnosed as having chronic progressive multiple sclerosis (MS) (Eldridge et al., 1984). MRI is the method of choice for this diagnosis of the white matter disorder and helps distinguish it from MS (Costello et al., 2009). ADLD patients show widespread diffuse, confluent, symmetrical white matter abnorm alities mainly involving the fronto–parietal and cerebellar white matter (Fig. 2A) (Brussino et al., 2009b; Coffeen et al., 2000; Melberg et al., 2006; Schwankhaus et al., 1988). Patients in whom the disease is more advanced showed involvement of the occipital and temporal lobes (Schwankhaus et al., 1988). The periventricular rim of white matter appeared to be spared (Brussino et al., 2009b). White matter changes were also seen in the posterior part of the corpus callosum of the Swedish patients studied (Melberg et al., 2006). MRI examination also revealed the involvement of the spinal cord in ADLD, with patients showing signal intensity changes suggesting white matter abnormalities along the whole cord (Sundblom et al., 2009). However, white matter changes seemed less marked in the spinal cord than the brain. Patients also showed size reduction indicative of mild cerebral and cerebellar hemispheric atrophy and atrophy of the medulla oblongata (Melberg et al., 2006; Schwankhaus et al., 1988). In the Swedish study, atrophy of the corpus callosum was also described (Melberg et al., 2006). C. Histopathological Features Histopathological analyses were carried out on autopsy brain samples from patients in the Irish American and Swedish kindreds and were similar in both cases (Coffeen et al., 2000; Melberg et al., 2006; Schwankhaus et al., 1988). Grossly, the brain samples showed striking white matter abnormalities involving the superiomedial por tions of the frontal and parietal lobes with diminishing severity toward the occipital lobes and some involvement of the corpus callosum. The affected white matter was gray, semitranslucent, and jellylike (Fig. 2B). The subcortical U fibers were spared. Cerebellar white matter was also severely affected with white matter abnormalities seen in both the cerebellar peduncles and the hemispheres. Patchy brain stem involve ment was also seen. Grey matter appeared normal (Coffeen et al., 2000; Melberg et al., 2006; Schwankhaus et al., 1988).

345

14. ADLD Caused by Lamin B1 Duplications

(A)

(C)

(B)

(D)

Fig. 2 Radiology and pathology findings in ADLD. (A) MRI scan of an ADLD patient showing high intensity signal of cerebral white matter, suggesting a pattern of symmetrical demyelination. (B) The white matter of the occipital lobe shows small and confluent gray, gelatinous patches of white matter depletion (arrows). Arcuate fibers and cortical gray matter are intact. (C) Immunostaining of cerebellar white matter with an antibody to MBP shows patchy depletion of myelinated fibers. Scale bars = 2 mm (D) Astrocytes (arrows) in an area of demyelination show foreshortened and beaded processes. Scale bars = 50 µm. Reprinted with permission from Coffeen et al. (2000). Copyright Oxford University Press.

Light microscopy showed extensive myelin loss in affected areas where the myelin appeared to be vacuolated with better preservation around blood vessels (Fig. 2C). Axons did not seem to be severely affected and there was no obvious neuronal pathology (Coffeen et al., 2000; Melberg et al., 2006; Schwankhaus et al., 1988). Oligodendrocytes did not appear to be reduced in number within the lesions and no astrogliosis was observed. No inflammatory infiltrates were observed. Astrocytes within the lesion showed abnormally beaded and shortened processes using the astrocyte marker (Fig. 2D) (Coffeen et al., 2000; Melberg et al., 2006; Schwankhaus et al., 1988). Lipid macrophages were absent, suggesting the lack of evidence for any well-defined lysosomal storage diseases (Coffeen et al., 2000; Melberg et al., 2006; Schwankhaus et al., 1988). Analysis of spinal

346

Quasar Saleem Padiath and Ying-Hui Fu

cord from an ADLD patient showed myelin loss only in high magnification images (Sundblom et al., 2009). The lack of astrogliosis and inflammation together with the lack of reduction in oligodendrocyte number distinguishes the pathological appearance from that of MS (Coffeen et al., 2000).

IV. Molecular Genetics of ADLD Coffeen et al. (2000) first localized the gene for ADLD to chromosome 5q31 in the large Irish American kindred described by Eldridge et al., using family-based linkage analysis. The region was further narrowed to a 1.5 mega base (Mb) within the same chromosomal region by Melberg et al. (2006). Analyzing the same kindred described by Coffeen et al. (2000) and another Irish American kindred, Padiath et al. (2006) were able to identify the underlying genetic defect as an extra copy (a duplication) of a genomic segment within this chromosomal region. No other mutations were identified in any of the other genes within the critical region. The presence of the duplication was confirmed by using quantitative PCR (QT-PCR) analysis and it was shown to be present only in patients and not in controls. This suggested that the duplication was indeed a mutation and not a common polymorph ism. Analysis of the extent of the duplication in different families revealed that in all cases, the duplication encompassed three protein-coding sequences (Fig. 3A). Of these genes, only lamin B1 was completely contained with the duplicated segment and was known to express in the brain. The duplicated segment was shown to be tandemly oriented in a head to-tail manner in all the families studied (Fig. 3A). As no other genetic mutation was identified in the families, the authors hypothesized that an increase in gene dosage of lamin B1 was responsible for the disease phenotype. This was confirmed by analysis of RNA and protein expression of lamin B1 from postmortem patient’s brain tissue. Patients showed higher levels of lamin B1 when compared to controls. Together with the Irish American kindreds, Padiath et al. (2006) were also able to identify lamin B1 duplications in independent families from Caucasian and Japanese ethnic groups. Subsequent to this, lamin B1 duplications were identified in Italian and French Canadian kindreds. Brussino et al. (2009b) described a family with ADLD-like features with the genetic defect localizing to Chr. 5q23, but without copy number mutations involving the lamin B1 genomic region. This suggested that the lamin B1 gene was not duplicated. They also did not find any point mutations in the lamin B1 coding region, untranslated region (UTR), and putative promoter region. Southern blotting analysis did not reveal any structural alteration in the lamin B1 region in patients. However, when lamin B1 mRNA expression levels were analyzed in lymphoblastoid cell lines, it was found that lamin B1 levels were significantly increased in patients versus controls. In these patients, lamin B1 mRNA levels were comparable to that of patients that had lamin B1 duplications. The increase in lamin B1 levels without any mutation or structural defect led the authors to hypothesize that a cis-acting regulatory mutation may have led to the increased mRNA levels. These findings further support the notion that the ADLD phenotypes may be a result of increased lamin B1 expression (Brussino et al., 2010).

347

14. ADLD Caused by Lamin B1 Duplications

(A)

Centromere

Chromosome 5

Telomere

126050000 126100000 126150000 126200000 126250000 126300000 126350000 126400000

Chromosomal position

Lamin B1

Genes

AK093561 MARCH 3

Lamin B1

Lamin B1

AK093561 MARCH 3

AK093561 MARCH 3

(B)

Centromere

Chromosome 5

Telomere

126050000 126100000 126150000 126200000 126250000 126300000 126350000 126400000

Chromosomal position

Lamin B1 AK093561 MARCH 3

Genes

Family 1 - 169,455 bp Family 2 - 341,064 bp 31 kb Family 3 - >150,000 bp

Duplication extent 69 kb

22 kb Family 4 - >150,000 bp 28 kb

Fig. 3 Lamin B1 duplications in ADLD patients. (A) Schematic representation showing genomic structure of the duplications. The genes LMNB1 and AK093561 are completely duplicated. Note that MARCH3 is only partially duplicated. Arrows above the genes represent the direction of transcription. (B) The size and extent of the duplication for different ADLD families. Families 1–3 were described by Padiath et al. (2006) and family 4 was described by Brussino et al. (2009b). Reprinted with permission from Padiath et al. (2006). Copyright Nature Publishing Group.

V. Molecular Mechanisms of ADLD Duplication Events The sizes of the ADLD duplications have been analyzed in four different families to date (Fig. 3B) (Brussino et al., 2009b; Padiath et al., 2006). The duplication break points were different in different families demonstrating that the duplications were nonrecurrent and independent mutational events. The size of the duplicated segment ranged from �140 to �340 kb in the different families analyzed (Fig. 3B). In all cases, the duplicated segment encompassed the lamin B1 gene (Brussino et al., 2009b; Padiath et al., 2006).

348

Quasar Saleem Padiath and Ying-Hui Fu

The precise duplication junctions have been determined in two families studied by Padiath et al. (2006) where a microhomology of two base pairs was observed at the junction between the duplicated and original genomic segments. The sequence at both ends of the duplicated segment matched the reference sequence perfectly with no deletions or insertions. The authors did not find any evidence of low complexity repeats or significant homology between the telomeric and centromeric ends of the duplicated segment that may have predisposed the region to genomic rearrangements. The authors suggested that duplications caused solely by a nonallelic homologous recombination mechanism may be unlikely The authors did observe that the ends of the duplicated segments were in regions that were rich in Alu repeats in both these families. They also observed that homopolymeric stretches of A or T nucleotides were found around the centromeric end in the two families studied (Brussino et al., 2009b; Padiath et al., 2006). The ADLD duplication is similar in many respects to the duplication that causes PMD. PMD duplications involve the myelin gene PLP1 and like ADLD they are also nonrecurrent and independent mutation events. An analysis by Woodward et al. (2005) of 13 duplication junction sequences from PMD patients also revealed microhomolo gies of 1–6 bp. The authors also noted that stretches of single or alternating tracts of purines and pyrimidines that may cause secondary structures and predispose to doublestranded breaks were common. The authors suggest that a mechanism involving coupled homologous and nonhomologous recombination completed by nonhomolo gous end joining may be responsible for PMD mutations (Woodward, 2008; Woodward et al., 2005). A detailed analysis of PMD duplications using array comparative genomic hybridi zation (CGH) and oligonucleotide array CGH revealed that some PMD duplications are not simple tandem duplications but may be more complex and involve simple duplications mixed with deletions, inverted duplications, and triplications (Lee et al., 2006, 2007). The authors suggested a replication-based model termed fork stalling and template switching (FoSTeS) as a possible mechanism for these complex rearrange ments (Lee et al., 2007; Zhang et al., 2009). Another replication-based model termed “microhomology-mediated break-induced replication” (MMIR) has also been put forward to explain the presence of the microhomology in the PMD duplication junctions and the complex nature of the rearrangements (Hastings et al., 2009). A more detailed examination of the region around the lamin B1 gene is essential to understand the nature of the duplication events and why the genomic architecture of this region is prone to instability. It is also essential that more ADLD duplication events are analyzed to determine whether these are simple tandem duplications, as initially reported, or whether they also exhibit the complexity that has been described for PMD duplications and rearrangements. An indicator that ADLD rearrangements may be more complex is the finding by Brussino et al. (2009b) of a family with ADLD-like features localized to chr5q23 but without lamin B1 duplications. Their findings that lamin B1 mRNA expression levels were increased may suggest that complex rearrangements or mutations involving the regulatory elements of the lamin B1 gene may play a part. Although duplications have

14. ADLD Caused by Lamin B1 Duplications

349

been the only confirmed mutational mechanism identified in ADLD to date, it remains to be seen whether deletions, point mutations, or other mutational events also play a role in disease causation. The identification and characterization of more ADLD families is essential to answer these questions.

VI. Molecular Mechanisms Underlying ADLD Disease Pathology ADLD is a disease that is characterized primarily by the loss of myelin in the CNS with a relatively late adult age of onset. The identification of the mutation causing ADLD implicates an increased dosage of lamin B1 gene expression in the disease phenotype. ADLD is also the first report of the involvement of lamin genes in disorders of the CNS. The mechanism of increased gene dosage seems to be a recurring theme in demyelinating disorders. PMD and Charcot–Marie–Tooth neuro pathy type I disease, which are characterized by central and peripheral demyelination respectively, are also caused by gene duplications (Lupski et al., 1991; Woodward, 2008). It may suggest that pathways involved in myelin formation and regulation may be uniquely susceptible to gene dosage. The late age of onset also poses interesting mechanistic questions in ADLD. Although it has not been addressed, it is possible that patients with ADLD have a normal initial development of myelin. This would suggest that the loss of myelin later in life is the result of a breakdown of the normal mechanisms involved in myelin maintenance. However, it is also possible that if there are defects in the initial development of myelin these may be mild enough not to exhibit a phenotype and do so only after an advanced age. An example of the variations in gene dosage and the age of phenotype is seen in PLP1 mutant mice. The mice with high PLP1 dosage (usually homozygous for the transgene) have severe early-onset dysmyelination, increased oligodendrocyte apoptosis, and premature death. By contrast, hemizygous mice with a lower transgene copy number develop normally and have no clinical signs until later in life. Late onset demyelination plus axonal swelling and degeneration becomes apparent at around 15–18 months of age, suggesting that the oligodendrocytes are unable to maintain their myelin sheaths (Woodward, 2008). In PMD, deletions involving the PLP1 gene also produce disease phenotypes (Woodward, 2008). Thus, reducing or increasing levels of the PLP1 have pathological effects. It is unclear if this is also the case with ADLD. It is possible that in the case of ADLD, lamin B1 duplications represent a novel gain of function mutation with phenotypes distinct from those caused by a loss of lamin B1. As the lamin B1 duplication has been identified only recently, there are relatively few studies that have directly studied the mechanistic effects of increased lamin B1 levels. It is thus only possible to speculate on the different cell types or cellular mechanisms that may underlie the disease pathology and some of these are discussed below. Our understanding of the functional role of the nuclear lamina has evolved rapidly in the last few years from it being regarded as a passive structural entity to identifying its

350

Quasar Saleem Padiath and Ying-Hui Fu

roles in diverse cellular functions such as chromatin organization, control of gene expression, and cell division (Parnaik, 2008; Worman et al., 2009). Perturbations of the nuclear lamina by alterations in lamin B1 gene dosage levels can thus affect any of these nonexclusive and overlapping processes, thereby leading to the ADLD disease phenotype. A. Cell Types Affected in ADLD The ubiquitous nature of lamin B1 gene expression makes it difficult to pinpoint the cell type or types that are affected by increased lamin B1 expression. As CNS demye lination is the primary pathology in ADLD, cells that are involved in this process are the obvious candidate target cell types. In the CNS, oligodendrocytes are the primary cells responsible for myelinating axons (Sherman and Brophy, 2005). However, signals from astrocytes and neurons, among other cells, are essential for proper myelin formation and maintenance (Barres, 2008; Barres and Raff, 1999). In histopatholgical analysis of ADLD patient brains, oligodendrocytes and neurons did not show any alternations in cell number or morphology. Astrocytes did, however, show an abnormal beading and shortening of their processes (Coffeen et al., 2000). It is unclear whether this is a cause or consequence of the ADLD pathology. Demyelination as a result of astrocyte dysfunction has been suggested before for Alexander disease and also in a disease caused by mutations in the gene called megalencephalic leukoencephalopathy with subcortical cysts gene 1 (MLC1) (Boor et al., 2005; Messing et al., 2001). One clue toward understanding the cell types and mechanisms involved in ADLD was the study by Lin and Fu (2009). Overexpression of lamin B1 in oligodendrocytes led to premature arrest of oligodendrocyte differentiation, which the authors suggest may be caused by reduced transcription of myelin genes and by mislocalization of myelin proteins such as MBP and PLP1. Conversely, increased lamin B1 did not exert significant effects on the expression of the astrocyte marker, GFAP, in mixed glial cultures and caused an increase in expression from a GFAP promoter in luciferase assays. This suggests that increased lamin B1 may have different effects on different CNS cell types. Lamin B1 was also shown to be developmentally regulated with levels peaking at birth or postnatal day 1, followed by a gradual decrease from postnatal day 1 to 10 months of age. This was inversely correlated with other myelin genes such as MBP and MAG. The authors conclude that their results indicate that regulation of lamin B1 is important for myelin maintenance (Lin and Fu, 2009).

B. Alterations in Nuclear Structure and Organization as a Disease Mechanism in ADLD A primary role for the nuclear lamina is to provide structural scaffolding for the cell nucleus (Cohen et al., 2008). It is therefore possible that increased expression of lamin B1 may in some way perturb nuclear integrity and affect cell function. Although a histopathological examination of brain tissue from ADLD patients did not reveal any abnormalities of nuclear morphology, it is possible that these were too subtle to be identified and these need to be reexamined in light of lamin B1 involvement.

14. ADLD Caused by Lamin B1 Duplications

351

An overexpression of lamin B1 in different cell types results in a marked increase in the surface area of the nuclear envelope accompanied by extensive folding and blebbing (Favreau et al., 2003; Padiath et al., 2006). How these alterations affect the functioning of the cells is unclear. Lin and Fu (2009) also showed that increased lamin B1expression results in disturbances of inner nuclear membrane proteins and nuclear pore transport in different cell types. Given the level of interaction between the different lamin proteins, it might be likely that maintaining stoichiometric amounts of these proteins is important for the proper assembly of the nuclear lamina. Altering relative amounts of lamin proteins may affect the structural integrity of the nuclear membrane and produce cellular dysfunction. The occurrence of nuclear abnormalities in cell culture models with mutant lamin A proteins was reduced when lamin B1 was coexpressed with mutant lamin A, emphasizing the functional interaction of the two types of lamins (Favreau et al., 2003; Padiath et al., 2006). C. Alterations in the Regulation of Gene Expression as a Disease Mechanism in ADLD The nuclear lamina plays an important role in the regulation of gene expression through a number of different mechanisms. These include modulation of gene expres sion by virtue of chromatin organization and chromosome positioning within the nucleus (Dauer and Worman, 2009). The nuclear lamina provides an anchor point for linking chromatin to the nuclear envelope. Regions of the chromosome that are associated with the lamina are thought to be transcriptionally silenced and exhibit characteristic histone methylation and acetylation signatures (Guelen et al., 2008; Towbin et al., 2009). An increased expression of lamin B1 can alter regions of the chromatin that are associated with the nuclear lamina and thereby affects gene that are important for myelin formation or maintenance. Defects in lamin B1 expression have been shown to affect the interphase chromosome position and gene expression (Malhas et al., 2007). Chromatin and histone changes were observed in fibroblasts overexpres sing lamin B1 (Lin and Fu, 2009). Nuclear organization has also been suggested to play a role in differentiating oligodendrocytes (Nielsen et al., 2002). The nuclear lamina and specifically lamin B1 have also been shown to play an important role in RNA synthesis by RNA polymerase I and II and thus alterations in lamin B1 dosage may affect the transcription of gene important for myelination (Tang et al., 2008). The interaction of lamin B1 with the transcription factor Oct-1 has been shown to be an important mechanism by which the nuclear envelope may regulate gene expression and contribute to the cellular response to oxidative stress (Malhas et al., 2009).

VII. Animal Models of Lamin B1 Mutations No animal models with increased lamin B1 expression recapitulating the ADLD phenotype have been generated to date. However, Padiath et al. (2006) showed that overexpression of either the human lamin B1 or the Drosophila ortholog, lamin Dm0, in the fly eye resulted in a degenerative phenotype. Expression in neurons or

352

Quasar Saleem Padiath and Ying-Hui Fu

glial cells using different drivers resulted in complete lethality. Overexpression of lamin Dm0 was also shown to have an effect on lifespan in flies when expressed in fat and muscle cells (Brandt et al., 2008). This suggests that increased levels of lamin B can indeed cause toxic and deleterious cellular effects. However, as flies do not have myelin, it is difficult to directly correlate these effects with the ADLD phenotype. Reducing levels of lamin B1 has significant consequences as homozygous mutant mice engineered with a lamin B1 protein lacking several key functional domains survived embryonic development but died at birth with defects in lung and bone. Fibroblasts from mutant embryos grew under standard cell culture conditions but displayed grossly misshapen nuclei, impaired differentiation, increased polyploidy, and premature senescence (Vergnes et al., 2004). It is unclear whether these mice also had CNS abnormalities or demyelination as these were not checked for. It is possible that the lamin B1 duplications represent a novel gain of function mutation with phenotypes distinct from those caused by a loss of lamin B1. Interestingly, flies with mutations in the Drosophila lamin B1 ortholog, lamin Dm0, that greatly reduce expression showed altered locomotor activity and impaired motor coordination (LenzBohme et al., 1997).

VIII. Aging and ADLD One of the most distinctive features of the ADLD phenotype is the relatively late onset of the disease symptoms. Whether this represents the time taken for the consequences of the lamin B1 mutation to manifest—a possible scenario would be that the accumulation of lamin B1 protein is toxic to the cell over time—or whether this indicates that changes during aging trigger the consequences of the lamin B1 duplication is unclear. It is possible that the nuclear architecture becomes more susceptible to perturbations at a later age. The pathological effects of lamin B1 overexpression may thus manifest itself only later in adulthood. The latter possibi lity is interesting as in the recent years there has been an explosion of data linking the nuclear lamina and aging (Broers et al., 2006; Worman et al., 2009). This includes the identification of mutations in lamin A that cause the premature aging syndrome, Hutchinson–Gilford progeria syndrome (HGPS), and the identification that nuclear architecture undergoes progressive age-related changes, in part influ enced by levels of lamin proteins (Broers et al., 2006; Eriksson et al., 2003; Haithcock et al., 2005; Scaffidi and Misteli, 2006). Aging is also one of the most profound factors affecting remyelination. A recent study has suggested that age-dependent remyelination is under epigenetic control which involves histone acetylation which in turn controls the expression of genes important for oligodendrocyte differentiation (Shen et al., 2008). It is possible that perturbations in the nuclear lamina caused by lamin B1 duplications affect this age-related epigenetic control of remyelination.

14. ADLD Caused by Lamin B1 Duplications

353

IX. ADLD and MS Although the clinical and radiological phenotypes of ADLD and MS are distinct, they do show some degree of similarity and overlap. Before the widespread use of MRI and CT techniques, ADLD was often misdiagnosed as chronic progressive MS (Eldridge et al., 1984; Schwankhaus et al., 1994). An analysis of MS patients, however, did not identify any lamin B1 mutations including copy number changes, suggesting that lamin B1 mutations may not play a direct role in MS disease causation (Brussino et al., 2009a). Although lamin B1 mutations may not cause MS, a number of intriguing mechan istic links are being identified between the two. MS is thought to be caused by an autoimmune inflammatory process. Interestingly, autoantibodies to lamin B have been found in the sera of patients with autoimmune disorders such as systemic lupus erythematosus, scleroderma, and autoimmune liver disease (Hill et al., 1996; Tan, 1988). Furthermore, the monoclonal antibody, J1-31, raised against plaque material from MS patient’s brain was shown to recognize lamin B (Garcia et al., 2003). A recent study has shown that neutrophils that undergo spontaneous apoptosis expressed LMNB1 on the cell surface. As neutrophils are thought to represent an important source of autoantigens in autoinflammatory diseases, it provides a further link between lamin B1 and autoimmune diseases (Moisan and Girard, 2006). MS results from destruction of the protective myelin sheath surrounding axons, which prevents the transmission of nerve impulses. Precursors of oligodendrocytes, the cells capable of myelinating axons, are preserved in demyelinating lesions; however, why these precursors do not differentiate into mature oligodendrocytes and remyelinate axons is unknown (Brosnan and John, 2009). A recent study has implicated a disturbance of the Notch signaling pathway in this failure to differ entiate. Nakahara et al. (2009) showed that the Notch1 intracellular domain failed to show evidence of nuclear translocation that is required for proper Notch signaling in MS brain samples. Instead, along with other proteins, it formed perinuclear cyto plasmic aggregates. One of the proteins that was found in the aggregates was lamin B1. The authors hypothesize that the presence of lamin B1in the aggregates found in MS oligodendrocyte precursor cells raised the possibility that similar pathophysiol ogies may underlie both MS and ADLD (Nakahara et al., 2009).

X. Summary The recent identification of lamin B1 duplications as the cause of the adult-onset demyelinating disorder, ADLD, defines the nuclear lamina as a new player in CNS myelin biology. The elucidation of the mechanisms underlying ADLD and the devel opment of animal models that recapitulate the phenotype may help in identifying additional levels of regulation in the complex program of myelin formation and oligodendrocyte development. Understanding these mechanisms may not only help

354

Quasar Saleem Padiath and Ying-Hui Fu

identify therapeutic avenues in ADLD but also provide an insight into more common demyelinating diseases such as MS.

Acknowledgments This work was supported by NIH grant NS062733 (YHF), a Sandler Neurogenetics fund (YHF), and a fellowship from the Larry L. Hillblowm Foundation (QSP). The authors would also like to acknowledge Louis J. Ptáček and other members of the Fu and Ptáček laboratories for helpful suggestions and critical reading of the manuscript.

References Aebi, U., Cohn, J., Buhle, L., and Gerace, L. (1986). The nuclear lamina is a meshwork of intermediate-type filaments. Nature 323, 560–564. Asahara, H., Yoshimura, T., Sada, S., Furuya, H., and Kobayashi, T. (1996). A Japanese family with probably autosomal dominant adult-onset leukodystrophy. Rinsho Shinkeigaku 36, 968–972. Barres, B. A. (2008). The mystery and magic of glia: A perspective on their roles in health and disease. Neuron 60, 430–440. Barres, B. A. and Raff, M. C. (1999). Axonal control of oligodendrocyte development. J. Cell Biol. 147, 1123–1128. Baumann, N., and Turpin, J. C. (2000). Adult-onset leukodystrophies. J. Neurol. 247, 751–759. Boor, P. K., de Groot, K., Waisfisz, Q., Kamphorst, W., Oudejans, C. B., Powers, J. M., Pronk, J. C., Scheper, G. C., and van der Knaap, M. S. (2005). MLC1: A novel protein in distal astroglial processes. J. Neuropathol. Exp. Neurol. 64, 412–419. Brandt, A., Krohne, G., and Grosshans, J. (2008). The farnesylated nuclear proteins KUGELKERN and LAMIN B promote aging-like phenotypes in Drosophila flies. Aging Cell 7, 541–551. Broers, J. L., Ramaekers, F. C., Bonne, G., Yaou, R. B., and Hutchison, C. J. (2006). Nuclear lamins: Laminopathies and their role in premature ageing. Physiol. Rev. 86, 967–1008. Brosnan, C. F., and John, G. R. (2009). Revisiting Notch in remyelination of multiple sclerosis lesions. J. Clin. Invest. 119, 10–13. Brussino, A., D’Alfonso, S., Cagnoli, C., Di Gregorio, E., Barberis, M., Padovan, S., Vaula, G., Pinessi, L., Squadrone, S., Abete, M. C., Collimedaglia, L., Guerini, F. R., et al. (2009a). Mutations in the lamin B1 gene are not present in multiple sclerosis. Eur. J. Neurol. 16, 544–546 Brussino, A., Vaula, G., Cagnoli, C., Mauro, A., Pradotto, L., Daniele, D., Di Gregorio, E., Barberis, M., Arduino, C., Squadrone, S., Abete, M. C., Migone, N. et al. (2009b). A novel family with Lamin B1 duplication associated with adult-onset leucoencephalopathy. J. Neurol. Neurosurg. Psychiatry 80, 237–240. Brussino, A., Vaula, G., Cagnoli, C., Panza, E., Seri, M., Di Gregorio, E., Scappaticci, S., Camanini, S., Daniele, D., Bradac, G. B., Pinessi, L., Cavalieri, S., et al. (2010). A family with autosomal dominant leukodystrophy linked to 5q23.2-q23.3 without lamin B1 mutations. Eur. J. Neurol. 17(4), 541–549. Burke, B., and Stewart, C. L. (2006). The laminopathies: The functional architecture of the nucleus and its contribution to disease. Annu. Rev. Genomics Hum Genet. 7, 369–405. Coffeen, C. M., McKenna, C. E., Koeppen, A. H., Plaster, N. M., Maragakis, N., Mihalopoulos, J., Schwankhaus, J. D., Flanigan, K. M., Gregg, R. G., Ptacek, L. J., and Fu, Y. H. (2000). Genetic localization of an autosomal dominant leukodystrophy mimicking chronic progressive multiple sclerosis to chromosome 5q31. Hum. Mol. Genet. 9, 787–793. Cohen, T. V., Hernandez, L., and Stewart, C. L. (2008). Functions of the nuclear envelope and lamina in development and disease. Biochem. Soc. Trans. 36, 1329–1334. Costello, D. J., Eichler, A. F., and Eichler, F. S. (2009). Leukodystrophies: Classification, diagnosis, and treatment. Neurologist 15, 319–328.

14. ADLD Caused by Lamin B1 Duplications

355

Dauer, W. T., and Worman, H. J. (2009). The nuclear envelope as a signaling node in development and disease. Dev. Cell 17, 626–638. Delbarre, E., Tramier, M., Coppey-Moisan, M., Gaillard, C., Courvalin, J. C., and Buendia, B. (2006). The truncated prelamin A in Hutchinson-Gilford progeria syndrome alters segregation of A-type and B-type lamin homopolymers. Hum. Mol. Genet. 15, 1113–1122. Eldridge, R., Anayiotos, C. P., Schlesinger, S., Cowen, D., Bever, C., Patronas, N., and McFarland, H. (1984). Hereditary adult-onset leukodystrophy simulating chronic progressive multiple sclerosis. N. Engl. J. Med. 311, 948–953. Eriksson, M., Brown, W. T., Gordon, L. B., Glynn, M. W., Singer, J., Scott, L., Erdos, M. R., Robbins, C. M., Moses, T. Y., Berglund, P., Dutra, A., Pak, E., et al. (2003). Recurrent de novo point mutations in lamin A cause Hutchinson-Gilford progeria syndrome. Nature 423, 293–298. Favreau, C., Dubosclard, E., Ostlund, C., Vigouroux, C., Capeau, J., Wehnert, M., Higuet, D., Worman, H. J., Courvalin, J. C., and Buendia, B. (2003). Expression of lamin A mutated in the carboxylterminal tail generates an aberrant nuclear phenotype similar to that observed in cells from patients with Dunnigan-type partial lipodystrophy and Emery-Dreifuss muscular dystrophy. Exp. Cell Res. 282, 14–23. Furukawa, K., and Hotta, Y. (1993). cDNA cloning of a germ cell specific lamin B3 from mouse sperma tocytes and analysis of its function by ectopic expression in somatic cells. EMBO J. 12, 97–106. Garcia, D. M., Weigum, S. E., and Koke, J. R. (2003). GFAP and nuclear lamins share an epitope recognized by monoclonal antibody J1-31. Brain Res. 976, 9–21. Gerace, L., and Burke, B. (1988). Functional organization of the nuclear envelope. Annu. Rev. Cell Biol. 4, 335–374. Goldman, R. D., Gruenbaum, Y., Moir, R. D., Shumaker, D. K., and Spann, T. P. (2002). Nuclear lamins: Building blocks of nuclear architecture. Genes Dev. 16, 533–547. Guelen, L., Pagie, L., Brasset, E., Meuleman, W., Faza, M. B., Talhout, W., Eussen, B. H., de Klein, A., Wessels, L., de Laat, W., and van Steensel, B. (2008). Domain organization of human chromosomes revealed by mapping of nuclear lamina interactions. Nature 453, 948–51. Haithcock, E., Dayani, Y., Neufeld, E., Zahand, A. J., Feinstein, N., Mattout, A., Gruenbaum, Y., and Liu, J. (2005). Age-related changes of nuclear architecture in Caenorhabditis elegans. Proc. Natl. Acad. Sci. USA 102, 16690–16695. Hastings, P. J., Ira, G., and Lupski, J. R. (2009). A microhomology-mediated break-induced replication model for the origin of human copy number variation. PLoS Genet. 5, e1000327. Hegele, R. A., Cao, H., Liu, D. M., Costain, G. A., Charlton-Menys, V., Rodger, N. W., and Durrington, P. N. (2006). Sequencing of the reannotated LMNB2 gene reveals novel mutations in patients with acquired partial lipodystrophy. Am. J. Hum. Genet. 79, 383–389. Hill, C., Roberts-Thomson, P., Pollard, A., Gillis, D., and Kirkham, B. (1996). Clinical associations of anti lamin autoantibodies. Aust. N Z J. Med. 26, 162–166. Hoger, T. H., Zatloukal, K., Waizenegger, I., and Krohne, G. (1990). Characterization of a second highly conserved B-type lamin present in cells previously thought to contain only a single B-type lamin. Chromosoma 100, 67–69. Lee, J. A., Inoue, K., Cheung, S. W., Shaw, C. A., Stankiewicz, P., and Lupski, J. R. (2006). Role of genomic architecture in PLP1 duplication causing Pelizaeus-Merzbacher disease. Hum. Mol. Genet. 15, 2250–2265. Lee, J. A., Carvalho, C. M., and Lupski, J. R. (2007). A DNA replication mechanism for generating nonrecurrent rearrangements associated with genomic disorders. Cell 131, 1235–1247. Lenz-Bohme, B., Wismar, J., Fuchs, S., Reifegerste, R., Buchner, E., Betz, H., and Schmitt, B. (1997). Insertional mutation of the Drosophila nuclear lamin Dm0 gene results in defective nuclear envelopes, clustering of nuclear pore complexes, and accumulation of annulate lamellae. J. Cell Biol. 137, 1001–1016. Lin, F., and Worman, H. J. (1993). Structural organization of the human gene encoding nuclear lamin A and nuclear lamin C. J. Biol. Chem. 268, 16321–16326. Lin, F., and Worman, H. J. (1995). Structural organization of the human gene (LMNB1) encoding nuclear lamin B1. Genomics 27, 230–236.

356

Quasar Saleem Padiath and Ying-Hui Fu Lin, F., and Worman, H. J. (1997). Expression of nuclear lamins in human tissues and cancer cell lines and transcription from the promoters of the lamin A/C and B1 genes. Exp. Cell Res. 236, 378–384. Lin, S. T., and Fu, Y. H. (2009). miR-23 regulation of lamin B1 is crucial for oligodendrocyte development and myelination. Dis. Model Mech. 2, 178–188. Lupski, J. R., de Oca-Luna, R. M., Slaugenhaupt, S., Pentao, L., Guzzetta, V., Trask, B. J., SaucedoCardenas, O., Barker, D. F., Killian, J. M., Garcia, C. A., Chakravarti, A., and Patel, P. I. (1991). DNA duplication associated with Charcot-Marie-Tooth disease type 1A. Cell 66, 219–232. Lyon, G., Fattal-Valevski, A., and Kolodny, E. H. (2006). Leukodystrophies: Clinical and genetic aspects. Top Magn. Reson. Imaging 17, 219–242. Malhas, A., Lee, C. F., Sanders, R., Saunders, N. J., and Vaux, D. J. (2007). Defects in lamin B1 expression or processing affect interphase chromosome position and gene expression. J. Cell Biol. 176, 593–603. Malhas, A. N., Lee, C. F., and Vaux, D. J. (2009). Lamin B1 controls oxidative stress responses via Oct-1. J. Cell Biol. 184, 45–55. Marklund, L., Melin, M., Melberg, A., Giedraitis, V., and Dahl, N. (2006). Adult-onset autosomal dominant leukodystrophy with automnomic symptoms restricted to 1.5 Mpb on chromosome 5q23. Am. J. Med. Genet. B Neuropsychiatr. Genet. 141B, 608–614. Meijer, I. A., Simoes-Lopes, A. A., Laurent, S., Katz, T., St-Onge, J., Verlaan, D. J., Dupre, N., Thibault, M., Mathurin, J., Bouchard, J. P., and Rouleau, G. A. (2008). A novel duplication confirms the involvement of 5q23.2 in autosomal dominant leukodystrophy. Arch. Neurol. 65, 1496–1501. Melberg, A., Hallberg, L., Kalimo, H., and Raininko, R. (2006). MR characteristics and neuropathology in adult-onset autosomal dominant leukodystrophy with autonomic symptoms. AJNR Am. J. Neuroradiol. 27, 904–911. Messing, A., Goldman, J. E., Johnson, A. B., and Brenner, M. (2001). Alexander disease: New insights from genetics. J. Neuropathol. Exp. Neurol. 60, 563–573. Mignot, C., Boespflug-Tanguy, O., Gelot, A., Dautigny, A., Pham-Dinh, D., and Rodriguez, D. (2004). Alexander disease: Putative mechanisms of an astrocytic encephalopathy. Cell Mol. Life Sci. 61, 369–385. Moisan, E., and Girard, D. (2006). Cell surface expression of intermediate filament proteins vimentin and lamin B1 in human neutrophil spontaneous apoptosis. J. Leukoc. Biol. 79, 489–498. Nakahara, J., Kanekura, K., Nawa, M., Aiso, S., and Suzuki, N. (2009). Abnormal expression of TIP30 and arrested nucleocytoplasmic transport within oligodendrocyte precursor cells in multiple sclerosis. J. Clin. Invest. 119, 169–181. Nielsen, J. A., Hudson, L. D., and Armstrong, R. C. (2002). Nuclear organization in differentiating oligodendrocytes. J. Cell Sci. 115, 4071–4079. Padiath, Q. S., Saigoh, K., Schiffmann, R., Asahara, H., Yamada, T., Koeppen, A., Hogan, K., Ptacek, L. J., and Fu, Y. H. (2006). Lamin B1 duplications cause autosomal dominant leukodystrophy. Nat. Genet. 38, 1114–1123. Parnaik, V. K. (2008). Role of nuclear lamins in nuclear organization, cellular signaling, and inherited diseases. Int. Rev. Cell Mol. Biol. 266, 157–206. Peter, M., Kitten, G. T., Lehner, C. F., Vorburger, K., Bailer, S. M., Maridor, G., and Nigg, E. A. (1989). Cloning and sequencing of cDNA clones encoding chicken lamins A and B1 and comparison of the primary structures of vertebrate A- and B-type lamins. J. Mol. Biol. 208, 393–404. Scaffidi, P., and Misteli, T. (2006). Lamin A-dependent nuclear defects in human aging. Science 312, 1059–1063. Schiffmann, R., and van der Knaap, M. S. (2004). The latest on leukodystrophies. Curr. Opin. Neurol. 17, 187–192. Schirmer, E. C., and Gerace, L. (2004). The stability of the nuclear lamina polymer changes with the composi tion of lamin subtypes according to their individual binding strengths. J. Biol. Chem. 279, 42811–42817. Schwankhaus, J. D., Patronas, N., Dorwart, R., Eldridge, R., Schlesinger, S., and McFarland, H. (1988). Computed tomography and magnetic resonance imaging in adult-onset leukodystrophy. Arch. Neurol. 45, 1004–1008. Schwankhaus, J. D., Katz, D. A., Eldridge, R., Schlesinger, S., and McFarland, H. (1994). Clinical and pathological features of an autosomal dominant, adult-onset leukodystrophy simulating chronic progres sive multiple sclerosis. Arch. Neurol. 51, 757–766.

14. ADLD Caused by Lamin B1 Duplications

357

Shen, S., Sandoval, J., Swiss, V. A., Li, J., Dupree, J., Franklin, R. J., and Casaccia-Bonnefil, P. (2008). Agedependent epigenetic control of differentiation inhibitors is critical for remyelination efficiency. Nat. Neurosci. 11, 1024–1034. Sherman, D. L., and Brophy, P. J. (2005). Mechanisms of axon ensheathment and myelin growth. Nat. Rev. Neurosci. 6, 683–690. Shimi, T., Pfleghaar, K., Kojima, S., Pack, C. G., Solovei, I., Goldman, A. E., Adam, S. A., Shumaker, D. K., Kinjo, M., Cremer, T., and Goldman, R. D. (2008). The A- and B-type nuclear lamin networks: Microdomains involved in chromatin organization and transcription. Genes Dev. 22, 3409–3421. Stuurman, N., Heins, S., and Aebi, U. (1998). Nuclear lamins: Their structure, assembly, and interactions. J. Struct. Biol. 122, 42–66. Sundblom, J., Melberg, A., Kalimo, H., Smits, A., and Raininko, R. (2009). MR imaging characteristics and neuropathology of the spinal cord in adult-onset autosomal dominant leukodystrophy with autonomic symptoms. AJNR Am. J. Neuroradiol. 30, 328–335. Tan, E. M. (1988). Autoantibodies to nuclear lamins. Ann. Intern. Med. 108, 897–898. Tang, C. W., Maya-Mendoza, A., Martin, C., Zeng, K., Chen, S., Feret, D., Wilson, S. A., and Jackson, D. A. (2008). The integrity of a lamin-B1-dependent nucleoskeleton is a fundamental determinant of RNA synthesis in human cells. J. Cell Sci. 121, 1014–1024. Towbin, B. D., Meister, P., and Gasser, S. M. (2009). The nuclear envelope—a scaffold for silencing? Curr. Opin. Genet. Dev. 19, 180–186. Vergnes, L., Peterfy, M., Bergo, M. O., Young, S. G., and Reue, K. (2004). Lamin B1 is required for mouse development and nuclear integrity. Proc. Natl. Acad. Sci. USA 101, 10428–10433. Woodward, K. J. (2008). The molecular and cellular defects underlying Pelizaeus-Merzbacher disease. Expert Rev. Mol. Med. 10, e14. Woodward, K. J., Cundall, M., Sperle, K., Sistermans, E. A., Ross, M., Howell, G., Gribble, S. M., Burford, D. C., Carter, N. P., Hobson, D. L., Garbern, J. Y., Kamholz, J. et al. (2005). Heterogeneous duplications in patients with Pelizaeus-Merzbacher disease suggest a mechanism of coupled homologous and nonhomo logous recombination. Am. J. Hum. Genet. 77, 966–987. Worman, H. J., Fong, L. G., Muchir, A., and Young, S. G. (2009). Laminopathies and the long strange trip from basic cell biology to therapy. J. Clin. Invest. 119, 1825–1836. Zhang, F., Gu, W., Hurles, M. E., and Lupski, J. R. (2009). Copy number variation in human health, disease, and evolution. Annu. Rev. Genomics Hum. Genet. 10, 451–481.

SUBJECT INDEX A Acetylation, 38, 40 See also Histones Acquired partial lipodystrophy (APL), 340 Actins cell culture models for mechanical induction of, 191–194 immunoprecipitation/solubilization conditions suitable for protein recovery from isolated HeLa cell nuclei, 111 nuclear, 100–102 promoter regulation, 190–191 a-smooth muscle, 190–191 See also Actomyosins; Lamins Actomyosins actomyosin vs, adherence surface cortical tensions, 300–301 cell genetic identity and cell sorting, 300–301 planar and apicobasal polarities patterning, 298–299 See also Cellular mechanosensing; Myosin–II dependent germ-band extension compression embryonic development Adhesion, 148–151 associated proteins identification, 195–196 focal (FA) mechanosensitivity models of FA, 149–151 stem cells and, 279–281 ADLD, See Autosomal dominant leukodystrophy (ADLD) Altered mechanical properties (nuclear), See Nuclear mechanical properties Animal models (lamin B1 mutations), 351–352 Anisotropy, fluorescence, 66–70 See also Chromatins; Nuclear plasticity Apico-basal polarity mechanical induction Fog endocytosis, 314 Twist expression into mesoderm, 315 Myo-II apical redistribution coordinated constrictions integrated by mechanical cues and leading to mesoderm invagination, 316 drosophila mesoderm invagination genetics analysis, 310–312

long-range and rapid cell–cell interactions through mechanical cues, 315 mechanotransduction hypothesis viability, 312 mechanotransduction testing, 313–314 See also Mechanical induction APL, See Acquired partial lipodystrophy (APL)

Apoptosis, 68

Aspiration, See Micropipette aspiration (MPA)

Assembly mechanisms, 8

See also Nuclear organization Atomic force microscopy (AFM) nuclear mechanical properties study

materials, 125

methods, 129–130

results and discussion, 135

prestressed nuclear organization and chromatin decompaction, 225 single-nucleus study and, 92 Autosomal dominant leukodystrophy (ADLD), 337–354

aging and, 352

animal models of lamin B1 mutations,

351–352

duplication events, 347–349

features

clinical, 341–344

histopathological, 344–346

radiological, 344

molecular genetics, 346–347

multiple sclerosis (MS) and, 353

pathology, 349–350

affected cell types, 350 gene expression expression alterations, 351 nuclear structure/organization alterations, 350–351

B Barraquer–Simons syndrome, See Acquired partial lipodystrophy (APL) Bulk extraction, 90 Bulk isolation harvesting nuclei, 217

imaging of isolated nuclei, 218

stock solutions, 217

See also Nuclear plasticity

359

360

Subject Index

C Cardiac fibrosis, 188 Adhesion-associated proteins identification, 195–196

cardiac interstitium, 189

cell transfection and promoter methods, 195

coating methods for beads, 194

gene expression regulation in mechanically

loaded cardiac cells, 190 mechanical induction of myofibroblasts, 189–190 a-smooth muscle actin cell culture models for mechanical induction, 191–194 promoter regulation, 190–191 See also Connective tissues Cell nucleus, 3 assembly mechanisms, 8 FCS and FRAP in nuclear cell biology, 28–29 measurements in cell nucleus, 25 FFM, 8

FCS, 11–13

FRAP, 8–11

RICS, 13–14

FRAP analyses, 8 and FCS in nuclear cell biology, 28–29 and mathematical modeling, 20 initial conditions for mathematical model, 21 local diffusion coefficient relation to net flux into ROI, 21–23 lessons for mathematical modeling, 29–30 measurements in cell nucleus and modeling, 25–28 numerical integration, initial conditions, and parameter fitting s, 24–25 numerical reaction–diffusion models for FRAP analysis, 23–24 kinetic analysis of fluorescence microscopy experiments, 30

mammalian, 4–8

materials and methods

cell culture and transfection, 14–15

FRAP modeling, 20–25

RISC thoery, 15–20

rationale behind, 14

subnuclear domains, 6–7

RISC

benefits, 29

diffusion coefficients and concentrations

determination, 19–20

in living cells, 19–20 measurements in cell nucleus and modeling, 28 measurements/general considerations, 17–19 thoery, 15–17 three-dimensional organization

cell nucleus, 4–8

subnuclear domains, 6

FRAP analyses, 8

Cellular mechanosensing, 145–148, 168 cell adhesion aspects, 148–149 mechanosensitivity models of FA, 149–151 cellular response to external mechanical stress, 159–167 active elastic dipole model, 160–164 cellular relaxation time, 164–167 homeostasis set point (stress vs, strain), 167 parallel vs, perpendicular orientations, 164 cytoskeletal

active and passive mechanics, 151–156

dipole polarization model, 154–156

semiflexible chain model, 153–154

tensegrity model, 153

structure reorganization aspects, 156–159

macromolecular components involved in, 145–148 mechanotransduction, 144–145 slow mechanical processes, 156 stress-fiber polarization, 156–159 See also Nuclear mechanical properties Chemical inhibitors and cytoskeletal tension modulation, 230–231

isolation method, 91

See also Prestressed nuclear organization

Chromatins, 4–5 binding protein, 9 condensation, 36–38 higher-order compaction, 65–69

decompaction, 222–228, 231–235

immunoprecipitation (ChIP), 36, 49–51 cross-linked chromatin preparation method, 44 immunoprecipitation method, 45–46 materials, 42 methods, 44–46 See also Histones organization, 58–59 in live cells, 59–75 See also Chromatins organization in live cells prestressed nuclear, 222–228, 231–235 types

euchromatin, 36, 58

hetero, 36, 40, 58

361

Subject Index Chromatins organization in live cells cell culture and tagged proteins, 60 chromatin dynamics, 59–60 fluorescence-based live-cell monitoring, 60–61 histones role, 59

embryonic stem (ES) cells, 69–71

FCS, 61–62

FCS results, 62–65

FRAP, 60–61

FRAP results, 64–65

Higher-order, 65–69

nuclear plasticity

cell culture, 69–70

fly lines, 70

results, 70–71

spatio-temporal organization of TFs and gene loci, 71 gene loci labeling strategies, 74 results, 74–75 single particle tracking and analysis, 73–74 TFs visualization by fluorescent UTPs, 72–73 Chromosome territories (CTs), 4–5 Circular ECM patterns, 253–254 See also Embryonic stem cells (ESCs) Circulating tumor cells (CTCs), 83–84 See also Embryonic development Coimmunoprecipitation (co-IP), 111 Compaction (chromatin) laser perturbation of differentially, 231–235 higher-order

cell cultures, 66

fluorescence anisotropy, 66–69

fluorescence polarization, 65–69

See also Decompaction (chromatin) Compression cell connective tissue mechanical cell stimulation method, 185 nuclear mechanical properties, 126, 130–131 stomodeal cell inhibition by photoablation, 305–306 quantitative rescue by ferrofluid injection and magnetic manipulation, 306–308 uniaxial, 130 Condensation, chromatin, 36, 38 Connectins, See Titins Connective tissues, 179 extracellular matrix environment, 181 mechanical cell stimulation methods compression, 185 shear forces, 184–185 static mechanical stimuli—substrate stiffness matters, 187

stretch, 185–186 subcellular mechanical stimulation, 187–188 mechanical loading aspects, 181 mechanical signaling cell contacts, 182–183 systems, 182 See also Cardiac fibrosis Coomassie brilliant blue (CBB) staining, 46–47, 49, 51 Creep curves, 211 in micropipette aspiration, 212–213 See also Nuclear plasticity Cross-linked chromatin preparation, 44 Cytochalasin D, 230 Cytoskeletons active and passive mechanics, 151

dipole polarization model, 154–156

elastic rigidity, 152

semiflexible chain model, 153–154

tensegrity model, 153

See also Cellular mechanosensing

stem cells, cell exerting forces on underlying substrate, 281–284 viscoelasticity, 155

D Deacetylation, histone, 38 Decompaction (chromatin) entropic force measurement, 225 optical trap and nucleus softening, 225–227 See also Compaction (chromatin) Deformation, 130, 136 See also Nuclear mechanical properties Dielectrophoresis, 84–86 Diffusion coefficient

FRAP modeling, 21–23

RICS in living cells, 19–20

FRAP analysis numerical reaction–diffusion models, 23–24 relation of local diffusion coefficient to net flux into ROI, 21–23 Digitonin permeabilization, 104 Dipole model active elastic cellular response to external mechanical stress, 159–164

dipole polarization model, 154–156

See also Cellular mechanosensing

Diseases, nucleus in, 121–124 See also Nuclear mechanical properties

362

Subject Index Dystrophy leuko, See Autosomal dominant leukodystrophy (ADLD) lipo, See Acquired partial lipodystrophy (APL)

E Elastic dipole model, 160–164 Elastic rigidity, 152 Elastic stress dipole polarization model, 154–156

semiflexible chain model, 153–154

tensegrity model, 153

Elasticity, See Viscoelasticity Electron beam lithography, 243 Electrophoresis, See Dielectrophoresis Electrospinning, 246–265 Embryogenesis, See Embryonic development Embryonic development, 296 from genes to shape actomyosin versus adherence surface cortical tensions, 300–301 contractile actomyosin planar and apicobasal polarities patterning, 298–299 molecular and cell biology of in vivo force generation, 298–301

genetics of morphogenesis, 298

mechanical induction aspects, 297

molecular mechanism and physiological function in development, 308–309 Twist gene, 303–304 Twist induction in future anterior gut cells, 305–308 Mechano-genetics active forces in morphogenetic movements (in vivo), measurement of, 301–302 primitive feeding response in morphogenetic invagination, 318 mechanotransduction in control of posttranslational morphogenetic events Fog endocytosis mechanical modulation, 314 in silico physical tools testing, 313–314 mechanical and genetic tools coupling, 313– 314 myo-II apical redistribution mechanical activation hypothesis and drosophila mesoderm invagination genetics, 310–312 myo-II apical redistribution mechanical induction in developmental biology, 315–316 Twist expression mechanical induction into mesoderm, 315

Embryonic stem cells (ESCs), 69–71, 249 circular ECM patterns, 253–254 grating-like structures, 250–251 pillar-like structures, 251–252 well-like structures, 254–256 Emerins, 101–102 immunoprecipitation, 111 solubilization conditions suitable for recovery from isolated HeLa cell nuclei, 111 Endogenous nucleoskeleton proteins localization at inner- versus outer-nuclear membrane, 109–110 by indirect immunofluorescence microscopy, 106 immunolabel cells cultured on glass-bottom dishes, 107–108 nuclei isolation prior to immunolabeling, 108–109 Entropic force measurement, 225 See also Decompaction (chromatin) Epithelial cells, transfection of adherent, 214 Extracellular matrix (ECM) connective tissues, 181

stem cell, 242

F Fabrication techniques (stem cells) nanoimprint lithography (NIL), 244 optical lithography, 243 reactive ion itching, 243 FCS, See Fluorescence correlation spectroscopy (FCS) Ferrofluid injection, 306–308 See also Stomodeal cell compression FFM, See Fluorescence fluctuation microscopy (FFM) Fibroblasts mechanical stretching of growth arrested, 332 mouse embryonic fibroblasts derivation, 326–331 See also Laminopathies Fibrosis, See Cardiac fibrosis Fluorescence anisotropy, 66–69 Fluorescence correlation spectroscopy (FCS), 11–13

chromatin organization in live cells, 62–65

FRAP and, 13

in nuclear cell biology, 28–29

measurements in cell nucleus, 25

Fluorescence fluctuation microscopy (FFM) FCS, 11–13 FRAP, 8–11 RICS, 13

363

Subject Index Fluorescence imaging, 60

chromatin organization (nuclear plasticity), 70

chromatin organization (spatio-temporal

organization of TFs and gene loci)

gene loci labeling strategies in live cells, 74

results, 74–75

single particle tracking and analysis, 74

TFs visualization by fluorescent UTPs, 72–73

Fluorescence In Situ Hybridization (FISH), 74

Fluorescence microscopy

indirect (endogenous nucleoskeleton proteins

localization), 106

immunolabel cells cultured on glass-bottom

dishes, 107–108

nuclei isolation prior to immunolabeling,

108–109

kinetic analysis, 30

Fluorescence redistribution after photobleaching

(FRAP), 8–11

chromatin organization in live cells, 60

FRAP results, 64–65

nuclear plasticity results, 71

FCS and, 13

in nuclear cell biology, 28–29

mathematical modeling

FRAP measurements in cell nucleus and modeling, 25–28

initial conditions, 21

lessons, 29–30

local diffusion coefficient to net flux relation

into ROI, 21–23 numerical integration, initial conditions, and parameter fitting, 24–25 numerical reaction–diffusion models, 23–24 See also Raster image correlation spectroscopy (ICS) Focal adhesions (FA), 279–281 See also Stem cells Fog endocytosis, 314 See also Mechanotransduction FRAP, See Fluorescence redistribution after photobleaching (FRAP)

G Gel electrophoresis, 48

Gene

altered expression in ADLD, 351

loci (spatio-temporal chromatin organization)

gene loci labeling strategies in live cells, 74

results, 74–75

regulation, mechanical manipulation of stem

cells, 284–287

See also Cardiac fibrosis Grating-like structures

embryonic stem cells (ESCs) on, 250–251

mesenchymal stem cells (MSCs) on, 258

neural progenitor cells (NPCs), 269

osteogenic progenitors on, 274–275

H Hematopoietic stem/progenitor cells (HSPCs), 271–274

High CpG Promoters (HCPs), 41

Higher-order chromatin organization

compaction

cell cultures, 66

fluorescence polarization, 65

fluorescence anisotropy, 66– 69

decompaction (prestressed nuclear organization)

entropic force measurement, 225

laser perturbation, 231–235

nuclear isolation from living cells, 222–225

optical trap and nucleus softening, 225–227

photo-bleaching and nuclear swelling

techniques, 227–228

Hippocampal progenitor cells (AHPCs), 269

Histones, 35

acid extraction, 47

acetyl transferases (HATs), 38

acetylation, 38, 40

chromatin organization and, 59

deacetylases (HDACs), 38

deacetylation, 38

methylation, 38–41

modifications, 35, 37

by ChIP, 36, 42, 44–46, 49–51

by immunoblotting/Coomassie staining, 46–51

histone code, 38

rationale behind, 42

phosphorylation, 39

ubiquitination, 40–41

Homeostasis set point

stress vs strain, 167

HSPCs, See Hematopoietic stem/progenitor cells

(HSPCs)

I

Image acquisition, 127

See also Nuclear mechanical properties

Image correlation spectroscopy (ICS) raster (RICS), 13–14 See also Fluorescence fluctuation microscopy (FFM)

364

Subject Index Immunoblotting histone modifications by, 51

acid extraction of histones, 47

material, 46–47

methods, 47–49

See also Immunoprecipitation Immunofluorescence (indirect), 106 cells cultured on glass-bottom dishes, 107–108 nuclei isolation prior to immunolabeling, 108–109 See also Nucleoskeleton proteins Immunolabeling, 107–109 Immunoprecipitation chromatin (ChIP), 36, 49–51, 45–46 nucleoskeletal proteins solubilization aspects lamins, 111–112

nesprins, 112

solubilization conditions suitable for

nucleoskeletal components recovery from isolated HeLa cell nuclei, 111 Immunostaining for lamin-A/C visualization, 214–215 See also Nuclear plasticity Inner nuclear membrane (INM) endogenous proteins localization at outer vs., 109–110 See also Nucleoskeleton proteins Integrins, 279–281 Intranuclear rheology, See Microrheology Isolation, See Nuclear isolation Itching, See Reactive ion etching

L Laminopathies, 98, 339 mechanosignaling, 323–334

equipment, 331–333

informatics-based analysis of, 323–326

mechanical stretching of growth arrested

fibroblast cell, 332–333

methods, 326–331

reagents, 331

mouse embryonic fibroblasts derivation embryonic fibroblast derivation, 326–327 mouse preparation, 326 mouse uteri essential equipment dissection, 326 PMEF cultures passaging and expansion, 328 primary mouse adult fibroblasts and essential equipment preparation, 328–329 primary mouse adult fibroblast and myoblast derivation, 329–331 Lamins ADLD and, 338–339 A-type, 89, 91, 98

B-type, 89, 91, 98 B1 mutations, animal models of, 351–352 immunoprecipitation of nucleoskeletal proteins solubilization, 112–114 solubilization conditions suitable for recovery from isolated HeLa cell nuclei, 111 knockdown with RNA interference immunostaining, 214–215 siRNA complex for transfection, 214 transfection of adherent epithelial cells, 214 western blotting, 215–216 See also Nuclear plasticity See also Actins Laser perturbation differentially compacted chromatin, 231–235 See also Prestressed nuclear organization Latrunculin, 230 Leukodystrophies ADLD, See Autosomal dominant leukodystrophy (ADLD) myelin and, 338–340 LINC complex, 325 nuclear, 103–105 Linker of nucleoskeleton and cytoskeleton, See LINC complex Lipodystrophy, See Acquired partial lipodystrophy (APL) Lithography, nanoimprint (NIL) optical, 243 stem cell fabrication technique, 244 See also Stem cells Loading connective tissues and mechanical, 181 See also Nuclear mechanical properties

M Manipulation microneedle/nuclear mechanical properties

materials, 126

methods, 132–133

results and discussion, 135

single-cell, 80–81

dielectrophoresis technique, 84–86

microfluidics technique, 8–84

micropipette aspiration technique, 87–89

optical-based techniques, 86–87

single-nucleus, 89–91 See also Nuclear isolation Mean square displacement (MSD), 73–74 Mechanical cell stimulation methods, See Connective tissues Mechanical induction

365

Subject Index cardiac fibrosis and gene expression, 189–196 myofibroblasts, 189–190 a-smooth muscle actin expression, 191–194 in embryonic development, 295 in future anterior gut cells, in response to Myo-II-dependent germ-band extension compression, 305–308 mechano-genetics network in perspective of evolution, 318 mechanotransduction molecular mechanism and physiological function in development, 308–309 testing by endogenous forces application, 302–309 Twist gene, 303–308

in tumor development, 316–317

See also Mechanotransduction

Mechanical isolation, 90 See also Nuclear isolation Mechanical manipulation, 284–287 See also Manipulation Mechanical properties (nuclear), See Nuclear mechanical properties Mechanical signaling connective tissues

cell contacts, 182–183

systems, 182

laminopathies, 323–334

equipment, 331–333

informatics-based analysis of, 323–326

mechanical stretching of growth arrested

fibroblast cell, 332–333

methods, 326–331

reagents, 331

Mechanosensing, See Cellular mechanosensing Mechanosignaling, See Mechanical signaling Mechanotransduction in cellular mechanosensing, 145–148 in embryonic development

Fog endocytosis, 314

molecular mechanism and physiological

function in, 308–309 myo-II apical redistribution, 310–312, 313–314 Twist expression into mesoderm, 315 in stem cells, 278–279

See also Mechanical induction

Mesenchymal stem cells (MSCs) complex structures, 261, 264–268 electrospinning, 265 grating-like structures, 258 nanofibrous material forming approaches electrospinning, 261

phase separation, 261

self-assembly, 261

pillar-like structures, 258–259, 261

topology, 256

topology, 257

Methylation, histones, 38–41 Microbead rheology, See Microrheology Microfluidics CTCs detection and, 83–84

Single-cell manipulation technique, 81–84

Microneedle manipulation materials, 126 methods, 132–133 results and discussion, 135 See also Nuclear isolation; Nuclear mechanical properties Micropipette aspiration (MPA) nuclear mechanical properties

materials for, 125

methods, 128–129

results and discussion, 136

nuclear plasticity study experimental method, 208–210 fractional differential models complexity, 212 nuclear creep, 212–213 physical responses mathematics, 210–213 nucleoskeletal proteins, 111–112

single-cell manipulation technique, 87–89

Microrheology active rheology, 126, 133–134 nuclear mechanical properties study materials for, 126

methods for, 133–134

passive rheology, 126, 133–134 Mitosis, 105–106 MPA, See Micropipette aspiration (MPA) MS, See Multiple sclerosis (MS) Multiple sclerosis (MS), 344 See also Autosomal dominant leukodystrophy (ADLD) Myelins, 340–341 See also Leukodystrophies Myofibroblasts mechanical induction of, 189–190 See also Cardiac fibrosis Myosin–actin interaction, See Cellular mechanosensing Myosin-II dependent germ-band extension compression mechanotransduction aspects Fog endocytosis, 314 hypothesis viability, 311 in myo-II apical redistribution, 310–312, 313–314 Twist expression, 315

366

Subject Index Myosin-II dependent germ-band extension compression (cont.) stomodeal cells inhibition by photoablation, 305–306 quantitative rescue by ferrofluid injection and magnetic manipulation, 306–308

Twist expression (mechanical rescue after

extension inhibition), 305–308

biophysical tools (magnetic tweezers), 305–308

genetic tools, 305

See also Actomyosins

N Nanofabrication techniques, 243–246 lithography, 243–245

electron beam, 243

nanoimprint (NIL), 243–244

photolithography, 243

reactice ion etching, 243, 245

See also Stem cells Nanoimprint lithography (NIL), 243–244 Nesprins nuclear, 104–105

immunoprecipitation, 111

solubilization, 112

Neural progenitor cells (NPCs), 268

hippocampal cells, 269

neuronal differentiation on grating-like structures,

269

pillar-like structures, 269

Neural stem cells (NSCs), 268–271

on complex structures, 271

well-like structures, 270–271

Nuclear isolation

bulk isolation protocol, 217–218

isolated nuclei preparation, 125

plasticity aspects, 216–218

prestressed nuclear organization (chromatin), 221

entropic force measurement, 225

isolation from living cells, 222–225

optical trap and nucleus softening, 225–227

photo-bleaching and nuclear swelling, 227–228

prior to immunolabeling, 108–109 single-nucleus, 79–80, 89

chemical isolation, 91

mechanical isolation, 90

mixed (chemical/mechanical) isolation, 91

nuclear isolation, 90

nucleus study application, 91

See also Manipulation; Nuclear mechanical

properties; Nuclear organization

Nuclear mechanical properties, 121–123 AFM

materials, 125

methods, 129–130

results and discussion, 136

analysis, materials for, 127

cell compression

materials, 126

methods, 130–131

image acquisition aspects, 127

intranuclear microrheology

materials, 126

methods, 133–134

isolated nuclei preparation, 125

methods, 127–134

microbead microrheology experiments, 133–134

microinjection of large, fluorescently labeled

molecules into nucleus, 134–135 microneedle manipulation

materials, 126

methods, 132–133

results and discussion, 135

micropipette aspiration (MPA)

materials, 125

methods, 128–129

results and discussion, 136

microrheology

active, 126, 133–135

passive, 126, 133–135

rationale behind study of, 124

results and discussion, 135–137

substrate strain experiments

materials, 126

methods, 131–132

time-lapse videomicroscopy, 134

See also Cellular mechanosensing; Nuclear

plasticity Nuclear organization alteration in ADLD, 348–349 in living cells, See Prestressed nuclear organization

nucleoskeletal proteins roles in, 105–106

See also Nuclear isolation; Nucleoskeleton

proteins Nuclear plasticity, 207

bulk isolation protocol, 217–218

chromatin organization in live cells

cell culture, 69–70

fly lines, 70

results, 70–71

isolation of individual nuclei, 216–218

micropipette aspiration (MPA)

367

Subject Index experimental method, 208–210 physical responses mathematics, 210–213 molecular mechanisms from reengineered nuclei immunostaining for lamin-A/C visualization, 214–215 lamin knockdown with RNAi, 213–216 siRNA complex for transfection, 214 transfection of adherent epithelial cells, 214 western blotting for lamin-A/C knockdown, 215–216 See also Nuclear mechanical properties Nucleoskeleton proteins, 114 endogenous proteins localization at inner– vs, outer-nuclear membrane, 109, 110 by indirect immunofluorescence microscopy, 107–109

immunolabeling, 107–109

immunoprecipitation of (solubilization aspects) lamins, 111–112 nesprins, 112 solubilization conditions suitable for nucleoskeletal proteins recovery from isolated HeLa cell nuclei, 111

internal, 98

lamins, 97–100

micropipette aspiration (MPA) and recoil,

112–114 mitosis and nuclear assembly roles, 105–106 nonlamin components actins, 100–102

LINC complexes, 103–105

nesprin, 104–105

spectrins, 102–103

titins, 103

peripheral, 98 See also Lamins Nucleus in disease, See Nuclear mechanical properties

O Optical lithography, 243 Optical stretcher, 86–87 Optical traps, 225–227 See also Prestressed nuclear organization Optical-based techniques cell manipulation technique, 86–87 Osteogenic progenitors on complex structures, 277 grating-like structures, 274–275 pillar-like structures, 275–276 well-like structures, 276

Outer nuclear membrane (ONM) endogenous proteins localization at inner vs., 109–110 See also Nucleoskeleton proteins

P Pelizaeus–Merzbacher disease (PMD), 341 See also Autosomal dominant leukodystrophy (ADLD) Phosphorylation, histone, 39 Photoablation (stomodeal cell compression inhibition), 305–306 Photo-bleaching fluorescence redistribution after (FRAP), 8–11 prestressed nuclear chromatin organization and, 227–228 Photolithography, 243 Pillar-like structures embryonic stem cells (ESCs) on, 251–252 mesenchymal stem cells (MSCs) on, 258–259, 261 neural progenitor cells (NPCs), 269 osteogenic progenitors on, 275–276 Plasticity, See Nuclear plasticity PMD, See Pelizaeus–Merzbacher disease (PMD) Point spread function (PSF), 15–16 Polarization dipole model, 154–156 stress-fiber, cytoskeletal structure reorganization, 156–159 See also Cellular mechanosensing Polyacrylamide gel electrophoresis, 48 Prestressed nuclear organization, 221 isolated nucleus mechanics (chromatin organization) entropic force measurement, 225 isolation from living cells and higher-order chromatin, 222–225 optical trap and nucleus softening, 225–227 photo-bleaching and nuclear swelling, 227–228 cellular context, 228–235 chemical inhibitors and cytoskeletal tension modulation, 230–231 laser perturbation of differentially compacted chromatin, 231–235 See also Nuclear mechanical properties Progenitor cells hematopoietic stem/progenitor cells (HSPCs), 271–274

neural (NPCs), 268–269

osteogenic, 274–277

Promyelocytic leukemia (PML), 6

368

Subject Index

R Raster image correlation spectroscopy (RICS), 13–14 benefits, 29 in living cells, diffusion coefficients and concentrations determination, 19–20 measurements

general considerations, 17–19

in cell nucleus, 28

theory, 15–17

See also Fluorescence redistribution after

photobleaching (FRAP) Reactive ion etching, 243 Region of interest (ROI) FRAP modeling, 20–25 RICS modeling and, 19–20 Relaxation time, cellular, 164–167 See also Cellular mechanosensing Rheology active

materials for, 126

methods, 133–134

results and discussion, 135

microbead microrheology experiments, 133–134 passive

materials for, 126

methods, 133–134

results and discussion, 135

See also Nuclear mechanical properties RICS, See Raster image correlation spectroscopy (RICS) RNA interference (RNAi), 213–214 lamin knockdown with

adherent epithelial cells transfection, 214

siRNA complex for transfection, 214

See also Nuclear plasticity ROI, See Region of interest (ROI)

S Semiflexible chain model, cytoskeleton, 153–154 See also Cellular mechanosensing Shear forces connective tissue cell stimulation method, 184–185 See also Stimulation (mechanical cell) Signal-to-noise ratio (SNR), 17–20 Single-cells, 79 future implications of study of, 92 manipulation, 80–81 dielectrophoresis, 84–86

microfluidics, 81–84

micropipette aspiration, 87–89

optical-based techniques, 86–87

Single-nucleus, 79 future implications of study of, 92 manipulation and isolation, 79–80, 89 chemical isolation method, 91 mechanical isolation method, 90 mixed (chemical and mechanical) isolation method, 91 nuclear isolation method, 90 nucleus study application, 91–92 Sodium dodecyl sulfate polyacrylamide gel electrophoresis, 48 Softening (nuclear), 225–227 See also Optical traps Solubilization lamins, 111–112 nesprins, 112 See also Immunoprecipitation Somatic cell nuclear transfer (SCNT), 208 Speckles, 6 Spectrins, 102–105 aI, 102

aII, 102

repeat proteins (nesprins), 104–105, 112

Spindle matrix, 105–106 Stem cells, 241 extracellular matrix (ECM), 242 reception to substrate topology, 247–249 embryonic, 249–256 hematopoietic stem/progenitor cells (HSPCs), 271–274 mesenchymal stem cells (MSCs), 256–259, 261, 264–268 neural stem cells/neural progenitor cells, 268–271 osteogenic progenitors, 274–277 shape and biological processes regulation, 277 cytoskeleton, 281–282, 284 integrins and focal adhesions, 279–281 mechanical manipulation of gene regulation, 284–287

mechanotransduction, 278–279

topological patterns chemical topology/protein patterning, 246–247 geometry, 247 nanofabrication techniques, 243–246 physical topography/surface texture, 242 Stiffness, 135–136 connective tissue, 187 nuclear, 131 See also Nuclear mechanical properties Stimulation (mechanical cell) compression, 185 shear forces, 184–185

369

Subject Index static mechanical stimuli—substrate stiffness matters, 187

stretch, 185–186

subcellular, 187–188

Stomodeal cell compression inhibition by photoablation, 305–306 quantitative rescue by ferrofluid injection and magnetic manipulation, 306–308 Strain biaxial, 131 cytoskeletal, 132–133 nuclear, 132–133 substrate (nuclear mechanical properties) materials for, 126 methods, 131–132

uniaxial, 131

vs, stress, 167

See also Cellular mechanosensing; Nuclear

mechanical properties Stress cellular mechanosensing cellular response to external mechanical stress, 159–167 stress-fiber polarization, 156–159 elastic

dipole polarization model, 154–156

semiflexible chain model, 153–154

tensegrity model, 153

vs, strain, 167

See also Prestressed nuclear organization

Stretching connective tissue, 185, 186 growth arrested fibroblast cell, 332–333 Subcellular mechanical stimulation, 187–188 Substrate strain, See Strain Swelling, 227–228

Titins, 103 Transcription factories (TFs), 59 chromatin spatio-temporal organization, 71 results, 74–75 single particle tracking and analysis, 73–74 visualization of transcription factories by fluorescent UTPs, 72–73 Transfection cardiac fibroblast, 195 cell culture and transfection, 14–15 nuclear plasticity aspects adherent epithelial cells, 214

siRNA complex, 214

Tumor development mechanical induction in, 316–317 See also Embryonic development Twist gene, 300 master mechanosensitive gene in early drosophila embryos, 303–304 mechanical induction in future anterior gut cells, in response to Myo-II-dependent germ-band extension compression, 305–308 into mesoderm, 315 mechanical rescue after mechanical inhibition of germ-band extension by two-photon local ablation, 305–308

T

W

Tensegrity model, cytoskeleton, 153 See also Cellular mechanosensing Tension, 230–231 Time-lapse imaging fluorescence imaging, 70–71

videomicroscopy, 134

Well-like structures embryonic stem cells (ESCs) on, 254–256 neural stem cells (NSCs), 270–271 osteogenic progenitors on, 276 Western blotting, 215–216 See also Nuclear plasticity

U Ubiquitination, histone, 40–41

V Viscoelasticity, 155