Beginning Of Last Century Cellular Behaviour

Published Date: 02 Nov 2017

1 Introduction

In the beginning of last century, cellular behaviour was found to be affected by the topographical pattern of the underlying substrate. In 1912, Harrison(RG 1912) reported that substrata with a specific linear arrangement, as in the spider web, influence the direction of the movement as well as the form and arrangement of the cells. Later Loeb and Fleisher(L and MS 1917) introduced the term stereotropism, which was described as the direction in which cells move, mainly governed by the contact with solids or very viscid bodies like fibers or fibrin. In 1945, Weiss(P 1945) called this cellular response to the topography of a substratum surface ‘contact guidance’, a term still in use today. Surprisingly, no further attention was paid to this guidance phenomenon until the early 1970s. It was Rovensky(Rovensky, Slavnaja et al. 1971) and Maroudas(Maroudas 1973) who rediscovered that cells are able to react on the topography of a substratum surface. From this moment on, research on this subject has expanded, resulting in many publications, which were thoroughly reviewed recently by Singhvi(Singhvi, Stephanopoulos et al. 1994).

Some researches show that various physicochemical and geometrical material surface properties can be used to modulate the accompanying host response(Meyle, Gultig et al. 1994; Singhvi, Stephanopoulos et al. 1994). This makes it possible to engineer future biomaterials that provoke a specific biological response, resulting in a unique healing process. Physicochemical properties that have an effect on tissue behaviour are surface charge, surface energy and surface oxidation(Singhvi, Stephanopoulos et al. 1994; den Braber, de Ruijter et al. 1995). Geometrical surface properties that can influence the cellular interactions are shape, size and topography of a surface. The latter is not only limited to surface conditions like roughness or curvature, but also includes microstructured surfaces with a standardized surface roughness. For example, in vitro experiments have already demonstrated that surfaces possessing microgrooves induce orientation of fibroblasts(den Braber, de Ruijter et al. 1995).

Now, microfabrication and materials-science-based approaches provide a powerful new set of tools to control the spatial organization and temporal presentation of cellular cues(Raghavan and Chen 2004). Spatial patterning of the behaviors of individual cells generates global changes in tissue architecture that drive morphogenesis. Several morphogenic mechanisms likely collaborate to direct tissue form, including local changes in cell adhesion, cell shape, and cell proliferation, and differentials in cell growth can locally alter tissue form(Goldin, Hindman et al. 1984).

What causes such localized patterns is one of the central puzzles of biology and has fascinated scientists from numerous disciplines for at least two millennia. Perhaps most well described are concentration gradients of diffusible factors, known as morphogens, which can drive spatial patterns of cellular behaviors(Wolpert 1969). In addition to soluble factors, adhesion to extracellular matrix and mechanical forces also are known to modulate cell functions, including proliferation(Huang, Chen et al. 1998).

Although spatial patterning of these cues can certainly explain spatial patterning of cellular behaviors, it remains unclear what initiates or maintains patterns. One theory suggests that these gradients (e.g., of morphogens) are entirely driven by prespecified genetic programs. A more tractable alternative suggests that the highly ordered architectures of mature tissues and the evolution of ever more complex structures from simpler ones arise as a result of feedback mechanisms, whereby tissue form regulates patterned growth to ensure that certain structures are encouraged and elaborated upon while others are eliminated(Shraiman 2005).

Some researchers reported that tissue can feed back to regulate patterns of proliferation(Nelson, Jean et al. 2005). Using microfabrication to control the organization of sheets of cells, they demonstrated the emergence of stable patterns of proliferative foci, that multicellular form could direct patterns of proliferation. Not only cells on the edges of the islands proliferated more than cells in the center but corners of the square islands proliferated more than edges. The patterns of mechanical stress could be generated by the contraction of cells within a monolayer. Thus a high imposed bending of cells can affect cell proliferation. Such bended regions can also be observed in our fibroblasts pattern.

1.1 CELL–TOPOGRAPHY INTERACTIONS

1.1.1 Natural cell–topography interactions

Extracellular matrix (ECM) proteins are hypothesized to play a role in cell–matrix signaling due to their abundant nanometer-size structure. The basement membranes of many tissues exhibit rich topographies that interact directly with neighboring cells(Goodman, Sims et al. 1996; Abrams, Goodman et al. 2000). Nanotopography is also present in individual ECM molecules, such as collagen molecules, which are approximately 300 nm long and 1.5 nm wide(Pamula, De Cupere et al. 2004). These molecules can form fibrils of several micrometers in length, diameters between 260 and 410 nm(Bozec, van der Heijden et al. 2007). Cell-topography interactions could happen in many ways, often recognized as contact guidance phenomenon. Contact guidance is characterized by the response of cells to structures on the micrometer and nanometer scale. It is quite essential in regulating cell migration, which is modulated by organized ECM proteins(Wolf, Muller et al. 2003). Migration can also be influenced by adjacent cells, as in the case of fibroblast migration in vivo(Sutherland, Denyer et al. 2005) and epithelial cells migration on a collagen substrate in vitro(Haga, Irahara et al. 2005). T-cell migration is also known to be highly dependent upon cell–substrate interactions with native ECM proteins(Friedl and Brocker 2000). Contact guidance can also play an important role in the migration of individual cells or groups of cells or tissue(Friedl 2004), as well as an important component in efficient organelle formation, such as axonal guidance and growth cone motility(Dent and Gertler 2003).

1.1.2 Cell–topography responses on synthetic substrates

With the help of micro- and nanofabrication techniques, it is enable for us to fabricate the patterns and length scale of natural topography in 2D substrates. Cells can respond to 2D synthetic structured substrates in different ways, including cell type, pattern size, geometry(Flemming, Murphy et al. 1999), and the physical properties of the bulk substrate material, such as substrate stiffness(Discher, Janmey et al. 2005). The effect of synthetic substrates with features in nano and micro scale has been explored intensively, including nanofibers(Pham, Sharma et al. 2006), electrospun fibrous mats, and substrates with nano size roughness(Dalby, Riehle et al. 2002; Lim, Hansen et al. 2005; Variola, Yi et al. 2008). Normally three basic topographic patterns are frequently used: nanogratings, nanoposts, and nanopits (Figure 1.1). Cell–topography interactions affect cell properties based on the effect of topography coupled with the physicochemical properties of the substrate.

Figure 1.1 Schematic depictions (a) and SEM images (b) of representative nanotopography geometries. Three basic nanotopography geometries include nanograting (458 tilt, scale bar 5 mm), nanoposts (158 tilt, scale bar 5 mm), and nanopits (08 tilt, scale bar 1 mm). Schematics not drawn to scale. Image got from (Bettinger, Langer et al. 2009)

1 Morphology

Many cell types typically respond to nanogratings by simultaneously aligning and elongating in the direction of the grating. This response has been observed across different cell types, including fibroblasts, endothelial cells, stem cells, smooth muscle cells, epithelial cells, and Schwann cells(Hsu, Chen et al. 2005). The morphological response is seen in cells cultured on substrates with feature widths down to 100 nm and depths down to 75 nm(Loesberg, te Riet et al. 2007). Cells response stronger in smaller pitch and larger depth. Other studies have demonstrated that some nanograting feature sizes induced alignment of cells both parallel and perpendicular to the nanograting direction(Teixeira, McKie et al. 2006). There are also several examples of cell types that do not respond to nanogratings, including human-derived leukocytes, keratinocytes, and monocytes(Meyle, Gultig et al. 1995). Hence, morphological effects of cell–topography interactions are not observed across all cell types.

2 Attachment and Adhesion

The size of synthetic topography is designed to mimic natural extracellular matrix environment(Abrams, Goodman et al. 2000), including collagen(Bozec and Horton 2005), which may substantiate the hypothesis that topography can enhance attachment and adhesion of mammalian cells. Nanogratings generally appear to enhance the adhesion in various cell–biomaterial geometry combinations(Teixeira, Abrams et al. 2003). Further studies must be aimed at elucidating the apparent dependence on feature size and geometry for differential adhesion.

3 Proliferation

Topography has also been shown to affect the proliferation of various cell types. In general, cells cultured on nanogratings exhibit lower proliferation rates than cells cultured on flat substrates(Gerecht, Bettinger et al. 2007; Bettinger, Zhang et al. 2008). Furthermore, there are currently no widely accepted hypotheses regarding the mechanism for the effect of cell–topography interactions on cell proliferation.

4 Migration

The effect of topography on migration is typically observed in cells cultured on nanogratings. Many cell types have exhibited biased migration in the direction of the grating axis and increased overall migration velocities. Cells exhibiting such behavior include endothelial cells(Bettinger, Zhang et al. 2008), epithelial cells(Dalton, Walboomers et al. 2001; Diehl, Foley et al. 2005; Rajnicek, Foubister et al. 2007), osteoblasts(Lenhert, Meier et al. 2005), and C6 glioma cells(Wang, Ohlin et al. 2008).

Topography also biases markers for directional migration, as shown in work by Yim et al. in which microtubule organization centers (MTOCs) were observed to be polarized as a direct consequence of the nanograting(Yim, Reano et al. 2005). Furthermore, the polarization of MTOCs was observed to supersede directional migration cues from wound healing. Enhanced migration is a response that is typically coupled with elongated morphology and alignment of the cell body with the nanograting axis(Lenhert, Meier et al. 2005; Bettinger, Zhang et al. 2008).

1.2 CELL–TOPOGRAPHY INTERACTIONS ON CELL FATE

Topography could affect the gene expression profiles of various cell types. These genetic profiles have been analyzed first through analysis of individual genes and subsequently by comprehensive gene analysis studies. A study by Chou et al. suggested that the mRNA levels and stability in human fibroblasts were influenced by nanogratings(Chou, Firth et al. 1995). More specifically, fibroblasts cultured on titanium nanogratings expressed higher levels of fibronectin mRNA with increased stability. Furthermore, nanogratings induced higher levels of fibronectin incorporation into cell-matrix proteins. More recent work has utilized gene array techniques to probe the effects of topography on gene expression in fibroblasts(Dalby, Riehle et al. 2003) and mesenchymal stem cells(Dalby, Gadegaard et al. 2007; Yim, Pang et al. 2007).

The concomitant impact of topography on both fundamental cell function and gene expression in many cells types suggests that topography could potentially be utilized as a signaling modality for directing differentiation. There has been significant progress in this direction, despite the fact that coordinated work in this specific application of topography has only recently been explored. Work by Yim et al. suggests that human mesenchymal stem cells (hMSCs) cultured on nanogratings can be preferentially differentiated into neuronal lineages as determined by the presence of synaptophysin, tuj1, and nestin markers as well as by the up regulation of MAP2(Yim, Pang et al. 2007). The results demonstrate the potential of topography to direct cell fate. Furthermore, the complementary findings of hMSCs cultured on nanogratings suggest the potential for selective, controllable differentiation based solely on the geometry of the topographic substrate.

1.3 SYNTHETIC TOPOGRAPHIC SUBSTRATES FOR TISSUE ENGINEERING

1.3.1 Fabrication

There are many state of the art fabrication methods available for making topographic substrates, which have been discussed in detail(Norman and Desai 2006). The effect of order and symmetry of features has been demonstrated in numerous studies(Curtis, Casey et al. 2001; Curtis, Gadegaard et al. 2004; Dalby, Gadegaard et al. 2007). These findings suggests that the reaction of cells to substrates with feature roughness of a characteristic length scale should not be assumed to be equivalent to those cells cultured on ordered topography of similar feature size. The nanofabrication of substrates with long-range order across a wide range of feature sizes and geometries has been pursued using a variety of methods, including traditional photolithography, electron beam lithography, and laser interference lithography. Although suitable for creating ordered arrays of features, these processes are expensive and time-consuming, and they require access to intricate equipment. Alternative approaches have been pursued to fabricate polymeric substrates containing structures with long-range order, including the use of diblock copolymers(Cheng, Mayes et al. 2004) for nanograting lamellae(Ruiz, Sandstrom et al. 2007). Advanced fabrication techniques and compatible materials must eventually be synergized as a means to integrate cell–topography interactions into advanced tissue-engineering scaffolds. There are several potential fabrication strategies for integrating topographic cues into 3D structures, including advanced two-photon polymerization(LaFratta, Li et al. 2006; Pikulin and Bityurin 2007) and microscale origami(In, Kumar et al. 2006; Seunarine, Meredith et al. 2008). The integration of topographic cues into 3D scaffold design and fabrication remains a challenging pursuit.

1.3.2 Topographical cues

Cell–topography interactions can control stem-cell differentiation and cellular superstructure, both of which have obvious implications for use in tissue engineering. However, topographical cues can be used to modulate fundamental cell function in scaffold design as well. For example, the influence of nanogratings on morphology can be used to form aligned populations of cells, which are important for the structure and function of smooth muscle cells and endothelial cells. Vascular tissue-engineering scaffolds are of particular interest because of the correlation between alignment and cell function for multiple cell types in close proximity to one another. Recent work has led to the fabrication of a tubular scaffold with multiple nanograting surfaces(Seunarine, Meredith et al. 2008). The influence of topography on adhesion could also serve as a method to create patterned arrays of cells without the need for direct patterning of proteins. Incorporating topographic cues directly into a 3D scaffold may therefore overcome the intrinsic planar limitations of micro contact printing(Xia and Whitesides 1998; Bernard, Renault et al. 2000; Lin, Co et al. 2005) and related methods(Kim, Xia et al. 1997; Chiu, Jeon et al. 2000). Observed enhanced migration on nanogratings also has potential implications for the design of guidance channels for peripheral nerve regeneration. For example, tubular conduits modified with nanogratings could enhance the migration of Schwann cells into the injury site to promote axonal regeneration(Torigoe, Tanaka et al. 1996). These structures could also potentially promote the rapid migration of neurites across the nerve gap.

1.3.3 Material selection

There has been substantial progress in the design and fabrication of topography in a wide spectrum of materials. These techniques are typically designed to be used in combination with materials that are directly adapted from or closely related to the semiconductor industry. Bulk materials processing and nanofabrication strategies for many topographic surfaces are typically fine-tuned for silicon, silicon oxide, polycrystalline silicon, and other inorganic material systems such as titanium. These substrates can be used directly or serve as masters for replica-molding of organic polymers such as poly(dimethylsiloxane) (PDMS), polystyrene (PS), poly-(methyl methacrylate) (PMMA), polycarbonate (PC), and poly(ethylene glycol) (PEG) for in vitro applications or biodegradable polymers such as poly(e-caprolactone) (PCL), poly(l-lactic acid) (PLA), poly(glycolic acid) (PGA), and poly(l-lactic-co-glycolic acid) (PLGA) for potential use in vivo. Although the aforementioned biodegradable candidate materials are ubiquitous in biomedical applications, they have significant drawbacks, including bulk degradation and rigid mechanical properties for PLA, PGA, and PLGA. Noncompliant materials can result in localized inflammation within the dynamic in vivo mechanical environment(Nijst, Bruggeman et al. 2007; Bettinger, Bruggeman et al. 2008; Bruggeman, de Bruin et al. 2008). Novel material selection is of critical importance as cell–topography interactions continue to be utilized in tissue-engineering applications. Synthetic and natural materials must not only be selected on the basis of cell–biomaterial interactions but also on the basis of compatibility with nanofabrication processes. Natural proteins exhibit many advantages, including favorable cell–biomaterial interactions.

However, difficulty in processing and the potential for immune response(Ellingsworth, DeLustro et al. 1986; Garcia-Domingo, Alijotas-Reig et al. 2000) may limit widespread adoption. Synthetic biodegradable elastomers(Wang, Ameer et al. 2002; Nijst, Bruggeman et al. 2007; Bettinger, Bruggeman et al. 2008; Bruggeman, de Bruin et al. 2008) offer advantages such as ease of processing, variety of physical and mechanical properties, favorable tissue response and biodegradation kinetics(Wang, Kim et al. 2003; Bettinger, Bruggeman et al. 2009), and compatibility with nanofabrication techniques(Mahdavi, Ferreira et al. 2008; Bettinger, Kulig et al. 2009).

1.4 MECHANISMS OF CELL-TOPOGRAPHY INTERACTIONS

Studies demonstrating significant influence of nanoscale topographic features have yet to elucidate well-defined mechanisms of cell–topography interaction. In particular, several characteristics of this field of study confound the pursuit of precise interaction mechanisms. There are a virtually infinite number of potential combinations of cell types, biomaterial compositions, and topographic feature arrangements. Cell–topography interactions are also transient, which increases the difficulty in extricating a mechanistic view of the contact-guidance response(Dalby, Riehle et al. 2003). For example, cells in long-term culture can secrete additional extracellular matrix proteins, which can lead to convoluted topographic signaling(Hamilton, Wong et al. 2006). The large potential set of experiments and cell-specific outputs has resulted in a primarily phenomenological approach to studying cell–topography interactions. Despite this large body of work, little is known about the origin or underlying mechanism of the effect of topographical cues on cell function. Various theories have been proposed to explain such phenomena as the alignment and elongation of cells along the grating axis in nanograting substrates. Recent work has begun to interrogate cell–topography interactions with a focus on elucidating the mechanism, including identifying relevant signal transduction pathways and the role of organelles (including the cytoskeleton). Nevertheless, there is a significant opportunity to further explore cell–topography interactions, which could lead to refinement and more comprehensive predictive models of cell–topography interactions(Kemkemer, Jungbauer et al. 2006).

1.4.1 Current theories

We suggest that the morphological response serves as both an indicator of relevant cell–topography interactions and a basis for second-order effects. The elongation and alignment of the nucleus is presumably another source for alteration of the gene profile as cells respond to substrate topography(Dalby, Riehle et al. 2004). The generalized consensus regarding the mechanism for the morphological response is that it arises from the generation of anisotropic stresses. However, the precise origin and specific role of the anisotropic stresses is still under debate. Theories for the basis of cell–topography interactions will be discussed in the context of nanogratings. Contact-guidance kinetics of fibroblasts on titanium nanogratings suggest that microtubules align within 20 min after attachment and that their alignment preceded alignment of the overall cell(Oakley and Brunette 1993). This cluster of events is followed by the alignment of microfilament bundles at 40–60 min and focal adhesion contacts after 3 h. From this study it is clear that there are numerous organelles that are responsible for initializing and transmitting the effect of surface topography throughout the cell to influence overall cell functions, such as stress fiber formation, lamellipodia, and filopodia. One critical organelle that is thought to play an instrumental role in the contact-guidance response is filopodia(Dalby, Gadegaard et al. 2004), which could be modulated through Cdc42 activation(Teixeira, Abrams et al. 2003). While this model can explain the mechanism of detection and transmission of cell–topography interactions, several theories have been proposed to explain the origin of this response.

1.4.1.1 Focal adhesion formation

Topography can induce the overall alignment and elongation of cells by first inducing the alignment of focal adhesions. The initial alignment of focal adhesions could result from asymmetric probability of focal adhesion formation owing to feature geometry or geometrically restricted focal adhesion morphology. The alignment of focal adhesions could then lead to an overall response in the cell morphology through the aforementioned intimate signaling connection between focal adhesions and cytoskeleton proteins. Although this theory may explain the connection between aligned focal adhesions and the aligned, elongated gross morphology, it does not sufficiently address the initial alignment of focal adhesions.

1.4.1.2 Actin polymerization

Actin polymerization dynamics involved in cytoskeleton rearrangement are essential for cell attachment(Pierres, Benoliel et al. 2008) and serve as a driving force for directional migration and morphological alterations(Wojciak-Stothard, Curtis et al. 1995; Walboomers, Monaghan et al. 1999). Filopodia are highly motile organelles involved in many cellular processes, including migration(Nemethova, Auinger et al. 2008) and sensing of local topography(Galbraith, Yamada et al. 2007). Filopodia formation perpendicular to the ridge–groove features is hypothesized to occur less frequently owing to unfavorable stress formation. Conversely, the formation of filopodia parallel to the ridge–groove features is more frequent, which leads to biased propagation, cytoskeleton rearrangement, polarization of the cell body, and ultimately a gross morphological effect of alignment and elongation. Highly dynamic filopodia serve as topographical sensors, which are able to detect the immediate surrounding environment. This theory is most consistent with the body of work on this topic(Bettinger, Orrick et al. 2006).

1.4.2 Role of small GTPases

The Rho family of GTPases has been shown to control the formation and organization of filaments that compose the actin cytoskeleton(Hall 1998). This activation of Rho, Rac, and Cdc42 GTPases controls a wide spectrum of cell functions, including cytoskeleton formation and remodeling, alterations in gene expression, cell-cycle progression, cell morphogenesis, and cell migration in many cell types(Jaffe and Hall 2005; Tzima 2006). Hence, these molecular switches likely play a vital role in the concerted response of cells to substrate topography.

Recent studies have only begun to explore the role of these signaling pathways in the context of cell–topography interactions(Gerecht, Bettinger et al. 2007; Dalby, Hart et al. 2008). One key function that could directly connect topographic signaling to cell responses is spatially biased focal adhesion formation(Uttayarat, Toworfe et al. 2005) through Rho activation, which could have dramatic downstream effects on cell migration and signaling(Wang, Clark et al. 2005). Focal adhesions influence cell morphogenesis(Jungbauer, Gao et al. 2008) and have been shown to be sensitive to mechanical forces(Bershadsky, Kozlov et al. 2006). Future work must be conducted to further elucidate the dynamics between topographic signaling and modulation of cell function.

1.4.3 Advanced biological techniques

The impact of topography on gene expression is a somewhat intuitive extension of the effect on basic cell function. Recent quantification of the precise impact of topography on genome-wide expression has provided an enormous body of data, but has yet to elucidate any clear mechanisms. Studies that investigate individual signaling pathways that are likely to be implicated in cell–topography interactions may provide greater utility in understanding the origin of these responses. These studies can be conducted by examining cell–topography interactions in the presence of other signaling modalities—a trend that is evident in more recent studies that investigate the coupled effect of soluble factors and substrate topography on cell function(Gerecht, Bettinger et al. 2007; Dalby, Hart et al. 2008). Future studies should ideally span multiple cell types and substrates to confirm the generality of subsequent findings. Genetic manipulation(Putnam 2006) and gene knockdown using siRNA(Mullard 2007) are other techniques that would serve to identify and investigate the role of specific organelles and signaling pathways implicated in cell–topography interactions.

2 TOPOGRAPHIC EFFECT ON STEM CELLS

2.1 INTRODUCTION

Since the initial derivation of mouse embryonic stem (ES) cells in the early 1980s(Evans and Kaufman 1981; Martin 1981), the controlled differentiation of ES cells has progressed from a novelty to a standard approach to understanding lineage segregation in the early embryo, a model of developmental stages not normally accessible for study, and a potential source of cells for replacement following injury or disease. Embryonic stem cells have also provided a platform to study pathways to differentiation and maintenance of pluripotency that are likely to have applications to the broad field of stem cell biology, cancer stem cell biology, and the understanding of how development may go awry. Application of array-based technologies, gene trap and modifications, such as cell trapping, as well as techniques to rapidly express and down regulate gene expression in ES cells, have significantly improved our understanding of stem cell biology and lineage differentiation(O'Shea 2001; Smith 2001; Loebel, Watson et al. 2003).

2.1.1 Stem cell differentiation

Stem cells are unspecialized precursor cells with the ability to self-renew and differentiate into specialized cells in response to the appropriate signals(Ding and Schultz 2004). The recent interest in stem cells is related to their promising applications as cell sources for cell therapy, tissue repair and implants, among others(Kajstura, Rota et al. 2005; Hwang, Varghese et al. 2008). Apart from therapeutic applications, stem cells are being used to understand the complex molecular and cellular events occurring during early development, disease progression, epigenetics and pathophysiology(Hwang, Varghese et al. 2008).

Reports from the last decade have shown that it is possible to obtain stem cells from the embryo, called embryonic stem cells (ESCs), or from different sources of adult human cells, such as bone marrow, umbilical cord blood, adipose and neural tissue. Adult stem cells are called multipotent cells, because they are able to differentiate into several cell phenotypes, if adequate stimuli are provided at the appropriate times. For example, mesenchymal stem cells (MSCs) were differentiated successfully into osteoblasts, chondrocytes, adipocytes, neurons and cardiac cells(Pittenger, Mackay et al. 1999; Kajstura, Rota et al. 2005; Moviglia, Varela et al. 2006). Clinical therapeutic applications are not far away; for instance, currently, bone marrow derived-MSCs are undergoing clinical trials for cardiac and orthopedic applications(Hwang, Varghese et al. 2008).

Figure 6. Differentiation of human tissues

Figure 2.1 Differentiation of human tissues. Human development begins when a sperm fertilizes an egg and creates a single cell that has the potential to form an entire organism, called the zygote. In the first hours after fertilization, this cell divides into identical cells. These cells then begin to specialize, forming a hollow sphere of cells, called a blastocyst. The blastocyst has an outer layer of cells (yellow), and inside this hollow sphere, there is a cluster of cells called the inner cell mass (ICM) (light blue). The inner cell mass can give rise to the germ cells—eggs and sperm—as well as cells derived from all three germ layers (ectoderm; mesoderm and endoderm), depicted in the bottom panel, including nerve cells, muscle cells, skin cells, blood cells, bone cells, and cartilage. Image taken from the National Institutes of Health.

ESCs are usually regarded as the ideal model for regenerative medicine applications. ESCs are pluripotent cells derived from the inner cell mass of blastocysts (Figure 2.1). They can be expanded indefinitely and they can differentiate into derivatives of the three germinal layers (ectoderm, mesoderm, and endoderm); therefore, they constitute a source for obtaining a large number of differentiated cells that can be used in cell therapy, tissue engineering and new drug-screening strategies. Mature hepatocytes obtained by the co-culture of ESCs with a mesodermal cell line, under selective culture conditions, can establish a source of cells for pharmaceutical studies and toxicology(Baharvand, Hashemi et al. 2008; Shiraki, Umeda et al. 2008).

Although human stem cells, both adult and embryonic, possess huge potential for biotechnology and regenerative medicine, before stem cell-based therapies can be transferred to clinics, many fundamental biological and engineering challenges need to be overcome. These factors include the control of the self-renewal process, directing the specific stem cell differentiation and cell in vivo delivery(Hwang, Varghese et al. 2008). In vivo, stem cell fate is controlled by intrinsic factors and the cellular environment, which constitute the so called cell niches(Fuchs, Tumbar et al. 2004; Moore and Lemischka 2006). In vitro, stem cells can also differentiate spontaneously but this differentiation process is inefficient and leads to highly heterogeneous cell populations(Ding and Schultz 2004). That is why, in vitro, typical differentiation strategies involve the culture of stem cells in tissue culture plates with a medium enriched by a combination of soluble differentiation factors and animal serum, which leads to the induction of predominant phenotypes for the desired fate after periods ranging from 1 to 3 weeks(Pittenger, Mackay et al. 1999).

However, the signals leading to stem cell differentiation into most adult cell types are largely unknown(Murray and Edgar 2004) and the specific response depends on the time, concentration and combination of applied factors, as well as a pure population of lineage-specific progenitors, which are difficult to obtain. Therefore, mostly, stem cell differentiation in vitro results in a mixture of the desired cell type with other cell types: there is no existing procedure for generating 100% differentiated cells. Nontarget cells develop aberrant tissue formation in transplants and undifferentiated cells present a large risk of teratocarcinoma formation. Therefore, a big challenge still exists to develop strategies for directed differentiation of stem cells to specialized functional cell types, with the stringent manufacturing practice routines needed in clinical therapies.

The problem of controlling the self-renewal of stem cells, while keeping them undifferentiated, is another issue requiring a solution. As yet, factors determining stem cell self-renewal have not been elucidated and feeder layers, which are sometimes nonhuman cells, are required. The ideal culture method for human ESCs based cell and tissue therapy would be a defined culture, free of animal components and a feeder layer(Hwang, Varghese et al. 2008). Efforts to create defined cell culture substrates free of animal components have been intensive. A polyamide-based 3D nanofibrillar porous matrix (Ultra-Web)TM to support ex vivo expansion of mouse ESCs in a LIF-based medium has been reported(Nur-E-Kamal, Ahmed et al. 2006). However, this technique does not work for human cells. Both drawbacks, of feeder-free cultures and directed differentiation, could be addressed by using synthetic extracellular matrix (ECM)-like supports with controlled biomechanical and chemical properties. These biomaterials could recreate microenvironments that include signals from specific topographies, to morphogenetic factors Hhgs (hedgehog proteins), Wnts proteins, Notch ligands, TGF-β, bone morphogenetic proteins (BMPs), FGFs and ECM proteins (collagen, fibronectin [FN], vitronectin and laminin) to modulate cellular functions(Lutolf and Hubbell 2005; Hwang, Varghese et al. 2008). The underlying idea relies on the fact that, once differentiation by the interpretation of morphogenetic signals has occurred, local cell–cell interactions establish boundaries between different populations of cells(Lutolf and Hubbell 2005).

2.1.2 Topographical substrates for stem cell differentiation

Structured 2D surfaces have been widely used to examine the effect of surface topography on the characteristics of stem cells by using inhomogeneous scaffolds found in nature, such as silk fibers(Fan, Liu et al. 2008), and synthetic, engineered materials, especially polymers(Karageorgiou and Kaplan 2005; Lutolf and Hubbell 2005). Substrate topography is, in principle, a highly attractive method for regulating cell function because it is a noninvasive, nonbiological system that does not involve biomolecules(Lim and Donahue 2007). Recent studies have examined cell-structure interactions in a systematic way, using 2D structures consisting of arrays of micro- or nanostructures(Zahor, Radko et al. 2007) or, in the case of 3D scaffold matrices, with increased control of pore size(Tsuruma, Tanaka et al. 2008). Physical structuring can be undertaken using serial fabrication methods, such as electron beam or ion beam lithographies(Vieu, Carcenac et al. 2000; Dalby, Gadegaard et al. 2007; Martinez, Engel et al. 2008), or parallel methods, such as nanoembossing, conventional UV or UV nanoimprinting lithographies(Gadegaard, Martines et al. 2006). Simpler and cheaper methods involving direct production of regular, aligned, nanoscale structures, especially in polymer materials, are being developed and may become useful in the future. For example, one such technique simply involves the fracturing of a polymer film to produce a 60 nm grating(Pease, Deshpande et al. 2007). The materials for such structures(Dawson, Mapili et al. 2008) include the inorganic materials already used in common surgical applications, mainly silicon based(Lipski, Pino et al. 2008), ceramic(Kasten, Beyen et al. 2008), or metallic(Lipski, Jaquiery et al. 2007). Increasingly, a variety of biocompatible organic materials, mainly polymers, are being used owing to their low cost and simple engineering of sub-micrometer structures(Dawson, Mapili et al. 2008).

The micro- and nanostructures designed for cell culture assays are mainly basic shapes as mentioned above, such as gratings, posts or pits, of different dimensions and are disposed in an ordered or random fashion, depending on the method of preparation. The new fabrication techniques have enabled substrates with controlled topographies to be produced that affect a range of stem cell types, such as neuronal stem cells(Tsuruma, Tanaka et al. 2008), MSCs(Zahor, Radko et al. 2007), murine embryonic stem cells(Rosenthal, Macdonald et al. 2007) and neural progenitor embryonic stem cells(Willerth, Rader et al. 2008), to name but a few. Such structures extend to the nanoscale, which also affects the gene expression(Andersson, Backhed et al. 2003) and differentiation(Popat, Chatvanichkul et al. 2007) of stem cells, as well as the physical properties(Johansson, Carlberg et al. 2006).

Grating structures in the microrange (from 1 to 20 um in size) are successful in directing stem cell differentiation to neural phenotypes. Grating polystyrene substrates chemically modified with laminin were used to enhance the neuronal differentiation of adult rat hippocampal progenitor cells in co-culture with astrocytes(Recknor, Sakaguchi et al. 2006). Kim et al. have also shown recently that microstructured poly(dimethylsiloxane) (PDMS) substrates, with features 1 um in size, significantly increase the differentiation of umbilical cord blood-derived cells towards neuronal cells with respect to flat surfaces and microstructured surfaces with larger features (2–4 um)(Kim, Lee et al. 2008). The mechanism behind gene-expression modification in cells by line topography has been related by some authors with the cell nuclei elongation and alignment. It has been suggested, and supported by several experimental works(Dalby, Biggs et al. 2007; Dalby, Gadegaard et al. 2007), that mechanical tension causing the alignment of cells rearranges centromeres through deformation of the nucleus and hence affect gene expression. The work of Maniotis points out that the nucleus is mechanically integrated with the physical entity of the cell through intermediate filaments(Maniotis, Chen et al. 1997); thus, active or passive cell extension can lead to passive nuclear deformation.

More recent works have shown that nanoscale grating structures are an even more useful strategy for neural cell differentiation. Yim and coworkers(Yim, Pang et al. 2007) investigated the influence of nanoscale topography on the transdifferentiation of hMSCs, (i.e., differentiation into specific nondefault pathways, such as neuronal cells). This transdifferentiation, which is not well understood, can be induced by neuronal induction medium or cell contact with neurons. In this work, hMSCs were cultured on PDMS samples structured with 350 nm wide lines and this showed that cells and nuclei were significantly elongated: neuronal and muscular gene markers were up regulated even in culture medium containing no biochemically inducing factors. The authors claimed that cell alignment and elongation needs lines at least one order of magnitude smaller than the cell body. Although the exact signaling pathways are still not known, it is stated that cytoskeleton rearrangement and nuclei elongation can have an important role in creating the appropriate signal transduction chain. The basal cell membrane, in contact with the nanostructured surface, will be subjected to mechanical forces that will open ion channels and rearrange cellular components, such as integrin clusters, transmembrane proteins, which affect cell adhesion, cytoskeleton organization and even gene expression(Curtis, Dalby et al. 2006; Dalby, Biggs et al. 2007; Dalby, Gadegaard et al. 2007). Further investigations will be needed to identify the components in the cytoskeleton and the signaling pathways that govern the topography-directed cell differentiation(Yim, Pang et al. 2007). It has already been observed that changes to the cytoskeleton lead to altered stress levels applied to the nucleus(Feldherr and Akin 1993) that could affect organelle and DNA organization and distribution, ultimately altering cell function.

The alteration of cellular function at micro and nanostructured interfaces may result from the direct influence of cellular response (as in the example of nuclei elongation) to local mechanical stimulus(McBeath, Pirone et al. 2004), but it has also been associated with an altered ECM layer deposited on the surface because topography at the nanoscale alters the functional behavior of adhesive and connective tissue proteins(Dalby, McCloy et al. 2006). Nanostructures produce size-dependent effects on cell cultures both at the cellular and subcellular levels. For example, substrates coated with nanoparticles induce cell-type specific effects on the morphology and metabolism of stem cells depending on the particle sizes(Kunzler, Huwiler et al. 2007; Lipski, Pino et al. 2008). Surfaces coated with nanoparticles of small diameter (<100 nm) induced high rates of proliferation (compared with a control), whereas cells grown on larger nanoparticles exhibited inhibited proliferation. Nanoparticle-modified surfaces, such as these, can enhance the differentiation of human marrow-derived mesenchymal progenitor cells towards an osteogenic lineage in the presence of soluble signaling molecules(Lipski, Jaquiery et al. 2007).

Similarly, in porous 3D scaffolds, the porosity and pore size of the scaffold influence osteogenic differentiation, as well as protein production in hMSCs(Kasten, Beyen et al. 2008). Osteogenic differentiation was enhanced in scaffolds with a high percentage of pores with diameters similar to that of the cell diameter. Results show that hMSCs cultured on a nanofibrous scaffold composed of a thermally responsive hydroxybutyl chitosan (HBC) are able to form aligned cell sheets. This scaffold provides nanometer-scale topographical cues to the cell with the potential to modulate cell behavior. hMSCs adapt their cytoskeleton and nuclear shape to the environment. The elongation experienced by the cells and their nuclei is associated with the activation of intracellular signaling cascades and the control of gene expression. hMSCs on HBC fibers and HBC/collagen-blended fibers present an up regulation of myogenic genes as a result of cell and nuclear elongation(Dang and Leong 2007).

The ability of progenitor cells to respond to ordered and random topographies at the nanoscale have been studied by Dalby et al.(Dalby, McCloy et al. 2006). In this work, the authors created substrates with semi-ordered topographies at the nanoscale by colloidal lithography and substrates with random nanoscale topographies by polymer demixing. They produced replicas of these substrates by nanoembossing to ensure the same surface chemistry and hMSCs were assayed for osteoblast differentiation. The nanotopographies increased cell spreading and expression markers for osteoblast phenotypes and proved that hMSCs react strongly to topographic surface features down to 10 nm in height, even if they have low aspect ratios. The authors suggest that the first reaction of cells is the formation of filopodia, which have the role of a sensory system, and it is the filopodial guidance that leads to changes in cell morphology and differentiation, a process that requires features smaller than 50 nm in size. The response of stem cells to nanoscale symmetries was also studied by the same authors(Dalby, Gadegaard et al. 2007), again with nanoembossed samples of PMMA, but obtained from moulds fabricated by electron- beam lithography with different but well controlled degrees of symmetry (square arrays, hexagonal arrays, distorted arrays and random pit arrays). hMSCs cultured in these substrates proved to have an osteoblast-differentiation behavior dependant on the structure symmetry; the topographies with controlled disorder giving the highest levels of osteoblast differentiation.