THE ROLE OF THE CYTOSKELETON IN PLURIPOTENT STEM CELL DIFFERENTIATION
AN ABSTRACT SUBMITTED ON THE 21ST DAY OF AUGUST, 2013 TO THE DEPARTMENT OF BIOMEDICAL ENGINEERING OF THE SCHOOL OF SCIENCE AND ENGINEERING AT TULANE UNIVERSITY IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY BY
______EMMA T. PINEDA FORTIN
APPROVED: ______TABASSUM AHSAN, Ph.D. (COMMITTEE CHAIR)
______WALTER LEE MURFEE, Ph.D.
______DAMIR KHISMATULLIN, Ph.D.
ABSTRACT
An understanding of the pathways responsible for differentiation in pluripotent stem cells (PSCs) would accelerate their translation to medical therapies. Specifically, studies that identify criteria for the better design of experiments targeting certain phenotypes would allow for the generation of cell sources adequate for transplantation. In this dissertation, we aimed at elucidating the role of the cytoskeleton in the spontaneous differentiation of PSCs in two dimensional (2D) and three dimensional (3D) microenvironments. First, we quantified the expression of the cytoskeleton in ESCs, iPSCs, and the iPSC source phenotype, showing that there were indeed differences in the expression of microfilaments and certain intermediate filaments among all three phenotypes.
Next, we found that there were inherent differences in ESC differentiation when cultured in 2D and 3D microenvironments. Lastly, alterations in the cytoskeleton were found to decrease mesodermal differentiation in 3D culture, while increase both mesodermal and endodermal differentiation in 2D culture. Taken together, we identified the cytoskeleton as a regulator of differentiation to the mesodermal and endodermal lineages in both 2D and 3D culture.
THE ROLE OF THE CYTOSKELETON IN PLURIPOTENT STEM CELL DIFFERENTIATION
A DISSERTATION SUBMITTED ON THE 21ST DAY OF AUGUST, 2013 TO THE DEPARTMENT OF BIOMEDICAL ENGINEERING OF THE SCHOOL OF SCIENCE AND ENGINEERING AT TULANE UNIVERSITY IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY BY
______EMMA T. PINEDA FORTIN
APPROVED: ______TABASSUM AHSAN, Ph.D. (COMMITTEE CHAIR)
______WALTER LEE MURFEE, Ph.D.
______DAMIR KHISMATULLIN, Ph.D.
ACKNOWLEDGEMENTS
I am grateful towards the Department of Biomedical Engineering at Tulane
University for allowing me to pursue my scientific career. I would especially like to thank my advisor, Dr. Taby Ahsan for helping me grow and become a better scientist and thinker. I also want to thank all of the members of the STEM cell lab through the years, undergraduates, high school students, and graduates. Thank you in particular, to Dr. Kristen Lynch and Dr. Russell Wolfe for all the good times and laughter. I would like to thank my committee members Dr. Lee Murfee and
Dr. Damir Khismatullin for their support, as well as the entire BME faculty. A big thanks to the BME staff, Lorrie McGinley, Megan Ohar, John Sullivan, and Cindy
Stewart who made everything work smoothly. I would like to thank my parents,
Angel & Guadalupe, my brother Angel, sisters Denise and Marcela, and my nephew Santiago for their love and constant support. Special thanks to Abuelita for her prayers and candles. Lastly, I thank my boyfriend Brendan who always believes in me!
ii
TABLE OF CONTENTS
Chapter 1: Introduction ...... 1
Chapter 2: Background ...... 7
2.1 Defining Pluripotency and iPSCs...... 7
2.2 Germ Lineage Commitment in vivo: Gastrulation ...... 11
2.2.1 Physical Force Regulation in vivo ...... 13
2.3 In vitro Differentiation Models ...... 13
2.4 Directed Differentiation ...... 16
2.5 The Microenvironment and Differentiation ...... 17
2.6 Mechanotransduction ...... 18
2.6.1 Cytoskeleton: A Continuous Mechanical Link to the Nucleus ...... 19
2.7 Actomyosin Contractility in Pluripotency and Differentiation ...... 23
Chapter 3: Cytoskeletal Protein Characterization of Induced Pluripotent Stem Cells ...... 28
3.1 Abstract ...... 28
3.2 Introduction ...... 29
3.3 Materials & Methods ...... 31
3.3.1 Pluripotent Cell Expansion ...... 31
3.3.2 Embryoid Body Differentiation ...... 32
3.3.3 Phase Microscopy ...... 32
3.3.4 Gene Expression...... 32
3.3.5 Protein Expression ...... 33
iii
3.3.6 Statistical Analysis ...... 34
3.4 Results ...... 34
3.4.1 Pluripotent stem cell morphology differed in ESCs and iPSCs ...... 34
3.4.2 Transcription factors of pluripotency were expressed similarly or at lower levels in iPSCs ...... 36
3.4.3 iPSCs expressed microfilaments and integrin α5 at a higher level than ESCs ...... 38
3.4.4 Embryoid Bodies from iPSCs and ESCs had similar levels of microfilaments, intermediate filaments, and microtubules at Day 6 ...... 42
3.5 Discussion ...... 44
Chapter 4: Differentiation Patterns of Embryonic Stem Cells in Two versus Three Dimensional Culture ...... 49
4.1 Abstract ...... 49
4.2 Introduction ...... 50
4.3 Materials and Methods ...... 52
4.3.1 Expansion of Mouse Embryonic Stem Cells ...... 52
4.3.2 Two and Three Dimensional Differentiation Systems...... 53
4.3.3 Gene Expression Analysis ...... 54
4.3.4 Statistical Analysis ...... 55
4.4 Results ...... 57
4.4.1 Cell Growth in Differentiation Systems ...... 57
4.4.2 Pluripotency and Germ Lineage Differentiation ...... 58
4.4.3 ESC Differentiation Patterns ...... 60
4.4.4 Functional Gene Groupings ...... 63
4.4.5 Matrix Remodeling ...... 67
4.5 Discussion ...... 70
4.6 Acknowledgements ...... 75
iv
Chapter 5: Actin and myosin II modulate mesodermal commitment of embryonic stem cells in 3D ...... 79
5.1 Abstract ...... 79
5.2 Introduction ...... 80
5.3 Materials and Methods ...... 81
5.3.1 Embryoid Body Culture ...... 81
5.3.2 Microscopy ...... 82
5.3.3 Gene Expression...... 82
5.3.4 Protein Expression ...... 83
5.3.5 Statistical Analysis ...... 83
5.4. Results ...... 84
5.4.1 CYTO-D and BLEBB altered cell and EB size ...... 84
5.4.2 Inhibitor treatment decreased commitment to the mesodermal lineage ...... 88
5.4.3 CYTO-D and BLEBB modulated early mesodermal differentiation ..... 91
5.5 Discussion ...... 95
Chapter 6: Cytochalasin-D treatment increased endodermal and mesodermal differentiation of stem cells on E-cadherin- and Fibronectin- coated surfaces ...... 105
6.1 Abstract ...... 105
6.2 Introduction ...... 106
6.3 Materials & Methods ...... 107
6.3.1 Embryonic Stem Cell Culture ...... 107
6.3.2 Gene Expression...... 108
6.3.3 Protein Expression ...... 109
6.3.4 Statistical Analysis ...... 109
6.4 Results ...... 110
v
6.4.1 Fibronectin and ECAD were expressed in embryoid bodies ...... 110
6.4.2 FN and ECAD coatings allowed for the localization of distinct focal complex proteins ...... 112
6.4.3 Early mesoderm and endoderm markers were similarly expressed in cells on FN and ECAD ...... 115
6.4.4 CYTO-D increased the expression of mesoderm and endoderm markers on both substrates ...... 117
6.4.5 Markers of endothelial and hematopoietic progenitor cells were upregulated by CYTO-D...... 119
6.5 Discussion ...... 121
Chapter 7: Discussion and Future Studies ...... 130
Appendix ...... 137
Appendix A: Effect of Y27632 on mesodermal and endodermal differentiation in 2D Culture ...... 137
Appendix B: Effects of RA and BMP4 on Mesodermal Differentiation ...... 140
Appendix C: Effect of tensile strain on ESC Cytoskeletal and Germ lineage Marker Expression ...... 147
Appendix D: Tail-tip iPSC Expansion & Pluripotency ...... 153
Appendix E: Effects of Tubulin Inhibition on EBs ...... 155
Appendix F: Cytoskeletal Inhibition in Static vs. Shear EBs ...... 157
Appendix G: SEM Protocol - Chemical Fixation & Dehydration ...... 160
Appendix H: SEM Protocol - Critical Point Drying & Gold Sputtering ...... 162
Bibliography ...... 164
Biography ...... 185
vi
LIST OF TABLES
Table 2.1 Cytoskeletal Elements & their Functions………………………………..21
Supplemental Table 4.1 Forward and reverse primers of genes analyzed using standard real-time PCR……………………………………………………………….76
Supplemental Table 4.2 Genes analyzed using the Mouse Embryonic Stem Cell RT2 Profiler ™ PCR Array……………………………………………………………77
vii
LIST OF FIGURES
Figure 2.1 Embryonic Stem Cell Pluripotency and Differentiation...... 9 Figure 2.2 Somatic cell reprogramming by defined factors...... 10 Figure 2.3 Germ layer determination during gastrulation at Day 6.5 of mouse development...... 12 Figure 2.4 Culture models for stem cell differentiation...... 15 Figure 2.5 Cytoskeletal components of the cell and their cell-cell and cell-ECM attachments...... 22 Figure 2.6 Rho-ROCK & Rho-mDia pathways...... 24 Figure 3.1 ESCs and iPSCs were morphologically similar to each other and distinct from MEFs...... 35 Figure 3.2 iPSCs expressed pluripotency markers at comparable or lower levels than ESCs...... 37 Figure 3.3 Cytoskeletal proteins were more highly expressed in iPSCs than ESCs...... 39 Figure 3.4 ECAD was similarly expressed in both cell types, while integrin α5 was more highly expressed in iPSCs...... 41 Figure 3.5 EBs generated from iPSCs and ESCs similarly expressed microfilaments, intermediate filaments, and adhesion proteins...... 43 Figure 4.1 Initial cell distribution and structures supported by the 2D and 3D models...... 56 Figure 4.2 Gene expression of pluripotent and lineage markers for SLIDE, GEL, and EB samples...... 59 Figure 4.3 Hierarchical clustering and heat maps based on gene expression from a PCR array...... 62 Figure 4.4 Relative gene expression of functional groups at Day 4 and 12...... 66 Figure 4.5 Collagen gel compaction from Day 8 to Day 12 of culture...... 68 Figure 4.6 Gene expression of cytoskeletal genes for SLIDE, GEL, and EB samples...... 69 Figure 5.1 CYTO-D and BLEBB altered EB size and ECAD expression...... 86 Figure 5.2 CYTO-D and BLEBB produced a rounded cell shape...... 87 Figure 5.3 Expression and distribution of BRACHY-T were altered with CYTO-D and BLEBB...... 90
viii
Figure 5.4 CYTO-D and BLEBB reduced early mesoderm marker expression at Day 7, not Day 4...... 93 Figure 5.5 Y27632 increased early mesodermal differentiation at Day 4, while decreased it at Day 7...... 94 Supplemental Figure 5.1 CYTO-D caused a rounded cell shape and an increase in size...... 101 Supplemental Figure 5.2 BLEBB did not alter mesodermal protein expression...... 102 Supplemental Figure 5.3 Endoderm and ectoderm markers were altered by CYTO-D and BLEBB...... 103 Supplemental Figure 5.4 RUNX1 gene expression was increased with CYTO-D treatment...... 104 Figure 6.1 Embryoid bodies expressed ECAD and FN...... 111 Figure 6.2 ESCs attached through distinct adhesion proteins on FN and ECAD...... 114 Figure 6.3 Cells cultured on FN or ECAD similarly expressed early mesodermal and endodermal markers...... 116 Figure 6.4 CYTO-D increased mesodermal and endodermal markers on FN and ECAD...... 118 Figure 6.5 CYTO-D upregulated the expression of endothelial and hematopoietic markers...... 120 Supplemental Figure 6.1 No change was detected in pluripotency markers when seeded on FN or ECAD substrates upon the removal of LIF...... 127 Supplemental Figure 6.2 By Day 4 of culture, ECAD samples expressed paxillin at focal adhesions...... 128 Supplemental Figure 6.3 Cells treated with CYTO-D were a heterogeneous population, some balled up and some spread...... 129 Figure A.1 Effects of Y27632 in 2D culture of ESCs...... 138 Figure B.1 Effects of RA on EB morphology and differentiation towards the mesoderm...... 141 Figure B.2 Effects of RA on EB morphology and differentiation towards the mesoderm when applied in early and late treatment windows...... 142 Figure B.3 Effects of BMP4 on EB morphology and differentiation towards the mesoderm...... 144 Figure B.4 Effects of BMP4 on EB morphology and differentiation towards the mesoderm when applied in early and late treatment windows...... 145
ix
Figure C.1 Gross images of ESCs seeded in collagen type I gels...... 148 Figure C.2 Gene expression analysis of ESCs grown in collagen type I gels and exposed to 1 day of strain...... 149 Figure C.3 Application of equibiaxial strain to ESCs...... 151 Figure D.1 Mouse primary iPSCs express NANOG, while have a low number of dome-like colonies and lower SOX2 expression as compared to ESCs...... 154 Figure E.1 Effects of Nocodazole on EB culture...... 156 Figure F.1 ESCs differentiated as embryoid bodies and treated with Y27632 and BLEBB were larger...... 158 Figure F.2 Treatment with inhibitors decreased expression of pluripotency markers...... 159
x
1
CHAPTER 1: INTRODUCTION
Regenerative medicine and tissue engineering research have the potential to address many of the leading causes of death including acute and chronic diseases such as heart disease, stroke, cancer, alzheimer’s and end-stage organ failure [1]. Organ failure alone accounted for more than 600,000 deaths in 2008
[1]. At present, allogeneic transplantation is the definitive treatment for end-stage organ failure, yet it has many drawbacks including donor availability, recipient compatibility and long term immunosuppression [2]. Current research indicates that tissue engineering approaches to resolve this issue will utilize a combination of cells, matrices, and factors to create complete functional tissues or organs. An adequate source of cells would need to be readily available in sufficient quantities, non-immunogenic, non-carcinogenic/tumurogenic, and present no ethical dilemmas. Possible cell sources for these applications range from unipotent cells (allogeneic or autologous) that are already at the desired terminal stage for therapeutic use and have been treated for immunogenicity, to cells that have the potential to become any other cell [3]. Use of terminal cell types is limited due to their low yields after recovery and their inability to proliferate for prolonged periods of time in vitro while remaining functional. Alternatively, embryonic stem cells (ESCs) have the ability to become any cell in
2 the adult body and self-renew. These cells can be isolated from the inner cell mass of a blastocyst stage embryo in several species including mouse [4, 5] and human [6] and can be maintained pluripotent with defined culture conditions in vitro. Studies with pluripotent cells of human origin have ensued ethical concerns, as a result many more have been done with an alternative mammalian model, mouse. Recently, the ethical dilemma of human embryonic stem cell use from a developing embryo has been circumvented with the creation of induced pluripotent stem cells from somatic cells that have the same self-renewal capabilities and differentiation potential as isolated ESCs [7]. An understanding of the pathways responsible for differentiation in pluripotent stem cells (PSCs) in general would accelerate their translation to medical therapies. Specifically, studies that identify criteria for the better design of experiments targeting certain phenotypes would allow for the generation of cell sources adequate for transplantation. This dissertation addresses the need to better understand the role of the cytoskeleton in PSC differentiation and its potential role in transducing signals from the physical microenvironment (including two and three dimensional culture).
The overall objective of this dissertation was to determine the expression of cytoskeletal elements in PSCs as well as to identify the role of the cytoskeleton in differentiation. The cytoskeleton of terminal cell types such as those from the cardiac and musculoskeletal systems has a highly dynamic role in their cellular functions. Hence, perhaps the cytoskeleton may play a role in the
3 differentiation of cells of the mesodermal linage in particular. Later on we focused on mesodermal and endodermal differentiation due to their proximity in their lineage determination in vivo. Our overarching hypothesis was that cytoskeletal expression in PSCs would be lower than that of adherent phenotypes such as fibroblasts, and that perturbation of the cytoskeleton either by culture in different microenvironments or with the use of protein inhibitors of actin and myosin II would alter differentiation towards the three lineages. To evaluate this hypothesis, we proposed the following four aims.
AIM 1: Determine the relative expression of cytoskeletal proteins in iPSCs, their source cell phenotype, and ESCs.
Somatic cells, such as mouse embryonic fibroblasts (MEFs), can be reprogrammed genetically to create induced pluripotent stem cells (iPSCs). iPSCs allow for the generation of autologous cell phenotypes for use in therapeutics. Although iPSCs hold great promise as a pluripotent cell source, they differ from ESCs in their abilities to differentiate towards specific cell types.
In the first study, we characterized the cytoskeleton in iPSCs compared to their cell source (MEFs) as well as ESCs in an attempt to identify inherent differences that may lead to differential capacities. The expression of microfilaments, and intermediate filaments (with the exception of KRT8) in iPSCs was found to be higher than that of ESCs and closer to MEF levels, yet upon 6 days of differentiation both PSC expression levels had dropped to similar levels. These results suggest that although iPSCs are morphologically similar to ESCs, their cytoskeletal state may not have been reprogrammed to baseline levels and that
4 both pluripotent stem cells express certain cytoskeletal proteins, namely microfilaments, at lower levels than MEFs. This work is addressed in Chapter 3.
AIM 2: Characterize overall differentiation and cytoskeletal protein expression of ESCs in 2D and 3D.
In these studies, we screened for overall differentiation of ESCs cultured in 2D and 3D microenvironments and implicated the cytoskeleton as a possible mediator of altered responses. We found that culture in monolayer, embedded in gels, and in suspension induced different differentiation patterns with increased time. Moreover, we found that culture in 3D, either in gels or in suspension, resulted in higher differentiation towards mesodermal phenotypes. Although, culture in 2D featured much higher expression of cytoskeletal proteins, often found in terminal mesodermal phenotypes. These results suggest that the physical microenvironment, specifically 2D and 3D culture, must be considered in the design of experiments for directed differentiation. This work is published as
Pineda et al, Cells Tissues Organs 2013;197(5):399-410 and is addressed in
Chapter 4.
AIM3: Identify the effect of actin polymers and actomyosin contractions on ESC differentiation in 3D.
Physical forces are known to play a role in mammalian morphogenesis, yet specific cues regulating differentiation to each lineage are not well understood. Here, we cultured ESCs as embryoid bodies (EBs) in the presence of inhibitors of actin polymerization and actomyosin contractions to elucidate their
5 role in mesodermal lineage commitment and differentiation. Inhibitors of actin polymerization and actomyosin contractions were found to decrease the expression of markers of mesodermal commitment when applied at later stages of differentiation. At earlier time points, only functional actin polymers were found to regulate mesodermal commitment. These results suggest that functional actin polymers, in part responsible for cell stiffness, are required at early and late stages of mesodermal commitment, while actomyosin contractions may be required only at later stages. This work is addressed in Chapter 5.
AIM4: Determine the effect of actin polymers on ESC differentiation in 2D.
Small molecule inhibitors may provide a facile method of directed differentiation. Use of CYTO-D, an inhibitor of actin polymerization, has been shown to promote adipogenic phenotypes by others, while decreasing mesodermal phenotypes in three dimensions. Here, we examined the role of
CYTO-D in differentiation towards mesodermal and endodermal phenotypes and found an opposite result in monolayer culture as in 3D. This effect was not shown to be dependent on the type of transmembrane protein used to attach to the physical microenvironment. These studies suggest that in 2D, actin depolymerization increased both mesodermal and endodermal differentiation.
Moreover, perhaps differences in response to actin depolymerization in 2D and
3D were not due to the adhesion protein used to attach to the physical microenvironment. This work is addressed in Chapter 6.
6
Taken together, this dissertation attempted to understand the role of the physical microenvironment on PSC differentiation. Aim 1 focused on the characterization of the cytoskeletal machinery in ESCs and iPSCs.
Subsequently, the effects of 2D and 3D culture configurations on early differentiation were quantified on their own (Aim 2) as well as probed by their response to cytoskeletal alterations (Aims 3&4). In Aim 2, we found inherent differences in differentiation by culture condition without any guidance cues. Aims
3 and 4 further supported that finding by producing an inverse response to the same treatment. Although adherent culture paradigms are predominantly used to identify relevant processes or cues to direct differentiation, our studies suggest that 2D results may not map to 3D models more pertinent to scale-up processes for use in medical therapeutics.
These studies increase the understanding of PSC differentiation in response to alterations in both the physical microenvironment and the physical properties of the cell. With this knowledge, we are able to improve the design of experiments for directed differentiation to specific phenotypes. An improved design of experiments will hopefully allow for adequate populations of stem cells for use in medical therapies.
7
CHAPTER 2: BACKGROUND
2.1 DEFINING PLURIPOTENCY AND IPSCS
The definition of cell pluripotency has evolved since the derivation of embryonic stem cells from the mouse blastocyst in 1981. Currently, pluripotent cells may be defined by their ability to generate all cell types of the adult organism as well as those of the trophectoderm (extraembryonic tissue) [8]. In addition, stem cells can be characterized by their expression of transcription factors required for self-renewal and maintenance of pluripotency: OCT4, SOX2, and NANOG (Figure 2.1). These factors work in unison to regulate gene expression in pluripotent cells, for example, OCT4 can heterodimerize with SOX2 to affect the expression of several genes in ESCs [9]. In addition to genomics, epigenetics also play a role in maintaining the pluripotent state. Epigenetic control is in part mediated by posttranslational modification of histone proteins, which in turn modulate chromatin structure [10, 11]. Therefore, in order to consider a cell as pluripotent, it must meet genetic, epigenetic, and functional requirements established by the ESC model.
The Yamanaka lab derived iPSCs in 2006 by the ectopic expression of transcription factors such as OCT4, SOX2, KLF4, and C-Myc in terminally differentiated cells of different origins [12, 13]. These factors progressively
8 activate and de-activate genes that cohesively aid in the de-differentiation of terminal cell types (Figure 2.2). The resultant reprogrammed cells are similar to
ESCs in morphology, gene expression, and their capacity to contribute to the three germ lineages when implanted into a developing embryo, as well as to form teratomas in immune-deficient mice. Due to their sourcing from adult tissues, while having the same potential as embryonically derived pluripotent cells, iPSCs have the potential to become an autologous cell source for a wide range of human medical therapies.
9
Figure 2.1 Embryonic Stem Cell Pluripotency and Differentiation. Pluripotent cells can be identified by transcription factor expression, which is lost with differentiation to the three germ lineages.
10
Figure 2.2 Somatic cell reprogramming by defined factors. Terminally differentiated cells may be reprogrammed to a pluripotent state through forced expression of pluripotency transcription factors
11
2.2 GERM LINEAGE COMMITMENT IN VIVO: GASTRULATION From fertilization to the creation of a full organism, cells undergo large transformations guided by spatiotemporally regulated genetic and epigenetic gradients [14]. In mouse development, the 3.5 day old blastocyst consists of a cavity and two distinct layers: the inner cell mass, from which embryonic stem cells are derived, and the trophectoderm. Shortly after implantation onto the uterine wall, the proximal-distal axis is formed through the reciprocal signaling of growth factors including Nodal, BMP, Wnt, and β-catenin [15]. On Day 7 of development, the primitive streak forms in the posterior epiblast and initiates the process of gastrulation; this gives rise to the three germ layers: endoderm, mesoderm, and ectoderm (Figure 2.3) [16]. Mesendodermal cells actively migrate through the primitive streak undergoing epithelial-to-mesenchymal transition (EMT), a process by which cells lose expression of E-Cadherin
(ECAD), simultaneously upregulating N-cadherin and adopting a mesenchymal phenotype. As cells migrate through the primitive streak, they are either incorporated as definitive endoderm into the pre-existing visceral-endoderm layer or emerge as mesoderm (reviewed in [17]). Cells from the definitive endoderm will become amongst others, tissues of the gastrointestinal and respiratory tracts, while those of the mesoderm will become tissues of the musculoskeletal and cardiac systems. The third germ layer, the ectoderm, will become skin and neural tissues [18].
12
Figure 2.3 Germ layer determination during gastrulation at Day 6.5 of mouse development. During gastrulation, cells migrate into the primitive streak and emerge as endoderm or mesoderm.
13
2.2.1 Physical Force Regulation in vivo
Embryonic development occurs in a mechanically dynamic environment in part caused by large cell movements and continuous matrix deposition. Recent studies of whole embryos have elucidated a feedback mechanism between mechanical forces and genome activity during development. Mechanical stimulation of Drosophila embryos induced expression of Twist, a regulator of the onset of gastrulation and EMT [19], while bending of early Xenopus embryos showed a spatially sensitive genetic response to the deformation, by which spatial arrangements of neural and mesodermal phenotypes were lost upon bending [20]. Numerous chemical signals have been elucidated using in vivo models and techniques such as chimaera formation, real-time imaging, and knock out studies. Yet, it is difficult to understand the physical forces governing germ lineage commitment in a three-dimensional, multi-layered model. Hence, in vitro studies provide a better-controlled system for assessing the role of the physical microenvironment on stem cell differentiation.
2.3 IN VITRO DIFFERENTIATION MODELS Embryonic stem cells have been differentiated to cells of all three germ lineages in a variety of culture models, including adherent monolayer surface [21,
22], embedded in protein gels [23-25] and suspension culture [26, 27] (Figure
2.4). ESCs and iPSCs seeded as single cells on non-adherent dishes will spontaneously aggregate into spheroids termed embryoid bodies (EBs). EB cultures have been shown to upregulate gene markers of the three germ
14 lineages, although their spatial arrangements are not paralleled to in vivo development [27].
Alternatively, pluripotent cells can be differentiated in two dimensions, where differentiation kinetics and potential are not the same as in EB culture [28].
Culture in a monolayer permits uniform growth factor delivery and application of well-defined external physical force. Unfortunately, unlike the EB model which allows for the generation of millions of cells, monolayer culture results in a lower number of cells that is limited by culture surface area.
15
Figure 2.4 Culture models for stem cell differentiation. Stem cells can be differentiated in monolayer, in suspension, or embedded in gels.
16
Other systems, such as protein or polymeric gels and microcarrier culture, combine two and three dimensional characteristics resulting in certain aspects of the EB model, while retaining the enhanced control of 2D culture. Cells seeded into polymeric or protein-based gels are able to form 3D structures. Furthermore, polymeric gels provide a scaffold for the tethering of proteins [24, 29, 30].
Microcarrier beads are a model transferred from vaccine development and have recently been used in stem cell expansion and differentiation. Cells are seeded onto protein-coated bead surfaces and the beads themselves are then grown in suspension. Human ESCs grown on microcarrier beads have been shown to have increased aggregate size and efficiency of cardiomyocyte differentiation compared to the EB model [31].
2.4 DIRECTED DIFFERENTIATION
For the translation of stem cell therapeutics to regenerative medicine and tissue engineering applications, a threshold number of cells (on the order of 109) of a desired phenotype is considered necessary [32, 33]. One way to create such a population is through the use of growth factors that can aid in directed differentiation. Growth factors supplemented in the medium have been used to efficiently guide differentiation towards bone [34] and myocardial [35] cells amongst others. Unfortunately, their use in scale up bioreactors for regenerative medicine applications can become cost-prohibitive. Secondary steps have also been used in the process of directed differentiation, where after some time in culture, progenitor cells are selected from the population using antibody-based techniques. Much investigation is done in finding adequate protein markers to
17 identify such progenitor phenotypes [36]. Yet, the use of antibodies is also not cost effective at a large scale. Alternatively, alterations in the stem cell microenvironment can provide a more cost-efficient way to generate an enhanced population of cells of a specific phenotype for use in medical therapies.
2.5 THE MICROENVIRONMENT AND DIFFERENTIATION
Alternative methods of directing differentiation include altering the physical microenvironment, such as protein presentation, cell-cell contacts, rigidity of the substrate, as well as the application of physical force [21, 24, 37, 38]. Human mesenchymal stem cells (MSCs) spread over a large area become osteocytes, while those restricted to a small area became adipocytes [39]. Application of uniaxial cyclic strain to MSCs induced an osteogenic phenotype in three dimensions, and a smooth muscle cell phenotype in two dimensions [37, 40].
Similar to MSCs, embryonic stem cells adjusted their differentiation patterns as a result of an altered physical microenvironment. Pointed shear stress (through magnetic twisting cytometry) on single ESCs promoted cell differentiation by
OCT3/4 loss [41], while whole population shear stress application induced an endothelial-like phenotype [42]. In addition, culturing ESCs in a more compliant substrate, rather than traditional tissue culture plastic promoted self-renewal and pluripotency, even with the removal of LIF [43]. Although, numerous studies detail genetic and functional outcomes of alterations in the physical microenvironment, a mechanistic understanding of how these physical changes are converted to chemical changes at the nucleus is poorly understood.
18
2.6 MECHANOTRANSDUCTION
To understand the mechanisms by which a physical input or an alteration of the physical microenvironment can induce changes in cell function, such as differentiation and morphogenesis, one can focus on the possible ways a cell translates physical signals. Mechanotransduction can be defined as the conversion of a physical input into a chemical or electrical response that affects whole cell function [44, 45]. Put into electrical engineering terminology, mechanotransduction may be broken down into three elements: signal or cue
(input), cell processing (sensors, transducers, and effectors), and cell outcome
(output). A signal, varying with respect to time and space, is first detected by a sensor, modulated (amplified or attenuated) by a transducer, and finally interpreted by an effector. These three components will determine the final cell output to the physical input. Studies in endothelial cells, neuronal cells, and smooth muscle cells have identified a series of sensory proteins, including cilia, cell membrane proteins, and nuclear membrane proteins able to induce a biochemical signal as a result of a physical deformation [46-50]. Cell membrane proteins include ion channels, G-coupled proteins, cell adhesion molecules, and integrins. Integrins are the primary sites of force transmission in cell-matrix interactions, and recent evidence suggests that cadherins may transmit mechanical cues at cell-cell sites [51, 52]. After initial sensing, molecular mechanisms of mechanotransduction may involve conformational changes, including the unfolding of epitopes or whole proteins to induce activation [53].
The conversion of physical signals into biochemical responses may occur at the
19 sensory site or at a distance. For example, centralized mechanical to chemical conversion may occur through conformational changes of proteins at the cell membrane site (i.e. integrins or cadherins), nuclear membrane or in the cytoplasm (i.e. activation of enzymes and kinases) by direct force application to these sites. Alternatively, de-centralized mechanotransduction involves force transfer across the cell via mechanical linkages, perhaps provided by the cytoskeleton [54, 55].
2.6.1 Cytoskeleton: A Continuous Mechanical Link to the Nucleus
The cytoskeleton is composed of microfilaments, intermediate filaments and microtubule elements (Table 2.1 & Figure 2.5). Microfilaments and intermediate filaments are present in the cell cytoplasm and the nucleus. Their primary functions are to provide structural support for cell shape, migration, molecule transport and also the generation of force [56]. The myosin II motor, a cytoskeleton related protein, cycles across actin microfilaments thus generating force [57]. Actin microfilaments have been shown to link to both integrins and cadherins at focal adhesion complexes, and β-catenin/α-catenin links, respectively [44, 57, 58]. Additionally, actin fibers can attach to the nuclear membrane [59, 60]. The nuclear envelope is internally lined by the intermediate filament lamin, which can interact with several transcription factors [59]. Hence, alteration in nuclear organization may end in transcription factor activation or repression which may alter gene expression and subsequent differentiation programs. Evidence supporting the cytoskeleton as a mechanotransducer
20 includes spreading studies of fibroblasts showing that traction forces at one end of the cells are counterbalanced symmetrically across the cytoplasm [61].
21
Table 2.1 Cytoskeletal Elements & their Functions
Approximate Category Diameter (nm) Example Proteins Function Motility [62] Microfilaments 6 Skeletal Actin Cell Shape [63] Smooth Muscle Actin Force Transfer [60] Keratin Support [64] Intermediate 10 Vimentin Force Transfer [65] Filaments Lamin Mechanical stability [66] Alpha-Tubulin Cell Shape [63] Microtubules 23 Beta-Tubulin Proliferation [67]
22
Figure 2.5 Cytoskeletal components of the cell and their cell-cell and cell- ECM attachments. Cytoskeletal elements span the cytoplasm and serve as a link between the extracellular space and the nucleus.
23
2.7 ACTOMYOSIN CONTRACTILITY IN PLURIPOTENCY AND DIFFERENTIATION
Cell traction forces have been shown to be generated by terminally differentiated cells and stem cells alike [43, 68]. These traction forces serve a number of purposes, including cell migration and possibly regulating a response to an external physical force [43]. These forces are produced by actomyosin contractility whereby myosin II heads cyclically bind and unbind to actin polymers. At static equilibrium, external changes in physical force are dynamically balanced by actomyosin forces. In a migrating cell, actomyosin contractility must overcome the bond forces of the integrins and cadherins attached to surrounding matrix and neighboring cells, respectively thus allowing microtubule expansion into the leading lamella [69].
The Rho family of GTPases including RhoA/C and Rac1 regulate multiple cytoskeletal elements including microfilaments and microtubules. RhoA/C targets
Rho Kinase (ROCK) and mDia, whose functions include stabilizing microfilaments and microtubules, respectively [70]. In turn, ROCK targets adducin [71], ERM proteins [71], Myosin Light Chain (MLC) [72] and LIM Kinase
[73] (Figure 2.6). Actomyosin contractions are promoted by the ROCK-mediated phosphorylation of MLC.
24
Figure 2.6 Rho-ROCK & Rho-mDia pathways. RhoA/C targets both ROCK [74] and mDia [75] which ultimately alter actin and microtubule stabilization, respectively. Moreover, ROCK has a number of targets including adducing, ERM proteins [71], MLC [72], and LIMK [73].
25
Small molecule inhibitors of actin and myosin have been utilized to elucidate the role of cytoskeletal generated force in a variety of cellular functions.
Mean traction force and cell adhesion were shown to decrease when human
ESCs were cultured with Blebbistatin, a myosin II ATPase inhibitor [76]. Cell morphology and cell spreading changes were also reported with the use of an inhibitor of actin polymerization, Cytochalasin-D, in mouse mesenchymal stem cells and ESCs [63, 77]. Use of the Rho Kinase (ROCK) inhibitor Y27632 has greatly improved the cloning efficiency of human ESCs that usually undergo apoptosis when dissociated to single cells [78-80]. Subsequent studies have shown that depletion of myosin II by silencing or with inhibition by blebbistatin also increases survival of human ESCs. Therefore, it has been proposed that actomyosin contractions may be responsible for the reduced viability of dissociated ESCs. Not only does actomyosin inhibition affect cell spreading and adhesion, but also the cell response to the physical microenvironment. In human
ESCs, a reduction in nanotopography-mediated responses was observed with
Cytochalasin-D treatment [77]. This same lack of response to a physical microenvironmental input was observed in mouse ESCs, where spreading caused by a targeted application of shear stress to integrins was blocked by blebbistatin [43].
Recently, inhibitors of myosin II ATPase and its upstream regulator ROCK have been used to elucidate the role of internally generated force in stem cell pluripotency. OCT4 and Nanog expression were upregulated in Myosin II
26 silenced human ESCs as compared to controls after 2 days in culture [81], while prolonged exposure to blebbistatin (5 days) resulted in a decrease of OCT4,
SOX2, and NANOG in human and mouse ESCs [76]. When grown in the presence of Rho Kinase inhibitor Y27632, human ESCs increased their expression of OCT3/4 [74].
There are a limited number of studies that have analyzed the role of actomyosin contractility in stem cell differentiation. The use of Y27632 in the directed (growth factor-guided) differentiation of p19 carcinoma cells led to an enhancement of ectodermal and mesodermal differentiation [82] and promoted the generation of endothelial cells from FLK1-sorted mouse ESCs [83]. Rho kinase has multiple targets including myosin light chain, LIM kinase and ERM proteins. Hence, to tease out the role of cytoskeletal contractility on differentiation, inhibitors of myosin II and actin should be used. Preliminary studies on human ESCs showed no statistical significance between untreated and blebbistatin-treated groups as quantified by gene analysis of ectoderm, mesoderm and endoderm markers [76]. Therefore, further studies directly targeting actin and myosin in the context of differentiation without the use of growth factors would aid in the translation of these methods to medical therapies.
As mentioned before, in addition to ROCK, RhoA targets mDia which stabilizes microtubules. Microtubules are heterodimeric cytoskeletal elements that serve as a rail on which motor proteins, such as kinesin and dynein, transport organelles [84]. Microtubules are polar structures with a plus end
27
(growing) and a minus end. The kinesin motor protein directs anterograde transport, while dynein directs retrograde transport [84]. Microtubules and actin polymers interact in a variety of cell functions including: migration [85], neuronal growth cone guidance [86], wound healing [87], and cell division [88]. Due to the close role of microfilaments and microtubules in the regulation of various cell processes, microtubules may also play a role in differentiation. In addition, their role in differentiation may differ to that of microfilaments since contrary to disruption of microfilaments, disruption of microtubules has been shown to enhance cellular contractility [89].
28
CHAPTER 3: CYTOSKELETAL PROTEIN CHARACTERIZATION OF INDUCED PLURIPOTENT STEM CELLS
iPSCs have the potential to be an autologous cell source for the treatment of maladies. iPSCs have been compared to ESCs in their overall genetic and epigenetic characteristics. Yet, their cytoskeletal protein expression has not yet been analyzed. This chapter addresses Aim 1: determine the relative expression of cytoskeletal proteins in iPSCs, their source cell phenotype, and ESCs.
3.1 ABSTRACT
Although induced pluripotent stem cells (iPSCs) hold great promise as a pluripotent cell source, they differ from embryonic stem cells (ESCs) in their abilities to differentiate to all cell types. For this reason, research has centered in understanding the cell state of iPSCs compared to ESCs. Insight into overall gene and protein expression, as well as the epigenetic state, has been obtained, however the cytoskeletal state of iPSCs is yet unknown. In this study, we compared the relative cytoskeletal gene and protein expression of undifferentiated mouse ESCs, MEF-derived iPSCs, and MEFs. Our results show that although both pluripotent cell types were morphologically similar, they varied both in pluripotency and particularly in cytoskeletal protein expression. iPSCs expressed microfilaments and integrins 3-7X higher than ESCs, at levels
29 closer to that of their cell source, MEFs. In the course of differentiation, these initial differences were lost and EBs from ESC and iPSC sources expressed cytoskeletal proteins at similar levels. This knowledge may be used to further the understanding of iPSCs in an effort to generate cells with identical differentiation capacities to ESCs.
3.2 INTRODUCTION
Currently, somatic cell reprogramming allows for the generation of autologous stem cells that are ideal candidates for therapeutic use. Somatic cell reprogramming has been investigated since the mid-20th century. The first nuclear transfer experiments by Briggs & King in 1952 [90] gave way to milestones including the cloning of tadpoles by Gurdon [91], reprogramming of fibroblasts to muscle cells by Davis [92], the first animal cloned from an adult cell
[93] and lastly the generation of induced pluripotent stem cells (iPSCs) from adult mouse [12] and human [7] fibroblasts by defined transcription factors as pioneered by Yamanaka. Reprogramming of adult phenotypes to pluripotent phenotypes with defined factors enabled the derivation of autologous stem cells that have the potential to be utilized in medical therapies. Functional terminal phenotypes such as cardiomyocytes [94, 95], blood cells [96], endothelial cells
[97], neurons [98, 99], and hepatic cells [100] have been derived from iPSCs.
Some derived cell types have been tested for their potential in treating disease.
Specifically, iPSCs have been used to alleviate symptoms of Parkinson’s disease
[98] and sickle cell anemia [101] in the mouse model.
30
Although iPSCs have many similarities with embryonic stem cells (ESCs), they do differ in some aspects. Wholistic analyses of human iPSCs (hiPSCs) and human ESCs (hESCs) resulted in the detection of 1,560 transcripts, 293 proteins and 292 phosphoisoforms that were differentially regulated [102]. iPSCs have also been shown to retain a methylation signature of their tissue of origin [103] as well as chromatin state [104] which can result in preferential differentiation.
These differences in cell state may be consequences of inefficiencies in reprogramming, in part due to cell source phenotype. Embryonic tissues have been shown to be most efficiently reprogrammed, progenitor cells follow, and lastly terminally differentiated phenotypes [103]. There are differences observed in efficiency of reprogramming even within terminally differentiated cells. For example, keratinocytes reprogram more readily than fibroblasts [105]. Analyses studying specific cell processes, such as cytoskeletal development, may aid in understanding differences in reprogramming and ultimately in induced pluripotent cell state.
One of the first noticeable changes during cell reprogramming is the transformation from a flat adherent phenotype into tightly packed clusters of rounded cells [7, 106]. In this shape shift, there may be structural cytoskeletal modifications that occur simultaneously to reprogramming. Yet, the cytoskeletal structure of iPSCs has yet to be characterized. Given the established role of the cytoskeleton in differentiation, we aimed at characterizing the cytoskeletal state of iPSCs and ESCs. Specifically, we compare cytoskeletal gene and protein
31 expression of undifferentiated mouse ESCs, mouse embryonic fibroblast (MEF)- derived iPSCs, and MEFs. These studies provide a basis for understanding the possible role of cytoskeletal state in the process of de-differentiation.
3.3 MATERIALS & METHODS
3.3.1 Pluripotent Cell Expansion
Mouse D3 embryonic stem cells (ESCs; ATCC™) and mouse primary induced pluripotent stem cells (iPSCs; STEMGENT®, WP5; catalogue # 08-
0007) were expanded on a mitotically inactivated mouse embryonic fibroblast
(MEF; ATCC) feeder layer. These iPSCs were reprogrammed from MEFs with the use of retroviruses to deliver OCT4, SOX2, KLF4, and C-MYC and verified by generation of chimeric mice [107]. For the feeder layer, MEFs were expanded in fibroblast medium consisting of Alpha modification of Eagle’s medium supplemented with 10% fetal bovine serum (FBS) and penicillin/streptomycin
(PS). When feeder cells reached confluence, they were treated with mitomycin C
(Sigma-Aldrich®) to inhibit further proliferation. After overnight acclimation, ESCs and iPSCs were plated onto the feeder layer in medium consisting of Dulbecco’s
Modification of Eagles Medium, 15% ESC-qualified FBS (Invitrogen), 2 mM L- glutamine, 0.1 mM non-essential amino acids, PS, and supplemented with 1000
U/ml leukemia inhibitory factor (LIF; EMD Millipore) that allows for maintenance of pluripotency. Before the fusion of adjacent colonies, cells were frozen down and stored in liquid nitrogen. To prepare the pluripotent cells for use, vial contents were thawed and expanded on gelatin-coated dishes in pluripotent
32 culture medium for 4-5 days. This culture markedly reduced (routinely to <5%) the presence of the mitotically inactive feeder cells.
3.3.2 Embryoid Body Differentiation
For differentiation, embryoid bodies (EBs) were generated from ESCs
(EBESCs) and iPSCs (EBiPSCs). Pluripotent cells were placed in non-TC treated dishes in culture medium without LIF (0.5x106 cells/10 mL) and kept in constant motion on a rotary shaker (40 RPM; New Brunswick©). Medium and dishes were changed daily after the second day by gravity separation.
3.3.3 Phase Microscopy
Pluripotent cell colonies and differentiated EBs were imaged using phase contrast microscopy. As an indication of size, cross sectional areas of EBs were calculated by analyzing calibrated phase images (ImageJ software).
3.3.4 Gene Expression
Samples were evaluated for gene expression as described previously [38].
Briefly, for each sample, 1 µg of RNA was isolated, converted into cDNA, and analyzed using real-time PCR on a StepOnePlus™ PCR System (Applied
Biosystems). SYBR® Green was used as a DNA reporter. Primers were designed to assess the expression of transcription factors of pluripotency
(Homeobox Protein NANOG: NANOG; Octamer-Binding Protein 4: OCT4; SRY
(Sex Determining Region Y)-Box 2: SOX2) and cytoskeletal proteins (Tubulin,
Alpha 1b: TUBA1B; Alpha-Cardiac Actin: ACTA2; Vimentin: VIM; Keratin 8:
33
KRT8; Lamin A/C: LMNA). Gene expression levels were determined using standard curves and reported normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH). ESC and iPSC samples are represented as n=3 experiments at Day 0 and Day 6 EBs, while MEF samples are preliminary and only represent n=1 experimental sample. In some plots, samples are plotted as fold changes where each experimental sample was normalized by its corresponding control.
3.3.5 Protein Expression
For protein expression, single cell suspensions were created by treating colonies with Accutase (Life Technologies™) for 6 minutes at 37°C. Cell suspensions were then fixed with 4% formaldehyde for 5 min at room temperature and then stored in buffer solution (0.3% bovine serum albumin and
0.001% polyoxyethylenesorbitanmonolaurate in PBS ++) at 4°C. Samples assessed for intracellular markers were permeabilized with 0.5% triton-X. All samples were blocked with 10% equine serum and incubated with a primary and then a secondary antibody (except for PE-conjugated anti-ACTA2 and anti-α5 integrin antibodies). Fluorescence was detected for each sample using a BD
FACSCanto II. Results are represented in fluorescence histogram and bivariate plots of fluorescence vs forward scatter (FSC) formats. Markers assessed included NANOG and SOX2 (pluripotency), ACTA2 (microfilament), VIM
(intermediate filament), TUBA1B (microtubule), e-cadherin (ECAD, cell-cell adhesion), and integrin α5 (cell-matrix adhesion).
34
3.3.6 Statistical Analysis
Results are presented as mean ± standard error of the mean. Since MEF samples consisted of an n=1, comparison between ESC and iPSC gene expression were analyzed via student’s t-test using Minitab® software.
Comparison between pluripotent cells and MEFs was not verified statistically.
3.4 RESULTS
3.4.1 Pluripotent stem cell morphology differed in ESCs and iPSCs
Expansion of pluripotent cells on a gelatin-coated substrate resulted in the formation of colonies for both ESCs and iPSCs. Phase images taken during expansion showed highly refractive dome-like colonies in both pluripotent cell types (Figure 3.1, single arrows). Although colonies were of similar size, there were noticeably less of them in iPSC cultures. Moreover, a higher percentage of more or less tightly packed, yet not dome-like colonies were present in iPSC cultures (Figure 3.1, double arrows). Both cultures also evidenced cell types that were not part of colonies, and were more spread. This phenotype was observed to be morphologically distinct to MEFs, suggesting they are not residual MEFs from prior expansion, but possibly a subpopulation of spontaneously differentiating cells (Figure 3.1). Hence, iPSCs and ESCs samples used in these studies have similar types of colonies, yet they may be distinguished by the relative amounts of dome-like and tightly packed colonies.
35
Figure 3.1 ESCs and iPSCs were morphologically similar to each other and distinct from MEFs. Representative phase images of ESCs, iPSCs, and MEFs at low (TOP) and high (BOTTOM) magnifications. Scale bars represent 200 μm. Two types of colonies are observed, dome-like (single arrow) and compact (double arrow).
36
3.4.2 Transcription factors of pluripotency were expressed similarly or at lower levels in iPSCs
Initial assessments focused on determining the relative levels of transcription factors known to be required for pluripotency. NANOG, OCT4, and
SOX2 gene expression of ESCs and iPSCs was quantified via real-time PCR. No difference was detected between groups in their expression of NANOG, yet both
OCT4 and SOX2 were significantly lower in iPSC cultures (Figure 3.2A). Despite significance, the drop in OCT4 expression was only of 0.2-fold. In contrast, SOX2 was 2-fold lower in iPSCs as compared to ESCs. Protein evaluation of NANOG and SOX2 resulted in similar trends as their gene counterparts, where NANOG protein was similarly expressed in both pluripotent cell types (Figure 3.2B) and
SOX2 expression in iPSCs shifted to lower levels of fluorescence as shown by fluorescence histograms and fluorescence vs FSC bivariate plots (Figure 3.2C).
Together, these data suggest that iPSCs and ESCs may be at different cell states, or that our source of iPSCs may not have been homogenously reprogrammed.
37
Figure 3.2 iPSCs expressed pluripotency markers at comparable or lower levels than ESCs. (A) Gene expression of ESCs (WHITE) and iPSCs (GRAY) presented normalized to ESC values. Statistical differences are indicated by asterisks (*p<0.05, ** p<0.01). (B) Fluorescence histograms of NANOG protein expression quantified by flow cytometry, where secondary only stains are in GRAY and stained samples in BLACK. (C) SOX2 protein expression represented in histograms (TOP) and fluorescence vs FSC bivariate contour plots (BOTTOM). For histograms, secondary only samples are denoted in each graph, stained samples are in BLACK for ESCs, RED for iPSCs, and GRAY for MEFs. In bivariate plots, BLUE denotes highest concentration of events, and RED the least.
38
3.4.3 iPSCs expressed microfilaments and integrin α5 at a higher level than
ESCs
Gene and protein expression of various cytoskeletal proteins including microtubules, microfilaments, and intermediate filaments were quantified to investigate the cytoskeletal state of iPSCs. TUBA1B, an alpha-type microtubule, was similarly expressed in both pluripotent cells and MEFs at the gene (Figure
3.3A) and protein levels (Figure 3.3B). In contrast, the microfilament ACTA2 and intermediate filaments (VIM, KRT8, LMNA) were significantly different between pluripotent cell types and at least 10X lower than MEFs. ACTA2 and VIM were expressed significantly (p<0.001) higher in iPSCs than ESCs by 7- and 4-fold respectively (Figure 3.3A), with a small but noticeable shift to higher fluorescence levels in protein expression (Figure 3.3B). KRT8 was also 5-fold higher in iPSCs
(p<0.001) and interestingly, the only cytoskeletal element that was more prevalent in pluripotent cells versus MEFs. The nuclear intermediate filament
LMNA followed the same trend as the cytoplasmic intermediate proteins KRT8 and VIM, where it was 3-fold higher (p<0.001) in iPSCs as compared to ESCs.
Together, these data suggest that in general, iPSCs expressed microfilament and intermediate cytoskeletal proteins at higher levels than ESCs, while both their expression levels were orders of magnitude lower than that of MEFs, except for the intermediate filament KRT8 and the microtubule, TUBA1B.
39
Figure 3.3 Cytoskeletal proteins were more highly expressed in iPSCs than ESCs. (A) Gene expression of cytoskeletal proteins of ESCs (WHITE), iPSCs (GRAY), and MEFs (BLACK). Statistical differences are indicated by asterisks (*** p<0.001). (B) Protein expression microtubules (TUBA1B), microfilaments (ACTA2), and intermediate filaments (VIM). Secondary only samples are denoted in each graph, stained samples are in BLACK for ESCs, RED for iPSCs, and GRAY for MEFs.
40
We further investigated the cytoskeletal state of pluripotent cell types by probing the expression of adhesion proteins that play a role in cell-cell and cell- matrix adhesion and have been shown to complex with the cytoskeleton. ECAD, a cell-cell adhesion protein, was similarly expressed in both pluripotent cell types at a higher level than that of MEFs (Figure 3.4A, LEFT). By contrast, α5 integrin was expressed at much higher levels in MEFs and a subpopulation of iPSCs expressed integrin α5 at higher levels than ESCs (Figure 3.4A, LEFT; Figure
3.4B). Taken together, cytoskeletal elements and cell-matrix adhesion proteins were more highly expressed in iPSCs as compared to ESCs.
41
Figure 3.4 ECAD was similarly expressed in both cell types, while integrin α5 was more highly expressed in iPSCs. (A) Quantification of cell adhesion proteins ECAD (LEFT) and α5 integrin (RIGHT) by flow cytometry. Secondary only stains are in GRAY and stained samples in BLACK. (B) Fluorescence versus FSC plots of ESCs, iPSCs, and MEFs. BLUE denotes highest concentration of events, and RED the least.
42
3.4.4 Embryoid Bodies from iPSCs and ESCs had similar levels of microfilaments, intermediate filaments, and microtubules at Day 6
To understand the progression of cytoskeleton development with differentiation, we investigated the cytoskeletal protein expression of Day 6 embryoid bodies generated from iPSCs (EBiPSCs) and ESCs (EBESCs). EBiPSCs were smaller (p<0.001) than EBESCs as seen in phase images and corresponding quantification of surface area (Figure 3.5A&B). Microtubule expression at Day 6 of differentiation was similar across EBESCs, EBiPSCs, and MEFs (data not shown).
In contrast to their pluripotent progenitors, there was no difference detected between microfilament and intermediate filament expression of EBESCs and
EBiPSCs after 6 days of EB differentiation (Figure 3.5C). Hence, it seems that although iPSCs had higher initial values, upon differentiation EBESCs and EBiPSCs levels of cytoskeletal protein expression were similar. Except for KRT8, cytoskeletal protein expression of pluripotent cells was much lower than that of
MEFs. Both ECAD and integrin α5 were expressed at very low levels in EBESCs and EBiPSCs (Figure 3.5D). Taken together, despite difference in overall initial cytoskeletal states, by Day 6 of differentiation EBs from both pluripotent cell types had similar expression levels of cytoskeletal elements and adhesion molecules. With the exception of KRT8, partially differentiated cells (EBs) expressed cytoskeletal elements at lower levels than fully differentiated adherent phenotypes such as MEFs.
43
Figure 3.5 EBs generated from iPSCs and ESCs similarly expressed microfilaments, intermediate filaments, and adhesion proteins. (A) Phase Images of EBESCs and EBiPSCs. Scale bar represents 200 μm. (B) Surface area of EBESCs (WHITE) and EBiPSC (GRAY). Statistical significance represented by asterisks (***p<0.001). (C) ACTA2, VIM, KRT8, and LMNA expression of EBESCs (WHITE) and EBiPSC (GRAY). (D) Quantification of cell adhesion proteins ECAD (LEFT) and α5 integrin (RIGHT) by flow cytometry. Secondary only stains are in GRAY and stained samples in BLACK.
44
3.5 DISCUSSION
Although iPSCs have similarities to ESCs in various modalities including morphology, tumorogenicity, and contribution to all three germ lineages of a chimera, subtle differences in both cell state and differentiation capabilities exist.
Studies characterizing the cell state of iPSCs may help to uncover their relative potential compared to ESCs. The cytoskeletal state of iPSCs as compared to
ESCs and its possible role in de-differentiation has yet to be elucidated. In these studies, we performed preliminary assessments to analyze the cytoskeletal gene and protein expression of iPSCs, their source cell, and ESCs. We found that although both pluripotent cell types were morphologically similar, they varied both in pluripotency and especially in cytoskeletal protein expression. With the exception of KRT8 and TUBA1B, iPSCs expressed cytoskeletal proteins and integrins 3-7X higher than ESCs, but at markedly lower levels than that of their cell source, MEFs. In the course of differentiation, however, these initial differences were lost and EBs from ESC and iPSC sources expressed cytoskeletal proteins at similar levels. Therefore, our studies indicate that, except for microtubules, cytoskeletal protein expression in iPSCs is not entirely reset to that of ESCs, but upon differentiation these differences are lost. Moreover, the expression of microfilaments and intermediate filaments, except for KRT8, of both pluripotent cell types was orders of magnitude lower than that of the iPSC source, MEFs.
45
Numerous studies have focused on characterizing iPSCs in order to understand their differentiation potential. Although studies have found many similarities between iPSCs and ESCs, differences exist in cell state, which may relate to their capabilities of differentiation. For example, epigenetic memory
[103, 104, 108], protein expression and phosphorylation [102], as well as mechanical properties [109] have been shown to be distinct in hiPSCs and hESCs. In a thorough analysis of pluripotent cells, Phansteil et al analyzed the transcriptome and proteome of four hESC and hiPSC lines in biological triplicate
[102]. Researchers found statistical differences in the cytoskeletal proteins KRT8
(1.5X higher in iPSC) and Tubulin, Beta 2A Class IIa. Our studies with murine cells support these findings and expand on them by assessing all three cytoskeletal protein subtypes and confirming that microfilaments and intermediate filaments alike may be more highly expressed in iPSCs derived from cytoskeleton-rich phenotypes such as fibroblasts. In addition to the cytoskeleton,
Phansteil et al identified proteins involved in muscle contraction and wound healing that are more highly expressed in iPSCs derived from human fibroblasts than their ESC counterparts [102]. Both muscle contraction and wound healing processes are known to require coordinated cytoskeletal activity, hence our results may provide reasoning for their upregulation in iPSCs.
When comparing hESCs and hiPSCs, Daniels et al demonstrated differences in the mechanical properties of pluripotent cells. hiPSCs cells were found to be purely viscous, while hESCs were predominantly viscous, but had
46 localized elastic regions [109]. Changes in cell stiffness have also been observed in the course of differentiation. AFM probing of individual cells resulted in a 2-3X increase in stiffness by Day 6 of differentiation as compared to undifferentiated cells [110]. Cell stiffness may in part be attributed to cytoskeletal elements [111,
112]. In fact, changes in F-actin have been reported with differentiation [43].
Moreover, inhibition of Rho kinase, an upstream regulator of actomyosin contractions, at Day 7 of reprogramming significantly increased the efficiency of reprogramming of hiPSCs [113]. In our studies, iPSCs started out at higher microfilaments and integrin expression levels than ESCs, but by Day 6 of differentiation they expressed similar levels of cytoskeletal and cell-matrix adhesion proteins. Furthermore, the cytoskeletal protein expression levels, except for the intermediate filament KRT8, of spontaneously differentiated cells were orders of magnitude lower than that of terminally differentiated adherent phenotypes such as MEFs.
Our studies probed the cytoskeletal state of iPSCs as compared to their cell source and ESCs. They provide a foundation for further analysis of the possible role of the cytoskeleton in de-differentiation, yet are limited in scope due to the low number of cytoskeletal proteins assessed and the low experimental n of MEF samples. More importantly, changes in pluripotent markers were observed in the iPSCs indicating that they may not have been completely reprogrammed. Subsequent studies should correlate cytoskeletal state to the reprogramming process in an attempt to elucidate its role in de-differentiation.
47
This knowledge may be used to further the understanding of iPSCs in an effort to generate cells with identical differentiation capacities to ESCs. In conclusion, our studies indicate that during de-differentiation, cytoskeletal protein expression in iPSCs may not be entirely reset to that of ESCs and that the expression of microfilaments of pluripotent cells was markedly lower than that of somatic cells from which iPSCs may be derived from.
The results found in this chapter demonstrate that PSCs had lower expression levels of microfilaments (ACTA2) and cytoplasmic (VIM) and nuclear
(LMNA) intermediate filaments, with the exception of KRT8. Yet, it was interesting to discover that the expression of microtubules did not change as a function of cell phenotype, whether pluripotent or terminally differentiated. This may be due to the role of microtubules in a wide variety of cell functions critical to cell viability such as mitosis, where they are responsible for the formation of bipolar spindles [114]. As cells differentiate, they undergo the process of EMT by which they adopt a mesenchymal shape, lose their expression of ECAD and increase their expression of N-CAD. During somatic cell reprogramming they undergo the opposite transition, MET, where ECAD expression is increased and
N-CAD expression is lost. We hypothesized that cytoskeletal changes may parallel these shape and adhesion protein changes, where with reprogramming a loss of cytoskeletal proteins would be observed. From our results, it seems that cytoskeletal elements may not be uniformly decreased upon reprogramming of
MEFs to iPSCs. Moreover, it seems that microfilaments were lower in PSCs,
48 microtubules were similarly expressed, and finally that overall intermediate filaments were lower in PSCs, yet keratin subtypes may have specialized roles.
49
CHAPTER 4: DIFFERENTIATION PATTERNS OF EMBRYONIC STEM CELLS IN TWO VERSUS THREE DIMENSIONAL CULTURE
(Published: Emma T. Pineda, Robert M. Nerem, Tabassum Ahsan. Cells Tissues Organs. 2013;197(5):399-410)
Culture conditions in two and three dimensions are commonly used for stem cell differentiation, yet a systematic comparison of these physical microenvironments in undirected differentiation is required. This chapter addresses Aim 1: characterize overall differentiation and cytoskeletal protein expression of ESCs in 2D and 3D.
4.1 ABSTRACT
Pluripotent stem cells are attractive candidates as a cell source for regenerative medicine and tissue engineering therapies. Current methods of differentiation result in low yields and impure populations of target phenotypes, with attempts for improved efficiency often comparing protocols that vary multiple parameters. This basic science study focused on a single variable to understand the effects of two- versus three- dimensional culture on directed differentiation.
We compared mouse embryonic stem cells (ESCs) differentiated on collagen type I-coated surfaces (SLIDEs), embedded in collagen type I gels (GELs), and in suspension as embryoid bodies (EBs). For a systematic analysis in these studies, key parameters were kept identical to allow for direct comparison across
50 culture configurations. We determined that all three configurations supported differentiation of ESCs and that the kinetics of differentiation differed greatly for cells cultured in 2D versus 3D. SLIDE cultures induced overall differentiation more quickly than 3D configurations, with earlier expression of cytoskeletal and extracellular matrix proteins. For 3D culture as GELs or EBs, cells clustered similarly, formed complex structures, and promoted differentiation towards cardiovascular phenotypes. GEL culture, however, also allowed for contraction of the collagen matrix. For differentiation towards fibroblasts and smooth muscle cells which actively remodel their environment, GEL culture may be particularly beneficial. Overall, this study determined the effects of dimensionality on differentiation and helps in the rational design of protocols to generate phenotypes needed for tissue engineering and regenerative medicine.
4.2 INTRODUCTION
Despite numerous advances in tissue engineering and regenerative medicine, cell sourcing remains a significant hurdle for the translation of therapies from the bench to clinic [115, 116]. Regenerative medicine and tissue engineering approaches, including the creation of tissues for transplantation and in vitro disease models, benefit from continuously available, functional, and pure cellular phenotypes. Pluripotent stem cells are considered good candidates for regenerative medicine applications due to their ability to self-renew and potential to become any cell in the adult body. For example, embryonic stem cells can spontaneously differentiate in vitro into the three germ lineages (ectoderm, mesoderm, or endoderm), from which will arise all somatic cell types [5, 6, 27].
51
Embryonic stem cells have been differentiated to cells of all three germ lineages in both two and three dimensional physical microenvironments. The common configurations of cells on a monolayer, embedded in protein gels [23,
24] and in suspension [26, 27] have both advantages and disadvantages. Stem cells cultured on adherent surfaces can be presented with bound proteins [117,
118] and well-controlled exogenous physical cues, such as cyclic tension [119,
120] and shear stress [38, 42, 121]. Yet, culture in this 2D configuration restricts cell growth to a single geometric plane. Suspension culture, which allows for the formation of cell clusters during spontaneous differentiation (or embryoid bodies:
EBs), can mimic cellular interactions reminiscent of in vivo development processes [35] but only allows external stimuli in the form of soluble factors or hydrodynamic forces [122]. Encapsulation of cells within scaffolds or hydrogels, however, enables both the presentation of proteins [24, 29, 30] and the application of mechanical cues [123, 124], while maintaining the cells in the rounded configuration often prevailing in vivo (Reviewed in [125]). Although these different 2D and 3D modalities have been used in conjunction with other exogenous cues to promote directed differentiation, there has not yet been a systematic analysis to determine the fundamental effects of dimensionality on differentiation.
This study characterizes the overall differentiation of pluripotent stem cells cultured in both two- and three- dimensions. In particular, mouse embryonic stem cells (ESCs) were differentiated in 2D on collagen type I-coated slides and
52 compared to both the 3D analog of ESCs embedded within collagen type I hydrogels and standard EB differentiation. Differentiation kinetics for the three culture configurations were evaluated by gene expression of germ lineage markers and cytoskeletal proteins, as well as higher throughput screens for general differentiation patterns. This type of systematic study of culture dimensionality enables more informed choices when targeting specific phenotypes in vitro for tissue engineering and regenerative medicine applications.
4.3 MATERIALS AND METHODS
4.3.1 Expansion of Mouse Embryonic Stem Cells
Mouse D3 embryonic stem cells (ESCs) and embryonic fibroblasts (MEFs) were purchased from ATCC and cultured as described previously [38, 42].
Briefly, ESCs were initially expanded on mitotically arrested MEFs and stored in liquid nitrogen. Prior to experiments, ESCs were thawed and cultured on gelatin- coated tissue culture plastic. Culture medium consisted of Dulbecco’s
Modification of Eagles Medium (DMEM) supplemented with 15% ES-qualified fetal bovine serum (Invitrogen), 2 mM L-glutamine, 0.1 mM non-essential amino acids, 1000 U/ml leukemia inhibitory factor (ESGRO® from EMD Millipore), and
0.1 mM penicillin/streptomycin (Thermo Scientific, Inc.).
53
4.3.2 Two and Three Dimensional Differentiation Systems
Cells were differentiated in three physical configurations: adherent, gel, and suspension culture (Figure 4.1A). Differentiation medium for all systems consisted of culture medium except without leukemia inhibitory factor. For adherent culture, considered a two-dimensional (2D) culture system (Figure
4.1A, TOP), glass slides were coated with 3.5 μg/cm2 collagen type I (MP
Biomedicals©) for at least one hour and were then seeded with 1 x104 ESCs/cm2
(SLIDE). SLIDE samples were maintained at 37°C/5% CO2 for 2, 4, or 6 days.
To present cells with collagen type I but in a three dimensional (3D) configuration, 0.75 x106 ESCs were embedded in collagen type I (2 mg/mL) that was base-neutralized to generate 0.75 mL gels (Figure 4.1A; MIDDLE). GEL samples were maintained in free-floating conditions for 4, 8, or 12 days with medium (25 mL/construct) changed every other day. Using a standard differentiation protocol, ESCs were cultured as embryoid bodies (Figure 4.1A;
BOTTOM) by initially plating 1 x106 cells in a 100 mm diameter non-tissue culture treated dish with 10 mLs of medium (EB). For EB culture, dishes and medium were changed daily after day 2 and maintained for up to 12 days. Phase images were taken during culture (SLIDE samples) or after histological processing (EB and GEL samples). Macroscopic pictures of GEL samples were quantified using
ImageJ software.
54
4.3.3 Gene Expression Analysis
At the end of culture, samples were analyzed for mRNA expression, utilizing either standard real-time PCR or PCR arrays. RNA was isolated using the Qiagen RNeasy Kit (for SLIDE and EB samples) or the RNeasy Lipit Tissue
Kit (for GEL samples) and then quantified using a Nanodrop® spectrophotometer for each sample. Standard analysis of mRNA levels for each sample was done on cDNA converted from 1 µg RNA (Invitrogen Superscript® III First-strand synthesis) and analyzed using SYBR® Green (Applied Biosystems) on a
StepOnePlus™ PCR System. Primers were custom designed (Primer Express®
Software v3.0) for octamer-binding protein 4 (OCT4), alpha-fetoprotein (AFP),
Brachyury (BRACHY-T), nestin (NES), actin alpha 1 (ACTA1), actin alpha 2
(ACTA2), tubulin (TUBA1B), keratin 8 (KRT8), vimentin (VIM), lamin (LMNA), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Forward and reverse primers are listed in Supplemental Table 4.1. Gene expression levels were quantitated using standard curves and are reported normalized to GAPDH expression.
The Mouse Embryonic Stem Cell RT2 Profiler™ PCR Array (SA
Biosciences, of Qiagen) was used to profile the expression of genes involved in the maintenance of pluripotency and the general differentiation of embryonic stem cells. Eight different groups were analyzed: undifferentiated ESCs, SLIDEs at Day 4; GELs at Day 4, 8, and 12; and EBs at Day 4, 8, and 12. For each sample, the 84 analyzed genes were normalized to a set of housekeeping genes
55
(listed in Supplemental Table 4.2). Analysis across groups (n=3 independent samples per group) was performed using Matlab (Mathworks) and Ingenuity®
Pathway Analysis (Redwood City, CA) software. A binary tree consisting of nested subsets to show relative similarities across experimental samples was created using an algorithm of Euclidean distance and average linkage [126].
Further analysis was performed clustering on experimental group and/or gene to produce colored heatmaps indicating relative expression levels. The mean expression levels of genes are indicated in black, with relative upregulation and downregulation shown in red and green, respectively, and color saturation (dark hues) used for values three or more standard deviations away from the mean.
4.3.4 Statistical Analysis
Results are presented as mean ± standard error of the mean.
Experimental samples were analyzed via student’s t-test when comparing within
1 variable (GEL compaction). A one-way analysis of variance (ANOVA) was used when comparing the three culture conditions at the same time point (SLIDE,
GEL, EB at Day 4). When comparing culture condition and time in culture, a two- way ANOVA was performed with a post-hoc Tukey test (GEL, EB at Day 4 & 12).
56
Figure 4.1 Initial cell distribution and structures supported by the 2D and 3D models. (A) ESCs were seeded on glass slides coated with collagen type I in the 2D model (SLIDE). In the 3D models, ESCs were either seeded in collagen type I constructs (GEL) or grown in suspension (EB) for 4 days. (B) Images of ESCs cultured for 1, 2, and 4 days as SLIDEs (i-iii), GELs (iv-vi), and EBs (vii-ix). (C) In GEL samples at days 8 and 12, cell bodies of defined boundaries (i) and those with cavities (ii) were found. Elongated bodies of cells (iii) and lumen-like structures (arrow in iv; insert shows higher magnification) were also observed. SLIDE and EB phase images were taken of live cells, while GEL images were histological sections stained with Hoechst to indicate nuclei. Scale bars represent 200 µm.
57
4.4 RESULTS
4.4.1 Cell Growth in Differentiation Systems
All three differentiation systems (SLIDE, GEL, and EB culture; Figure
4.1A) successfully supported cell growth over a period of days. Initial cell density was sufficiently low to ensure that conditions started as single or few cells for all models, as illustrated by images at Day 1 (Figure 4.1Bi, iv, vii). With subsequent culture over four days, cell clusters on SLIDEs, in GELs, and as EBs all continued to increase in size and number indicating support for cell proliferation
(Figure 4.1B). Clusters in 2D culture generated multiple cell layers but preferentially grew along the plane of the slide. SLIDE samples were limited to 6 days of differentiation because cell confluency ultimately became such that the initially dominant trait of cell-matrix binding was overcome by cell-cell interaction.
Clusters in either GEL or EB samples grew radially, as would be expected from
3D culture. Higher magnification images of fluorescently labeled nuclei in histology cross-sections provided a more detailed view of the range of morphological arrangements of the clusters and cell-cell interactions in GEL samples (Figure 4.1C). Cell organization in GEL samples was similar to that seen in standard EB culture, commonly including bodies with defined boundaries
(Figure 4.1Ci) that occasionally contained cavities (Figure 4.1Cii). Unique to cells grown in GELs, however, were elongated bodies along the edges of the gel construct (Figure 4.1Ciii, arrows). A few instances also included lumen-like structures where cells were cuboidal in shape arranged in a circular arrangement
(Figure 4.1Civ, arrow & insert).
58
4.4.2 Pluripotency and Germ Lineage Differentiation
Representative markers of pluripotency and germ lineage specification were assessed using real time rtPCR (Figure 4.2). Here the expression patterns of OCT4 (pluripotency), NESTIN (ectoderm), and BRACHY-T (mesoderm) were largely similar across culture conditions. For SLIDE, GEL, and EB culture, OCT4 expression markedly decreased after 4 days indicating loss of pluripotency. All three groups also showed dynamic levels of NESTIN and BRACHY-T, for which expression was highest at day 4. Expression of AFP, a marker of endoderm, instead steadily increased with time for all groups, but with levels at all timepoints
1-2 orders of magnitude higher in EB samples. Taken together, this data indicated that ESCs differentiate in SLIDE, GEL, and EB cultures. To better characterize the breadth of differentiation, a higher throughput approach was subsequently used.
59
Figure 4.2 Gene expression of pluripotent and lineage markers for SLIDE, GEL, and EB samples. Samples were evaluated for pluripotency (OCT4) and specification towards the three germ lineages (ectoderm – NESTIN; mesoderm – BRACH-T; and endoderm – AFP). Real time rtPCR values for genes of interest were normalized to the housekeeping gene GAPDH. SLIDE samples (RED, triangle) were assessed at days 2, 4, and 6, while GEL (BLUE, circle) and EB samples (BLACK, square) were evaluated at days 4, 6, and 8. Data presented are mean±SEM for n=3.
60
4.4.3 ESC Differentiation Patterns
To obtain an overall sense for the patterns of differentiation of ESCs grown in SLIDE, GEL, and EB culture systems, the Mouse Embryonic Stem Cell
PCR array (SA Biosciences of Qiagen) was used to determine gene expression of eighty four genes of pluripotency and differentiation (listed in Supplemental
Table 4.2). Hierarchical cluster analysis showed that samples from the same group generally clustered together (Figure 4.3A), indicating that the chosen PCR array was suitable for interpreting effects of culture system and duration. The hierarchical tree indicated that effects due to culture duration dominated more during early differentiation, while samples from the same culture paradigm clustered together at later timepoints. Specifically, at day 4 it was seen that for cells grown on SLIDEs, differentiation was greater (further from ESCs) than for cells grown in GELs or as EBs. Closer inspection of the branching points then showed that cells cultured in 3D (either in GELs or as EBs) clustered on a separate branch than those cultured on 2D SLIDEs (indicated in bold and with an asterisk in Fig 3A). At later time points, sample associations in 3D configurations were stronger based on culture system (GEL vs EB) rather than culture duration
(Day 8 vs Day 12). Overall, this indicates that culture configuration and duration strongly define the general differentiation pattern of cells and that configuration effects become increasing important for extended culture durations.
Heat maps of the averages of normalized gene expression levels (to values for undifferentiated ESCs) of SLIDE, GEL, and EB samples were created
61 to spatially represent general differentiation patterns. Evaluation of all three differentiation systems at Day 4 showed a distinct difference in patterning between 2D and 3D culture (Figure 4.3B). At this time point, 35 of 84 genes were upregulated only in SLIDE samples (Figure 4.3B-i), while a separate set of genes
(Figure 4.3B-ii) were similarly highly expressed in both GEL and EB samples.
These results expound on the tree analysis showing the relative differentiation of
SLIDE, GEL and EB samples by revealing that the differentiation pattern, as reflected by individual genes, are more closely related between GEL and EB samples than to SLIDE samples.
Heat maps for the independent 3D culture systems displayed similar patterns of differentiation kinetics. When GEL samples at days 4, 8, and 12 were analyzed, groups aligned temporally and, as one would expect, genes fell in one of three categories: highly expressed genes at day 4 that were then downregulated with increased culture time (Figure 4C-I); low expressing genes at day 4 that were only transiently upregulated (Figure 4.3C-II); and genes that were monotonically upregulated with culture time over 12 days (Figure 4.3C-III).
When the gene order was fixed to maintain those same categories and used to analyze EB samples, groups again aligned temporally. Furthermore, the overall differentiation pattern was similar between GEL and EB samples though certain individual genes transitioned differently. Taken together, clustering analysis indicated that all systems supported differentiation but that differentiation patterns were distinct for 2D versus 3D cultures.
62
Figure 4.3 Hierarchical clustering and heat maps based on gene expression from a PCR array. Expression of 84 genes were grouped using a hierarchical algorithm to clusters values according to similarity. (A) A hierarchical tree indicating relative similarity between D0 ESCs, SLIDEs at Day 4, GELs at Day 4, 8, and 12, and EBs at Day 4, 8, and 12. (B) A heat map displaying gene–clustered data for SLIDE, GEL and EB samples after 4 days of culture. Distinct groupings of genes: (i) genes classified as those upregulated in SLIDE samples only and (ii) genes similarly regulated in GEL and EB samples. (C) Heat maps indicating differentiation over time for GEL and EB samples. Distinct groupings of genes for GEL samples: (I) genes most highly expressed at day 4 with subsequent loss in expression with time, (II) genes transiently expressed with highest levels at day 8, and (III) genes for which expression increases with time from day 4 to 12. The EB heat map is shown with the gene order of the GEL heat map. The colors indicate level of gene expression: BLACK is the mean across samples, RED indicates expression higher than the mean, and GREEN indicates expression below the mean. For each group n=3 independent samples.
63
4.4.4 Functional Gene Groupings
Specific groupings of genes were identified using the PCR array data and
IPA software. Functional categories that were further analyzed included
Developmental Morphogens, Cardiovascular Differentiation, Neural
Differentiation, and Matrix Proteins (Figure 4.4). Fold regulation is represented for early (Day 4) and later (Day 12) differentiation compared to undifferentiated
ESCs, such that bar length above the x axis indicates level of higher expression and the bar length below the x-axis indicates level of lower expression.
(Acronyms for genes are listed in Supplemental Table 4.2).
Development relies on creating morphogenic gradients, in part by expression of SFRP2, NODAL, FOXA2, and NOG (Figure 4.4A). SFRP2, soluble frizzled-related protein 2 which modulates Wnt signaling, was upregulated after 4 days of culture in both 3D culture conditions (GEL and EB samples) with respect to undifferentiated ESCs, with significant greater increases in expression observed after 12 days in EB culture. NODAL, a member of the TGF-β superfamily essential for mesodermal formation and axial patterning in the developing embryo [127, 128], was instead only transiently upregulated in GEL samples at day 4, with a significant downregulation observed at day 12 in both
GEL and EB samples. FOXA2, a transcription factor required for notochord formation and endodermal differentiation [129-131], was upregulated in all groups at day 4 and increased another 7-fold in EB cultures by day 12. NOG, an antagonist of the TGF-β superfamily, instead was only significantly elevated in
64
2D slide culture at day 4, but was significantly upregulated in both 3D culture conditions at day 12. The three separate culture conditions distinctly regulated expression of morphogens, but with more similar trends in overall expression between the 3D culture conditions. The ability to continue culture of the 3D systems for extended durations is likely valuable in developing and sustaining morphogen spatial gradients during differentiation in vitro. The choice of the specific 3D culture system would then be application specific as the GEL cultures maintained similar gene expression levels for up to 12 days and EB cultures allowed for a more dynamic expression profile with time.
Expression of cardiovascular markers was evaluated to determine differentiation towards a mesodermal phenotype (Figure 4.4A). At day 4 in both
2D and 3D culture systems, CD34 and PECAM1, markers strongly associated with the hematopoietic and endothelial phenotypes respectively, were similarly regulated by the different culture conditions. At day 4, expression was only significantly affected by culture on SLIDEs. With added culture time, however, both markers were subsequently significantly upregulated in both GELs and EBs.
Flt1 was found to be upregulated only in GEL and EB culture and was elevated at both day 4 and 12. ENDRB, an endothelin receptor, was modestly upregulated in EB samples at day 4, but was markedly upregulated by ~350- and ~1000-fold in GEL and EB samples at day 12. GATA4, a transcription factor implicated in myocardial differentiation, was expressed an order of magnitude higher on slide culture at day 4 compared to undifferentiated ESCs, though was expressed at
65 even higher levels in GEL and EB samples but not until day 12. Overall, it was found that 3D culture is supportive of cardiovascular differentiation, particularly with extended culture durations. EB culture, however, seemingly promotes the greatest level of cardiovascular differentiation.
Neural differentiation was assessed by looking at NEUROD1, PAX6, NES, and OLIG2 (Figure 4.4A). In general, neural differentiation was not significant at day 4 under any culture condition as most markers were expressed at levels either similar to, or even markedly lower than, undifferentiated ESCs. NEUROD1 and NES levels remained very low even at day 12, but PAX6 and OLIG2 were markedly upregulated in EB samples. Thus, neural differentiation of ESCs is not readily supported on glass slides or in GELs, but instead requires sustained culture in EBs.
Collagen type I (COL1A1), Laminin (LAMB1-1), and Fibronectin (FN1) expression was evaluated as an indication of the overall expression of matrix proteins (Figure 4.4B). ESCs on collagen type I-coated glass slides significantly expressed higher levels of both LAMB1-1 and FN1, without any change in
COL1A1 expression, compared to undifferentiated ESCs. While at day 4 expression levels were not changed in the 3D culture conditions, by day 12 all three genes were significantly upregulated in both GEL and EB samples. Thus, all three culture models support expression of matrix proteins though the kinetics of expression were delayed in 3D compared to 2D configurations.
66
Figure 4.4 Relative gene expression of functional groups at Day 4 and 12. Genes of the PCR Array results were grouped based on function: (A) Developmental signals, Cardiovascular differentiation, and Neural differentiation, as well as (B) Extracellular Matrix Proteins. Relative gene expression compared to Day 0 ESCs are shown for Day 4 and 12, for which the length of the bar corresponds to the magnitude of expression. For the genes of interest, bars above the y=1 axis represent an upregulation and below represent a downregulation. Asterisks (*) indicate a significant difference (p ≤ 0.05) compared to the corresponding Day 0 ESC expression level. Crosses (†) on the Day 12 plots indicate a significant difference (p≤0.05) between the Day 4 and Day 12 expression levels for that gene. Data presented are mean±SEM for n=3.
67
4.4.5 Matrix Remodeling
Collagen gels serve as a scaffold that allows for cell-based remodeling during culture. ESCs cultured in gels were observed to significantly compact the volume after approximately a week. Macroscopic images were taken and surface area was quantified using image analysis (Figure 4.5). A significant (p<0.001) decrease (50%) in surface area was observed between days 8 and 12. Since cytoskeletal proteins are often implicated in cell-generated traction forces, we used standard PCR to quantify gene expression of microfilaments, intermediate filaments, and microtubules over time for samples from all three culture models
(Figure 4.6). Cytoskeletal protein expression in all models was extremely low at early differentiation time points, with the exception of TUBA1B. Expression of microfilaments ACTA1 and ACTA2 were differentially regulated, where ACT1A was only upregulated in SLIDE samples and ACTA2 was upregulated with time in both 2D and 3D culture systems. All intermediate filaments (KRT8, VIM, and
LMNA) were expressed earlier for cells on SLIDEs than in 3D culture. For both
GEL and EB samples, expression of the assessed intermediate filaments were markedly higher at days 8 and 12 compared to day 4, though EB culture always had the promoted higher expression levels at day 12. Conversely, microtubule expression, as indicated by TUBA1B, was not affected by either culture condition or time. Taken together, these results indicate that 2D SLIDE culture promotes accelerated expression of cytoskeletal proteins compared to 3D culture and that increases in expression for GEL samples is concomitant with observed changes in hydrogel compaction.
68
Figure 4.5 Collagen gel compaction from Day 8 to Day 12 of culture. (A) Image of GELs cultured for 8 and 12 days show a decrease in surface area. (B) Quantification of the images reveal a significant difference in surface area across days (***, p<0.001). Data presented are mean±SEM for n=6.
69
Figure 4.6 Gene expression of cytoskeletal genes for SLIDE, GEL, and EB samples. SLIDE samples (RED, triangle) were assessed at days 2, 4, and 6, while GEL (BLUE, circle) and EB samples (BLACK, square) were evaluated at days 4, 6, and 8. Gene expression are shown for microfilaments (ACTA1 and ACTA2), intermediate filaments (KRT8, VIM, and LMNA), and microtubules (TUBA1B) all normalized to the housekeeping gene GAPDH. Data presented are mean±SEM for n=3.
70
4.5 DISCUSSION
In these studies, we compared mouse embryonic stem cells (ESCs) differentiated on collagen type I-coated surfaces (SLIDEs), embedded in collagen type I gels (GELs), and in suspension as embryoid bodies (EBs) to determine the effects of culture dimension on differentiation. The extracellular protein collagen type I was selected due to its capability to induce embryonic stem cells to mesodermal phenotypes and its potential to mitigate tumorigenicity of cells delivered in vivo [132, 133]. Many parameters (e.g. initial single cell seeding, days of expansion, and media formulation) were kept identical to allow for direct comparison across the three culture configurations. Based on this systematic approach, we were able to determine that all three configurations supported differentiation of ESCs and that the kinetics of differentiation differed greatly for cells cultured in two versus three dimensions. Two dimensional adherent cultures induced overall differentiation more quickly than 3D configurations; in particular, expression of cytoskeletal and extracellular matrix proteins emerged sooner on
SLIDES. For 3D culture as GELs or EBs, the differentiation patterns and cell cluster morphology were similar. There were some differences, however in the rate and level of cardiovascular differentiation and expression of developmental morphogens. Additionally, the ability of differentiated ESCs to markedly contract the collagen hydrogels is likely indicative of differences in the ability to generate traction forces between GEL and EB cultures.
EB and GEL culture largely supported morphogen expression and cardiovascular differentiation. Both 3D culture systems are based on cell
71 expansion in clusters, where organization is not homogenous within the cluster.
This allows for distinct spatial regions for simultaneous but heterogenous differentiation, which is also a key aspect in gastrulation when distinct germ layers form during early development [134]. Furthermore, GELs may allow for more complex structures since the encapsulation step does not preferentially select for cells that are able to bind to matrix or other cells, as in the case of 2D
SLIDE and 3D EB culture, respectively. In both 3D models, however, the observed expression of developmental markers (e.g. SFRP2, NODAL, and NOG) is consistent with the timing of germ lineage specification both in our studies and those by others for embryoid bodies [135, 136]. Extension of culture in GELs also allowed for asymmetric organization into elongated bodies and lumen-like structures. Such self-assembly may be useful in the formation of complex functional tissues or vascular networks.
For cells in cluster configurations, multi-directional cell-cell connections are present. As these types of connections are important regulators of vascular and cardiac cell homeostasis [137, 138], this may contribute to the support of cardiovascular differentiation in 3D. In addition, cluster configurations have implications in the number of cell-cell connections and cadherin-mediated binding, both of which have been implicated in stem cell fate decisions [22, 139].
While overconfluent adherent cultures can also be dominated by cell-cell interactions, the switch from cell-matrix to cell-cell dominated interactions may complicate data interpretation and even alter differentiation kinetics. Thus, for
72 cells cultured in 3D as either GELs or EBs, differentiation may be mediated by similar mechanisms.
The differentiation patterns between GELs and EBs were found to be similar without the addition of exogenous factors. The use of additional soluble cues, including cytokines, are readily presented to cells either as EBs or in GELs through changes in medium composition or presentation of degradable microspheres [140]. In hydrogels, however, there is also the potential to present bound proteins and create gradients to promote differentiation to challenging phenotypes. While our differentiation modalities, which were unbiased towards any particular phenotype, did not induce meaningful neural differentiation, researchers have used tethering of biomolecules to direct neural cell migration in a polymeric scaffold [141-143]. In addition to spatial gradients, hydrogels also allow for the use of other physical factors that are known to influence differentiation, such as substrate stiffness [21, 144] and applied physical forces
[37, 42, 145, 146]. Cell-generated contraction forces also add a dynamic mechanical environmental factor in collagen gels. These differentiating pluripotent cells were able to remodel the volume and density of the gels, as has been observed with more mature phenotypes such as fibroblasts [147, 148], smooth muscle cells [149, 150], and adult stem cells [151]. Such a contractile permissive environment may in fact facilitate differentiation towards mechanically active cell phenotypes. Combining the above aspects, hydrogels readily allow for
73 the control of numerous elements of the microenvironment, which can then provide higher efficiency in differentiation towards target phenotypes.
SLIDES, of the three groups tested, had the most accelerated loss of pluripotency and overall differentiation. Adherent slide culture, as a means of promoting cell-matrix interactions, is often used to study mechanotransduction. In this modality, applied fluid shear stress [152, 153] or tensile strain [154] impose well-defined mechanical forces to terminally differentiated cell populations.
Recent studies in stem cell differentiation have used similar systems [37, 146,
155, 156], as well as ones that modified the underlying substrate [21]. All of these systems alter the force balance across the membrane (via transmembrane proteins such as integrins) to modulate cytoskeletal tension, known to play a role in mechanotransduction to the nucleus [157, 158]. Especially noteworthy in this system is that changes in expression of cytoskeletal elements were observed to be highly dynamic, indicating the potential to be mechanoresponsive. Thus, use of adherent cultures systems with pluripotent stem cells may be a valuable model to study mechanotransduction in pluripotent stem cells during early stages of differentiation.
Cells respond to the initially presented microenvironment by secretion of molecules and synthesis of proteins that then help to remodel the cellular environment. Cells in adherent cultures, with collagen type I presented on only a single surface, were found here to have high expression levels of cytoskeletal elements and extracellular matrix proteins. These categories of proteins not only
74 change the surrounding protein composition but also change the mechanisms by which the cell can interact with the microenvironment. When cells were instead surrounded with collagen type I as embedded in gels, the gene expression for synthesis of matrix proteins was dampened, perhaps due to a decrease in the need to establish a protein-based microenvironment. In addition, it was observed that as culture time progressed to 12 days the differences between EB and GEL culture became greater. This may be a reflection of an increasingly different microenvironment between the two groups since the cells embedded in collagen were actively contracting the gel and dramatically remodeling their microenvironment. Thus, the selection of culture paradigm may need to consider the process of active remodeling, which may impact both autocrine signals in cell differentiation and matrix properties for tissue engineering.
Current methods of differentiation result in low yields and impure populations of target phenotypes and attempts for improved efficiency have been largely through comparison of discrete protocols. This basic science study focused on a single variable to understand the effects of dimensionality on directed differentiation. Various markers of differentiation were evaluated and reported for both 2D and 3D culture conditions. Due to differences in stem cell phenotype and differentiation mechanisms, these studies need to be replicated in human embryonic and induced pluripotent stem cells for translation to medical therapies. Our results, however, will help in the rational design of protocols to generate a number of specific phenotypes, which ultimately may require complex
75 paradigms that include the sequential use of multiple distinct culture configurations and include growth factors, heterotypic co-culture, and physical force.
4.6 ACKNOWLEDGEMENTS
This work was initiated at the Georgia Institute of Technology and completed at Tulane University. Research reported in this publication was supported by the National Institute of General Medical Sciences of the National
Institutes of Health under Award Number P20GM103629. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
THIS SECTION WAS NOT INCLUDED IN THE PUBLISHED MANUSCRIPT.
The force balance across the cell membrane may differ in cells cultured in
2D vs. 3D. In 2D culture, cells were cultured on rigid glass slides where substrate stiffness would be higher than in either of the 3D culture models where their underlying substrate would be primarily constituted of ECM. Moreover, in 2D culture, connections to the extracellular environment would be primarily dominated by cell-matrix interactions, while in 3D culture cell-cell connections would dominate. The results obtained in this chapter demonstrate that ESCs grown in three distinct physical microenvironments have overall changes in cytoskeletal protein expression and differentiation. Specifically, they provided a comparative study characterizing the development of the cytoskeleton in 2D and
76
3D culture, whereby in general microfilaments and intermediate filaments were more highly expressed in 2D than 3D culture. These results build on Aim 1, since they identify an increase in the expression of intermediate filaments, and the microfilament smooth muscle actin with time. Yet, microtubules and the microfilament skeletal actin were not altered with time. Therefore, across aim 1 and aim 2, microtubules were not only uniformly expressed across cells of varying pluripotent capacities, but also across time in culture in different microenvironments. Therefore, this chapter supports our hypothesis in that an alteration in the physical microenvironment resulted in differences in cytoskeletal expression and concomitantly in gene expression to the mesodermal and endodermal lineages.
77
78
79
CHAPTER 5: ACTIN AND MYOSIN II MODULATE MESODERMAL COMMITMENT OF EMBRYONIC STEM CELLS IN 3D
The cytoskeleton has been shown to play a role in multiple cell processes, including motility, proliferation, and recently differentiation. It is not yet known what effects actin polymers and actomyosin contractions have on embryonic stem cell differentiation. This chapter addresses Aim 3: identify the effect of actin polymers and actomyosin contractions on ESC differentiation in 3D.
5.1 ABSTRACT
Physical forces are known to play a role in mammalian morphogenesis, yet specific cues regulating differentiation to each lineage are not well understood. Here, we cultured embryonic stem cells as embryoid bodies (EBs) in the presence of inhibitors of actin polymerization (cytochalasin-D) and actomyosin contractions (blebbistatin) to elucidate their role in mesodermal lineage commitment and differentiation. Treatment with cytochalasin-D or blebbistatin induced a rounded cell shape and disrupted e-cadherin expression in differentiating EBs. BRACHY-T, a marker of mesendodermal commitment, was downregulated by both inhibitors when applied from Day 4-7 and only by cytochalasin-D when applied from Day 2-4. Similarly, gene expression of early markers of mesodermal differentiation, PAX2 and FLK1, were significantly downregulated in the later treatment window by both inhibitors, while only by
80 cytochalasin-D in the earlier application window. However, FLK1 protein was seen to be significantly upregulated along with RUNX1 gene expression in cytochalasin-D samples at Day 7, suggesting that disruption of actin polymers may promote suspension cell phenotypes within the mesodermal lineage.
Inhibition of Rho Kinase with Y27632 resulted in an upregulation of early mesodermal markers when applied early on, and a downregulation when applied at later time points. Perhaps then, the ROCK-myosin II pathway may play a role in the decrease in mesodermal differentiation at later time points. Day 4 of EB differentiation has been shown to correlate to the onset of gastrulation, indicating that after this time point a failure to adopt an extended shape decreases mesodermal commitment. Contrastingly, prior to the onset of gastrulation the ability to generate actomyosin contractions may not be required for mesodermal commitment.
5.2 INTRODUCTION
During germ lineage commitment, populations of cells undergo morphological and migratory changes resulting in differences in spatial distribution and expression of cytoskeletal proteins [159, 160]. Of the three germ layers, cells that arise from the mesodermal germ layer tend to be cytoskeleton- rich. It is yet to be understood if the cytoskeleton plays a key role in the generation of mesodermal cell phenotypes. Embryonic stem cells cultured in suspension as embryoid bodies (EBs) are a useful in vitro model of development that temporally expresses similar gene and protein markers to those expressed in vivo [161, 162]. This model may aid in understanding the role of the
81 cytoskeleton in mesodermal lineage specification. Here, we investigated the role of actin polymers and actomyosin contractions in mesodermal commitment and early differentiation in the EB model of mammalian development.
5.3 MATERIALS AND METHODS
5.3.1 Embryoid Body Culture
Mouse D3 embryonic stem cells (ESCs; ATCC™) were cultured as described previously [38, 42]. Prior to experiments, cells were maintained on gelatin-coated tissue culture plastic with culture medium that consisted of
Dulbecco’s Modification of Eagles Medium (DMEM), 15% ESC-qualified fetal bovine serum (Invitrogen), 2 mM L-glutamine, 0.1 mM non-essential amino acids, antibiotics, and supplemented with 1000 U/ml leukemia inhibitory factor (LIF;
EMD Millipore). To generate EBs, ESCs were placed in non-TC treated dishes in culture medium without LIF (0.5x106 cells/10 mL) and kept in constant motion on a rotary shaker (40 RPM; New Brunswick©). Medium and dishes were changed daily after the second day by gravity separation. For experimental samples, the medium was supplemented with either cytochalasin-D (CYTO-D; 0.2 μM; Sigma
Aldrich®), Blebbistatin (BLEBB; 24 μM; Sigma Aldrich), or Y27632 (10 µm; EMD
Millipore) during Days 2-4 or 4-7 of EB culture. For controls (CON), DMSO solvent was added in equal volume to CYTO-D or BLEBB, and water to Y27632 samples.
82
5.3.2 Microscopy
EBs were imaged using phase contrast microscopy for overall morphology and size and scanning electron microscopy (SEM) for cell morphology. As an indication of size, cross sectional areas of EBs were calculated by analyzing calibrated phase images (ImageJ software). For SEM analysis, samples were fixed in 2.5% glutaraldehyde and a 1% osmium tetroxide (Sigma Aldrich®) / 0.1M sodium cacodylate solution prior to storage in saline. Finally, samples were fractured, dehydrated, and sputter-coated with gold for imaging on a Hitachi 4800
High-resolution SEM©. In certain imaging studies, a control without any solvent
(CONTROL) is shown, since there were no observed morphological differences between corresponding CONs for CYTO-D and BLEBB.
5.3.3 Gene Expression
Samples were evaluated for gene expression as described previously [38].
Briefly, for each sample, 1 µg of RNA was isolated, converted into cDNA, and analyzed using standard real-time PCR with SYBR® Green on a StepOnePlus™
PCR System (Applied Biosystems). Primers were designed to assess mesodermal commitment (T-homeobox domain: BRACHY-T), and mesodermal differentiation (mesenchyme homeobox 1: MEOX1; paired box gene 2: PAX2; vascular endothelial growth factor receptor 2: FLK1). Additional analyses included ectodermal (nestin: NES) and endodermal (alpha-fetoprotein: AFP) commitment, as well as early hematopoiesis (runt-related transcription factor 1:
RUNX1). Gene expression levels were determined using standard curves and
83 reported normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH).
Samples are plotted as fold changes where each experimental sample was normalized by its corresponding control. When CYTO-D and BLEBB samples are plotted on the same graph, CON represents each treatment’s solvent-matched control.
5.3.4 Protein Expression
Flow cytometry (FCM) and immunohistochemistry were used to quantitate changes in protein expression. E-Cadherin (α-E-cadherin, R&D Systems®),
BRACHY-T (BRACHY-T, abcam®), and FLK1 (Santa Cruz Biotechnology, Inc.) expression in EBs was detected using standard protocols and fluorescently tagged (AF488 & AF546, Molecular Probes). For flow cytometry, fluorescence was detected for each sample using a BD FACSCanto II. Results are represented in fluorescence histogram and bivariate plots of fluorescence vs forward scatter formats. Immunohistological staining was visualized with a standard fluorescent microscope (Olympus®).
5.3.5 Statistical Analysis
Results are presented as mean ± standard error of the mean. Comparison between inhibitor-treated samples and their corresponding solvent controls were analyzed via student’s t-test to determine differences of a given treatment. We did not directly compare across different inhibitors for early versus late application.
84
5.4. RESULTS
5.4.1 CYTO-D and BLEBB altered cell and EB size
Initial studies explored the effects of CYTO-D and BLEBB on overall EB morphology at Day 2- Day 4 and Day 4- Day 7 treatment windows. At Day 4, quantification of EB cross sectional areas resulted in a significant decrease
(p<0.001) with CYTO-D and BLEBB treatment, indicating a decrease in EB size
(Figure 5.1A, TOP). Similarly to Day 2-4, CYTO-D treatment between Day 4 and
7 of differentiation resulted in a significant decrease (p<0.05) in EB size. In contrast, BLEBB treatment resulted in an increase in EB size in the later treatment window (Figure 5.1A, BOTTOM). Changes in EB size could result from alterations in the ability to adhere to neighbors, proliferation, apoptosis, individual cell size or shape, presence of cavitations, amongst other factors.
Immunohistological staining of E-Cadherin (ECAD) protein, a cell-cell adhesion molecule, revealed continuous areas of disruption along the outer layer of inhibitor-treated EBs (Figure 5.1B). SEM was used to investigate cell and matrix distribution within EBs. Both low and high magnification SEM images revealed cell rounding of cells treated with CYTO-D and BLEBB at Day 4 and Day 7
(Figure 5.2, Supplemental Figure 5.1A). In addition, BLEBB treatment induced an increase in the number and size of cavitations throughout the EB at Day 7 not
Day 4, while CYTO-D treated EBs showed no difference to controls at any time point. Forward scatter (FSC) and side scatter (SSC) in flow cytometry analyses have been shown to relate to cell size and granularity, respectively. Both CYTO-
D and BLEBB-treated cells seemed to increase FSC and SSC as compared to
85
CON (Supplemental Figure 5.1B). Hence, an increase in EB size could be due to increased number and size of cavitations or cell size. While, a decrease in EB size with concomitant increase in cell size may indicate the loss of cells upon inhibitor application due to a decrease in ECAD attachment capabilities. Taken together, changes in EB size may be caused by variations in ECAD expression, cell shape, and size of cavitations.
86
Figure 5.1 CYTO-D and BLEBB altered EB size and ECAD expression. (A) Phase images of EBs at Day 4 (TOP) and Day 7 (BOTTOM) with corresponding quantification of surfaces area for CYTO-D and BLEBB treatment groups and their corresponding controls (CON). Scale bar applies to all phase images. Data presented as mean ± SEM where statistical differences between CON and treatment groups is indicated by asterisks (*p<0.05, *** p<0.001). (B) Immunohistological staining of ECAD protein (RED) and nuclear stain (BLUE) for CON, CYTO-D and BLEBB treatment groups. Arrows (WHITE) show areas of disruption along the EB edge. Scale bar represents 100 μm.
87
Figure 5.2 CYTO-D and BLEBB produced a rounded cell shape. SEM images of CONTROL, CYTO-D, and BLEBB treatment groups at Day 4 (TOP) and Day 7 (BOTTOM) of differentiation. White arrows indicate cavitations at day 7. Scale bar lengths indicated in each image.
88
5.4.2 Inhibitor treatment decreased commitment to the mesodermal lineage
To determine the effects of CYTO-D and BLEBB on mesodermal commitment, BRACHY-T gene expression was assessed after the Day 2-4 and
Day 4-7 application windows. At Day 4, BRACHY-T was significantly downregulated (p<0.001) in CYTO-D samples, yet no change was detected in
BLEBB samples (Figure 5.3A). In contrast, at Day 7 BRACHY-T was significantly downregulated in both CYTO-D (p<0.001) and BLEBB samples (p<0.01; Figure
5.3A). Protein analysis of CYTO-D samples at both Day 4 and Day 7 demonstrated a wider distribution of cells expressing BRACHY-T. Bivariate plots indicated that a sub-population of cells expressed BRACHY-T at lower levels than CON (Figure 5.3B, TOP). No change was observed in the quantity of
BRACHY-T protein in BLEBB-treated samples as assessed by flow cytometry
(Supplemental Figure 5.2A). However, histological staining of BRACHY-T protein resulted in patchy expression in both treatment groups, contrasting the homogenous expression throughout controls (Figure 5.3C). The early endodermal marker, AFP, was uniformly upregulated across all treatment windows and inhibitors (Supplemental Figure 5.3). In contrast, NES was differentially regulated when CYTO-D or BLEBB were applied from Day 2-4 or
Day 4-7. CYTO-D treatment resulted in a significant downregulation (p<0.001) at
Day 4, and an upregulation (p<0.001) at Day 7. BLEBB significantly (p<0.05) decreased NES expression at Day 4 and did not alter its expression at Day 7
(Supplemental Figure 5.3). Taken together, mesodermal commitment was decreased by both CYTO-D and BLEBB at Day 7 and only by CYTO-D at Day 4,
89 while endodermal commitment was uniformly increased at both treatment windows.
90
Figure 5.3 Expression and distribution of BRACHY-T were altered with CYTO-D and BLEBB. (A) BRACHY-T gene expression of CYTO-D and BLEBB treatment groups and their corresponding controls (CON) at Day 4 and Day 7 of differentiation. CON samples shown in WHITE, CYTO-D in BLACK and BLEBB in BLUE. Data presented as mean ± SEM normalized to their corresponding CON (fold change) where statistical differences between CON and treatment groups is indicated by asterisks (*p<0.05, ** p<0.01, *** p<0.001). (B) BRACHY-T protein expression as measured by FCM of CYTO-D and CON at Day 4 (TOP) and Day 7 (BOTTOM). In histograms, 2nd antibody only controls are shown in GRAY for CON samples and light PINK for CYTO-D, while stained samples are shown in BLACK for CON and RED for CYTO-D (see arrows. Contour plots for FITC vs FSC are shown (RIGHT), where BLUE denotes highest concentration of events, and RED the least. (C) BRACHY-T protein expression in histological sections of CONTROL, CYTO-D and BLEBB EBs. Scale bar represents 100 μm.
91
5.4.3 CYTO-D and BLEBB modulated early mesodermal differentiation
Early mesodermal differentiation was evaluated through the quantification of MEOX1, PAX2, and FLK1 gene expression. No change was detected in
MEOX1 expression with CYTO-D or BLEBB supplementation at Day 2-4 (Figure
5.4A). When supplemented at later time points, BLEBB induced a 2.5X decrease in MEOX1 expression, while CYTO-D effects on MEOX1 were highly variable across 3 separate experiemnts (Figure 5.4C). At Day 4, CYTO-D upregulated
PAX2 (p<0.001), yet downregulated FLK1 (p<0.001) gene expression (Figure
5.4A). No change was detected with BLEBB treatment in either PAX2 or FLK1 during the early application window. At Day 7, both BLEBB and CYTO-D downregulated PAX2 (p<0.001) and FLK1 (p<0.001) gene expression (Figure
5.4C). For CYTO-D treated samples, protein analysis of FLK1 at Day 4 and 7 resulted in a broad distribution of fluorescence, as seen earlier for BRACHY-T
(Figure 5.4B). At Day 4, there was a small shift in mean expression to lower values, although it was not statistically significant. Interestingly, FLK1 analysis of
Day 7 samples resulted in a significant (p<0.05) increase in FLK1 protein expression of CYTO-D samples (Figure 5.4D). Moreover, CYTO-D samples had a significantly (p<0.01) higher expression of RUNX1, a hematopoietic progenitor marker (Supplemental Figure 5.4). BLEBB-treated samples did not show a shift in FLK1 protein expression (Supplemental Figure 5.2B). Y27632, an inhibitor of
Rho kinase (ROCK) was used to explore possible regulation by the ROCK-
Myosin II pathway in the observed results. MEOX1 (p<0.01), PAX2 (p<0.01), and
FLK1 (p<0.05) were significantly upregulated by Y27632 when applied from Day
92
2-4 (Figure 5.5). When Y27632 was applied at later time points, the effects on
MEOX1 and FLK1 were lost, and interestingly BRACHY-T (p<0.05) and PAX2
(p<0.05) were downregulated. Therefore, ROCK may be an upstream regulator of the effects by CYTO-D and BLEBB only at later time points. Taken together, early mesodermal differentiation was regulated by CYTO-D and BLEBB at later time points, while only by CYTO-D prior to Day 4 of EB differentiation.
93
Figure 5.4 CYTO-D and BLEBB reduced early mesoderm marker expression at Day 7, not Day 4. (A, C) Mesoderm marker expression of CYTO-D and BLEBB treatment groups and their corresponding controls (CON) at Day 4 (A) and Day 7(C). CON samples shown in WHITE, CYTO-D in BLACK and BLEBB in BLUE. Data presented as mean ± SEM normalized to their corresponding CON (fold change) where statistical differences between CON and treatment groups is indicated by asterisks (*** p<0.001). (B, D) FLK1 protein expression at Day 4 (B) and Day 7 (C). In histograms, 2nd antibody only controls are shown in GRAY for CON samples and light PINK for CYTO-D, while stained samples are shown in BLACK for CON and RED for CYTO-D (see arrows). Contour plots for FITC vs FSC are shown (RIGHT), where BLUE denotes highest concentration of events, and RED the least.
94
Figure 5.5 Y27632 increased early mesodermal differentiation at Day 4, while decreased it at Day 7. Markers of early mesodermal commitment were quantified (BRACH-T< MEOX1, PAX2, FLK1). CONTROL samples shown in WHITE and Y27632 in RED, data presented as mean ± SEM where statistical difference is indicated by asterisks (*p<0.05, ** p<0.01).
95
5.5 DISCUSSION
During in vivo development, mesodermal specification requires an orchestration of chemical and physical changes. It is not yet understood if changes in cytoskeletal organization and force generation are solely an effect, or perhaps a signaling cue for differentiation towards mesodermal phenotypes. In these studies, we investigated the role of actin polymers and internal force generation on the commitment and early differentiation of embryonic stem cells towards the mesodermal lineage. We found that treatment with either CYTO-D or
BLEBB induced changes in EB size and cell shape, with concomitant alteration in
ECAD expression. When mesodermal commitment was probed by quantifying
BRACHY-T, CYTO-D uniformly decreased BRACHY-T expression when dosed at both early (Day 2-4) and later (Day 4-7) stages of EB differentiation.
Contrastingly, BLEBB only mitigated BRACHY-T expression at the later treatment window. Early markers of mesodermal differentiation were downregulated by both CYTO-D and BLEBB from Day 4-7, while effects on mesodermal specification varied with CYTO-D and no change was detected with
BLEBB treatment from Day 2-4. Interestingly, ROCK inhibition resulted in similar effects to CYTO-D and BLEBB treatments at later time points. Hence, from Day
4-7, a reduction in both actin polymers and internal force generation decreased differentiation towards the mesoderm, while prior to Day 4 internal force generation may not be required for mesodermal commitment and differentiation.
Actin and myosin II have been shown to play a role in a number of cell functions, including cell shape. Actin polymers and microtubules have been
96 identified as being the major contributors establishing cell shape [163, 164], while actin and myosin II are together responsible for internally generated force [61,
165]. Alterations in cell shape can be accomplished through cell seeding on patterned substrates [39, 166, 167], alterations in adhesion to the underlying matrix [168], or through the use of small molecule inhibitors, such as CYTO-D
[169] and lantriculin B [170] for actin polymerization, as well as of nocodazole for microtubule polymerization [171]. Our early studies characterizing the effects of
CYTO-D and BLEBB on EB culture found that treatment with either inhibitor resulted in cell rounding. Hence, it appears that cell shape can be modulated by decreasing actin polymerization or inhibiting myosin II phosphorylation. While differentiation may induce changes in cell shape, cell shape in turn has been shown to regulate cellular differentiation [172, 173]. In particular, cell shape- regulated differentiation has been studied in adult [39, 174] and embryonic [169,
175] stem cells with particular applications to the directed differentiation of osteogenic [176] and adipogenic [169, 175] phenotypes. In our model of spontaneous differentiation, we found a significant increase in FLK1 protein and
RUNX1 gene expression when EBs were treated with CYTO-D at Day 4-7.
Hence, it seems that although overall mesodermal differentiation is decreased by a rounded cell shape, perhaps differentiation towards terminal cell types that are rounded, such as blood cells, benefit.
Cytoskeletal tension, determined by actomyosin contractions, has been found to play a role in stem cell pluripotency. Use of Y27632, an inhibitor of
97
ROCK, an upstream regulator of myosin II, has greatly improved the cloning efficiency of human embryonic stem cells (hESCs) that ordinarily undergo apoptosis when dissociated to single cells [78-80]. Subsequent studies have shown that depletion of myosin II by silencing or with inhibition by BLEBB also increased survival of hESCs [79, 81]. There are a limited number of studies that have studied the role of actomyosin contractility in stem cell differentiation.
Preliminary studies on hESCs showed no statistical significance between untreated and BLEBB-treated groups as quantified by gene analysis of ectoderm, mesoderm and endoderm markers [76]. However, use of Y27632 in the directed differentiation of p19 carcinoma cells led to an enhancement of ectodermal and mesodermal differentiation [82] and promoted the generation of endothelial cells from FLK1-sorted mouse ESCs [83]. We found that the effect of Y27632 on EB differentiation depended on time of application. At earlier time points, ROCK inhibition was found to promote mesodermal marker expression and at later time points it seemed to lower their expression. Hence, these results indicate that the decrease in mesodermal differentiation at later time points by both actin disassembly and Myosin II inhibition may be regulated by ROCK. Differences in response with point of application of the ROCK inhibitor may be due to its multiple targets, including MLCK (reviewed in [177]), LIM kinase [178] and ERM proteins [179]. Our studies focused on mesodermal commitment and differentiation and indicate that both actin polymers and cytoskeletal contraction are likely involved in mesodermal specification. Of particular interest are the differences observed prior to and after Day 4 of EB differentiation.
98
In vivo, the process of gastrulation gives rise to the three germ layers: endoderm, mesoderm, and ectoderm [16]. During gastrulation, mesodermal cells undergo cell shape changes, adopting a mesenchymal phenotype, as well as migrate through the primitive streak. Distinct mesodermal plates become allocated according to the time and site of ingression through the primitive streak
[180]. The earliest, most posterior population gives rise to extraembryonic tissues, such as the mesodermal layer of the chorion and blood islands [181].
Lateral plate, paraxial, and cardiac mesoderm emerge after from the intermediate and anterior levels of the streak, while the most anterior tip gives rise to the axial mesoderm (prechordal plate, notchord and node) and the definitive endoderm
[180]. In the mouse embryo, peak BRACHY-T and MIXL-1 expression coincides with the formation of the primitive streak and the onset of gastrulation [182] around day 6.5 of development [181, 183]. Peak expression of these genes is paralleled in the EB model of differentiation on Day 4 as shown by others [184-
187]. We found that prior to Day 4 of EB differentiation, inhibition of actin polymerization decreased mesodermal commitment (BRACHY-T) and specification to the lateral plate (FLK1) mesoderm, while increasing intermediate mesoderm (PAX2). While, after 4 days of differentiation both actin polymerization and myosin II-based contraction decreased mesodermal commitment and specification. Given the temporal similarity of the EB model and in vivo development, our results may suggest that myosin II contractions may be critical during the onset of gastrulation. Future studies should focus on targeting mesodermal phenotypes, perhaps by treating EB cultures with a ROCK inhibitor
99 prior to Day 4 and with a contractile agonist such as lysophosphatidic acid after
Day 4 of differentiation.
Creation of particular phenotypes, including those of the mesodermal lineage, for therapeutic purposes may benefit from a mechanistic understanding of early differentiation events. Our studies contribute to these efforts by identifying actin polymers and actomyosin contractions as regulators of mesodermal commitment and differentiation. Our results show that both CYTO-D and BLEBB alter cell shape. Therefore, we postulate that at the onset of gastrulation, failure to adopt an extended shape decreases mesodermal commitment. Contrastingly, prior to the onset of gastrulation it is unclear if actomyosin contractions play a role in mesodermal commitment and differentiation. Not only do these insights aid in the understanding of the physical regulation of early differentiation events, but will also serve to realize the full potential of stem cells as a viable option for therapeutic purposes.
In this chapter, we identified actin polymers, myosin II, and ROCK as regulators of mesodermal differentiation. Specifically, disassembly of actin polymers regulated commitment to different mesodermal plates when applied early and at later time points seemed to decrease mesodermal differentiation.
Myosin II inhibition only regulated commitment when applied at later time points.
We hypothesized that mesodermal and endodermal differentiation would be affected by alterations in both actin and myosin II, since the specification of both these lineages requires coordinated, temporally specific migration across the
100 primitive streak in vivo and along with microtubules, actin and myosin play significant role in cell migration. These findings in part support our overall hypothesis in that by perturbing the microfilament actin and the ability of myosin
II to generate force, we observed a decrease in mesodermal commitment. Yet, it seems that the role of actin and myosin II in mesodermal differentiation is not uniform across all time points. Modulation of actin and myosin II at particular points in differentiation led to different effects.
101
Supplemental Figure 5.1 CYTO-D caused a rounded cell shape and an increase in size. (A) Whole EB SEM pictures of CYTO-D samples showed a lack of a smooth epithelial layer and a rounded cell shape when compared to CON samples at Day 4 and 7. Scale bar length as depicted. (B) FSC vs SSC contour plots of CYTO-D, BLEBB, and their corresponding CON.
102
Supplemental Figure 5.2 BLEBB did not alter mesodermal protein expression. Histograms of BRACHY-T (A) and FLK1 (B) protein as measured by FCM.
103
Supplemental Figure 5.3 Endoderm and ectoderm markers were altered by CYTO-D and BLEBB. AFP (endoderm) and NES (ectoderm) gene expression at Day 4 and Day 7 of differentiation. CON samples shown in WHITE, CYTO-D in BLACK and BLEBB in BLUE. Data presented as mean ± SEM normalized to their corresponding CON (fold change) where statistical differences between CON and treatment groups is indicated by asterisks (*p<0.05, ** p<0.01, *** p<0.001).
104
Supplemental Figure 5.4 RUNX1 gene expression was increased with CYTO-D treatment. CONTROL samples shown in WHITE and CYTO-D in BLACK, data presented as mean ± SEM where statistical difference is indicated by asterisks (** p<0.01).
105
CHAPTER 6: CYTOCHALASIN-D TREATMENT INCREASED ENDODERMAL AND MESODERMAL DIFFERENTIATION OF STEM CELLS ON E-CADHERIN- AND FIBRONECTIN-COATED SURFACES
In the previous aim, we found that actin polymer disassembly and a decrease in cytoskeletal tension decreased mesodermal differentiation. In this chapter, we explored the effects of CYTO-D in 2D culture paradigms that distinguished cell-cell and cell-matrix dominated adhesion. This chapter addresses Aim 4: determine the effect of actin polymers on ESC differentiation in
2D.
6.1 ABSTRACT
Small molecule inhibitors may provide a facile method of directed differentiation. Use of CYTO-D, an inhibitor of actin polymerization, has shown to promote adipogenic phenotypes by others, while decreasing mesodermal phenotypes in three dimensions. It is not yet known whether these effects will occur in two dimensions. These studies investigated the effects of CYTO-D on mesodermal and endodermal differentiation of mouse embryonic stem cells
(ESCs) cultured in monolayer. In doing so, they attempted to elucidate the role of cell-cell and cell-matrix interactions in this response. ESCs were seeded on fibronectin- (FN) or e-cadherin- (ECAD) coated slides and assessed for early
106 markers of mesodermal (FLK1, PAX2) and endodermal (SOX17, FOXA2) differentiation. First, our model was validated using fluorescent tagging of adhesion molecules shown to accumulate at focal adhesions (paxillin) and cell- cell contacts (β-catenin). At junctions, FN- and ECAD-coated samples expressed primarily paxillin and β-catenin, respectively. Without CYTO-D treatment, there was no change detected in overall differentiation to the mesoderm and endoderm lineages. With CYTO-D treatment, cells on both substrates increased their expression of mesoderm and endoderm markers significantly. Markers of endothelial (TIE2) and hematopoietic (RUNX1) differentiation were increased by
CYTO-D on both substrates, although ECAD further increased RUNX1. Hence, independent of the type of junction molecules, culture in monolayer resulted in an increase in endodermal and mesodermal differentiation in response to CYTO-D treatment. These data aid in the understanding of force-regulated differentiation and may be used to design experiments targeting mesodermal and endodermal phenotypes.
6.2 INTRODUCTION
Uses of inhibitors of actin polymerization, such as cytochalasin-D (CYTO-
D), have elucidated the role of actin polymers in cell migration, proliferation, and differentiation. Disruption of actin polymers in mouse embryonic stem cells
(ESCs) grown in suspension as embryoid bodies (EBs) has been shown to decrease early mesodermal differentiation (refer to Chapter 5) and increase adipogenic differentiation [169]. In addition, EB-derived cells directed towards chondrogenesis showed enhanced differentiation with CYTO-D treatment [188].
107
The specific role of the physical microenvironment (2D vs 3D) in CYTO-D mediated differentiation is not yet known. Cell shape, nutrient availability, cell- cell, and cell-matrix interactions are among the factors that differ between 2D and
3D culture [189-191]. Our studies specifically focus on identifying the role of cell- cell and cell-matrix interactions in CYTO-D mediated differentiation. For this purpose, we seeded ESCs on fibronectin (FN) to engage integrins in cell-matrix junctions, and e-cadherin (ECAD) to engage homophylic ECAD junctions. This model allowed us to investigate the role of differential adhesion proteins on cytoskeletal-mediated differentiation towards the mesoderm and endoderm.
6.3 MATERIALS & METHODS
6.3.1 Embryonic Stem Cell Culture
Mouse D3 embryonic stem cells (ESCs; ATCC™) were cultured as previously described [38]. Cells were maintained on tissue culture plastic coated with gelatin in culture medium that consisted of Dulbecco’s Modification of Eagles
Medium (DMEM), 15% ESC-qualified fetal bovine serum (Invitrogen™), 2 mM L- glutamine, 0.1 mM non-essential amino acids, antibiotics, and supplemented with
1000 U/ml leukemia inhibitory factor (LIF; EMD Millipore©). To generate control
EBs, ESCs were placed in untreated dishes in culture medium without LIF
(0.5x106 cells/10 mL) and kept in constant motion on a rotary shaker (40 RPM;
New Brunswick©). For experimental samples, ESCs were dissociated using
Accutase (Life Technologies™) and seeded as single cells on glass slides coated with either fibronectin (FN; 3.5 µg/cm2; BD) or e-cadherin/FC chimera
108
(ECAD; 0.5 µg/cm2; R&D Systems®) per manufacturer’s protocols. Samples were cultured in an incubator (37°C, 5% CO2) for 18 hours, 24 hours, 48 hours,
2, 4, or 6 days in medium without LIF (20 mLs). For baseline mesodermal and endodermal differentiation studies, cells were seeded at 8,000 cells/cm2. To analyze the effect of actin depolymerization on mesoderm and endoderm differentiation, cells were seeded at 5,000 cells/cm2 in medium supplemented with cytochalasin-D (CYTO-D; 0.2 μM; Sigma Aldrich®) at Day 2 and changed every day until Day 4 of culture. For controls (CON), DMSO solvent was added in equal volume to CYTO-D.
6.3.2 Gene Expression
Experimental samples were evaluated for gene expression as described previously [38]. Briefly, for each sample 1 µg of RNA was isolated, converted into cDNA, and analyzed using standard real-time PCR with SYBR® Green on a
StepOnePlus™ PCR System (Applied Biosystems). Primers were designed to assess mesodermal commitment (vascular endothelial growth factor receptor 2:
FLK1; paired box gene 2: PAX2) and endodermal commitment (Sex Determining
Region Y)-Box 17: SOX17; Forkhead Box A2:FOXA2). Mesodermal markers for progenitors of endothelial (TEK Tyrosine Kinase, Endothelial: TIE2) and hematopoietic (runt-related transcription factor 1: RUNX1) phenotypes were assessed to probe further differentiation to the mesodermal lineage. Markers of pluripotency (Nanog Homeobox: NANOG; POU Class 5 Homeobox 1: OCT4,
109
Sex Determining Region Y-Box 2: SOX2) were quantified to assess potential retention of pluripotency of cells seeded on ECAD-coated substrates.
6.3.3 Protein Expression
Immunohistochemistry and immunocytochemistry were used to quantitate changes in protein expression. FN (Santa Cruz) and α-e-cadherin (R&D
Systems) proteins were visualized using standard histology protocols and fluorescently tagged (AF546, Molecular Probes™). For immunocytochemistry, samples were washed with PBS and fixed in 4% formaldehyde. For staining, samples were blocked with 1% horse serum for 1 hour at room temperature (RT) and incubated with primary antibodies for paxillin (EMD Millipore) and beta- catenin (EMD Millipore) for 1 hr at RT. After rinsing with PBS, samples were incubated for 1 hour at RT with a fluorescently tagged secondary antibody
(AF546, Molecular Probes) as well as AFP488-Phalloidin (Invitrogen) for actin polymer visualization. Experimental samples were imaged using phase microscopy and fluorescence microscopy. Confocal microscopy allowed for the visualization of adhesion molecules and was performed on both Nikon and Leica systems.
6.3.4 Statistical Analysis
Quantitative results are presented as mean ± standard error of the mean.
For kinetics data, we ran a two-way analysis of variance (ANOVA) on substrate
(FN vs ECAD) and time (0,2,4,6 Days) for each gene of interest. In studies using
110
CYTO-D, we compared treated samples vs. solvent controls using a student’s t- test.
6.4 RESULTS
6.4.1 Fibronectin and ECAD were expressed in embryoid bodies
Consistent with results found by others [192, 193], we found that by day 7 of differentiation, embryoid bodies expressed both FN and ECAD heterogeneously throughout the entirety of the EB. In particular, ECAD was expressed at the outer cell layer, as was FN (Figure 6.1A&B). Previous studies by us have shown that in this model of differentiation, perturbations of actomyosin contractions by CYTO-D or Blebbistatin led to a decrease in early mesodermal marker expression (see Chapter 5). Therefore, to understand the possible role of cell-matrix and cell-cell adhesions on CYTO-D-mediated differentiation, we seeded ESCs on substrates coated with FN or ECAD.
111
Figure 6.1 Embryoid bodies expressed ECAD and FN. (A) Immunohistological staining of FN (GREEN) and the nucleus (BLUE) in Day 7 EBs. (B) Immunohistological staining of ECAD (RED) and the nucleus (BLUE) in Day 7 EBs. Scale bars represent 200 µm.
112
6.4.2 FN and ECAD coatings allowed for the localization of distinct focal complex proteins
By using FN and ECAD coatings, we attempted to isolate cell-matrix and cell-cell interactions governed by integrins and cadherins, respectively. Since cells attach to fibronectin via multiple integrin subtypes, we stained cells seeded on FN- and ECAD-coated slides for paxillin, a focal adhesion protein specifically recruited to integrin-matrix junctions [194]. Similarly, we stained for beta-catenin, due to their engagement in homotypic ECAD junctions. To understand the time course of focal adhesion formation, samples were analyzed at 18, 24, and 48 hours of differentiation. As early as 18 hours, stress fibers had formed in a subpopulation of cells on both substrates (Figure 6.2). Overall, stress fibers were expressed heterogeneously: throughout the cell, cortically, or not at all. No differences were seen in the number of cells with well-developed stress fibers in either substrate (data not shown). Paxillin staining of FN-coated slides at all 3 time points showed its localization at the ends of stress fibers, indicating that cells were in fact attaching via integrins (Figure 6.2A, TOP). On the other hand, beta-catenin was expressed primarily at the nucleus and some in the cytoplasm in those same samples (Figure 6.2A, BOTTOM). Paxillin was primarily found to be expressed in the cytoplasm of cells on ECAD-coated substrates with no localization at the end of stress fibers (Figure 6.2B, TOP). Contrastingly, beta- catenin was observed to be expressed at the ends of stress fibers, much like paxillin was in FN-coated samples (Figure 6.2B, BOTTOM). Therefore, through
113 the differential coatings we were able to predominantly activate integrin or ECAD attachment.
114
Figure 6.2 ESCs attached through distinct adhesion proteins on FN and ECAD. (A) Immunohistological staining of stress fibers (GREEN) and paxillin or beta- catenin (RED) of FN samples at 18, 24, and 48 hours. (B) Immunohistological staining of stress fibers (GREEN) and paxillin or beta-catenin (RED) of ECAD samples at 18, 24, and 48 hours. Scale bars represent 30 µm.
115
6.4.3 Early mesoderm and endoderm markers were similarly expressed in cells on FN and ECAD
Markers of early mesodermal and endodermal differentiation were quantified to investigate baseline expression in cells cultured on FN and ECAD substrates. A two-way ANOVA was run for substrate (ECAD, FN) and time (2, 4,
6, days) and for each gene p-values were reported as psubstrate >0.05, ptime<0.05, and psubstrate*time >0.05 (Figure 6.3). Similarly, no change was detected in the expression of three pluripotency markers: OCT4, SOX2, and NANOG
(Supplemental Figure 6.1). Therefore, in the absence of guiding cues, cells seeded on FN and ECAD had similar mesoderm and endoderm marker expression up to Day 6 of differentiation. This finding allowed us to use this model for the mechanistic understanding of the effects of CYTO-D on ESC differentiation and the possible mediating role of integrins and cadherins.
116
Figure 6.3 Cells cultured on FN or ECAD similarly expressed early mesodermal and endodermal markers. Samples were evaluated for the expression of FLK1 and PAX2 (mesoderm) as well as SOX17 and FOXA2 (endoderm) at Day 0, 2, 4, and 6 of differentiation on FN (BLUE, squares) or ECAD (BLACK, circles).
117
6.4.4 CYTO-D increased the expression of mesoderm and endoderm markers on both substrates
CYTO-D treatment induced a rounded cell morphology and increased mesoderm and endoderm marker expression. Most cells in samples treated with
CYTO-D adopted a rounded cell morphology and lost stress fibers as seen in nuclear and stress fiber staining (Figure 6.4A&B MIDDLE, Supplemental Figure
6.3). Moreover, expression of paxillin and beta-catenin was no longer localized to stress fibers in FN- and ECAD-coated samples respectively (Figure 6.4A&B,
RIGHT). By Day 4 of differentiation, paxillin was seen at focal adhesions of
ECAD-substrates, while beta-catenin was not localized in FN-coated focal adhesions (Supplemental Figure 6.2). The early mesoderm marker, FLK1, was significantly increased in CYTO-D samples of both FN (p<0.01) and ECAD
(p<0.001) substrates, while PAX2 was significantly increased (p<0.05) in FN- coated samples, although the fold difference was lower than that of FLK1 (Figure
6.4C). Similarly to mesoderm markers, SOX17 and FOXA2, markers of endodermal differentiation, were significantly increased in both FN- and ECAD- coated substrates. SOX17 was upregulated 1.5-fold in FN-coated samples and
1.6-fold in ECAD-coated samples. FOXA2 was expressed at much lower levels and increased by similar fold changes in both substrates, yet with increased variability. Taken together, both mesodermal and endodermal markers were similarly upregulated by CYTO-D in both FN- and ECAD-coated substrates.
118
Figure 6.4 CYTO-D increased mesodermal and endodermal markers on FN and ECAD. (A) Phase (LEFT) and immunocytochemical (MIDDLE & RIGHT) images of cells cultured on FN. Stress fibers in GREEN, paxillin in RED, and nuclei in BLUE. (B) Phase (LEFT) and immunocytochemical (MIDDLE & RIGHT) images of cells cultured on ECAD. Stress fibers in GREEN, beta-catenin in RED, and nuclei in BLUE. Scale bars represent 200 µm in phase images (LEFT) and 30 µm in fluorescent images (MIDDLE & RIGHT). (C) Gene expression of early mesoderm (FLK1, PAX2) and endoderm (SOX17, FOXA2) markers in CON (WHITE) and CYTO-D (BLACK) samples on FN (LEFT) or ECAD (RIGHT). Expression was normalized to each treatment sample’s CON. Asterisks are used to represent significance (*p<0.05; **p<0.01; ***p<0.001).
119
6.4.5 Markers of endothelial and hematopoietic progenitor cells were upregulated by CYTO-D
Markers of endothelial and hematopoietic differentiation were explored to investigate downstream effects of CYTO-D on mesodermal differentiation. Both endothelial and hematopoietic progenitor markers were significantly increased by
CYTO-D in either substrate, yet the relative magnitudes differed. An early endothelial marker, TIE2, was increased by CYTO-D 2.1-fold (p<0.001) on FN- substrates and 1.6-fold (p<0.001) on ECAD-substrates (Figure 6.5A&B).
Conversely, the hematopoietic progenitor marker RUNX1 was increased by 1.7X
(p<0.01) on FN-substrates and 2.7X (p<0.001) on ECAD substrates. Therefore, later markers of mesodermal differentiation were increased by CYTO-D on both substrates, yet the relative differences between endothelial and hematopoietic markers differed.
120
Figure 6.5 CYTO-D upregulated the expression of endothelial and hematopoietic markers. (A) Expression of TIE2 in CON (WHITE) and CYTO-D (BLACK) samples on FN (LEFT) or ECAD (RIGHT). (B) Expression of RUNX1 in CON (WHITE) and CYTO-D (BLACK) samples on FN (LEFT) or ECAD (RIGHT). Significance is represented by asterisks (**p<0.01; ***p<0.001).
121
6.5 DISCUSSION
Actin polymers play a role in a variety of cell processes, including cell migration, adhesion, force transfer, and differentiation. In these studies, we investigated the role of cell-matrix and cell-cell adhesions on the stress fiber- mediated mesodermal and endodermal differentiation of ESCs. First, we verified that FN and ECAD protein coatings would allow for differential attachment of
ESCs by evaluating signaling molecules found at cell-matrix and cell-cell junctions. Paxillin, found at integrin-matrix focal contacts [194] and beta-catenin, linking alpha-catenin and the actin cytoskeleton at cell-cell contacts [51, 195,
196], were found to localize at focal adhesions formed on FN- and ECAD-coated substrates respectively. Then, we quantified the kinetics of mesodermal and endodermal differentiation. With intact actin polymers, ESCs had similar differentiation patterns of early mesodermal and endodermal markers up to Day
6 of differentiation. Upon addition of CYTO-D, both mesodermal (FLK1, PAX2) and endodermal (SOX17, FOXA2) markers were similarly upregulated by CYTO-
D in FN- and ECAD-coated substrates. Furthermore, later markers of mesodermal differentiation, TIE2 and RUNX1, were increased with CYTO-D treatment. Taken together, these data indicate that the CYTO-D-mediated upregulation of early mesodermal and endodermal markers in 2D does not depend on the type of adhesion protein used to attach to the physical microenvironment.
Both integrins and cadherins have been shown to regulate stem cell pluripotency and differentiation. For instance, culture of ESCs on collagen type I
122 and type IV maintained pluripotency, while on fibronectin and laminin did not
[197]. Cells attach to extracellular matrices through different integrin subtypes, including α5, αv, β1, β3 to fibronectin [198], while α1 and α2 to collagen type I
(reviewed in [199]). Therefore, differences in differentiation may be due to the identity of cell adhesion proteins used for attachment to the physical microenvironment, or perhaps the signaling molecules that complex at these junctions. Various integrin subunits have been implicated in differentiation processes. Integrin α5-null embryos had pronounced defects in mesodermal structures [200], while integrin β1-deficient ESCs were able to differentiate to blood phenotypes, yet their homing capacities (in vivo migration) were limited
[201]. Cadherins, a class of type 1 transmembrane proteins responsible for intracellular adhesion, also play a role in self-renewal [202, 203]. In fact, ECAD expression is required for acquiring and maintaining pluripotency in induced pluripotent stem cells [202] and its absence results in a wide range of alterations at the gene level [204]. Culture of mesenchymal stem cells on an ECAD protein substrate, similar to the one used in our studies, increased proliferation and decreased spontaneous differentiation [205]. In human ESCs, ECAD substrates allowed for the maintenance of pluripotency [206], while increasing their proliferation [207]. In the context of differentiation, a substratum combining ECAD and N-cadherin, allowed for the highly homogenous derivation of ectodermal phenotypes [208]. Furthermore, ECAD protein substrates enhanced the derivation of endodermal phenotypes from ESCs using activin A [209]. Our findings showed that in the absence of guidance cues towards the endoderm,
123
ECAD substrate does not promote endodermal differentiation. Yet, CYTO-D mediated differentiation towards hematopoietic progenitors may be enhanced in
ESCs seeded on ECAD substrates. In parallel, studies by others have found that abrogation of ECAD in ESCs modified the cell response to external factors [210].
Hence, perhaps ECAD substrates may allow for ESCs to respond to chemical cues differently than when attached via integrins.
Both cadherins and integrins are capable of transmitting force across the cell membrane [51, 138]. Internally generated force via actomyosin contractions is transferred to the underlying substrate to generate traction. In the same fashion, externally applied forces such as shear stress and tensile strain are transferred to the intracellular environment at both these junctions. However, a study comparing force transfer across integrins versus ECAD in ESCs found that although cell stiffening occurs in both systems, differentiation and cell spreading responses were not seen in ECAD-transferred force [41]. In our studies, alterations in internally generated force resulted in an increase in endodermal and mesodermal when cells were attached to FN or ECAD, although effects on hematopoietic progenitor differentiation by CYTO-D were in fact amplified on
ECAD versus FN. Therefore, in our studies the identity of the adhesion protein did not alter overall differentiation. Perhaps, by breaking the connection between junctions and the nucleus, signaling molecules at both junctions were similarly disrupted and the dominating response is due to the lack of cellular tension. We observed an inverse response in 2D and 3D when comparing current results to
124 previous ones that used CYTO-D in a 3D model (refer to Chapter 5).
Mesodermal commitment was decreased by CYTO-D in 3D and increased in 2D.
Perhaps, instead of ligand identity, it is the initial state of adhesion that is responsible for differences in 2D and 3D response to cytoskeletal inhibition. In
2D, cells are more spread out and are able to develop stress fibers in a more defined way than in 3D. This increase in stress fiber formation may lead to increased cellular tension. Moreover focal adhesions [211] and cell-cell contacts
(reviewed in [212]) have been shown to increase in size with increased tension, hence the combination of altered tension and differences in the maturity of contacts may account for 2D and 3D disparities in response to actin disassembly.
In our studies, FN- and ECAD-coated substrates were used to investigate the role of cell-matrix and cell-cell junction molecules on endodermal and mesodermal differentiation. Our results are limited in that by Day 4 of differentiation, the cell-matrix adhesion marker paxillin was observed in ECAD- coated samples (Supplemental Figure 6.2). This is possibly due to the fact that cells deposit their own matrix and start using integrins for attachment. Another limitation is that definitive endoderm markers were not evaluated. We ended culture at 4 days to prevent overconfluence and excessive matrix deposition that could confound results. Future studies should focus on investigating the role of cell-cell versus cell-matrix interactions on force-mediated differentiation. In particular, shear stress has been shown to increase endothelial and hematopoietic phenotypes in ESCs grown on Collagen type IV [213]. Use of an
125
ECAD substrate with the synergistic application of shear stress and CYTO-D may further promote hematopoietic differentiation.
Modulation of stress fibers may provide a facile method of directed differentiation. In contrast to earlier studies in 3D, these studies indicate that actin depolymerization in ESCs cultured on FN or ECAD resulted in an increase of early mesodermal and endodermal differentiation markers. Perhaps then, it is the initial stress fiber state or substrate elasticity that may have caused the inverse response to actin fiber depolymerization in 2D and 3D. Taken together, we found that stress fiber dissociation led to an increase in mesodermal and endodermal differentiation independently of adhesion molecules used to bind to the microenvironment. These findings contribute to the understanding of force- regulated differentiation and may be used to design experiments targeting mesodermal and endodermal phenotypes.
This chapter identified a role for actin polymers in the differentiation of
ESCs grown in 2D towards mesodermal and endodermal phenotypes, by which a decrease in actin polymers led to an increase in markers of these two lineages.
In aim 2, we found that culture in 2D resulted in an overall higher expression of cytoskeletal proteins as compared to 3D. We also found that the adhesion proteins used to attach to the microenvironment (cell-cell vs. cell-matrix) did not seem to play a role in the increase in mesodermal and endodermal differentiation by actin disassembly. Perhaps then, it might be the initial level of microfilament expression as well as the initial level of cytoskeletal tension (force balance
126 across the membrane) that may affect the directionality of the differentiation response to alteration in actin polymerization. Similar to Aim 3, these results further support our overall hypothesis since a perturbation in microfilament structure resulted in an alteration of markers of the mesodermal and endodermal lineages.
127
Supplemental Figure 6.1 No change was detected in pluripotency markers when seeded on FN or ECAD substrates upon the removal of LIF. Samples were evaluated for the expression of pluripotency markers (OCT4, SOX2, NANOG) at Day 0, 2, 4, and 6 of differentiation on FN (BLUE, squares) or ECAD (BLACK, circles).
128
Supplemental Figure 6.2 By Day 4 of culture, ECAD samples expressed paxillin at focal adhesions. (A) Paxillin (RED) immunostaining of cells grown on ECAD. (B) Beta-catenin (RED) immunostaining of cells grown on FN. Stress fibers are shown in GREEN and nuclei in BLUE.
129
Supplemental Figure 6.3 Cells treated with CYTO-D were a heterogeneous population, some balled up and some spread. Projection of a z-stack of cells grown on FN and treated with CYTO-D. Stress fibers are in GREEN and nuclei in BLUE.
130
CHAPTER 7: DISCUSSION AND FUTURE STUDIES
Even with the aid of expensive growth factors, small molecules, and/or selection techniques, current differentiation protocols produce a heterogeneous population of cells. In order to better direct differentiation to a specific phenotype, a basic understanding of the mechanisms that regulate pluripotent cell differentiation would be quite beneficial. In the presented works, we aimed at elucidating the role of the cytoskeleton in the spontaneous differentiation of PSCs in 2D and 3D microenvironments. First, we quantified the expression of the cytoskeleton in ESCs, iPSCs, and the iPSC source phenotype, showing that there were indeed differences in the expression of microfilaments and certain intermediate filaments among all three phenotypes. Next, we found that there were inherent differences in PSC differentiation when cultured in 2D and 3D microenvironments. Lastly, alterations in the cytoskeleton were found to decrease mesodermal differentiation in 3D, while increase both mesodermal and endodermal differentiation in 2D. Taken together, we identified the cytoskeleton as a regulator of differentiation to the mesodermal and endodermal lineages in both 2D and 3D.
There are limitations to the results found in these studies. First off, in comparing culture systems, the ratio of medium to cells was not maintained constant. This fact is particularly important in the context of cytoskeletal inhibition
131 by small molecules, since the amount of inhibitor added to each sample is based on the volume of media in culture. Yet, there is no reason to suspect that the concentration was not sufficient in each of the culture paradigms. Other differences include possible diffusional limitations in cells cultured in suspension or in gels that are not found in cells cultured as a monolayer. Although these data provide an idea of the function of the cytoskeleton in baseline differentiation, it does not provide information on its role in directed differentiation with additional agents. For this reason, future experiments should incorporate additional cues for directed differentiation such as growth factors and applied physical forces
(Appendix C). The synergistic application of these cues with the modulation of the cytoskeleton may prove to enhance the generation of a targeted cell population. In particular, the application of shear stress and CYTO-D to cells seeded on an ECAD-coated substrate may lead to the enhancement of hematopoietic phenotypes.
These basic science studies need to be adapted to scale-up culture systems in order generate enough cells for clinical applications. Such systems use large volumes and it may be cost-prohibitive to use high concentrations of small molecules. Therefore, the minimal dose of inhibitors should be identified for these applications. In addition, these studies were performed in the mouse model. In order for human use, they will need to be replicated in human PSCs, preferably hiPSCs.
132
There are various hurdles to stem cell research translation to medical therapeutics, including use of human embryonic tissues. With the advent of iPSCs, this hurdle may now be circumvented. Yet, iPSCs have been shown to somewhat differ in their differentiation potential to ESCs. In our first set of studies, we found that the expression of microfilaments and certain intermediate filaments in iPSCs was higher than that of ESCs. This fact may be one of the reasons for differences in differentiation capacities found by others [214].
Therefore, further studies should characterize the differentiation potential of iPSC compared to ESCs to make use of current knowledge on directed differentiation, as well as on their own. These efforts will aid in the realization of stem cells in medical therapeutics.
Currently, a combination of suspension and adherent culture are used for the generation of specific cell phenotypes, yet a mechanistic understanding of the physical microenvironment as a modulator of differentiation is lacking. Our initial studies maintained a number of variables constant to allow for the direct comparison of different culture paradigms. Hence, we were able to ascertain the role of 3D culture in cardiovascular differentiation. Studies similar to these, which systematically identify design criteria for differentiation to specific phenotypes, will hopefully accelerate stem cell therapeutics.
Basic science understanding of the mechanisms by which PSCs relate to the physical microenvironment will aid the continuous effort to better control their differentiation. Specifically, knowledge of the functions of the cytoskeleton in the
133 transfer of signals to and from the nucleus is of particular interest. Our later studies focused on identifying the role of proteins at the membrane that link the cell to its environment, as well as microfilaments and the myosin II motor on early differentiation. We found that both intact actin polymers and actomyosin contractions played a role in differentiation towards mesodermal phenotypes.
Although actin polymer dissociation was also found to modulate differentiation to the mesoderm in monolayer culture, its effect was reversed. It seems that modulation of the microfilaments induced separate outputs depending on a cell’s initial state. To clarify these findings, it would be beneficial to characterize the cytoskeletal morphology of cells in 2D and 3D to assess the initial conditions of that which we inhibit. These studies provide a baseline understanding of the role of the cytoskeleton in differentiation, yet a complete understanding of the pathways by which they regulate gene expression remain unclear. Although microfilaments and microtubules are regulated by similar pathways, in these studies we were unable to identify the role of microtubules in differentiation since use of Nocodazole, an inhibitor of microtubule polymerization, resulted in mitotic arrest and disruption of EBs (Appendix E).
In two or three dimensional culture, a terminal cell state would be determined by factors that alter the expression of key genes at the nucleus. One approach for future work would be to systematically silence elements at the cellular membrane (integrins, cadherins), the cytoskeleton (actin, intermediate filaments), and elements at the nuclear membrane (Nesprin, SYNE2). In this
134 manner, for a particular physical input (such as shear stress or tensile strain) one could potentially follow an input signal from membrane to the nucleus. Another approach would be to modulate signaling molecules at the cell membrane that are able to translocate to the nucleus. For example, β-catenin has been shown to complex at cell-cell connections as well as translocate to the nucleus and serve as a transcription factor (reviewed in [215]). Hence, future works should focus on understanding the dependence of signaling molecules (such as beta-catenin) that regulate gene expression and protein synthesis and modification on an intact cytoskeleton.
Taken together, immediate studies should focus on understanding the mechanisms by which cytoskeletal inhibition alters differentiation as well as the design of experiments to further promote particular lineages. First, in order to obtain a baseline cytoskeletal state of cells in 2D and 3D microenvironments, studies should focus on the immunostaining of microfilaments, intermediate filaments and microtubules. Secondly, cytoskeletal contractile agonists should be applied in a 3D model after a period of ROCK inhibition to further promote differentiation towards the mesodermal lineage. In these studies, markers of mature mesodermal phenotypes should be quantified at the gene and protein level. Thirdly, silencing of key signaling molecules at the cell (β-catenin) and nuclear (SYNE2) membranes known to interact with the cytoskeleton would allow for the exploration of mechanotransduction pathways to the nucleus. Later studies should focus both on the translation of these findings to human iPSCs as
135 well as understanding the role of the cytoskeleton in somatic cell reprogramming itself.
The overall objective of this dissertation was to determine the expression of cytoskeletal elements in PSCs as well as identify the role of the cytoskeleton in differentiation. Our overarching hypothesis was that cytoskeletal expression in
PSCs would be lower than that of adherent phenotypes such as fibroblasts, and that perturbation of the cytoskeleton either by culture in different microenvironments or with the use of protein inhibitors of actin and myosin II would alter differentiation towards the three lineages. Our results indicate that our initial hypothesis was too simplistic. In fact, when PSCs were compared to MEFs, they had different levels of microfilaments, similar levels of microtubules, and that cytokeratin levels changed depending on the specific subtype. In addition, it seems that perturbation of either actin or myosin II in fact altered differentiation, yet the directionality of the change was dependent both on time and the initial state of the proteins being perturbed. Taken together, our new hypothesis is that cytoskeletal elements have unique expression levels across cell types of varying pluripotent capacities and that microfilaments and myosin II play a temporally and microenvironment-specific role in the differentiation of ESCs towards mesodermal and endodermal lineages. Future studies should aim at elucidating the mechanisms by which alterations of microfilaments and myosin II result in both gene and protein changes that determine differentiation. In addition, other proteins other than actin and myosin II should be considered as regulators of
136 differentiation, including intermediate filaments and microtubules. Future studies should perturb upstream regulators of microtubules to elucidate their role in differentiation, due to the inability of their direct perturbation due to their role in cell proliferation and viability (Appendix E).
The physical microenvironment provides critical signals both for the maintenance of pluripotency and in differentiation. Cytoskeletal inhibitors have been widely used to investigate the role of the cytoskeleton in mediating the response to externally applied physical force. In these studies, we used inhibitors to identify the role of cytoskeletal tension in differentiation. In conclusion, our studies identify the role of the cytoskeleton in modulating differentiation towards mesodermal and endodermal phenotypes. Moreover, they highlight the need to carefully design experiments with dimensionality and culture conditions in mind.
Taken together, although most studies are done in a 2D microenvironment, our studies suggest that translation of parameters ascertained in 2D may not map to those in 3D for a desired result. Moving forward, an understanding of the mechanisms by which bidirectional signaling from the extracellular environment and the nucleus via the cytoskeleton (microfilaments, intermediate filaments, and microtubules) is converted into genetic and protein changes will aid in the design of experiments that target specific phenotypes for medical therapeutics.
137
APPENDIX
APPENDIX A: EFFECT OF Y27632 ON MESODERMAL AND ENDODERMAL DIFFERENTIATION IN 2D CULTURE
Rho kinase (ROCK) was investigated as a possible upstream signaling pathway that may be responsible for the increase in mesoderm and endoderm differentiation with application of CYTO-D. Y27632 (10 µM), an inhibitor of
ROCK, was supplemented to samples grown on FN or ECAD substrates for 2 days (Day 2-4). Y27632 induced an increase in BRACHY-T and FOXA2 on FN.
Both effects are lost when seeded on ECAD. No changes were observed in
MEOX1, PAX2, or SOX17 on either substrate. FLK1 was significantly downregulated by Y27632 on ECAD, yet on FN controls are too variable to know if this effect is mirrored. Perhaps the increase in the mesendodermal marker
BRACHY-T corresponds to an increase in endoderm, not mesoderm.
138
Figure A.1 Effects of Y27632 in 2D culture of ESCs. (A) Phase images (4X) of cells on FN and ECAD at 4 days of culture, either treated with Y27632 (BOTTOM) or H2O CON (TOP). (B) Expression of BRACHY-T at Day 4. (C) Expression of early mesodermal markers (MEOX1, PAX2, FLK1). (C) Endodermal marker expression. CON samples in WHITE, Y27632 samples in BLACK. Data expressed as mean +/- SEM. Significance represented with asterisks (*p<0.05, **p<0.01).
139
Therefore, ROCK may specifically increase endodermal differentiation, while mesodermal differentiation remains unchanged. In conclusion, it seems that the inhibition of ROCK, an upstream regulator of actomyosin contractions, does not recapitulate the effects observed in Chapter 6. Perhaps then, it is not the reduction of actomyosin contractions that induces large increases in both endodermal and mesodermal differentiation.
140
APPENDIX B: EFFECTS OF RA AND BMP4 ON MESODERMAL DIFFERENTIATION
Growth factors are commonly supplemented in the media to direct differentiation. The following studies are preliminary in that they evaluate the effects of two factors: Retinoic Acid (RA; 1 µM) and Bone Morphogenetic Protein
(BMP4; 10 ng/mL) on the early mesodermal differentiation of ESCs. ESCs were differentiated as EBs (0.5E6 cells/100 mm dish) and the factors applied continuously from Day 0- Day 6 to understand their effects with respect to time.
Moreover, their effects were analyzed at early (2-4) or late (4-6) treatment windows.
Continuous application of RA at 1 µM did not allow for proper EB shape and size (Figure B.1A). This may be due to apoptosis or arrest of proliferation. At
Day 2, samples treated with RA had higher expression levels of MEOX1 (80X) and FLK1 (4X) (Figure B.1B). Application of RA in treatment windows spanning 2 days allowed for EBs to retain their shape, but may have decreased their size when applied early on (Figure B.2A). BRACHY-T was downregulated by RA application at 2-4 and 4-6 treatment windows by 20X and 75X, respectively
(Figure B.2B). Whereas, MEOX1 was upregulated by RA at 2-4 (600X) and 4-6
(20X). Interestingly, FLK1 response to RA was opposite at early (upregulation,
3X) and late (downregulation, 5X) treatment windows.
141
Figure B.1 Effects of RA on EB morphology and differentiation towards the mesoderm. (A) Phase images (4X) of EBs at Day 2, 3, and 6 with and without RA supplementation. (B) Gene expression levels of BRACHY-T, MEOX1, and FLK1 mesoderm markers of Control (GREEN) and RA-treated (BLACK) samples. N=1 per time point.
142
Figure B.2 Effects of RA on EB morphology and differentiation towards the mesoderm when applied in early and late treatment windows. (A) Phase images (4X) of EBs at Day 4 and 6 with and without RA supplementation. (B) Gene expression levels of BRACHY-T, MEOX1, and FLK1 mesoderm markers. n=1 per sample.
143
Taken together, it seems that RA may be a useful avenue for generating cells of the mesodermal lineage in EB culture when applied at early time points.
Future studies should investigate the reduction of RA concentration by 10-100X, since higher doses of RA have shown to promote both mesodermal and endodermal differentiation [216].
Continuous application of BMP4 did not alter EB size or morphology
(Figure B.3A). BMP4 accelerated the expression of BRACHY-T, MEOX1, and
FLK1, yet by Day 6 levels were similar to controls (Figure B.3B). When applied for 2 days early and late in EB differentiation, BMP4 did not alter EB size or shape (Figure B.4A). BRACHY-T (3X), MEOX1 (5X), and FLK1 (10X) were upregulated with BMP4 treatment from Day 2-4 (Figure B.4B). When BMP4 was applied at later time points these effects were lost or reversed. BRACHY-T was downregulated by 2X, while MEOX1 and FLK1 levels of CON and treatment were similar. Therefore, BMP4 may drive mesodermal differentiation when applied at early time points only.
144
Figure B.3 Effects of BMP4 on EB morphology and differentiation towards the mesoderm. (A) Phase images (4X) of EBs at Day 2, 3, and 6 with and without BMP4 supplementation. (B) Gene expression levels of BRACHY-T, MEOX1, and FLK1 mesoderm markers. n=1 per time point.
145
Figure B.4 Effects of BMP4 on EB morphology and differentiation towards the mesoderm when applied in early and late treatment windows. (A) Phase images (4X) of EBs at Day 4 and 6 with and without BMP4 supplementation. (B) Gene expression levels of BRACHY-T, MEOX1, and FLK1 mesoderm markers. n=1 per sample.
146
These data prove the possible use RA & BMP4 to drive mesodermal differentiation. Therefore, future studies may focus on understanding the effect of the cytoskeleton and cytoskeletal tension in the development of more downstream phenotypes.
147
APPENDIX C: EFFECT OF TENSILE STRAIN ON ESC CYTOSKELETAL AND GERM LINEAGE MARKER EXPRESSION
Tensile strain has been shown to affect differentiation of stem cells. In
ESCs, tensile strain has recently been shown to decrease pluripotency [217].
These preliminary studies examine tensile strain as a possible driver of cytoskeletal expression and mesodermal differentation. ESCs were seeded in collagen type I gels (1E6 cells/gel; 2 mg/mL collagen) for a 3D model and on a collagen type IV-coated flexible membrane (5,000 cells/cm2).
Collagen Type I gels deformed after application of strain application (2-
12%, sine) for 1 day at either 0.1 or 0.5 hz (Figure C.1). Cyclic tensile strain decreased BRACHY-T expression when applied at Day 3 and 6 of collagen type I gel culture (Figure C.2A LEFT). At Day 6, this response was observed at 0.1 hz as well as 0.5 hz (Figure C.2A RIGHT). Skeletal actin (ACTA1) on the other hand was increased by tensile strain at Day 3 and Day 6 of culture at both 0.1 and 0.5 hz (Figure C.2B). Delayed application of strain on Day 10 led to a loss of both
BRACHY-T and ACTA1 responses. A small, but significant increase in the nuclear cytoskeleton (LMNA) (Figure C.2C) was observed, while no change was detected in the early mesodermal marker FLK1 (Figure C.2D).
148
Figure C.1 Gross images of ESCs seeded in collagen type I gels. ESCs seeeded in gels were allowed to grow for 6 days prior to application of tensile strain. Cyclical uniaxial strain (2-12%) was applied for 24 hours at 1 hz. After the strain regimen, gels were seen to slack.
149
Figure C.2 Gene expression analysis of ESCs grown in collagen type I gels and exposed to 1 day of strain. (A) BRACHY-T expression of samples strained at Day 3, 6, and 10 of polymerization (LEFT) and at 0.1 and 0.5 hz at Day 6 (RIGHT). (B) ACTA1 expression of samples strained at Day 3, 6, and 10 of polymerization (LEFT) and at 0.1 and 0.5 hz at Day 6 (RIGHT). (C) Expression of LMNA (LEFT) and FLK1 (RIGHT) after strain application at Day 6 at 0.5 hz. STATIC samples in WHITE, equibiaxial in BLACK and uniaxial in DASHED. All data expressed as mean +/- SEM. Statistical significance represented by asterisks (*p<0.05, **p<0.01, ***p<0.001).
150
Cells were seeded on PDMS flexible membranes using H2SO4 acid etching. Cells attached and proliferated in both STATIC and STRAIN conditions
(Figure C.3A). Overall, equibiaxial strain did not alter the expression of cytoskeletal or germ lineage markers (Figure C.3B&C). Although, there were small, but significant (p<0.05) decreases in ACTA2 and VIM (Figure C.3B). When samples were strained using uniaxial strain for comparison and ACTA2 was evaluated via PCR (Figure C.3D). It appears that results were more dramatic when uniaxial strain is applied instead of equibiaxial.
151
Figure C.3 Application of equibiaxial strain to ESCs. (A) Phase images (4X) of ESCs pre-treated for 2 days and treated under STATIC or STRAIN conditions for 2 days. (B) Cytoskeletal protein marker expression at Day 4. (C) Germ lineage marker expression at Day 4. (D) Comparison of ACTA2 expression of samples strained with uniaxial and equibiaxial strain regimens. STATIC samples in WHITE, equibiaxial in BLACK and uniaxial in DASHED. All data expressed as mean +/- SEM. Statistical significance represented by asterisks (*p<0.05, **p<0.01).
152
Taken together, tensile strain did not dramatically alter expression of cytoskeletal protein or germ lineage markers. Moreover, the 3D collagen gel model may introduce inhomogeneity of force transfer with time due to the
“slacking” of the gels. Future studies should use uniaxial strain to further investigate the response to strain in 2D.
153
APPENDIX D: TAIL-TIP IPSC EXPANSION & PLURIPOTENCY
iPSCs may be obtained from an array of source phenotypes including adult fibroblasts. These preliminary studies expanded tail-tip iPSCs
(STEMGENT, Cat #08-0006) on MEF feeder layers as explained in Chapter 3.
These initial studies showed a low number of dome-like highly refractive colonies, yet similar NANOG protein expression (Figure D.1A). Similar to iPSCs derived from MEFs used in Chapter 3, they had lower expression of SOX2 protein, although it was higher than that of Day 6 EBESC.
154
Figure D.1 Mouse primary iPSCs express NANOG, while have a low number of dome-like colonies and lower SOX2 expression as compared to ESCs. (A) Phase pictures (4X) of tt-iPSCs grown for 5 days in gelatin-coated flasks. (B) Flow cytometry results of NANOG and SOX2 protein, where 2nd only samples are in solid colors, tt-iPSCs in BLUE, ESCs in GREEN, and Day 6 EBESC in BLACK.
155
APPENDIX E: EFFECTS OF TUBULIN INHIBITION ON EBS
Microtubules are cytoskeletal elements with multiple cell functions including structural support and proliferation. In these preliminary studies we investigated their effects on EB culture. Nocodazole (1 µg/mL; Sigma©; Cat #
M1404) was used as a microtubule inhibitor and applied daily from Day 2-4 of differentiation. Treated EBs were smaller in size compared to their matched controls at both Day 3 and Day 4 (Figure E.1A). Moreover, there were a lower number of cells in culture after 4 days (Figure E.1B). Cell viability was quantified
(Invitrogen™) to investigate the possibility of cell death due to Nocodazole treatment (Figure E.1C). ESCs were cultured in monolayer for 24 hours and then assessed for viability. No difference was detected in the number of dead cells
(RED), yet a lower number of live cells (GREEN) were present. In order to pursue further analyses with Nocodazole, a range of lower doses should be tested (0.01, 0.1 µg/mL) for their effects on proliferation and EB size. It may however, not be possible to study differentiation with this inhibitor, since differentiation occurs in the order of days while proliferation in the order of hours.
156
Figure E.1 Effects of Nocodazole on EB culture. (A) Phase images (4X) of EBs before inhibition at Day 2 (TOP) and at Day 4 (CONTROL & Nocodazole). (B) Cell number in 1 dish of EBs at Day 4 with or without Nocodazole treatment. (C) Live/dead assay of cells grown on adherent surfaces and treated with Nocodazole for 24 hours. Phase images (LEFT) and fluorescent images (RIGHT). Live cells are in GREEN and dead cells in RED.
157
APPENDIX F: CYTOSKELETAL INHIBITION IN STATIC VS. SHEAR EBS
Initial studies investigated the cytoskeletal inhibition of EBs grown statically (STATIC) and under constant motion on a rotary shaker (SHEAR; 40
RPM) for 7 days and treated with inhibitors of ROCK (Y27632; 10 μM) and
Myosin II (BLEBB; 24 μM) (Figure F.1A). Morphological differences were observed between STATIC and SHEAR CONTROL samples, but not in STATIC and SHEAR treated samples (Figure F.1B, LEFT). Both STATIC and SHEAR treated samples evidenced areas of translucence and numerous cavitations not present in CONTROL samples (Figure F.1B, RIGHT). When EB size was quantified, only SHEAR EBs were significantly (p<0.001) larger (Figure F.1 C).
Markers of pluripotency were similarly downregulated with inhibition of ROCK and Myosin II from Day 4-7 of EB culture (Figure F.2A). In addition, ECAD protein was similarly disrupted in both STATIC and SHEAR groups (Figure F.2B).
Unfortunately, these studies have a major drawback since they were not completed with a corresponding solvent control; rather CONTROL samples represent culture medium with a lack of the small molecule inhibitor.
158
Figure F.1 ESCs differentiated as embryoid bodies and treated with Y27632 and BLEBB were larger. (A) Single cells were seeded on non-adherent dishes and cultured for the duration of 7 days under STATIC or SHEAR conditions. Media was supplemented with Y27632 (10μM, Millipore) and BLEBB (24 μM, Sigma Aldrich®) for inhibition of ROCK and myosin II respectively, in treated samples. (B) More areas of translucence were seen both STATIC and SHEAR inhibitor- treated EBs as seen in phase images, as indicated by arrows (LEFT). SEM images of SHEAR groups evidenced larger and more numerous cavitations in treated samples (RIGHT, ARROWS) (C) EB size quantification showed a significant decrease in CONTROL STATIC vs. SHEAR EBs and significantly higher area of inhibitor-treated EBs in SHEAR conditions CONTROL(BLACK; n=20), Y27632 (WHITE; n=20), and BLEBB (HASHED; n=20) groups are presented. *** indicates p<0.001 compared to matched CONTROL group. ‡ indicates p<0.05 for STATIC vs SHEAR CONTROL groups.
159
Figure F.2 Treatment with inhibitors decreased expression of pluripotency markers. (A) Gene expression of pluripotency markers NANOG, OCT4, and SOX2 in CONTROL (BLACK), Y27632-(WHITE), and BLEBB-(HASHED) treated EBs cultured under STATIC or SHEAR conditions. ** indicates p<0.01 & *** indicates p<0.001 compared to matched CONTROL group. (B) Histological sections of EBs fluorescently stained for ECAD protein (RED) with a nuclear counterstain (BLUE) showed a disruption in outer expression of ECAD in EBs treated with Y27632 or BLEBB under either STATIC or SHEAR conditions, as indicated by arrows.
160
APPENDIX G: SEM PROTOCOL - CHEMICAL FIXATION & DEHYDRATION
Specialized Materials: Glutaraldeyhde 8% solution in 2ml vials EMS 16019 Osmium Tetroxide 4% solution in 2ml vials EMS 19150 Sodium Cacodylate Buffer 0.2M EMS 11650 1X PBS w/ Ca++ w/ Mg++
Misc Reagents: 1X PBS w/ Ca++ w/ Mg++
Supplies: Microcentrifuge tubes (can use same as FCM) RO Water Blades for cutting samples if necessary Wide bore pipet tips
______
Working Solutions
2.5% Glutaraldehyde1 2ml 8% glutaraldehyde 3.2ml Sodium Cacodylate 1.2ml RO water)
1% Osmium tetroxide2 2ml 4% osmium tetroxide 4ml sodium cacodylate 2ml RO water ______
Work Area Work should be done on chemical fume hood. Protective gloves and clothing should be used at all times.
1 Glutarldehyde can be diluted in water and disposed of in sink 2 OSMIUM TETROXIDE MUST BE NEUTRALIZED WITH CORN OIL (2 parts oil:1 part glutaraldehyde) AND DISPOSED OF AS HAZARDOUS WASTE
161
DATE:______EXPERIMENT:
SAMPLES: ______
Methods
Prepare EBs o As quickly as possible, collect EBs (or other sample) for fixation.
Sample Fixing o Fix cells in 2.5% glutaraldehyde in 0.1M sodium cacodylate buffer for 1 hour o Wash 3x with PBS for 5’ each o Fix cells in 1% osmium tetroxide in 0.1M sodium cacodylate buffer for 1 hour o Wash 3x with PBS for 5’ each o Can be stored in buffer 1-2 weeks in PBS +/+ before critical point drying
Sample Fractioning3 o On a cap of a culture dish, place PBX+/+ to ensure that EBs do not dry out o Use a microscope at low magnification to fraction EBs with surgical blade. o Try to cut in half, not disintegrate by chopping. o The microscopes with a computer screen are particularly helpful. o Using a wide bore pipet transfer cut pieces to new tube with PBS +/+
Sample Dehydration DATE:______o 30% acetone or ethanol for 10’ o 50% acetone or ethanol for 10’ o 70% acetone or ethanol for 10’ o 90% acetone or ethanol for 10’ o 100% acetone or ethanol for 15’ (x2)
3 If necessary.
162
APPENDIX H: SEM PROTOCOL - CRITICAL POINT DRYING & GOLD SPUTTERING
Specialized Materials: Microporus Specimen Capsule EMS 70187-20
Supplies: Gloves Forceps Wide bore tips ______
Work Area Work is done in the 3rd floor of Stern. Everyone should be trained by Jibao He at least once, prior to processing by self.
DATE:______EXPERIMENT: ______
SAMPLES: ______
______
163
Methods Critical Point Drying o Transfer samples from microcentrifuge tube to microporous capsule o Follow instructions of CPD machine located beside it
Gold Sputtering4 o Turn pump ON (MAINS) o Switch metal knob to forward position o Wait until vacuum reaches 50 MT (~30-45 mins) o Open Argon tank o Switch knob from OFF to Control o Open leak valve slowly to 100- 200 mtorr o Set voltage to 2.4 (see arrow) & time to 60-90s o Push Start o Use leak valve to keep mA at 20 o Once done, set voltage to 0 o Knob from Control to OFF Close leak valve o Push metal knob to back position o MAINS OFF o Close Argon tank o Vent slowly o Close vent once it reaches 1 atm
SEM Microscope1 o Sample should be mounted and measured to ensure it will not hit any of the microscope components prior to imaging o Press AIR o Push then pull knob to access rod o Press EVAC o Press OPEN o Push sample in o Press CLOSE
4 Users need to be trained by Jibao He.
164
BIBLIOGRAPHY
1. Mininio A. Death in the United States, 2007; 2009 Contract No.: Document Number|.
2. Badylak SF, Weiss DJ, Caplan A, Macchiarini P. Engineered whole organs and complex tissues. Lancet. 2012;379(9819):943-52.
3. Callihan P, Mumaw J, Machacek DW, Stice SL, Hooks SB. Regulation of stem cell pluripotency and differentiation by G protein coupled receptors. Pharmacol Ther. 2011;129(3):290-306.
4. Evans MJ, Kaufman MH. Establishment in culture of pluripotential cells from mouse embryos. Nature. 1981;292(5819):154-6.
5. Martin GR. Isolation of a pluripotent cell line from early mouse embryos cultured in medium conditioned by teratocarcinoma stem cells. Proc Natl Acad Sci U S A. 1981;78(12):7634-8. PMCID: 349323.
6. Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, Swiergiel JJ, Marshall VS, et al. Embryonic stem cell lines derived from human blastocysts. Science. 1998;282(5391):1145-7.
7. Takahashi K, Tanabe K, Ohnuki M, Narita M, Ichisaka T, Tomoda K, et al. Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell. 2007;131(5):861-72.
8. Niwa H. How is pluripotency determined and maintained? Development. 2007;134(4):635-46.
9. Botquin V, Hess H, Fuhrmann G, Anastassiadis C, Gross MK, Vriend G, et al. New POU dimer configuration mediates antagonistic control of an osteopontin preimplantation enhancer by Oct-4 and Sox-2. Gene Dev. 1998;12(13):2073-90.
10. O'Carroll D, Erhardt S, Pagani M, Barton SC, Surani MA, Jenuwein T. The Polycomb-group gene Ezh2 is required for early mouse development. Mol Cell Biol. 2001;21(13):4330-6.
165
11. Silva J, Mak W, Zvetkova I, Appanah R, Nesterova TB, Webster Z, et al. Establishment of histone H3 methylation on the inactive X chromosome requires transient recruitment of Eed-Enx1 Polycomb group complexes. Developmental Cell. 2003;4(4):481-95.
12. Takahashi K, Yamanaka S. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell. 2006;126(4):663-76.
13. Park IH, Zhao R, West JA, Yabuuchi A, Huo HG, Ince TA, et al. Reprogramming of human somatic cells to pluripotency with defined factors. Nature. 2008;451(7175):141-U1.
14. Shi L, Wu J. Epigenetic regulation in mammalian preimplantation embryo development. Reprod Biol Endocrinol. 2009;7:59. PMCID: 2702308.
15. Kimura-Yoshida C, Nakano H, Okamura D, Nakao K, Yonemura S, Belo JA, et al. Canonical Wnt signaling and its antagonist regulate anterior- posterior axis polarization by guiding cell migration in mouse visceral endoderm. Developmental Cell. 2005;9(5):639-50.
16. Fernandez-Sanchez ME, Serman F, Ahmadi P, Farge E. Mechanical induction in embryonic development and tumor growth integrative cues through molecular to multicellular interplay and evolutionary perspectives. Methods Cell Biol. 2010;98:295-321.
17. Arnold SJ, Robertson EJ. Making a commitment: cell lineage allocation and axis patterning in the early mouse embryo. Nat Rev Mol Cell Bio. 2009;10(2):91-103.
18. Camp E, Munsterberg A. Ingression, migration and early differentiation of cardiac progenitors. Front Biosci. 2012;17:2416-26.
19. Farge E. Mechanical induction of Twist in the Drosophila foregut/stomodeal primordium. Curr Biol. 2003;13(16):1365-77.
20. Kornikova ES, Troshina TG, Kremnyov SV, Beloussov LV. Neuro- mesodermal patterns in artificially deformed embryonic explants: a role for mechano-geometry in tissue differentiation. Dev Dyn. 2010;239(3):885-96.
21. Engler AJ, Sen S, Sweeney HL, Discher DE. Matrix elasticity directs stem cell lineage specification. Cell. 2006;126(4):677-89.
166
22. Rodriguez JP, Gonzalez M, Rios S, Cambiazo V. Cytoskeletal organization of human mesenchymal stem cells (MSC) changes during their osteogenic differentiation. J Cell Biochem. 2004;93(4):721-31.
23. Bosnakovski D, Mizuno M, Kim G, Takagi S, Okumura M, Fujinaga T. Chondrogenic differentiation of bovine bone marrow mesenchymal stem cells (MSCs) in different hydrogels: influence of collagen type II extracellular matrix on MSC chondrogenesis. Biotechnol Bioeng. 2006;93(6):1152-63.
24. Gerecht S, Burdick JA, Ferreira LS, Townsend SA, Langer R, Vunjak- Novakovic G. Hyaluronic acid hydrogel for controlled self-renewal and differentiation of human embryonic stem cells. Proc Natl Acad Sci U S A. 2007;104(27):11298-303. PMCID: 2040893.
25. Zoldan J, Karagiannis ED, Lee CY, Anderson DG, Langer R, Levenberg S. The influence of scaffold elasticity on germ layer specification of human embryonic stem cells. Biomaterials. 2011;32(36):9612-21. PMCID: 3313669.
26. Dang SM, Kyba M, Perlingeiro R, Daley GQ, Zandstra PW. Efficiency of embryoid body formation and hematopoietic development from embryonic stem cells in different culture systems. Biotechnol Bioeng. 2002;78(4):442- 53.
27. Itskovitz-Eldor J, Schuldiner M, Karsenti D, Eden A, Yanuka O, Amit M, et al. Differentiation of human embryonic stem cells into embryoid bodies compromising the three embryonic germ layers. Mol Med. 2000;6(2):88- 95. PMCID: 1949933.
28. Baharvand H, Hashemi SM, Kazemi Ashtiani S, Farrokhi A. Differentiation of human embryonic stem cells into hepatocytes in 2D and 3D culture systems in vitro. Int J Dev Biol. 2006;50(7):645-52.
29. Trappmann B, Gautrot JE, Connelly JT, Strange DG, Li Y, Oyen ML, et al. Extracellular-matrix tethering regulates stem-cell fate. Nat Mater. 2012;11(8):742.
30. Oh SA, Lee HY, Lee JH, Kim TH, Jang JH, Kim HW, et al. Collagen three- dimensional hydrogel matrix carrying basic fibroblast growth factor for the cultivation of mesenchymal stem cells and osteogenic differentiation. Tissue Eng Part A. 2012;18(9-10):1087-100.
167
31. Lecina M, Ting S, Choo A, Reuveny S, Oh S. Scalable platform for human embryonic stem cell differentiation to cardiomyocytes in suspended microcarrier cultures. Tissue Eng Part C Methods. 2010;16(6):1609-19.
32. Tzanakakis ES, Hess DJ, Sielaff TD, Hu WS. Extracorporeal tissue engineered liver-assist devices. Annu Rev Biomed Eng. 2000;2:607-32.
33. Lock LT, Tzanakakis ES. Stem/Progenitor cell sources of insulin- producing cells for the treatment of diabetes. Tissue Eng. 2007;13(7):1399-412.
34. Wang YK, Yu X, Cohen DM, Wozniak MA, Yang MT, Gao L, et al. Bone morphogenetic protein-2-induced signaling and osteogenesis is regulated by cell shape, RhoA/ROCK, and cytoskeletal tension. Stem Cells Dev. 2012;21(7):1176-86. PMCID: 3328763.
35. Boheler KR, Czyz J, Tweedie D, Yang HT, Anisimov SV, Wobus AM. Differentiation of pluripotent embryonic stem cells into cardiomyocytes. Circ Res. 2002;91(3):189-201.
36. Bijou F, Ivanovic Z, Boiron JM, Nicolini F. [Hematopoietic stem cells mobilization: state of the art in 2011 and perspectives]. Transfus Clin Biol. 2011;18(5-6):503-15.
37. Sumanasinghe RD, Bernacki SH, Loboa EG. Osteogenic differentiation of human mesenchymal stem cells in collagen matrices: effect of uniaxial cyclic tensile strain on bone morphogenetic protein (BMP-2) mRNA expression. Tissue Eng. 2006;12(12):3459-65.
38. Wolfe RP, Leleux J, Nerem RM, Ahsan T. Effects of shear stress on germ lineage specification of embryonic stem cells. Integr Biol (Camb). 2012;4(10):1263-73.
39. McBeath R, Pirone DM, Nelson CM, Bhadriraju K, Chen CS. Cell shape, cytoskeletal tension, and RhoA regulate stem cell lineage commitment. Dev Cell. 2004;6(4):483-95.
40. Kurpinski K, Chu J, Hashi C, Li S. Anisotropic mechanosensing by mesenchymal stem cells. Proc Natl Acad Sci U S A. 2006;103(44):16095- 100. PMCID: 1637542.
41. Uda Y, Poh YC, Chowdhury F, Wu DC, Tanaka TS, Sato M, et al. Force via integrins but not E-cadherin decreases Oct3/4 expression in embryonic stem cells. Biochem Biophys Res Commun. 2011;415(2):396-400. PMCID: 3221912.
168
42. Ahsan T, Nerem RM. Fluid shear stress promotes an endothelial-like phenotype during the early differentiation of embryonic stem cells. Tissue Eng Part A. 2010;16(11):3547-53. PMCID: 2992398.
43. Chowdhury F, Na S, Li D, Poh YC, Tanaka TS, Wang F, et al. Material properties of the cell dictate stress-induced spreading and differentiation in embryonic stem cells. Nat Mater. 2010;9(1):82-8. PMCID: 2833279.
44. Ingber DE. Cellular mechanotransduction: putting all the pieces together again. FASEB J. 2006;20(7):811-27.
45. Liu B, Kim TJ, Wang Y. Live cell imaging of mechanotransduction. J R Soc Interface. 2010;7 Suppl 3:S365-75. PMCID: 2943879.
46. Carattino MD, Sheng S, Kleyman TR. Epithelial Na+ channels are activated by laminar shear stress. J Biol Chem. 2004;279(6):4120-6.
47. Gottschalk KE, Gunther R, Kessler H. A three-state mechanism of integrin activation and signal transduction for integrin alpha(v)beta(3). Chembiochem. 2002;3(5):470-3.
48. Luo BH, Carman CV, Takagi J, Springer TA. Disrupting integrin transmembrane domain heterodimerization increases ligand binding affinity, not valency or clustering. Proc Natl Acad Sci U S A. 2005;102(10):3679-84. PMCID: 553322.
49. Malone AM, Batra NN, Shivaram G, Kwon RY, You L, Kim CH, et al. The role of actin cytoskeleton in oscillatory fluid flow-induced signaling in MC3T3-E1 osteoblasts. Am J Physiol Cell Physiol. 2007;292(5):C1830-6. PMCID: 3057612.
50. Satlin LM, Sheng S, Woda CB, Kleyman TR. Epithelial Na(+) channels are regulated by flow. Am J Physiol Renal Physiol. 2001;280(6):F1010-8.
51. Maruthamuthu V, Aratyn-Schaus Y, Gardel ML. Conserved F-actin dynamics and force transmission at cell adhesions. Curr Opin Cell Biol. 2010;22(5):583-8. PMCID: 2948584.
52. Schwartz MA, DeSimone DW. Cell adhesion receptors in mechanotransduction. Curr Opin Cell Biol. 2008;20(5):551-6. PMCID: 2581799.
53. Dahl KN, Ribeiro AJ, Lammerding J. Nuclear shape, mechanics, and mechanotransduction. Circ Res. 2008;102(11):1307-18. PMCID: 2717705.
169
54. Ingber DE. Tensegrity I. Cell structure and hierarchical systems biology. Journal of Cell Science. 2003;116(7):1157-73.
55. Mazumder A, Shivashankar GV. Emergence of a prestressed eukaryotic nucleus during cellular differentiation and development. Journal of the Royal Society Interface. 2010;7:S321-S30.
56. Fletcher DA, Mullins RD. Cell mechanics and the cytoskeleton. Nature. 2010;463(7280):485-92. PMCID: 2851742.
57. Yao L, Romero MJ, Toque HA, Yang G, Caldwell RB, Caldwell RW. The role of RhoA/Rho kinase pathway in endothelial dysfunction. J Cardiovasc Dis Res. 2010;1(4):165-70. PMCID: 3023892.
58. Drees F, Pokutta S, Yamada S, Nelson WJ, Weis WI. Alpha-catenin is a molecular switch that binds E-cadherin-beta-catenin and regulates actin- filament assembly. Cell. 2005;123(5):903-15.
59. Pekovic V, Hutchison CJ. Adult stem cell maintenance and tissue regeneration in the ageing context: the role for A-type lamins as intrinsic modulators of ageing in adult stem cells and their niches. J Anat. 2008;213(1):5-25. PMCID: 2475560.
60. Maniotis AJ, Chen CS, Ingber DE. Demonstration of mechanical connections between integrins, cytoskeletal filaments, and nucleoplasm that stabilize nuclear structure. Proc Natl Acad Sci U S A. 1997;94(3):849- 54. PMCID: 19602.
61. Cai Y, Rossier O, Gauthier NC, Biais N, Fardin MA, Zhang X, et al. Cytoskeletal coherence requires myosin-IIA contractility. J Cell Sci. 2010;123(Pt 3):413-23. PMCID: 2816186.
62. Olson EN, Nordheim A. Linking actin dynamics and gene transcription to drive cellular motile functions. Nat Rev Mol Cell Biol. 2010;11(5):353-65. PMCID: 3073350.
63. Feng TS, Szabo E, Dziak E, Opas M. Cytoskeletal Disassembly and Cell Rounding Promotes Adipogenesis from ES Cells. Stem Cell Rev Rep. 2010;6(1):74-85.
64. Cooper GM. The Cell: A Molecular Approach. 2 ed. Sunderland: Sinauer Associates; 2000.
170
65. Eckes B, Dogic D, Colucci-Guyon E, Wang N, Maniotis A, Ingber D, et al. Impaired mechanical stability, migration and contractile capacity in vimentin-deficient fibroblasts. J Cell Sci. 1998;111 ( Pt 13):1897-907.
66. Dechat T, Adam SA, Taimen P, Shimi T, Goldman RD. Nuclear lamins. Cold Spring Harb Perspect Biol. 2010;2(11):a000547. PMCID: 2964183.
67. Xu K, Schwarz PM, Luduena RF. Interaction of nocodazole with tubulin isotypes. Drug Develop Res. 2002;55(2):91-6.
68. Balaban NQ, Schwarz US, Riveline D, Goichberg P, Tzur G, Sabanay I, et al. Force and focal adhesion assembly: a close relationship studied using elastic micropatterned substrates. Nat Cell Biol. 2001;3(5):466-72.
69. Even-Ram S, Doyle AD, Conti MA, Matsumoto K, Adelstein RS, Yamada KM. Myosin IIA regulates cell motility and actomyosin-microtubule crosstalk. Nat Cell Biol. 2007;9(3):299-309.
70. Mammoto A, Huang S, Moore K, Oh P, Ingber DE. Role of RhoA, mDia, and ROCK in cell shape-dependent control of the Skp2-p27kip1 pathway and the G1/S transition. J Biol Chem. 2004;279(25):26323-30.
71. Fukata Y, Oshiro N, Kaibuchi K. Activation of moesin and adducin by Rho- kinase downstream of Rho. Biophys Chem. 1999;82(2-3):139-47.
72. Kimura K, Ito M, Amano M, Chihara K, Fukata Y, Nakafuku M, et al. Regulation of myosin phosphatase by Rho and Rho-associated kinase (Rho-kinase). Science. 1996;273(5272):245-8.
73. Maekawa M, Ishizaki T, Boku S, Watanabe N, Fujita A, Iwamatsu A, et al. Signaling from Rho to the actin cytoskeleton through protein kinases ROCK and LIM-kinase. Science. 1999;285(5429):895-8.
74. Harb N, Archer TK, Sato N. The Rho-Rock-Myosin signaling axis determines cell-cell integrity of self-renewing pluripotent stem cells. PLoS One. 2008;3(8):e3001. PMCID: 2500174.
75. Palazzo AF, Cook TA, Alberts AS, Gundersen GG. mDia mediates Rho- regulated formation and orientation of stable microtubules. Nat Cell Biol. 2001;3(8):723-9.
76. Li D, Zhou J, Wang L, Shin ME, Su P, Lei X, et al. Integrated biochemical and mechanical signals regulate multifaceted human embryonic stem cell functions. J Cell Biol. 2010;191(3):631-44. PMCID: 3003326.
171
77. Gerecht S, Bettinger CJ, Zhang Z, Borenstein JT, Vuniak-Novakovic G, Langer R. The effect of actin disrupting agents on contact guidance of human embryonic stem cells. Biomaterials. 2007;28(28):4068-77.
78. Zhang L, Valdez JM, Zhang BY, Wei L, Chang JA, Xin L. ROCK Inhibitor Y-27632 Suppresses Dissociation-Induced Apoptosis of Murine Prostate Stem/Progenitor Cells and Increases Their Cloning Efficiency. Plos One. 2011;6(3).
79. Chen G, Hou Z, Gulbranson DR, Thomson JA. Actin-myosin contractility is responsible for the reduced viability of dissociated human embryonic stem cells. Cell Stem Cell. 2010;7(2):240-8. PMCID: 2916864.
80. Baharvand H, Salekdeh GH, Taei A, Mollamohammadi S. An efficient and easy-to-use cryopreservation protocol for human ES and iPS cells. Nat Protoc. 2010;5(3):588-94.
81. Walker A, Su H, Conti MA, Harb N, Adelstein RS, Sato N. Non-muscle myosin II regulates survival threshold of pluripotent stem cells. Nat Commun. 2010;1:71.
82. Krawetz RJ, Taiani J, Greene A, Kelly GM, Rancourt DE. Inhibition of Rho kinase regulates specification of early differentiation events in P19 embryonal carcinoma stem cells. PLoS One. 2011;6(11):e26484. PMCID: 3227584.
83. Joo HJ, Choi DK, Lim JS, Park JS, Lee SH, Song S, et al. ROCK suppression promotes differentiation and expansion of endothelial cells from embryonic stem cell-derived Flk1+ mesodermal precursor cells. Blood. 2012;120(13):2733-44.
84. Hirokawa N. Kinesin and dynein superfamily proteins and the mechanism of organelle transport. Science. 1998;279(5350):519-26.
85. Salmon WC, Adams MC, Waterman-Storer CM. Dual-wavelength fluorescent speckle microscopy reveals coupling of microtubule and actin movements in migrating cells. J Cell Biol. 2002;158(1):31-7. PMCID: 2173033.
86. Forscher P, Smith SJ. Actions of cytochalasins on the organization of actin filaments and microtubules in a neuronal growth cone. J Cell Biol. 1988;107(4):1505-16. PMCID: 2115246.
172
87. Mandato CA, Bement WM. Actomyosin transports microtubules and microtubules control actomyosin recruitment during Xenopus oocyte wound healing. Curr Biol. 2003;13(13):1096-105.
88. Silverman-Gavrila RV, Forer A. Evidence that actin and myosin are involved in the poleward flux of tubulin in metaphase kinetochore microtubules of crane-fly spermatocytes. J Cell Sci. 2000;113 ( Pt 4):597- 609.
89. Danowski BA. Fibroblast contractility and actin organization are stimulated by microtubule inhibitors. J Cell Sci. 1989;93 ( Pt 2):255-66.
90. Briggs R, King TJ. Transplantation of Living Nuclei From Blastula Cells into Enucleated Frogs' Eggs. Proc Natl Acad Sci U S A. 1952;38(5):455- 63. PMCID: 1063586.
91. Gurdon JB. The developmental capacity of nuclei taken from intestinal epithelium cells of feeding tadpoles. J Embryol Exp Morphol. 1962;10:622- 40.
92. Davis RL, Weintraub H, Lassar AB. Expression of a single transfected cDNA converts fibroblasts to myoblasts. Cell. 1987;51(6):987-1000.
93. Wilmut I, Schnieke AE, McWhir J, Kind AJ, Campbell KH. Viable offspring derived from fetal and adult mammalian cells. Nature. 1997;385(6619):810-3.
94. Dambrot C, Passier R, Atsma D, Mummery CL. Cardiomyocyte differentiation of pluripotent stem cells and their use as cardiac disease models. Biochem J. 2011;434(1):25-35.
95. Mauritz C, Schwanke K, Reppel M, Neef S, Katsirntaki K, Maier LS, et al. Generation of functional murine cardiac myocytes from induced pluripotent stem cells. Circulation. 2008;118(5):507-17.
96. Choi KD, Yu J, Smuga-Otto K, Salvagiotto G, Rehrauer W, Vodyanik M, et al. Hematopoietic and endothelial differentiation of human induced pluripotent stem cells. Stem Cells. 2009;27(3):559-67. PMCID: 2931800.
97. Narazaki G, Uosaki H, Teranishi M, Okita K, Kim B, Matsuoka S, et al. Directed and systematic differentiation of cardiovascular cells from mouse induced pluripotent stem cells. Circulation. 2008;118(5):498-506.
98. Wernig M, Zhao JP, Pruszak J, Hedlund E, Fu D, Soldner F, et al. Neurons derived from reprogrammed fibroblasts functionally integrate into
173
the fetal brain and improve symptoms of rats with Parkinson's disease. Proc Natl Acad Sci U S A. 2008;105(15):5856-61. PMCID: 2311361.
99. Dimos JT, Rodolfa KT, Niakan KK, Weisenthal LM, Mitsumoto H, Chung W, et al. Induced pluripotent stem cells generated from patients with ALS can be differentiated into motor neurons. Science. 2008;321(5893):1218- 21.
100. Sullivan GJ, Hay DC, Park IH, Fletcher J, Hannoun Z, Payne CM, et al. Generation of functional human hepatic endoderm from human induced pluripotent stem cells. Hepatology. 2010;51(1):329-35. PMCID: 2799548.
101. Hanna J, Wernig M, Markoulaki S, Sun CW, Meissner A, Cassady JP, et al. Treatment of sickle cell anemia mouse model with iPS cells generated from autologous skin. Science. 2007;318(5858):1920-3.
102. Phanstiel DH, Brumbaugh J, Wenger CD, Tian S, Probasco MD, Bailey DJ, et al. Proteomic and phosphoproteomic comparison of human ES and iPS cells. Nat Methods. 2011;8(10):821-7. PMCID: 3432645.
103. Kim K, Doi A, Wen B, Ng K, Zhao R, Cahan P, et al. Epigenetic memory in induced pluripotent stem cells. Nature. 2010;467(7313):285-90. PMCID: 3150836.
104. Bar-Nur O, Russ HA, Efrat S, Benvenisty N. Epigenetic memory and preferential lineage-specific differentiation in induced pluripotent stem cells derived from human pancreatic islet beta cells. Cell Stem Cell. 2011;9(1):17-23.
105. Maherali N, Ahfeldt T, Rigamonti A, Utikal J, Cowan C, Hochedlinger K. A high-efficiency system for the generation and study of human induced pluripotent stem cells. Cell Stem Cell. 2008;3(3):340-5.
106. Qin D, Li W, Zhang J, Pei D. Direct generation of ES-like cells from unmodified mouse embryonic fibroblasts by Oct4/Sox2/Myc/Klf4. Cell Res. 2007;17(11):959-62.
107. Meissner A, Wernig M, Jaenisch R. Direct reprogramming of genetically unmodified fibroblasts into pluripotent stem cells. Nat Biotechnol. 2007;25(10):1177-81.
108. Lister R, Pelizzola M, Kida YS, Hawkins RD, Nery JR, Hon G, et al. Hotspots of aberrant epigenomic reprogramming in human induced pluripotent stem cells. Nature. 2011;471(7336):68-73. PMCID: 3100360.
174
109. Daniels BR, Hale CM, Khatau SB, Kusuma S, Dobrowsky TM, Gerecht S, et al. Differences in the microrheology of human embryonic stem cells and human induced pluripotent stem cells. Biophys J. 2010;99(11):3563-70. PMCID: 2998615.
110. Pillarisetti A, Desai JP, Ladjal H, Schiffmacher A, Ferreira A, Keefer CL. Mechanical phenotyping of mouse embryonic stem cells: increase in stiffness with differentiation. Cell Reprogram. 2011;13(4):371-80.
111. Pelling AE, Dawson DW, Carreon DM, Christiansen JJ, Shen RR, Teitell MA, et al. Distinct contributions of microtubule subtypes to cell membrane shape and stability. Nanomedicine. 2007;3(1):43-52.
112. Zhou J, Kim HY, Wang JH, Davidson LA. Macroscopic stiffening of embryonic tissues via microtubules, RhoGEF and the assembly of contractile bundles of actomyosin. Development. 2010;137(16):2785-94. PMCID: 2910388.
113. Lai WH, Ho JC, Lee YK, Ng KM, Au KW, Chan YC, et al. ROCK inhibition facilitates the generation of human-induced pluripotent stem cells in a defined, feeder-, and serum-free system. Cell Reprogram. 2010;12(6):641-53. PMCID: 2993021.
114. Rodriguez OC, Schaefer AW, Mandato CA, Forscher P, Bement WM, Waterman-Storer CM. Conserved microtubule-actin interactions in cell movement and morphogenesis. Nat Cell Biol. 2003;5(7):599-609.
115. Gimble JM, Katz AJ, Bunnell BA. Adipose-derived stem cells for regenerative medicine. Circ Res. 2007;100(9):1249-60.
116. Peister A, Woodruff MA, Prince JJ, Gray DP, Hutmacher DW, Guldberg RE. Cell sourcing for bone tissue engineering: amniotic fluid stem cells have a delayed, robust differentiation compared to mesenchymal stem cells. Stem Cell Res. 2011;7(1):17-27. PMCID: 3137122.
117. Nishikawa SI, Nishikawa S, Hirashima M, Matsuyoshi N, Kodama H. Progressive lineage analysis by cell sorting and culture identifies FLK1+VE-cadherin+ cells at a diverging point of endothelial and hemopoietic lineages. Development. 1998;125(9):1747-57.
118. Schenke-Layland K, Angelis E, Rhodes KE, Heydarkhan-Hagvall S, Mikkola HK, Maclellan WR. Collagen IV induces trophoectoderm differentiation of mouse embryonic stem cells. Stem Cells. 2007;25(6):1529-38.
175
119. Doyle AM, Nerem RM, Ahsan T. Human mesenchymal stem cells form multicellular structures in response to applied cyclic strain. Ann Biomed Eng. 2009;37(4):783-93.
120. Saha S, Ji L, de Pablo JJ, Palecek SP. Inhibition of human embryonic stem cell differentiation by mechanical strain. J Cell Physiol. 2006;206(1):126-37.
121. Nikmanesh M, Shi ZD, Tarbell JM. Heparan sulfate proteoglycan mediates shear stress-induced endothelial gene expression in mouse embryonic stem cell-derived endothelial cells. Biotechnol Bioeng. 2012;109(2):583- 94. PMCID: 3228881.
122. Fuchs C, Scheinast M, Pasteiner W, Lagger S, Hofner M, Hoellrigl A, et al. Self-organization phenomena in embryonic stem cell-derived embryoid bodies: axis formation and breaking of symmetry during cardiomyogenesis. Cells Tissues Organs. 2012;195(5):377-91.
123. Cullen DK, Lessing MC, LaPlaca MC. Collagen-dependent neurite outgrowth and response to dynamic deformation in three-dimensional neuronal cultures. Ann Biomed Eng. 2007;35(5):835-46.
124. Powers MJ, Domansky K, Kaazempur-Mofrad MR, Kalezi A, Capitano A, Upadhyaya A, et al. A microfabricated array bioreactor for perfused 3D liver culture. Biotechnol Bioeng. 2002;78(3):257-69.
125. Devolder R, Kong HJ. Hydrogels for in vivo-like three-dimensional cellular studies. Wiley Interdiscip Rev Syst Biol Med. 2012;4(4):351-65.
126. Herrero J, Valencia A, Dopazo J. A hierarchical unsupervised growing neural network for clustering gene expression patterns. Bioinformatics. 2001;17(2):126-36.
127. Nostro MC, Cheng X, Keller GM, Gadue P. Wnt, activin, and BMP signaling regulate distinct stages in the developmental pathway from embryonic stem cells to blood. Cell Stem Cell. 2008;2(1):60-71. PMCID: 2533280.
128. Willems E, Leyns L. Patterning of mouse embryonic stem cell-derived pan-mesoderm by Activin A/Nodal and Bmp4 signaling requires Fibroblast Growth Factor activity. Differentiation. 2008;76(7):745-59.
129. Ang SL, Rossant J. HNF-3 beta is essential for node and notochord formation in mouse development. Cell. 1994;78(4):561-74.
176
130. Weinstein DC, Ruiz i Altaba A, Chen WS, Hoodless P, Prezioso VR, Jessell TM, et al. The winged-helix transcription factor HNF-3 beta is required for notochord development in the mouse embryo. Cell. 1994;78(4):575-88.
131. Yamanaka Y, Tamplin OJ, Beckers A, Gossler A, Rossant J. Live imaging and genetic analysis of mouse notochord formation reveals regional morphogenetic mechanisms. Dev Cell. 2007;13(6):884-96.
132. Liu Y, Goldberg AJ, Dennis JE, Gronowicz GA, Kuhn LT. One-step derivation of mesenchymal stem cell (MSC)-like cells from human pluripotent stem cells on a fibrillar collagen coating. PLoS One. 2012;7(3):e33225. PMCID: 3310052.
133. Krawetz RJ, Taiani JT, Wu YE, Liu S, Meng G, Matyas JR, et al. Collagen I scaffolds cross-linked with beta-glycerol phosphate induce osteogenic differentiation of embryonic stem cells in vitro and regulate their tumorigenic potential in vivo. Tissue Eng Part A. 2012;18(9-10):1014-24.
134. Chenoweth JG, McKay RD, Tesar PJ. Epiblast stem cells contribute new insight into pluripotency and gastrulation. Dev Growth Differ. 2010;52(3):293-301.
135. Kurosawa H, Imamura T, Koike M, Sasaki K, Amano Y. A simple method for forming embryoid body from mouse embryonic stem cells. J Biosci Bioeng. 2003;96(4):409-11.
136. Wang H, Gilner JB, Bautch VL, Wang DZ, Wainwright BJ, Kirby SL, et al. Wnt2 coordinates the commitment of mesoderm to hematopoietic, endothelial, and cardiac lineages in embryoid bodies. J Biol Chem. 2007;282(1):782-91.
137. Bershadsky A. Magic touch: how does cell-cell adhesion trigger actin assembly? Trends Cell Biol. 2004;14(11):589-93.
138. Chen CS, Tan J, Tien J. Mechanotransduction at cell-matrix and cell-cell contacts. Annu Rev Biomed Eng. 2004;6:275-302.
139. Xu Y, Zhu X, Hahm HS, Wei W, Hao E, Hayek A, et al. Revealing a core signaling regulatory mechanism for pluripotent stem cell survival and self- renewal by small molecules. Proc Natl Acad Sci U S A. 2010;107(18):8129-34. PMCID: 2889586.
140. Carpenedo RL, Bratt-Leal AM, Marklein RA, Seaman SA, Bowen NJ, McDonald JF, et al. Homogeneous and organized differentiation within
177
embryoid bodies induced by microsphere-mediated delivery of small molecules. Biomaterials. 2009;30(13):2507-15. PMCID: 2921510.
141. Curley JL, Moore MJ. Facile micropatterning of dual hydrogel systems for 3D models of neurite outgrowth. J Biomed Mater Res A. 2011;99(4):532- 43. PMCID: 3213030.
142. Elisseeff J, McIntosh W, Anseth K, Riley S, Ragan P, Langer R. Photoencapsulation of chondrocytes in poly(ethylene oxide)-based semi- interpenetrating networks. J Biomed Mater Res. 2000;51(2):164-71.
143. Lutolf MP, Hubbell JA. Synthetic biomaterials as instructive extracellular microenvironments for morphogenesis in tissue engineering. Nat Biotechnol. 2005;23(1):47-55.
144. Sun Y, Villa-Diaz LG, Lam RH, Chen W, Krebsbach PH, Fu J. Mechanics regulates fate decisions of human embryonic stem cells. PLoS One. 2012;7(5):e37178. PMCID: 3353896.
145. McMahon LA, Reid AJ, Campbell VA, Prendergast PJ. Regulatory effects of mechanical strain on the chondrogenic differentiation of MSCs in a collagen-GAG scaffold: experimental and computational analysis. Ann Biomed Eng. 2008;36(2):185-94.
146. Yamamoto K, Sokabe T, Watabe T, Miyazono K, Yamashita JK, Obi S, et al. Fluid shear stress induces differentiation of Flk-1-positive embryonic stem cells into vascular endothelial cells in vitro. Am J Physiol Heart Circ Physiol. 2005;288(4):H1915-24.
147. Grinnell F. Fibroblast biology in three-dimensional collagen matrices. Trends Cell Biol. 2003;13(5):264-9.
148. Hinz B, Celetta G, Tomasek JJ, Gabbiani G, Chaponnier C. Alpha-smooth muscle actin expression upregulates fibroblast contractile activity. Mol Biol Cell. 2001;12(9):2730-41. PMCID: 59708.
149. Kanda K, Matsuda T. Mechanical stress-induced orientation and ultrastructural change of smooth muscle cells cultured in three- dimensional collagen lattices. Cell Transplant. 1994;3(6):481-92.
150. Lee RT, Berditchevski F, Cheng GC, Hemler ME. Integrin-mediated collagen matrix reorganization by cultured human vascular smooth muscle cells. Circ Res. 1995;76(2):209-14.
178
151. Awad HA, Butler DL, Harris MT, Ibrahim RE, Wu Y, Young RG, et al. In vitro characterization of mesenchymal stem cell-seeded collagen scaffolds for tendon repair: effects of initial seeding density on contraction kinetics. J Biomed Mater Res. 2000;51(2):233-40.
152. Li YS, Haga JH, Chien S. Molecular basis of the effects of shear stress on vascular endothelial cells. J Biomech. 2005;38(10):1949-71.
153. Shyy JY, Chien S. Role of integrins in endothelial mechanosensing of shear stress. Circ Res. 2002;91(9):769-75.
154. Lee AA, Delhaas T, McCulloch AD, Villarreal FJ. Differential responses of adult cardiac fibroblasts to in vitro biaxial strain patterns. J Mol Cell Cardiol. 1999;31(10):1833-43.
155. Kearney EM, Farrell E, Prendergast PJ, Campbell VA. Tensile strain as a regulator of mesenchymal stem cell osteogenesis. Ann Biomed Eng. 2010;38(5):1767-79.
156. Riha GM, Wang X, Wang H, Chai H, Mu H, Lin PH, et al. Cyclic strain induces vascular smooth muscle cell differentiation from murine embryonic mesenchymal progenitor cells. Surgery. 2007;141(3):394-402.
157. Mendez MG, Janmey PA. Transcription factor regulation by mechanical stress. Int J Biochem Cell Biol. 2012;44(5):728-32.
158. Wang N, Tytell JD, Ingber DE. Mechanotransduction at a distance: mechanically coupling the extracellular matrix with the nucleus. Nat Rev Mol Cell Biol. 2009;10(1):75-82.
159. Treiser MD, Yang EH, Gordonov S, Cohen DM, Androulakis IP, Kohn J, et al. Cytoskeleton-based forecasting of stem cell lineage fates. Proc Natl Acad Sci U S A. 2010;107(2):610-5. PMCID: 2818905.
160. Solnica-Krezel L, Sepich DS. Gastrulation: Making and Shaping Germ Layers. Annu Rev Cell Dev Bi. 2012;28:687-717.
161. Baron MH. Concise Review: Early Embryonic Erythropoiesis: Not so Primitive After All. Stem Cells. 2013;31(5):849-56.
162. Rust WL, Sadasivam A, Dunn NR. Three-dimensional extracellular matrix stimulates gastrulation-like events in human embryoid bodies. Stem Cells and Development. 2006;15(6):889-904.
179
163. Wessells NK, Spooner BS, Ash JF, Bradley MO, Luduena MA, Taylor EL, et al. Microfilaments in cellular and developmental processes. Science. 1971;171(3967):135-43.
164. Vasiliev JM. Actin cortex and microtubular system in morphogenesis: cooperation and competition. J Cell Sci Suppl. 1987;8:1-18.
165. Vicente-Manzanares M, Ma X, Adelstein RS, Horwitz AR. Non-muscle myosin II takes centre stage in cell adhesion and migration. Nat Rev Mol Cell Biol. 2009;10(11):778-90. PMCID: 2834236.
166. Watt FM, Jordan PW, O'Neill CH. Cell shape controls terminal differentiation of human epidermal keratinocytes. Proc Natl Acad Sci U S A. 1988;85(15):5576-80. PMCID: 281801.
167. Chen CS, Alonso JL, Ostuni E, Whitesides GM, Ingber DE. Cell shape provides global control of focal adhesion assembly. Biochem Biophys Res Commun. 2003;307(2):355-61.
168. Wang N, Ingber DE. Control of cytoskeletal mechanics by extracellular matrix, cell shape, and mechanical tension. Biophys J. 1994;66(6):2181-9. PMCID: 1275944.
169. Feng T, Szabo E, Dziak E, Opas M. Cytoskeletal disassembly and cell rounding promotes adipogenesis from ES cells. Stem Cell Rev. 2010;6(1):74-85.
170. Wakatsuki T, Schwab B, Thompson NC, Elson EL. Effects of cytochalasin D and latrunculin B on mechanical properties of cells. J Cell Sci. 2001;114(Pt 5):1025-36.
171. Kallas A, Pook M, Maimets M, Zimmermann K, Maimets T. Nocodazole treatment decreases expression of pluripotency markers Nanog and Oct4 in human embryonic stem cells. PLoS One. 2011;6(4):e19114. PMCID: 3084750.
172. Spiegelman BM, Ginty CA. Fibronectin Modulation of Cell-Shape and Lipogenic Gene-Expression in 3t3-Adipocytes. Cell. 1983;35(3):657-66.
173. Thomas CH, Collier JH, Sfeir CS, Healy KE. Engineering gene expression and protein synthesis by modulation of nuclear shape. P Natl Acad Sci USA. 2002;99(4):1972-7.
180
174. Kilian KA, Bugarija B, Lahn BT, Mrksich M. Geometric cues for directing the differentiation of mesenchymal stem cells. Proc Natl Acad Sci U S A. 2010;107(11):4872-7. PMCID: 2841932.
175. Szabo E, Feng T, Dziak E, Opas M. Cell adhesion and spreading affect adipogenesis from embryonic stem cells: the role of calreticulin. Stem Cells. 2009;27(9):2092-102.
176. Arnsdorf EJ, Tummala P, Kwon RY, Jacobs CR. Mechanically induced osteogenic differentiation--the role of RhoA, ROCKII and cytoskeletal dynamics. J Cell Sci. 2009;122(Pt 4):546-53. PMCID: 2714434.
177. Liao JK, Seto M, Noma K. Rho kinase (ROCK) inhibitors. J Cardiovasc Pharmacol. 2007;50(1):17-24. PMCID: 2692906.
178. Linseman DA, Loucks FA. Diverse roles of Rho family GTPases in neuronal development, survival, and death. Front Biosci-Landmrk. 2008;13:657-76.
179. Fukata Y, Oshiro N, Kinoshita N, Kawano Y, Matsuoka Y, Bennett V, et al. Phosphorylation of adducin by rho-kinase plays a crucial role in cell motility. Journal of Cell Biology. 1999;145(2):347-61.
180. Lawson KA. Fate mapping the mouse embryo. Int J Dev Biol. 1999;43(7):773-5.
181. Winnier G, Blessing M, Labosky PA, Hogan BL. Bone morphogenetic protein-4 is required for mesoderm formation and patterning in the mouse. Genes Dev. 1995;9(17):2105-16.
182. Gadue P, Huber TL, Nostro MC, Kattman S, Keller GM. Germ layer induction from embryonic stem cells. Exp Hematol. 2005;33(9):955-64.
183. Kaufman MH, Bard JBL. The anatomical basis of mouse development. San Diego: Academic Press; 1999.
184. Sajini AA, Greder LV, Dutton JR, Slack JM. Loss of Oct4 expression during the development of murine embryoid bodies. Dev Biol. 2012;371(2):170-9. PMCID: 3477512.
185. Sargent CY, Berguig GY, Kinney MA, Hiatt LA, Carpenedo RL, Berson RE, et al. Hydrodynamic modulation of embryonic stem cell differentiation by rotary orbital suspension culture. Biotechnol Bioeng. 2010;105(3):611- 26.
181
186. Pereira LA, Wong MS, Mossman AK, Sourris K, Janes ME, Knezevic K, et al. Pdgfralpha and Flk1 are direct target genes of Mixl1 in differentiating embryonic stem cells. Stem Cell Res. 2012;8(2):165-79.
187. Evans AL, Faial T, Gilchrist MJ, Down T, Vallier L, Pedersen RA, et al. Genomic targets of Brachyury (T) in differentiating mouse embryonic stem cells. PLoS One. 2012;7(3):e33346. PMCID: 3316570.
188. Zhang Z, Messana J, Hwang NS, Elisseeff JH. Reorganization of actin filaments enhances chondrogenic differentiation of cells derived from murine embryonic stem cells. Biochem Biophys Res Commun. 2006;348(2):421-7.
189. Pineda ET, Nerem RM, Ahsan T. Differentiation patterns of embryonic stem cells in two- versus three-dimensional culture. Cells Tissues Organs. 2013;197(5):399-410.
190. Lee GY, Kenny PA, Lee EH, Bissell MJ. Three-dimensional culture models of normal and malignant breast epithelial cells. Nat Methods. 2007;4(4):359-65. PMCID: 2933182.
191. Schindler M, Nur EKA, Ahmed I, Kamal J, Liu HY, Amor N, et al. Living in three dimensions: 3D nanostructured environments for cell culture and regenerative medicine. Cell Biochem Biophys. 2006;45(2):215-27.
192. Lyashenko N, Winter M, Migliorini D, Biechele T, Moon RT, Hartmann C. Differential requirement for the dual functions of beta-catenin in embryonic stem cell self-renewal and germ layer formation. Nature Cell Biology. 2011;13(7):753-U365.
193. Goh SK, Olsen P, Banerjee I. Extracellular Matrix Aggregates from Differentiating Embryoid Bodies as a Scaffold to Support ESC Proliferation and Differentiation. PLoS One. 2013;8(4).
194. Brown MC, Perrotta JA, Turner CE. Identification of LIM3 as the principal determinant of paxillin focal adhesion localization and characterization of a novel motif on paxillin directing vinculin and focal adhesion kinase binding. J Cell Biol. 1996;135(4):1109-23. PMCID: 2133378.
195. Aplin AE, Howe AK, Juliano RL. Cell adhesion molecules, signal transduction and cell growth. Curr Opin Cell Biol. 1999;11(6):737-44.
196. Aberle H, Schwartz H, Kemler R. Cadherin-catenin complex: protein interactions and their implications for cadherin function. J Cell Biochem. 1996;61(4):514-23.
182
197. Hayashi Y, Furue MK, Okamoto T, Ohnuma K, Myoishi Y, Fukuhara Y, et al. Integrins regulate mouse embryonic stem cell self-renewal. Stem Cells. 2007;25(12):3005-15.
198. Danen EH, Sonneveld P, Brakebusch C, Fassler R, Sonnenberg A. The fibronectin-binding integrins alpha5beta1 and alphavbeta3 differentially modulate RhoA-GTP loading, organization of cell matrix adhesions, and fibronectin fibrillogenesis. J Cell Biol. 2002;159(6):1071-86. PMCID: 2173988.
199. Barczyk M, Carracedo S, Gullberg D. Integrins. Cell Tissue Res. 2010;339(1):269-80. PMCID: 2784866.
200. Yang JT, Rayburn H, Hynes RO. Embryonic mesodermal defects in alpha 5 integrin-deficient mice. Development. 1993;119(4):1093-105.
201. Hirsch E, Iglesias A, Potocnik AJ, Hartmann U, Fassler R. Impaired migration but not differentiation of haematopoietic stem cells in the absence of beta1 integrins. Nature. 1996;380(6570):171-5.
202. Chen T, Yuan D, Wei B, Jiang J, Kang J, Ling K, et al. E-cadherin- mediated cell-cell contact is critical for induced pluripotent stem cell generation. Stem Cells. 2010;28(8):1315-25.
203. Takeichi M. Morphogenetic roles of classic cadherins. Curr Opin Cell Biol. 1995;7(5):619-27.
204. Soncin F, Mohamet L, Ritson S, Hawkins K, Bobola N, Zeef L, et al. E- cadherin acts as a regulator of transcripts associated with a wide range of cellular processes in mouse embryonic stem cells. PLoS One. 2011;6(7):e21463. PMCID: 3136471.
205. Xu J, Zhu C, Zhang Y, Jiang N, Li S, Su Z, et al. hE-cadherin-Fc fusion protein coated surface enhances the adhesion and proliferation of human mesenchymal stem cells. Colloids Surf B Biointerfaces. 2013;109:97-102.
206. Nagaoka M, Koshimizu U, Yuasa S, Hattori F, Chen H, Tanaka T, et al. E- cadherin-coated plates maintain pluripotent ES cells without colony formation. PLoS One. 2006;1:e15. PMCID: 1762325.
207. Nagaoka M, Si-Tayeb K, Akaike T, Duncan SA. Culture of human pluripotent stem cells using completely defined conditions on a recombinant E-cadherin substratum. BMC Dev Biol. 2010;10:60. PMCID: 2896937.
183
208. Haque A, Yue XS, Motazedian A, Tagawa Y, Akaike T. Characterization and neural differentiation of mouse embryonic and induced pluripotent stem cells on cadherin-based substrata. Biomaterials. 2012;33(20):5094- 106.
209. Haque A, Hexig B, Meng Q, Hossain S, Nagaoka M, Akaike T. The effect of recombinant E-cadherin substratum on the differentiation of endoderm- derived hepatocyte-like cells from embryonic stem cells. Biomaterials. 2011;32(8):2032-42.
210. Soncin F, Mohamet L, Eckardt D, Ritson S, Eastham AM, Bobola N, et al. Abrogation of E-cadherin-mediated cell-cell contact in mouse embryonic stem cells results in reversible LIF-independent self-renewal. Stem Cells. 2009;27(9):2069-80.
211. Wolfenson H, Bershadsky A, Henis YI, Geiger B. Actomyosin-generated tension controls the molecular kinetics of focal adhesions. J Cell Sci. 2011;124(Pt 9):1425-32. PMCID: 3078811.
212. Gomez GA, McLachlan RW, Yap AS. Productive tension: force-sensing and homeostasis of cell-cell junctions. Trends Cell Biol. 2011;21(9):499- 505.
213. Wolfe RP, Ahsan T. Shear stress during early embryonic stem cell differentiation promotes hematopoietic and endothelial phenotypes. Biotechnol Bioeng. 2013;110(4):1231-42.
214. Bilic J, Izpisua Belmonte JC. Concise review: Induced pluripotent stem cells versus embryonic stem cells: close enough or yet too far apart? Stem Cells. 2012;30(1):33-41.
215. Heuberger J, Birchmeier W. Interplay of cadherin-mediated cell adhesion and canonical Wnt signaling. Cold Spring Harb Perspect Biol. 2010;2(2):a002915. PMCID: 2828280.
216. Okada Y, Shimazaki T, Sobue G, Okano H. Retinoic-acid-concentration- dependent acquisition of neural cell identity during in vitro differentiation of mouse embryonic stem cells. Dev Biol. 2004;275(1):124-42.
217. Teramura T, Takehara T, Onodera Y, Nakagawa K, Hamanishi C, Fukuda K. Mechanical stimulation of cyclic tensile strain induces reduction of pluripotent related gene expressions via activation of Rho/ROCK and subsequent decreasing of AKT phosphorylation in human induced
184 pluripotent stem cells. Biochem Biophys Res Commun. 2012;417(2):836- 41.
185
BIOGRAPHY
Emma Pineda was born in San Pedro Sula, Honduras. She received her
B.S. in Chemical Engineering from the University of Notre Dame in 2008, and went on to obtain her PhD in Biomedical Engineering at Tulane University in
2013. While at Tulane, her work focused on understanding the role of the cytoskeleton on stem cell differentiation. This work will help improve the design of experiments for the generation of specific cell types for use in medical therapies, which ultimately may accelerate the translation of stem cell research from bench top to bedside.