Topological Control of Cell Behavior

Independent of biochemistry, topographical cues from the extracellular matrix may have significant effects on cellular behavior. Studies have demonstrated that substratum topography has direct effects on the ability of cells orientation, migration, and organization of cytoskeleton. Basement membranes are thin sheets which composed of extracellular matrix (ECM) proteins and spread through the vertebrate body, serving as substrates for constructing cellular structures. The topography of basement membranes is a composite meshwork by pores, fibers, and ridges in features of nanometer sized dimensions. Synthetic surfaces fabricated with micro-

and nanoscaled topographical features have been shown to influence cell behavior.

These accounts lead to the hypothesis that in regulating cellular behavior, the topography of the basement membrane may plays an important role independent from that of the chemistry of the basement membrane.

In addition to chemical and physical-mechanical properties, basement membranes possess a complex, three-dimensional topography which consist of micro- and nanometer sized features. Physical topography is known to affect cell behavior. Paul Weiss, among others, pioneered the field of “contact guidance” during the 1930s, 1940s and 1950s [38]. As early as 1962 and 1963, Rosenberg claimed that nanometer sized features influence cellular behavior [39]. Despite recognition of the importance of substratum topography, relatively little is known about the effects of topographical features of nanometer scale on cell behavior.

A lot of reports have indicate that interactions between substrate topography and cells were changed with cell types and substratum features including ridges, fibers, steps, pores, nodes, grooves, wells and adsorbed proteins. Table 2.2 summarizes the literature, providing a list of feature type, fabrication technique used, substratum material, feature size and spacing, cell type studied, and the cellular effect generated by the surface features.

Table 2.2 Interactions between substrate topography and cells [16]

Feature type Cell type studied Cellular effect Grooves/

PDMS cast of silicon original 2, 5, 10 μm width

0.5μm depth

Rat dermal fibroblasts

Microfilaments and vinculin aggregates oriented along 2 μm grooves after 1, 3, 5, and 7 days, but was less oriented on 5 and 10 μm grooves; vinculin located primarily on surface ridges; bovine and

endogeneous fibronectin and

vitronectin were oriented along

V-shaped grooves 35-165 μm width 30, 60, and 120 lm depth V-shaped pits 35-270 μm width and 30, 60, 120 μm depth

Rat parietal bone implant model

Mineralization occurred often on grooved or pitted surfaces, but rarely on smooth control surfaces; frequency of formation of bonelike foci

increased decreased as groove depth increased; frequency of mineralization increased as depth of pit increased;

bonelike foci oriented along long axis of grooves [41]

Grooves and chemical pattern / Ti, Au-coated polyurethane

treated with fibronectin, alkane thiols

V-shaped grooves 25, 50 μm width depth not listed

Bovine capillary endothelial cells

Cells adhered to regions coated with fibronectin, which adsorbed to regions silanized with methyl but not

tri(ethylene glycol)-terminated silanes;

cells attached to either grooves or ridges, depending on which possessed the methyl-terminated silane and fibronectin coatings [42]

Ridges/

Polystyrene cast of silicon original

observed for ridges or plateaus 0.5 μm high; ridges higher than 1.0 μm or smaller than 0.25 μm were not effective signals; ridge spacing of 0.5-6.7 μm caused high degree of orientation of the fungus [43]

Waves/

PDMS gels of varying softness Softer gels had smaller waves while hard gel had larger waves

Human dermal fibroblasts and keratinocytes

Fibroblasts proliferated equally on all substrates; keratinocytes spread more and secreted more ECM on soft gels than on hard gel [22]

Wells and nodes/

PDMS cast of silicon original Square nodes or wells 2, 5, 10 μm diameter

ATCC human abdomen fibroblasts

Cells on 2 and 5 μm nodes showed increased rate of proliferation and increased cell density compared to cells on 2 and 5 μm wells; 10 μm nodes and wells did not differ

statistically from smooth surfaces [44]

Pillars and pores/

PMMA, PET, polystyrene Circular pillars and pores 1, 5, 10, 50 μm diameter

Human osteoblasts and amniotic epithelial cells

Cells engulfed pillars or stretched between adjacent 1 and 5 μm pillars;

cells attached to edges of pores, especially on 10 μm pores; texture caused increase in cell adhesion on all materials but PMMA; greatest

increase in adhesion was on 50 μm PET pillars [45]

Pores/

Uncoated and silicon coated filters

0.2–10 μm diameter depth not listed

In vivo canine model

Non-adherent, contracting capsules around implants with pores smaller than 0.5 μm; implants with 1.4-1.9 μm pores showed adherent capsules but no inflammatory cells; pores bigger than 3.3 μm were infiltrated with inflammatory tissue; pores 1-2 μm allowed for fibroblast attachment [23]

Spheres/

Poly(NIPAM) particles on polystyrene surface

0.86-0.63 μm diameter when temperature raised from 25 to 37

℃

Neutrophil-like induced HL-60 cells

Cells loosely adhered but did not spread on spherecoated surface and could roll easily; excess active oxygen released when temperature was

increased on spherecoated surface, but not on poly(NIPAM) grafted

surface[24]

General roughness/

Ti, Ti/Al/V alloy, TiTa alloy 0.04, 0.36, and 1.36 μm peak-to-valley heights

Human gingival fibroblasts

Cells aligned to grinding marks: 10%

of cells oriented on surface with 0.04 μm roughness, 60% on 0.36 μm roughness, and 72% on 1.36 μm roughness [27]

Protein tracks/

Glass coated with fibronectin

BHK cells, rat tendon fibroblasts, rat dorsal root ganglia cells,

Fibers increased spreading and alignment in direction of fiber; actin aligned in fibroblasts; alignment of

0.2–5 μm width P388D1 macrophages focal contacts in fibroblasts and macrophages; increased

polymerization of F-actin; fibers increased speed and persistence of cell movement and rate of neurite

outgrowth; macrophages had actin-rich microspikes and became polarized and migratory [46]

Microtextured surface/

Polyurethane positive cast of PMMA negative

Micron and nanometer scale topography

Bovine aortic endothelial cells

Cells grown on replicas of ECM spread faster and spread areas at confluence which appeared more like cells in their native arteries than cells grown on untextured control surfaces [30]

Grooves are the most common feature type employed in studies for the influence of surface structure on cells. In general, investigations of grooved surfaces revealed that most cells aligned (Fig. 2.1) accompany with organization of actin and cytoskeleton elements paralleling to the grooves.

Figure 2.1 SEM images of cells cultured with 4 μm pitch pattern. (A) Cell aligned along the groove direction. (B) At the cell edges, lamellipodia were perpendicular to the patterns and able to adhere to the floor of the grooves [47].

Oakley and Brunette found that microtubules were the first element to align to grooves, followed by actin assembly, 20 minutes after cell plating [48]. In order to maximize their contact area, oblong focal adhesions (1-10 μm long) [49] orientated along the direction of the ridges, leading to an alignment of cytoskeleton elements and of the cell body as a whole. Many studies found that the depth of grooves was more important than their width in determining cell orientation [50]. Orientation often increased with increasing depth, but decreased with increasing groove width. In other words, as ridge width or groove width increased, the orientation phenomena of cells on grooves diminished.

In the literature, there are some studies investigating the behavior of cells on other synthetic features, including wells and pores, nodes, and spheres. Green and coworkers found that, compared to 10 μm nodes and smooth surfaces, nodes of 2 and

5 μm resulted in increased cell proliferation [44]. Fujimoto et al. investigated the behavior of cells on spheres and observed that cells responded to a change in sphere size which produced by an increasing in temperature [24]. The cells released excess active oxygen when sphere diameters shrunk as a result of the temperature being increased from 25 to 37℃. Chen et al. thought cell spreading also varied while changing the spacing between multiple islands by maintaining the total cell-matrix contact area constant [51]. No matter what the type of integrin antibody or matrix protein is used to mediate adhesion, cell shape was found to determine whether individual cells grow or die (Fig. 2.2).

Figure 2.2 Spreading had an effect on cell growth and apoptosis. (A) Schematic diagram show the initial patterns containing different size of square. (B) Apoptotic and DNA synthesis index after 24 hours culturing [51].

The fact that basement membranes are incorporated into cell adhesion and extension processes which composed of unique and intricate topographies. It can be seen that coupled with topographical features have been shown to influence cell behavior, leads to the hypothesis that the topography of the basement membrane

plays an important role in regulating cellular behavior distinct from the chemistry of the basement membrane.

2.3 Substrates With Micro- and Nanofabricated Surfaces Affect