Overlooked? Underestimated? Effects of Substrate Curvature on Cell Behavior
2019; Elsevier BV; Volume: 37; Issue: 8 Linguagem: Inglês
10.1016/j.tibtech.2019.01.006
ISSN0167-9430
AutoresDanielle Baptista, Liliana Moreira Teixeira, Clemens van Blitterswijk, Stefan Giselbrecht, Roman Truckenmüller,
Tópico(s)Silk-based biomaterials and applications
ResumoThere is increasing evidence that substrate curvature on a (near‐)cell scale affects cell fate. High-resolution rapid prototyping/additive manufacturing technologies – including stereolithography, two-photon polymerization (2PP) laser lithography, and digital mirror device-based digital light processing – can create structures with defined, complex (out-of-plane) curvature. 2PP technology can create smooth structures or structures with defined superimposed surface roughness, texture, or topography. Curvature chip technologies are about to drastically ease systematic studies on cell–curvature interactions, and to enable the (re)creation of microanatomically shaped cellular microenvironments in tissues/organs on chips. These new techniques are expected to change how cell–biomaterial interfaces in vitro and in vivo will be engineered in the future. In biological systems, form and function are inherently correlated. Despite this strong interdependence, the biological effect of curvature has been largely overlooked or underestimated, and consequently it has rarely been considered in the design of new cell–material interfaces. This review summarizes current understanding of the interplay between the curvature of a cell substrate and the related morphological and functional cellular response. In this context, we also discuss what is currently known about how, in the process of such a response, cells recognize curvature and accordingly reshape their membrane. Beyond this, we highlight state-of-the-art microtechnologies for engineering curved biomaterials at cell-scale, and describe aspects that impair or improve readouts of the pure effect of curvature on cells. In biological systems, form and function are inherently correlated. Despite this strong interdependence, the biological effect of curvature has been largely overlooked or underestimated, and consequently it has rarely been considered in the design of new cell–material interfaces. This review summarizes current understanding of the interplay between the curvature of a cell substrate and the related morphological and functional cellular response. In this context, we also discuss what is currently known about how, in the process of such a response, cells recognize curvature and accordingly reshape their membrane. Beyond this, we highlight state-of-the-art microtechnologies for engineering curved biomaterials at cell-scale, and describe aspects that impair or improve readouts of the pure effect of curvature on cells. In living systems, geometric form and biological function are inherently linked together on all scales. The diversity of such systems or organisms is expressed in a plethora of forms or shapes, but with a striking prevalence of one major class of shapes: The outer appearance of organisms is dominated by round(ed) shapes or curved surfaces, a phenomenon which continues inside at interfaces between tissues or at boundaries between tissues and body lumens (or the fluids or air contained therein); curvature also manifests itself under microscopic evaluation (Figure 1, Key Figure). An example of the relationship between curved form and biological or physiological function at a macroscopic level is the biomechanical damping contribution of the double S-shape of the human spine. Concomitantly, there is strong evidence that the loss of original shape is a cause or consequence of a disease. For example in keratoconus, an eye disorder, the curved cornea thins out and bulges like a cone, resulting in blurry and distorted vision. At a microscopic, cellular level, though, the curved form–biological function relationship is still widely unexplored. Over several decades, numerous studies have shown the influence of cellular- and subcellular-scale topography of (flat) culture substrates on cell fate, such as in a landmark paper by Dalby and colleagues [1Dalby M.J. et al.The control of human mesenchymal cell differentiation using nanoscale symmetry and disorder.Nat. Mater. 2007; 6: 997-1003Crossref PubMed Scopus (1994) Google Scholar]. Other substrate properties such as substrate chemistry have been investigated similarly extensively [2Anderson D.G. et al.Nanoliter-scale synthesis of arrayed biomaterials and application to human embryonic stem cells.Nat. Biotechnol. 2004; 22: 863-866Crossref PubMed Scopus (683) Google Scholar]; more recently, confined cell adhesiveness [3Chen C.S. et al.Geometric control of cell life and death.Science. 1997; 276: 1425-1428Crossref PubMed Scopus (4181) Google Scholar, 4McBeath R. et al.Cell shape, cytoskeletal tension, and RhoA regulate stem cell lineage commitment.Dev. Cell. 2004; 6: 483-495Abstract Full Text Full Text PDF PubMed Scopus (3393) Google Scholar] and matrix elasticity or stiffness [5Discher D.E. Tissue cells feel and respond to the stiffness of their substrate.Science. 2005; 310: 1139-1143Crossref PubMed Scopus (4806) Google Scholar] have also been studied. By contrast, far fewer studies have investigated the effect of substrate curvature on cell behavior. Early studies were conducted on glass fibers, as in 1964 when Curtis and Varde cultured chick heart fibroblasts on such substrates [6Curtis A.S.G. Varde M. Control of cell behavior: topological factors.J. Natl. Cancer Inst. 1964; 33: 15-26PubMed Google Scholar]. Other studies around that time were performed on glass beads [7Maroudas N.G. Anchorage dependence: correlation between amount of growth and diameter of bead, for single cells grown on individual glass beads.Exp. Cell Res. 1972; 74: 337-342Crossref PubMed Scopus (42) Google Scholar] or on rounded grooves/ridges copied into polyvinylchloride plates using stamps originating from modified discs for sound recording [8Rovensky Y.A. et al.Behaviour of fibroblast-like cells on grooved surfaces.Exp. Cell Res. 1971; 65: 193-201Crossref PubMed Scopus (86) Google Scholar]. In these studies, the effect of curvature often was not fully considered, or was at least not the main focus of the investigation. In addition to largely overlooking or underestimating the curvature effect on cell behavior for a long time, the lack of available methods to engineer the required complex substrate geometries in a controlled way might have contributed to the further delay of corresponding studies. The maximum curvature radius that can still be sensed by a cell stands in relation to the size of the cell and cannot be too different from it. Consequently, substrate engineering must occur somewhere at the milli- or micrometer range, or at a smaller scale. The aforementioned lack of engineering methods can be traced back to the fact that micromachining is based on 2½D (see Glossary) processes that have their origin in photolithographic patterning processes from the early semiconductor industry. With the advent of new, 3D-capable micro-/nanotechnologies such as two-photon polymerization (2PP) laser lithography (Figure 1), systematic studies screening for the cellular response to substrate curvature of different types at near-cell scales have become possible. This in turn can be expected to boost the development of a next generation of biomedical interfaces on and in devices ranging from biomaterial scaffolds for tissue engineering to microfluidic in vitro tissue or organ model systems for pharmaceutical testing. The review summarizes current knowledge and understanding of the effect of substrate curvature on cell response. Translating curved substrate geometry to the inherent molecular machinery of the cell as a consequence of mechanosensing and mechanotransduction includes events such as bending of the cell/plasma membrane and induction of cell polarity. We also review the state of the art of microtechnologies for both explicitly and implicitly engineering anatomically or biomimetically curved biomaterials at a microscale or at the cellular level. This condensed and structured information will help the readers to design and conduct their own advanced fundamental cell studies, or to develop and create innovative materials and devices with wide implications in the field of applied biosciences, such as in the areas of tissue engineering and regenerative medicine. Although still an unexplored field, cell behavior in 3D matrices is completely different from behavior in 2D/planar substrates of the same material [9Cukierman E. et al.Taking cell–matrix adhesions to the third dimension.Science. 2001; 294: 1708-1712Crossref PubMed Scopus (2436) Google Scholar]. Moreover, cells can discriminate between planar, convex, and concave surfaces (Figure 1). For example, fibroblasts can differentiate spherical convex substrate curvature up to a curvature diameter of 2 mm, above which they showed responses similar to those for a planar surface [10Lee S.J. Yang S. Micro glass ball embedded gels to study cell mechanobiological responses to substrate curvatures.Rev. Sci. Instrum. 2012; 83094302Crossref PubMed Scopus (11) Google Scholar]. So far, no general dimensional threshold for curvature sensing, such as the ratio between the size of a cell and the diameter of a curved surface, has been determined. This is probably because such a general curvature threshold would depend on (too) many assumed factors such as cell type or superimposed surface topography/roughness of the curved substrate, in each case leading to different results. Depending on cellular and substrate-related factors, cells are able to reshape and adapt to a given curved surface to different extents (Box 1 Figure IA, and Figure 2). Mechanotransduction of cells on convex surfaces is mediated by the BAR (Bin/amphiphysin/Rvs) domain proteins which can recognize and induce a corresponding bending of the cell membrane (Box 1 Figure IC, top). Upon contact of a cell membrane with a convex surface, the BAR domain releases small GTPases and binds to the membrane, inducing curvature [11Frost A. et al.The BAR domain superfamily: membrane-molding macromolecules.Cell. 2009; 137: 191-196Abstract Full Text Full Text PDF PubMed Scopus (447) Google Scholar]. It was found that various effectors of small GTPases participate in cell-cycle regulation and actin dynamics [12Nobes C.D. Hall A. Rho, Rac, and Cdc42 GTPases regulate the assembly of multimolecular focal complexes associated with actin stress fibers, lamellipodia, and filopodia.Cell. 1995; 81: 53-62Abstract Full Text PDF PubMed Scopus (3728) Google Scholar, 13Olson M.F. et al.An essential role for Rho, Rac, and Cdc42 GTPases in cell cycle progression through G1.Science. 1995; 269: 1270-1272Crossref PubMed Scopus (1058) Google Scholar]. Consequently, these actively regulate proliferation, cell shape, polarity, and locomotion. Thus, it is suggested that convex surfaces have a crucial effect on the cell cycle and the cytoskeleton.Box 1Bending Cell MembranesAccording to McMahon and Gallop, five mechanisms of inducing cell-membrane deformation have been reported (Figure IB) [82McMahon H.T. Gallop J.L. Membrane curvature and mechanisms of dynamic cell membrane remodelling.Nature. 2005; 438: 590-596Crossref PubMed Scopus (1605) Google Scholar]. These mechanisms are based on lipid composition modification (by conical lipids), clustering of shaped (trans)membrane proteins, cytoskeletal scaffolding, protein scaffolding including oligomerization of BAR domain proteins (Figure IC), and protein motif/amphipathic helix insertions. Their function in the process is not independent of each other; it is rather the combinatorial effect of all these mechanisms that leads to drastic changes in cell shape. The cellular membrane has a spontaneous shape (unstressed state) that is characterized by the spontaneous curvature of the membrane bilayer [83Helfrich W. Elastic properties of lipid bilayers: theory and possible experiments.Z. Naturforsch. C. 1973; 28: 693-703Crossref PubMed Scopus (4962) Google Scholar]. This curvature depends on the spontaneous curvature of the inner and outer layers of the membrane. The curvature of each layer in turn is governed by the composition (acyl chains and/or headgroups) of the lipids in the layer. The spatial and temporal lipid profile can be analyzed via mass spectrometry [84Harkewicz R. Dennis E.A. Applications of mass spectrometry to lipids and membranes.Annu. Rev. Biochem. 2011; 80: 301-325Crossref PubMed Scopus (149) Google Scholar].When modifications in the lipid profile are insufficient to bend the membrane, scaffolding membrane proteins such as from the BAR domain protein family are recruited, which deform a membrane by bracing it as a scaffold [85Zimmerberg J. Kozlov M.M. How proteins produce cellular membrane curvature.Nat. Rev. Mol. Cell Biol. 2006; 7: 9-19Crossref PubMed Scopus (1015) Google Scholar]. These proteins change the membrane curvature by applying pulling and bending forces to the membrane surface. The BAR and F-BAR domain proteins (Figure IC, top and middle) form a banana-shaped dimer of a three-helix coiled coil [86Habermann B. The BAR-domain family of proteins: a case of bending and binding?.EMBO Rep. 2004; 5: 250-255Crossref PubMed Scopus (241) Google Scholar]. The inverse BAR (I-BAR) domain proteins (Figure IC, bottom) are α-helical antiparallel dimers which display remote structural homology to BAR and F-BAR domains; however, the I-BAR domain has a zeppelin-shaped structure [14Zhao H. et al.I-BAR domain proteins: linking actin and plasma membrane dynamics.Curr. Opin. Cell Biol. 2011; 23: 14-21Crossref PubMed Scopus (136) Google Scholar, 87Linkner J. et al.The inverse BAR domain protein IBARa drives membrane remodeling to control osmoregulation, phagocytosis and cytokinesis.J. Cell Sci. 2014; 127: 1279-1292Crossref PubMed Scopus (24) Google Scholar]. Their natural conformation defines the type of curvature that they are able to recognize and induce. Therefore, BAR domain proteins are involved in sensing convexity and bend the membrane in a convex way, while I-BAR proteins are involved in sensing concavity and force the cellular membrane into a concave shape [11Frost A. et al.The BAR domain superfamily: membrane-molding macromolecules.Cell. 2009; 137: 191-196Abstract Full Text Full Text PDF PubMed Scopus (447) Google Scholar].Figure 2Impact of Curvature on Cell Morphology. Cell response to curved surfaces is dependent on cellular and substrate-related factors. Hemispherical cavities can induce different effects: (A) salivary gland cells (SIMS) formed a perfect monolayer in PLGA nanofiber-coated cavities with a diameter of 30 μm, whereas (B1) human mesenchymal stem/stromal cells (hMSCs) proliferated more on flat PDMS regions (B2) compared with PDMS cavities 200 μm in diameter (scale bars, 100 μm). (C1) Primary porcine aortic endothelial cells (PAECs) cultured in 600 μm diameter circular cylindrical channels did not show any effect of curvature because cells appear to be randomly organized, (C2) completely lining the channel (scale bars, 200 μm). Regarding convex substrates, there is a clear difference versus the previously mentioned concave examples. Fibroblasts cultured on PLGA fibers showed an inverse relationship between fiber diameter and alignment/elongation. Maximum elongation was registered with fibers of smaller diameters, such as (D1) 10 μm and (D2 and D4) 30 μm, whereas for fibers of (D3) 242 μm cell behavior was similar to that on flat surfaces. Fibroblasts were able to discriminate not only between grooved/ridged and flat substrates but also between sharp and rounded/curved ridges. (E1) Cells on sharp grooved substrates elongated and aligned (scale bar, 20 μm), (E3) cells on flat substrates were mostly uniformly spread, and (E2) cells on rounded grooved substrates showed a morphology between those of cells on sharp and flat substrates. Panels (A–E) reproduced/adapted, with permission, from [17Soscia D.A. et al.Salivary gland cell differentiation and organization on micropatterned PLGA nanofiber craters.Biomaterials. 2013; 34: 6773-6784Crossref PubMed Scopus (49) Google Scholar], [16Park J.Y. et al.Study of cellular behaviors on concave and convex microstructures fabricated from elastic PDMS membranes.Lab Chip. 2009; 9: 2043-2049Crossref PubMed Scopus (96) Google Scholar], [46Fiddes L.K. et al.A circular cross-section PDMS microfluidics system for replication of cardiovascular flow conditions.Biomaterials. 2010; 31: 3459-3464Crossref PubMed Scopus (131) Google Scholar], [57Hwang C.M. et al.Controlled cellular orientation on PLGA microfibers with defined diameters.Biomed. Microdevices. 2009; 11: 739-746Crossref PubMed Scopus (111) Google Scholar], and [19Mathur A. et al.The role of feature curvature in contact guidance.Acta Biomater. 2012; 8: 2595-2601Crossref PubMed Scopus (41) Google Scholar], respectively. Abbreviations: PDMS, polydimethylsiloxane; PLGA, poly(lactic-co-glycolic acid).View Large Image Figure ViewerDownload Hi-res image Download (PPT) According to McMahon and Gallop, five mechanisms of inducing cell-membrane deformation have been reported (Figure IB) [82McMahon H.T. Gallop J.L. Membrane curvature and mechanisms of dynamic cell membrane remodelling.Nature. 2005; 438: 590-596Crossref PubMed Scopus (1605) Google Scholar]. These mechanisms are based on lipid composition modification (by conical lipids), clustering of shaped (trans)membrane proteins, cytoskeletal scaffolding, protein scaffolding including oligomerization of BAR domain proteins (Figure IC), and protein motif/amphipathic helix insertions. Their function in the process is not independent of each other; it is rather the combinatorial effect of all these mechanisms that leads to drastic changes in cell shape. The cellular membrane has a spontaneous shape (unstressed state) that is characterized by the spontaneous curvature of the membrane bilayer [83Helfrich W. Elastic properties of lipid bilayers: theory and possible experiments.Z. Naturforsch. C. 1973; 28: 693-703Crossref PubMed Scopus (4962) Google Scholar]. This curvature depends on the spontaneous curvature of the inner and outer layers of the membrane. The curvature of each layer in turn is governed by the composition (acyl chains and/or headgroups) of the lipids in the layer. The spatial and temporal lipid profile can be analyzed via mass spectrometry [84Harkewicz R. Dennis E.A. Applications of mass spectrometry to lipids and membranes.Annu. Rev. Biochem. 2011; 80: 301-325Crossref PubMed Scopus (149) Google Scholar]. When modifications in the lipid profile are insufficient to bend the membrane, scaffolding membrane proteins such as from the BAR domain protein family are recruited, which deform a membrane by bracing it as a scaffold [85Zimmerberg J. Kozlov M.M. How proteins produce cellular membrane curvature.Nat. Rev. Mol. Cell Biol. 2006; 7: 9-19Crossref PubMed Scopus (1015) Google Scholar]. These proteins change the membrane curvature by applying pulling and bending forces to the membrane surface. The BAR and F-BAR domain proteins (Figure IC, top and middle) form a banana-shaped dimer of a three-helix coiled coil [86Habermann B. The BAR-domain family of proteins: a case of bending and binding?.EMBO Rep. 2004; 5: 250-255Crossref PubMed Scopus (241) Google Scholar]. The inverse BAR (I-BAR) domain proteins (Figure IC, bottom) are α-helical antiparallel dimers which display remote structural homology to BAR and F-BAR domains; however, the I-BAR domain has a zeppelin-shaped structure [14Zhao H. et al.I-BAR domain proteins: linking actin and plasma membrane dynamics.Curr. Opin. Cell Biol. 2011; 23: 14-21Crossref PubMed Scopus (136) Google Scholar, 87Linkner J. et al.The inverse BAR domain protein IBARa drives membrane remodeling to control osmoregulation, phagocytosis and cytokinesis.J. Cell Sci. 2014; 127: 1279-1292Crossref PubMed Scopus (24) Google Scholar]. Their natural conformation defines the type of curvature that they are able to recognize and induce. Therefore, BAR domain proteins are involved in sensing convexity and bend the membrane in a convex way, while I-BAR proteins are involved in sensing concavity and force the cellular membrane into a concave shape [11Frost A. et al.The BAR domain superfamily: membrane-molding macromolecules.Cell. 2009; 137: 191-196Abstract Full Text Full Text PDF PubMed Scopus (447) Google Scholar]. Similarly to convex surfaces, several proteins, such as inverse BAR (I-BAR) domain proteins, have been identified to play a role in mechanotransduction of cells on concave surfaces (Box 1 Figure IC, bottom). Whether the function of I-BAR domain proteins is to sense membrane (and substrate) curvature or to promote membrane bending is not fully understood. Thereby, these two functions do not need to be mutually exclusive. Both mechanisms may act simultaneously to efficiently sense and support membrane deformation [14Zhao H. et al.I-BAR domain proteins: linking actin and plasma membrane dynamics.Curr. Opin. Cell Biol. 2011; 23: 14-21Crossref PubMed Scopus (136) Google Scholar]. Possibly, membrane curvature sensing and/or generation is highly dependent on the local concentration of the activated I-BAR domain proteins in the cell. For example, at low concentrations, these proteins might predominantly have a sensory function. Curvature sensing could also lead to opening of mechanogated ion channels [15Patel A.J. et al.Lipid and mechano-gated 2P domain K+ channels.Curr. Opin. Cell Biol. 2001; 13: 422-427Crossref PubMed Scopus (245) Google Scholar]. These have been also considered as part of the mechanotransduction machinery of curved surfaces. Potentially before and instead of other, forced morphological and functional cell responses, when the substrate design allows this, sensing or probing the substrate might lead to escape from a particular curved location rather than to seek for it. In a study by Park and colleagues, the behavior of fibroblasts on concave and convex spherical microstructures made from polydimethylsiloxane (PDMS) was investigated, and fibroblasts were not reluctant to climb on the convex structures. Conversely, the same cells avoided concave surfaces, or entered the microwells briefly (< 10 h) before escaping to the surrounding flat region [16Park J.Y. et al.Study of cellular behaviors on concave and convex microstructures fabricated from elastic PDMS membranes.Lab Chip. 2009; 9: 2043-2049Crossref PubMed Scopus (96) Google Scholar]. However, in a contrasting study, cells of two immortalized salivary gland epithelial cell lines (ductal and acinar) were seeded inside hemispherical craters created from PDMS and coated with poly(lactic-co-glycolic acid) (PLGA) nanofibers where the cells (stayed and) successfully formed curved confluent monolayers lining the concavities [17Soscia D.A. et al.Salivary gland cell differentiation and organization on micropatterned PLGA nanofiber craters.Biomaterials. 2013; 34: 6773-6784Crossref PubMed Scopus (49) Google Scholar]. As already has been the case in the historical studies with fibroblasts, cylindrically curved structures such as fibers, tubes, and rounded ridges are often found to induce cell-body elongation and alignment along the longitudinal axis of the structure. Together with a corresponding directional organization of cellular stress fibers, this can be partly assigned to the well-known contact-guidance phenomenon (Figure 2E) [18Dunn G.A. Heath J.P. A new hypothesis of contact guidance in tissue cells.Exp. Cell Res. 1976; 101: 1-14Crossref PubMed Scopus (307) Google Scholar, 19Mathur A. et al.The role of feature curvature in contact guidance.Acta Biomater. 2012; 8: 2595-2601Crossref PubMed Scopus (41) Google Scholar]. For example, human fetal osteoblasts (HFObs) were reported to orient along microchannels copied into hydroxyapatite from parallel densely packed round metal wires [20Pilia M. et al.Influence of substrate curvature on osteoblast orientation and extracellular matrix deposition.J. Biol. Eng. 2013; 7: 23Crossref PubMed Scopus (23) Google Scholar]. On day 6 of culture, the strongest nuclear alignment was found for 250 μm diameter channels, while on day 18, the strongest alignment was found for the 100 μm diameter channels. The cells in the (less curved) 500 μm channels were always less organized. However, Levina and colleagues reported that rat epithelial cells of the IAR-2 line formed straight actin microfilament bundles and (extracellular) fibronectin- or laminin-positive fibrils that were predominantly oriented transversely to the cylinder axis of glass fibers with a diameter of 32 μm on which they were cultured [21Levina E.M. et al.Cylindrical substratum induces different patterns of actin microfilament bundles in nontransformed and in ras-transformed epitheliocytes.Exp. Cell Res. 1996; 229: 159-165Crossref PubMed Scopus (21) Google Scholar]. By contrast, the majority of their N-Ras-transformed descendants, IAR-Ras-c4 cells, on acquiring a polarized cell morphology, formed microfilament bundles and extracellular matrix (ECM) fibrils oriented approximately longitudinally to the fiber axes, similarly to normal polarized cells such as fibroblasts. In another study, endothelial colony-forming cells (ECFCs) cultured on electrospun scaffolds with fiber diameters of 5–11 μm were documented to align their cytoskeleton along the fiber axes, whereas human umbilical vein endothelial cells (HUVECs) cultured on the same scaffolds developed a cytoskeleton organized circumferentially around the fibers [22Fioretta E.S. et al.Differential response of endothelial and endothelial colony forming cells on electrospun scaffolds with distinct microfiber diameters.Biomacromolecules. 2014; 15: 821-829Crossref PubMed Scopus (44) Google Scholar]. Ye and coworkers reported that human brain microvascular endothelial cells (HBMECs) cultured on glass rods with diameters of 10–500 μm 'resist' elongation in response to the curvature of the rod. The authors hypothesize that the phenotype of HBMECs may have evolved to minimize the length of tight junctions per unit length of capillary, and hence minimize paracellular transport into the brain [23Ye M. et al.Brain microvascular endothelial cells resist elongation due to curvature and shear stress.Sci. Rep. 2014; 4: 1-6Google Scholar]. By contrast, HUVECs this time elongated along the axes of the rods instead of wrapping around them, thereby minimizing the curvature effect. In summary, for anisotropically curved substrate surfaces such as circular cylindrical surfaces, in the first instance, anisotropic morphological responses of cells such as their elongation and alignment can be expected and could clearly be demonstrated. For isotropically curved surfaces such as spherical surfaces, in the absence of directed stimuli such as matrix-mediated or fluidic (shear) forces, and/or substrate-bound or soluble molecular (gradient) signals, isotropic or random anisotropic cell responses can be anticipated. However, this does not exclude events such as spontaneous local self-alignment, as found with myoblasts [24Junkin M. et al.Cellular self-organization by autocatalytic alignment feedback.J. Cell Sci. 2011; 124: 4213-4220Crossref PubMed Scopus (45) Google Scholar]. Morphological differences between cells on less curved, or flat, and more curved substrates can then still be found as scalar variations such as cell area or aspect ratio. For fibroblasts grown on glass balls and plates, for instance, the cell spread area increased with increasing ball diameter and reached its maximum for the flat substrates [10Lee S.J. Yang S. Micro glass ball embedded gels to study cell mechanobiological responses to substrate curvatures.Rev. Sci. Instrum. 2012; 83094302Crossref PubMed Scopus (11) Google Scholar]. Cells such as epithelial cells, neurons, and migrating cells are naturally polarized due to an asymmetrical distribution of proteins and lipids along the cell-membrane leaflets that impose directionality in their different functions. Polarized cells within an epithelial monolayer exhibit a 'nonadhesive' apical domain, and an 'adhesive' basolateral surface, the latter characterized by interactions between cells and the ECM/basement membrane beneath, and between neighboring cells, such as (by) tight junctions [25Nelson W.J. Remodelling epithelial cell organization: transitions between front-rear and apical basal polarity.Cold Spring Harb. Perspect. Biol. 2009; 1: 1-19Crossref Scopus (204) Google Scholar, 26Cao X. et al.Polarized sorting and trafficking in epithelial cells.Cell Res. 2012; 22: 793-805Crossref PubMed Scopus (95) Google Scholar]. Curved surfaces are thought to facilitate the formation of such tight junctions not only by stimulating the production of occludins, functional components of tight junctions [17Soscia D.A. et al.Salivary gland cell differentiation and organization on micropatterned PLGA nanofiber craters.Biomaterials. 2013; 34: 6773-6784Crossref PubMed Scopus (49) Google Scholar], but also by inducing a specific localization of distinct actin-based cytoskeletal structures in adherent cells [27James J. et al.Subcellular curvature at the perimeter of micropatterned cells influences lamellipodial distribution and cell polarity.Cell Motil. Cytoskeleton. 2008; 65: 841-852Crossref PubMed Scopus (80) Google Scholar]. In neurons, polarity is essential for the propagation of electrical signals through the axon in a unidirectional manner. It was demonstrated that by varying a simple topographical parameter – the width of substrate ridges – the orientation and maturation of focal adhesions could be modulated, yielding independent control over the final number and direction of neurite outgrowths [28Ferrari A. et al.Nanotopographic control of neuronal polarity.Nano Lett. 2011; 11: 505-511Crossref PubMed Scopus (113) Google Scholar, 29Rajnicek A. et al.Contact guidance of CNS neurites on grooved quartz: influence of groove dimensions, n
Referência(s)