Cell Migration Gateway Logo

Biomaterials - Approaches

This group is designing, fabricating and characterizing materials that control both chemical and physical aspect of receptor-mediated interactions with the adhesion/migration environment. Input from other initiatives (Modeling, Signaling and Structure) contributes to the design, development and refinement of these materials, but the generic approach is to create surfaces that present well-defined ligands covalently linked to an underlying non-adhesive substrate.

Introduction

Experimental systems commonly used to investigate cell adhesion and motility rely on proteins adsorbed from solution onto the culture substrate. Prototypical adhesion molecules such as fibronectin are large, multimeric, glycoproteins, containing multiple sites for interaction with adhesion receptors and other molecules and, not surprisingly, they adopt a diverse array of unpredictable conformations when adsorbed onto surfaces (Lewandowska et al. 1992). Cell binding sites may be more accessible than in the native protein, due to surface denaturation and exposure of cryptic sites, or conversely they may be less accessible. Fibronectin, for example, is typically more active toward cell adhesion and towards binding of epitope-specific antibodies when adsorbed on relatively hydrophilic compared to hydrophobic surfaces(Groth & Altankov 1995; Garcia et al. 1999; Altankov et al. 2000). Further, surfaces that are able to adsorb a matrix protein of interest are also likely to adsorb proteins in serum or proteins secreted by cells after the culture is initiated, adding additional adhesion sites that are not controlled. This phenomenon can sometimes be mitigated by passivating the additional binding sites on the surface with albumin after the protein is initially coated, but the overall surface remains relatively undefined. Thus, while adsorbed proteins are useful to illustrate general features of biophysical and biochemical processes involved in migration, a more controlled presentation of molecular ligands is desired to parse the underlying molecular contributions (Griffith 2002). To better understand the underlying molecular contributions to the process of migration a more controlled presentation of molecular ligands is needed. The Bohn and Griffith laboratories are generating synthetic substrates that present well-defined peptide adhesion ligands targeted toward specific surface receptors. These ligands are covalently linked to the substrate in defined densities, orientations and spatial arrangements, with the latter sometimes also varying with time.

Design Features

Selective Substrates: PEO brushes modified with adhesion ligands and other motifs
A generic approach to creating selective surfaces for controlling molecular recognition comprises two steps:

  1. create an "inert" surface which repels proteins and is not modified in the presence of cells and
  2. attach ligands to the substrate to engender cell adhesion or other molecular recognition events.

Of course, no substrate is completely inert! Decades of research in biomaterials have shown that dense interfacial polyethylene oxide (PEO) brushes are highly resistant to protein adsorption over time scales of several days to several weeks (Prime & Whitesides 1993; Irvine et al. 1996; Harris 1997; Irvine et al. 1998). PEO, which has the chemical structure -(O-CH2-CH2)n-OH, is an extremely versatile polymer. It is highly water soluble and flexible in aqueous solution. "Brushes" of PEO at interfaces can be created using short (<20-mer) chains attached to the surface at spacings closer than the radius of gyration (Rg) of the PEO chains. (Figure 1) A PEO chain with 10 monomer units is roughly 3 nm long when fully extended.

Such dense, short-chain PEO brushes are virtually inert toward protein adsorption over time scales of hours to days and do not support cell adhesion in the absence of covalent modification by adhesion ligand. Ligands that mediate migration can be covalently linked to some of the PEO chain ends, and thus presented to cells in a "tethered" format that enhances ligand mobility and accessibility. Ligands used in the biomaterials core include both minimal adhesion domains (fibronectin-derived RGD-containing peptides of various affinities as well as non-RGD peptides reported to be specific for integrins alpha4 beta1 and alpha9 beta1) as well as the whole protein fibronectin (Fn). Alternatively, the PEO chain ends can be modified with biotin/streptavidin to provide a generic substrate for tethering any molecule that has been modified with biotin. This approach is useful in tethering proteins, such as antibodies, to the surface.

Existing biophysical models of cell migration generate two key parameters that provide the starting points for design of ligand presentation (Lauffenburger & Horwitz, 1996). These are (i) the average adhesion ligand density L0 (# ligands per micrometer squared) and (ii) the affinity of the ligand-receptor bond, identified as a dissociation constant, KD. With these two parameters, the number and strength of individual bonds formed with substrate - and thus the overall cell adhesion strength can be systematically varied.

Dense PEO brush in aqueous medium
PEO brush with attached ligands
Figure 1. Dense, non-adhesive PEO brush (upper) and brush with covalent attached (tethered) ligands (lower) showing dimensions of PEO tethers.

Two complementary approaches are used to generate these ligand presenting PEO brushes PEO-polymethyl methacrylate (PMMA) comb copolymers (Benerjee et al, 2000; Mayes & Kumar, 1997; Irvine et al., 2001(a); Irvine et al., 2001(b); Koo et al., 2002) and PEO-terminated alkanethiol self-assembled monolayers (SAMs) (Prime & Whitesides, 1993; Singhvi et al., 1994; Mrksich et al, 1997; Pakalns et al., 1999; Zhang et al., 1999; ;Terrill et al., 2000 Plummer et al., 2003; Reyes & Garcia, 2003; Wang & Bohn, 2003; Wang et al., 2004).

PEO-PMMA comb polymers - Using this approach, the PMMA provides a hydrophobic backbone with the PEO oligomer side chains (Figure 2).When composed from a ratio of 1 PEO side chain per 5 PMMA backbone units it forms a stable water-insoluble film that can be made completely resistant to cell adhesion (Walton et al, 1997). The polymer a 2D disc-like conformation at the interface, with dimensions of 30-50 nm (Figure 3). This has important consequences for spatial organization of ligands at the surface (Irvine et al. 2001(a); Irvine et al. 2001(b); Koo et al. 2002)

 
Figure 2Comb Co-polymers have a hydrophobic backbone and many PEO side chains that can be modified with ligands; the distance between ligands (dt) and the tether length (Lt) can be systematically modified providing flexibility in determining biophysical effects of ligand presentation on cell migration properties (speed, persistence.)

Combs that are modified with ligand can be mixed with inert combs to create islands of clustered ligand in an inert background (Figure 3).

 
Figure 3 Comb Co-Polymers exist confined discs at the interface (left) and allow presentation of clusters of ligand (right)

 

PEO-terminated alkanethiol SAMs - These self-assembled monolayers comprise PEO-oligomer-terminated alkanethiols which offer complementary control over ligand spatial organization on length scales of 1-1000 microns, which are relevant for haptotactic gradients. An additional advantage of this approach is the ability to move and rearrange an initial surface spatial organization of ligands, hence superimposing temporal patterns over spatial ones. The discovery that the thin silver (Au) films, which are used to support the alkanethiol SAMs, can be used to manipulate the spatial organization of the SAM components dynamically in real time via tuning of surface electrochemical properties (Zhang et al, 1999) provides the additional dimension of "time" to the controllable parameters (Figure 4).

Figure 4. (Top) Schematic of the gradient formation concept. A potential applied in-plane to an electrochemical working electrode generates a spatially varying distribution of electrochemical potentials (given by equation at top). (Middle) Since electrochemical reactions occur at characteristic potentials, and potential varies in space reactions can be mapped onto electrodes differentially. In the example oxidative adsorption (of thiols) occurs at locations on the left, and reductive desorption occurs at positions on the right. In the middle the composition is intermediate between bare and fully covered. (Bottom) The bare spots created in the above step can then be filled in with a second thiol component to create a two-component gradient. In the cartoon the gradient components are chosen to be thiols terminated with hydrophilic (right) and hydrophobic (left) groups.
return to top

Physical receptor interactions

Nanoscale Clustering of Adhesion Ligands - Integrins within focal adhesions exist as aggregates, and accumulating data support the idea that both occupancy and clustering of integrins are required to elicit full cellular responses mediated by integrins. Several aspects of integrin aggregation may be influenced by the spatial arrangement of integrin ligands in the extracellular environment, including the size of aggregates, the rate aggregates form and dissolve, and the total number of bound integrins for a given average ligand density. These factors, in turn, would be expected to influence cell migration behavior. Therefore, it is important to be able to develop substrates where cognate ligand clustering can be spatial controlled on the scale of 10-1000 nm, with ligand density and cluster size being controlled independently. The comb copolymer approach offers great flexibility in controlling parameters such as ligands per cluster, spacing between ligands and tether length/compliance (Figure 5).

average density

 

low

 

med

 

high

 
Figure 5. Ligand density can be controlled in both a random and formatted manner.

For example, a precursor to the PEO-PMMA comb, star PEO, was used to cluster the very low-affinity integrin ligand YGRGD, which does not support cell migration when presented in a random configuration at surface densities as high as 30,000 ligands m-2. Strikingly, YGRGD ligand presented in nanoscale (~35 nm) clusters of 5 YGRGD/cluster supports a maximal migration speed (comparable to that on fibronectin) at an average ligand density of ~30,000 ligands m-2, and ligand presented in clusters of 9 supports maximal migration at a 10-fold lower average surface density, 3000 ligands m-2 (Maheshwari et al. 2000). Fibroblasts are able to form focal adhesions on clustered YGRGD but are relatively unable to form focal adhesions on unclustered YGRGD at the same density (Figure 6).
Figure 6. NR6 fibroblasts form focal adhesions on clustered RGD but not on random RGD at comparable average ligand densities.

Further, adhesion (measured by a distraction force) depends on the average peptide density (Figure 7). Clustered ligand (5.4 ligands/cluster) enables greater cell adhesion strength than near-random ligand (1.7 ligands/cluster) - the % cells adherent after centrifugation is not a monotonically decreasing function of adhesion strength, but shows a peak at intermediate detachment forces for clustered ligand (Koo et al. 2002).

Figure 7: Ligand clustering enhances cell adhesion and is required for adhesion strengthening. Cell adhesion was measured using centrifugal detachment.

Ultimately, however, the size of an individual comb molecule may limit the number of ligands that can be presented in a single cluster. Comb polymers, including RGD-modified combs, can be used as stabilizers in the synthesis of latexes, to generate particle diameters in the 200-1000 nm range. The latexes can be imbedded in an inert polymer film, creating tiny adhesive islands that can readily be visualized by including fluorescent monomer in the latex synthesis. Latexes can also be "painted" onto a surface and coalesced to create an adhesive substrate, providing a facile route to practical application (Banerjee et al, 2000)

In the context of the Cell Migration Consortium and other projects, we have tested a wide variety of cell types for adhesion to both unmodified and peptide-modified combs. Cell types that do not adhere to unmodified comb include several strains of fibroblasts (3T3, NR6, primary liver stellate cells, mouse embryo fibroblasts), hepatocytes, PC12, endothelial cells, primary mouse bone marrow (whole), pig mesenchymal stem cells, HOS osteosarcoma cells, and CHO cells. Cells present in whole human bone marrow, seeded in serum and allowed to attach for 48 hours, do exhibit some adhesion if seeded at high density (>250,000 cells cm-2), but no spreading. Cells that adhere to comb substrates that present linear RGD ligands include fibroblasts, hepatocytes, and mesenchymal stem cells; these cells are non-adhesive to RGE substrates. Cells that adhere to ligands reported in the literature as specific for alpha4 beta1 include mesenchymal stem cells; fibroblasts fail to adhere to these substrates.

Current project areas: integrin signaling as a function of clustering, isolation of focal adhesion components (with Burridge Lab), creation of substrates for clustering 4 1 ligands (with Keeley Lab).

Heterotypic receptor interactions

The physical interaction of different receptors (between different types of adhesion receptors or between growth factor receptors and adhesion receptors) is important in determining signal transduction (Sastry & Horwitz 1996). By specifically controlling the relative placement of ligands for these receptors, the limits for physical interaction can be determined and either allowed or inhibited. A particular focus is the interactions between epidermal growth factor (EGF) receptor and the integrin alpha5 beta1. As a first step in controlling physical interactions between these receptors, the comb polymer was tested for its ability to present activate EGF. A non cell-resistant comb with only 20% PEO content was used to tether EGF by the N-terminus, and then fibronectin was absorbed on the substrate to provide adhesion. The EGF molecule remains active to stimulate migration and proliferation when tethered to the substrate (Figure 8). A challenge in generating substrates with minimal adhesion ligands such as YGRGDSP and the growth factor EGF is that EGF is antiadhesive, and thus very high affinity adhesion ligands must be used. We are currently developing substrates with defined EGF and alpha5 beta1 ligands by using a higher affinity adhesion ligand that includes the fibronectin synergy site PHSRN.

Figure 8. Comparison of NR6 fibroblasts motility on 20% PEO comb substrates coated with fibronectin and either unstimulated (top), or stimulated with soluble (left) or tethered (right) EGF. Cells on tethered EGF substrates respond with both migration and proliferation.
return to top

Haptotactic gradients and spatiotemporal evolution of migration environment

This component of the Biomaterials Initiative seeks to design and fabricate surfaces with quantitatively characterized spatial gradients of adhesion/motility ligands. For example, thiols have been synthesized which are terminated with the protein adhesion-resistant moiety poly(ethylene oxide), PEO, and others have been prepared where the termination contains the tripeptide RGD (Arg-Gly-Asp), the active epitope for binding of a wide variety of extracellular matrix (ECM) proteins, e.g. fibronectin to the integrin superfamily of cell surface receptors. Two-component gradients fabricated from these thiols yield dramatic differences in cell morphology and motility. The local concentration of the active peptide has a major effect on the ability of 3T3 fibroblasts to adhere to a surface, and therefore cellular motility is directly modulated by the presence or absence of the RGD epitope.

While the "small" molecule peptides, such as the tetradecapeptide thiol used here are competent to direct the adhesion and motility of prototypical motile cells, there is a need to connect these studies to the actual components of the ECM. A strategy to prepare gradients of the ECM protein fibronectin (Fn) is shown below (Figure 9).
Figure 9. (Left) Reaction sequence used to produce Fn gradients from gradients in an amine-reactive thiol. (Right) Fluorescence micrograph of the gradient and the corresponding line scan and fit of the line scan to a sigmoidal spatial distribution. Adapted from Plummer et al, 2003.

Here a gradient is initially prepared in an amine-reactive thiol. Once this is obtained the gradient is derivatized with Fn by reacting with solvent-accessible Lys residues. In order to visualize the gradient, the Fn is exposed to a sandwich antibody assay in which the primary antibody recognizes Fn, and a secondary, fluorescently tagged antibody subsequently recognizes the primary antibody. The gradient is then visualized by fluorescence microscopy, the resulting fit of the line scan then yielding values for the center position and width of the gradient in both physical and potential space.

Other approaches in this area include using light energy to reversibly activate immobilized ligands in a monolayer substrate (Blonder et al, 1997a,b) or making use of the thermal-phase transition of poly(N-isopropylacrylaminde) films to regulate the adhesiveness of substrates (Okano et al, 1995). Protein engineering approaches are also being used to create well-defined elastomeric substrates containing adhesion ligands(Liu et al. 2004). Many of these effort have recently been reviewed (Mrksich 2002; Yeo & Mrksich 2003).

return to top

Pattern & Format Options

The ability to pattern ligands at submicron resolutions enables studies related to the mechanistic issues associated with clustering of focal adhesions, because the sizes of focal adhesions range from 50 nm to 500nm making it impossible to explore the relationships of focal adhesion size and distribution on cell adhesion in the absence of nanopatterning approaches. In the Griffith laboratory (Koo et al, 2002) this has taken the form of mixing polymers that form a non-adhesive film (unmodified PEO) with polymers modified with adhesion ligands such as RGD. Another approach is to use dip-pen nanolithography to directly pattern the formation of monolayers into nanometer-scale protein adsorptive islands, surrounded by strictly inert areas (Lee et al, 2002).

return to top

Text References

Altankov G, Groth T, Krasteva N, Albrecht W, Paul D. 1997. Morphological evidence for a different fibronectin receptor organization and function during fibroblast adhesion on hydrophilic and hydrophobic glass substrata. J Biomater Sci Polym Ed. 8(9):721-40. PubMed

Altankov G, Thom V, Groth T, Jankova K, Jonsson G, Ulbricht M. 2000. Modulating the biocompatibility of polymer surfaces with poly(ethylene glycol): effect of fibronectin. J Biomed Mater Res. 52(1):219-30. PubMed

Banerjee P, Irvine DJ, Mayes AM, Griffith LG. 2000. Polymer latexes for cell-resistant and cell-interactive surfaces. J Biomed Mater Res. 50(3):331-9. PubMed

Blonder R, Katz E, Willner I, Wray V, Bueckmann AF. 1997a. J Am Chem Soc. 119:11747.

Blonder R, Levi S, Tao G, Ben-Dov I, Willner I. 1997b. J Am Chem Soc. 119:10467.

Chothia C, Jones EY. 1997. The molecular structure of cell adhesion molecules. Annu Rev Biochem. 66:823-62. PubMed

Garcia AJ, Vega MD, Boettiger D. 1999. Modulation of cell proliferation and differentiation through substrate-dependent changes in fibronectin conformation.Mol Biol Cell. 10(3):785-98. PubMed

Geiger B, Bershadsky A, Pankov R, Yamada KM. 2001. Transmembrane crosstalk between the extracellular matrix--cytoskeleton crosstalk. Nat Rev Mol Cell Biol. 2(11):793-805. PubMed

Griffith, LG. 2000. Polymeric biomaterials. Acta Mater. 48: 263-277.

Griffith LG. 2002. Emerging design principles in biomaterials and scaffolds for tissue engineering. Ann N Y Acad Sci. 961:83-95. PubMed

Groth T, Altankov G. 1995. Fibroblast spreading and proliferation on hydrophilic and hydrophobic surfaces is related to tyrosine phosphorylation in focal contacts. J Biomater Sci Polym Ed. 7(3):297-305. PubMed

Groth T, Altankov G, Kostadinova A, Krasteva N, Albrecht W, Paul D. 1999. Altered vitronectin receptor (alphav integrin) function in fibroblasts adhering on hydrophobic glass. J Biomed Mater Res. 44(3):341-51. PubMed

Harris, JM, Ed. 1997. Poly(Ethylene Glycol): Chemistry and Biological Applications. Acs Symposium Series, No 680. San Francisco, American Chemical Society Division of Polymer Chemistry.

Irvine, DJ, Hue KA, et al. 2002 Simulations of cell surface integrin binding to nanoscale clustered adhesion ligands. Biophys J. 82:120-32. PubMed

Irvine, DJ, Mayes AM, et al. 2001. Nanoscale clustering of RGD peptides at polymer surfaces using comb polymers: I. synthesis and characterization of comb thin films. Biomacromolecules. 2(1):85-94. PubMed

Irvine, DJ, Mayes AM, et al. 1996. Self-consistent field analysis of grafted star polymers. Macromolecules 29: 6037- 6043.

Irvine, DJ, Mayes AM, et al. 1998. Comparison of tethered star and linear poly (ethylene oxide) for control of biomaterials surface properties. J. Biomed. Mat. Res. 40:498-509. PubMed

Irvine DJ, Ruzette AV, Mayes AM, Griffith LG. 2001. Nanoscale clustering of RGD peptides at surfaces using comb polymers. 2. Surface segregation of comb polymers in polylactide. Biomacromolecules. 2(2):545-56. PubMed

Koo LY, Irvine DJ, Mayes AM, Lauffenburger DA, Griffith LG. 2002. Co-regulation of cell adhesion by nanoscale RGD organization and mechanical stimulus. J Cell Sci. 115(Pt 7):1423-33. PubMed

Lee KB, Park SJ, Mirkin CA, Smith JC, Mrksich M. 2002. Protein nanoarrays generated by dip-pen nanolithography. Science. 295(5560):1702-5. PubMed

Lauffenburger, DA and Horwitz AF. 1996. Cell migration: a physically integrated molecular process. Cell 84: 359-369. PubMed

Lewandowska, K., E. Pergament, et al. 1992. Cell-type-specific adhesion mechanisms mediated by fibronectin adsorbed to chemically derivatized substrata. J Biomed Mater Res 26(10):1343-63. PubMed

Liu JC, Heilshorn SC, et al. 2004. Comparative cell response to artificial extracellular matrix proteins containing the RGD and CS5 cell-binding domains. Biomacromolecules 5(2):497-504. PubMed

Maheshwari G, Brown G, Lauffenburger DA, Wells A, Griffith LG. 2000. Cell adhesion and motility depend on nanoscale RGD clustering. J Cell Sci. 113 ( Pt 10):1677-86. PubMed

Mayes AM and Kumar SK. 1997. Tailored Polymer Surfaces. MRS Bulletin 22:43.

Mrksich M. 2002. What can surface chemistry do for cell biology? Curr Opin Chem Biol. 6(6):794-7. PubMed

Mrksich M, Dike LE, Tien J, Ingber DE, Whitesides GM. 1997. Using microcontact printing to pattern the attachment of mammalian cells to self-assembled monolayers of alkanethiolates on transparent films of gold and silver. Exp Cell Res. 235(2):305-13. PubMed

Okano T, Yamada N, Okuhara M, Sakai H, Sakurai Y. 1995. Mechanism of cell detachment from temperature-modulated, hydrophilic-hydrophobic polymer surfaces. Biomaterials. 16(4):297-303. PubMed

Pakalns T, Haverstick KL, et al. 1999. Cellular recognition of synthetic peptide amphiphiles in self-assembled monolayer films. Biomaterials 20: 2265-2279. PubMed

Plummer ST, Wang Q, Bohn PW, Stockton R, Schwartz MA. 2003. Electrochemically-derived gradients of the extracellular matrix protein fibronectin on gold. Langmuir 19, 7528-36.

Prime KL, Whitesides GM. 1993. Self-assembled organic monolayers: model systems for studying adsorption of proteins at surfaces. Science. 252(5010):1164-7. PubMed

Reyes CD, Garcia AJ. 2003. Engineering integrin-specific surfaces with a triple-helical collagen-mimetic peptide. J Biomed Mater Res 65A(4): 511-23.

Saoncella S, Echtermeyer F, Denhez F, Nowlen JK, Mosher DF, Robinson SD, Hynes RO, Goetinck PF. 1999. Syndecan-4 signals cooperatively with integrins in a Rho-dependent manner in the assembly of focal adhesions and actin stress fibers. Proc Natl Acad Sci U S A. 96(6):2805-10. PubMed

Sastry SK, Horwitz AF. 1996. Adhesion-growth factor interactions during differentiation: an integrated biological response. Dev Biol. 180(2):455-67. PubMed

Singhvi R, Kumar A, et al. 1994. Engineering cell shape and function. Science 264:696-8. PubMed

Terrill RH, Balss KM, et al. 2000. Dynamic monolayer gradients: Active spatiotemporal control of alkanethiol coatings on thin gold films. J Am Chem Soc 122:988-9.

Walton DG, Soo PP, Mayes AM, Sofia Allgor SJ, Fujii JT, et al. 1997. Creation of stable hydrophilic surfaces with branched copolymers. Macromolecules, 30:6947-56.

Wang Q, Bohn PW. 2003. Active spatiotemporal control of Arg-Gly-Asp-containing tetradecapeptide organomercaptans on gold with in-plane electrochemical potential gradients. J Phys Chem B 107:12578-84.

Wang Q, Jakubowski JA, et al. 2004. Quantitative submonolayer spatial mapping of Arg-Gly-Asp-containing peptide organomercaptan gradients on gold with matrix-assisted laser desorption/ionization mass spectrometry. Anal Chem 76(1):1-8.

Yeo WS, Mrksich M. 2003. Self-assembled monolayers that transduce enzymatic activities to electrical signals. Angew Chem Int Ed Engl. 42(27):3121-24. PubMed

Zhang Y, Terrill RH, et al. 1999. Chemisorption and chemical reaction effects on the resistivity of ultrathin gold films at the liquid-solid interface. Analyt. Chem. 74: 119-25.

return to top