Session W39: Focus Session: Materials and Functional Structures for Biological Interfaces: Cells

Controlling Cell Function with Geometry

This presentation will describe the use of patterned substrates to control cell shape with examples that illustrate the ways in which cell shape can regulate cell function. Most cells are adherent and must attach to and spread on a surface in order to survive, proliferate and function. In tissue, this surface is the extracellular matrix (ECM), an insoluble scaffold formed by the assembly of several large proteins---including fibronectin, the laminins and collagens and others---but in the laboratory, the surface is prepared by adsorbing protein to glass slides. To pattern cells, gold-coated slides are patterned with microcontact printing to create geometric features that promote cell attachment and that are surrounded by inert regions. Cells attach to these substrates and spread to adopt the shape defined by the underlying pattern and remain stable in culture for several days. Examples will be described that used a series of shapes to reveal the relationship between the shape of the cell and the structure of its cytoskeleton. These geometric cues were used to control cell polarity and the tension, or contractility, present in the cytoskeleton. These rules were further used to control the shapes of mesenchymal stem cells and in turn to control the differentiation of these cells into specialized cell types. For example, stem cells that were patterned into a ``star'' shape preferentially differentiated into bone cells whereas those that were patterned into a ``flower'' shape preferred a fat cell fate. These influences of shape on differentiation depend on the mechanical properties of the cytoskeleton. These examples, and others, reveal that shape is an important cue that informs cell function and that can be combined with the more common soluble cues to direct and study cell function.   

Role of fiber functionality and angle on cell migration.

In order to determine the role of surface interactions on cell migration we compared the cell velocity on electrospun PMMA fibers which were either etched with UV/ozone plasma, had pre-adsorbed Fibronectin or both. It shows that dermal fibroblasts (CF-29, ATCC) did not adhere to the fibers without treatment, and the migration of cells was fastest on with both etching and pre-coat FN fibers. Vinculin was used to stain for the focal adhesion points and the largest number per cell were found on the FN pre-incubated samples, and nearly none on the plasma etched surface, despite good proliferation and migration. The results indicate that the migration velocity need not directly correlate to the cell adhesion. Using FN coated fibers we also studied the effect of angle on crossed fibers. We found that there was a clear preference by the cells for crossing a matrix where the fibers were oriented at 30 degrees. At this angle the migration velocity was slowest. Movies of the migrating cells indicate that the residence time of the cells at junctions with this angle is the longest cut to the interactions of the motion between fibers. The largest speed was observed for fibers placed at 90 degree.   

Anomalous cell migration properties on electrospun fibers

We have studied the influence of substrate morpholiogy on the en-mass cell migration from an agarose droplet. On flat surfaces, the cell velocity decreases asymptotically towards the single cell value as the radial distance increases, and remains constant thereafter. On fibers, the velocity remains constant at the single cell limit for the first 24 hours and then begins to increase continuously for the next four days. On flat surfaces we have shown that migration was triggered by nuclear deformation [Pan Z. et al, 2009], whereas on fibers the nucleus is constantly deformed as the cell assumers the shape of the fiber and hence does not seem to play as major a role. Vinculin and paxillin immunofluorescent staining were performed to determine the role of traction forces. We found that whereas polarization remains constant on flat surfaces with time, it increases on the fiber surfaces after the first 24 hours, and may explain the increased migration speed.   

Control cell adhesion with dynamic bilayer films

Interfacially-directed assembly of amphiphilic block copolymers was employed to create ultrathin films having the potential to correlate the dynamics of ECM cues with cell adhesion and cytoskeletally-generated forces. The mobility of the polymeric bilayer films were tuned by the incorporation of hydrophobic homopolymer chains, which are thought to reduce interlayer friction. Labeling of the block copolymer chains with an adhesive peptide ligand (RGD) provided a specific means to study integrin-mediated cellular processes and the corresponding mechanotransduction. By seeding anchorage-dependent cells on ``dynamic'' (laterally mobile) and ``static'' films that display the same amount of RGD, we have found that cells recognize the difference in RGD diffusivity and develop distinct responses over time. We intend to examine changes in cell response by controlling the extent of cytoskeletally-generated forces and the assembly dynamics of focal adhesion complexes. Such films provide a unique platform to unveil the biomechanical signals related with ECM dynamics, and may ultimately facilitate a deeper understanding of cellular processes.   

Diblock Copolymer Foams with Adhesive Nano-domains Promote Stem Cell Differentiation

Adhesions play an important role in cell behavior, including differentiation. Substrates are typically modified with homogeneous protein coatings; extracellular matrices \textit{in vivo} provide heterogeneous adhesive sites. To mimic adhesive heterogeneity, internal phase emulsion foams were polymerized with polystyrene-polyacrylic acid (PAA) and polystyrene-polyethylene oxide (PEO) to determine if interface de-mixing would form~patch-like surfaces. PEO/PAA mole ratios were~confirmed by XPS and water contact angle while spatial distribution was measured~by chemical force spectroscopy. This method confirmed the presence of patch-like PAA domains. Protein differentially adsorbs on PEO and PAA, so adsorption on foam mixtures was copolymer ratio dependent. Bone marrow-derived~mesenchymal stem cell (BMSC) adhesion was ratio dependent, but the highest density and vinculin expression was observed for 75PEO/25PAA. BMSCs appeared to change lineage expression the most on this composition, suggesting that this foam, which exhibits small adhesive PAA domains, may be more biomemetic than uniformally adhesive scaffolds, e.g. 0PEO/100PAA.   

Dental Pulp Stem Cell Differentiation on Poly-4-vinyl-pyridine surfaces

In the regeneration of a natural tissue, the mechanics and the chemical properties of the artificial substrate play a critical role. In this study, the influence of poly-4-vinyl-pyridine scaffold morphology on dental pulp stem cell differentiation was analyzed. Cells were plated on spun cast films and electrospun fibers with diameters ranging from nano to micrometers. Confocal microscopy showed the presence of various cell morphologies: on microfibers cells conform precisely to the main axis of elongation, while on nanometric scaffolds they result spread and in contact with several fibers. Even if the surface chemistry was identical, a great variation in the curvature was present. From day 9 of incubation, spontaneous biomineralization in the absence of induction agents occurred only on the fibrous structures. The SEM revealed template deposits directly on the microfibers, while on the nanofibers large spherical islands were also present. EDAX determined hydroxyl apatite nature of the deposits. RT-PCR indicated upregulation of osteogenic markers, confirming differentiation. SEM also revealed the presence of ECM fibers covering the polymer structure, which may enhance the expression of focal adhesion sites on the cell membrane.   

Spontaneous Differentiation of Dental Pulp stem cells on Dental polymers

Dental pulp stem cells were plated on two dentally relevant materials i.e. PMMA commonly used for denture and Titanium used for implants. In both cases, we probed for the role of surface interaction and substrate morphology. Different films of PMMA were spun cast directly onto Si wafers; PMMA fibers of different diameters were electro spun onto some of these substrates. Titanium metal was evaporated onto Si surfaces using an electron beam evaporator. In addition, on some surfaces, P4VP nanofibers were spun cast. DPSC were grown in alpha-MEM supplemented with 10\% fetal bovine serum, 0.2mM L-ascorbic acid 2-phosphate, 2mm glutamine and 10mM beta-glycerol phosphate either with or without 10nM dexamethasone. After 21 days samples were examined using confocal microscopy of cells and by scanning electron microscopy (SEM) and Energy dispersive X-ray Analysis (EDAX). In the case of Titanium biomineralization was observed independent of dexamethasone, where the deposits were templated along the fibers. Minimal biomineralization was observed on flat Titanium and PMMA samples. Markers of osteogenesis and specific signaling pathways are being evaluated by RT-PCR, which are up regulated on each surface, to understand the fundamental manner in which surfaces interact with cell differentiation.   

Interaction of Substrate Mechanics with Dental Pulp Stem Cells (DPSCs) differentiation to generate a scaffold for Bone regeneration

This work investigates the interaction of the substrate mechanics with the differentiation in the absence of chemical induction and only resulting from the stimuli of the substrate mechanics and chemistry. We chose enzymatically cross-linked gelatin hydrogels substrates of different stiffness varying from 8KPa to 100Pa. DPSCs were cultured and differentiated on the substrates for 7, 14 and 21 days with and without dexamethasone induction media. SEM and EDX analysis after 21 days indicate that cells produced a sheet of biomineralized deposits, several tenths of mm thick on the hard substrate irrespective of chemical induction. Modulli of the cells was independent of the induction and stiffness of the hydrogels. RT-PCR assays indicated that cells expressed more osteocalcin when cultured in non-induction media and harder substrate. The shape of the deposits was more uniform and in close packing on the harder substrate with a higher Ca:P ratio. On soft substrate the deposits were more flat with less Ca:P ratio. Further experiments indicated that conformational change due to the crosslinking of gelatin could be the reason for biomineralization.   

Neuron Growth on Carbon Nanotube Thread Bio-Scaffolds for Repair of Central Nervous System Damage

Approximately 11,000 new spinal cord injuries occur each year. Repairing such central nervous system (CNS) damage has proven to be very difficult. We report on\textit{ in vitro} experiments using carbon nanotube (CNT) threads as a bio-scaffold for promoting CNS repair via directed neuron regrowth along the CNT material. Mouse brain neurospheres, containing neuronal stem cells, neurons and support glia, were observed to attach to and grow along laminin-coated CNT threads \textit{in vitro}. However, due to their limited mobility, only neurospheres close to the threads attach. To increase cellular attachment to the threads, we exploit the fact that these cells can exhibit enhanced, directed migration along an externally applied electric field. Recent\textit{ in vitro} cell growth was carried out in chambers containing several parallel CNT threads with electrical connections extending out of the incubator so that a voltage applied across adjacent threads established an appropriate electric field. Electrochemical Impedance Spectroscopy, Cyclic Voltammetry and dc and ac IV measurements were used to monitor cell growth and attachment as a function of applied electric field and time. Cell migration and attachment were also investigated using time lapse photography in a separate growth chamber mounted on the stage of an optical microscope.   

Stiffness nanotomography of human epithelial cancer cells

The mechanical stiffness of individual cells is important in both cancer initiation and metastasis. We present atomic force microscopy (AFM) based nanoindentation experiments on various human mammary and esophagus cell lines covering the spectrum from normal immortalized cells to highly metastatic ones. The combination of an AFM with a confocal fluorescence lifetime imaging microscope (FLIM) in conjunction with the ability to move the sample and objective independently allow for precise alignment of AFM probe and laser focus with an accuracy down to a few nanometers. This enables us to correlate the mechanical properties with the point of indentation in the FLIM image. We are using force-volume measurements as well as force indentation curves on distinct points on the cells to compare the elastic moduli of the nuclei, nucleoli, and the cytoplasm, and how they vary within and between individual cells and cell lines. Further, a detailed analysis of the force-indentation curves allows study of the cells' mechanical properties at different indentation depths and to generate 3D elasticity maps.   

Snakeskin tribology: How snakes generate large frictional anisotropy

The limbless locomotion of snakes relies fundamentally upon friction. Snakes can adjust their spatial and temporal frictional properties in order to get friction anisotropies of around 2. Ventral scales play a major role in friction adjustment. In this combined experimental and theoretical study we measure the mechanical and frictional properties of snakeskin. We report the effect of substrate roughness and compliance on snake frictional anisotropy. We numerically model a snake scale interacting with an elastic rough substrate using contact mechanics. The scale is modeled as an isotropic rigid material attached to the body using a torsional spring. We find that the combined effect of the scales geometry and angle of attack leads to the scale high frictional anisotropy. Our results suggest that fabricating engineering surfaces such as artificial snakeskin with optimized geometry and orientation of scales will improve the efficiency of snake-like robots.