 |
Chemical Biology of Cell Adhesion and the Extracellular Matrix
Summary: Milan Mrksich's laboratory combines chemistry and materials science to create model surfaces that mimic complex biological surfaces. These approaches are used to study the adhesion of cells to the extracellular matrix and to construct biochip arrays for the analysis of biological function.
Most mammalian cells are adherent. They must attach to and spread on an underlying matrix in order to carry out normal metabolism, proliferation, and differentiation. The biological matrix that serves this role comprises a collection of insoluble proteins and glycosaminoglycans that are collectively referred to as the extracellular matrix (ECM). In addition to maintaining the organization and mechanical properties of tissue, the ECM presents many peptide and carbohydrate ligands that are recognized by cellular receptors. These receptor-ligand interactions are critical to maintaining cell function and enabling cells to respond appropriately to biochemical and mechanical cues in their environments.
The identification of ligand-receptor interactions that mediate the influence of ECM on cellular activities, and studies of the signaling pathways by which these influences operate, remain challenging. In laboratory experiments, the ECM is mimicked by coating plastic culture dishes with the resident proteins. These substrates have a number of limitations that make interpretation of experiments difficult. For example, cells can attach to surfaces by nonspecific interactions that do not depend on the ligands present in the ECM; ECM proteins are large macromolecules and contain a variety of ligands that can operate simultaneously; on adsorption, proteins can undergo substantial denaturation, with a loss of activity of the ligands; and adherent cells actively remodel the ECM, making it difficult to control the ligands that are presented to a cell during experiments.
Model Substrates
To address these challenges, we have pioneered the development of model substrates that mimic the ECM and permit complete control over the identity of ligands that are presented to a cell and over the densities and patterns of those ligands. At the same time, these substrates prevent unwanted protein adsorption and therefore remodeling of the matrix. Our approach is based on the self-assembly of alkanethiols onto a gold-coated glass slide to give a 2-nm-thick monolayer of chains that are forced into an extended conformation, much like one leaflet of a lipid bilayer. We use synthetic chemistry to introduce ligands at the ends of these chains. The resulting surfaces present those ligands in a context that permits interactions with cell-surface receptors. Early studies with monolayers that present the fibronectin peptide Arg-Gly-Asp show that fibroblast cells adhere, spread, and assemble cytoskeletal and focal adhesion structures. This result validates the notion that the model substrates represent simple, yet biologically relevant, models of the ECM and frames the opportunity to investigate the roles by which discrete ligands influence cell adhesion and related activities.
Applications of ECM Mimics
We have applied these model substrates to several questions in cell adhesion. One example was motivated by recent work by several groups that has identified a pentapeptide ligand in fibronectin, Pro-His-Ser-Arg-Asn, which serves as a synergy ligand: it is inactive when presented alone but promotes more efficient adhesion and spreading when presented in combination with the Arg-Gly-Asp peptide. A comparison of cell adhesion on model substrates that present each peptide alone or a mixture of both peptides gave the clear result that the "synergy" peptide could mediate adhesion on its own, but with poor spreading. Furthermore, inhibition experiments with soluble ligands revealed that the two peptides bind competitively to the integrin receptor, and therefore ruled out prior models that called for simultaneous binding of the two peptides to separate regions of the integrin receptor.
In another example, we found that the protein angiopoietin—which is primarily understood to serve as an angiogenic factor by activating the Tie2 receptor—may also function as an ECM molecule by promoting the adhesion of endothelial cells. To identify the peptide ligand within the Ang-1 primary sequence that is responsible for adhesion, we prepared a peptide array from overlapping sequences of the protein. We found that cells attached only to a single peptide in the array, and we went on to define the consensus sequence for this adhesion motif and to employ affinity purification to identify the cell-surface receptor. To investigate the influence of affinity as a factor that serves to activate the cell-surface receptors, we have used model substrates presenting peptide ligands that have either high or low affinity for integrin receptors. This work demonstrates that affinity alone can promote receptor activation and has provided a strategy by which receptor activation can be differentially controlled in the cell. To control the positions of subcellular activation, we use surfaces that are patterned into regions presenting either the high- or low-affinity ligand, and we use dynamic substrates that can be switched to reveal the high-affinity ligand, as described below. These examples illustrate the value of tailored substrates for mechanistic studies of cell adhesion.
Dynamic Substrates
We have explored an innovative approach to develop dynamic substrates that can switch the activities of immobilized ligands in situ. These substrates offer an unprecedented opportunity to understand the responses of adherent cells to changes in the ECM, including changes caused by binding of proteins to matrix, proteolytic action on matrix, and cellular forces that unfold matrix proteins. Our strategy has taken advantage of the presence of the gold film underlying the monolayer, and the use of applied electrical potentials to effect electrochemical conversions of the monolayer. We have designed a series of monolayers that respond to these applied potentials by revealing or releasing ligands at the interface, and therefore altering the ECM ligands with which an adherent cell interacts. We demonstrated the use of these dynamic substrates for studies of cell migration and for patterning multiple cell types into defined cocultures.
Biochip Arrays
The surface chemistries described above offer important benefits to the design and application of biochips. The use of inert surfaces and the development of a range of well-controlled methods for immobilizing biomolecules offer effective strategies for performing assays of soluble proteins with peptides, carbohydrates, and proteins on the surface. These benefits combine to give a high and uniform activity of the immobilized biomolecules, and permit quantitative assays directly on the chip. In an important advance, we have shown that mass spectrometry methods can be used to directly characterize the biomolecules on the surface. Post-translational modifications lead to changes in the mass of the immobilized molecules and therefore do not require the use of fluorescent or radioisotopic labels to observe the activity. This development addresses an important limitation in the use of microarrays. Although arrays can be used to discover entirely new enzyme activities, the need to use labels to observe these activities limits applications of arrays to observing known activities. The development of this method, which combines the biochips with mass spectrometric detection, now makes possible the identification of unanticipated activities. We have demonstrated the generality of this approach by performing assays of kinase, protease, methyltransferase, glycosyltransferase, and protein-binding activities, and we are applying this method to map protein-protein interactions within large sets of recombinant proteins.
Matrices from Protein Assembly
The development of general strategies to create well-defined three-dimensional environments around cells would permit model systems for studying cell-ECM interactions in a context that is more relevant to the in vivo setting. We are pursuing an approach that relies on the design and preparation of proteins that can be stimulated to self-assemble into ordered networks. These protein-building blocks can be engineered to present adhesion motifs and other functional domains from ECM proteins and therefore may provide a biomimetic, but controlled, scaffold for three-dimensional cell cultures. In one example, we created a hydrogel by cross-linking the calmodulin protein into a poly(ethylene glycol) network. The calmodulin protein undergoes a substantial conformational change in the presence of calcium ion, which can be used to effect a macroscopic change in the volume of the gel. We are applying these materials to the development of mechanically actuated matrices for cell culture.
This research is also supported by the National Institutes of Health, the National Science Foundation, and the Department of Defense.
Last updated October 01, 2007
|
|
|
|
