Multiphoton excited (MPE) photochemistry is used to fabricate model tissue engineering scaffolds directly from types I, II, and IV collagen. A modified benzophenone dimer (BPD) provides the photoactivation and becomes incorporated into the resulting collagen matrixes. Unlike xanthene photochemistries, the benzophenone dimer can be used in acidic environments, where most forms of collagen have the greatest solubility. The minimum feature sizes are investigated by using two- and three-photon excitation, where the latter provides for superior "resolution" and suggests that collagen structures can be fabricated on the size scales of focal contacts. The resulting structures display excellent retention of bioactivity as evidenced by highly specific cell adhesion as well as immunofluorescence labeling. Structural and chemical aspects of the collagen matrixes are probed through measuring the enzymatic degradation through specific and nonspecific proteases, as the resulting relative rates are consistent with the activity of these enzymes. The degradation rates can also be controlled through varying the cross-link density in the matrixes, which is achieved through tuning the exposure dose during the fabrication process. The degradation rates are also found to be consistent with swelling/shrinking measurements and thus the average mesh size of the matrixes. In all cases the enzymatic degradations are well-fit single exponentials, suggesting that the matrixes can be fabricated with a priori knowledge of their structural properties. These results coupled with the resulting bioactivity suggest that the multiphoton fabrication process may be a powerful tool for the creation of cell-sized tissue engineering scaffolds.

We use multiphoton excited (MPE) photochemistry to fabricate patterned extracellular matrices (ECM) and to investigate the morphology of human dermal fibroblasts adhered to the resulting photocrosslinked linear structures of fibronectin (FN), fibrinogen (FG), and bovine serum albumin (BSA). These proteins were chosen to systematically investigate the roles of topography and ECM biochemistry on cell spreading, as fibroblasts bind directly to both FN and FG at RGD sites through known integrins, whereas BSA provides no comparable ECM cues for cell binding. MPE crosslinked patterns are created from parallel linear structures 600 nm in width, 200 microm in length, and spaced by either 10 or 40 microm. Immunofluorescence staining of FN and FG was used to assay the functionality of crosslinked proteins. The metrics of orientation, elongation, and cell perimeter were used to quantitate the resulting cellular behavior on the crosslinked protein patterns. These parameters all reflect statistical differences for cells on BSA, relative to the similar statistical behavior on fibronectin and fibrinogen. Cells on the BSA patterns are constrained by physical guidance and orientation between linear structures. In contrast, cells adhered on both FN and FG had a greater propensity to spread across adjacent structures, indicating the importance of cell matrix interactions. Focal adhesion staining of cells adhered to the protein structures revealed similar trends. These findings are consistent with our hypothesis that these crosslinked matrix protein structures are expected to direct cell adhesion and spreading and that the topography and ECM cues lead to different forms of guidance.