The role of ERK signaling in protein hydrogel remodeling by vascular smooth muscle cells

Abstract

Collagen type I and fibrin hydrogels have been used for cell-based therapies and tissue engineering. These matrices can be broken down and remodeled by cells, but the effects that these proteins have on cell function are not completely understood. We examined activation of the extracellular signal-regulated kinase (ERK) signaling pathway by vascular smooth muscle cells (VSMC) in response to 2D and 3D matrices of type I collagen, fibrin, or a 1:1 composite mixture of these proteins. After 3 days of culture, ERK phosphorylation, osteopontin secretion, and MMP-2 activation were all markedly increased in 3D matrices, compared with 2D substrates. A strong positive correlation existed between these protein markers of the synthetic phenotype and phosphorylated ERK levels, and this relationship persisted across matrix geometries and compositions. Cell proliferation in 3D matrices was inversely correlated to ERK activation, while on 2D substrates a modest positive correlation was observed. Pharmacologic inhibition of ERK signaling confirmed that this pathway was involved in the observed phenotype shifts. This study suggests that contextual activation of the ERK pathway results in different effects on cell phenotype, depending on the geometry and composition of the ECM. These findings add to our understanding of cell function and remodeling in protein-based hydrogel biomaterials.

Figure 1

Macro- and micro- scale images of 3D ECM constructs. Panels A-C show images of collagen (A), collagen-fibrin (B) and fibrin (C) constructs floating in culture medium after 3 days in culture. Panels D-G show SEM images of collagen (D), collagen-fibrin (E) and fibrin (F) matrices. Panel G shows an SEM image of VSMC on the surface of a collagen-fibrin scaffold. Collagen matrices were characterized by thin filaments, whereas fibrin matrices had a fine mesh-like morphology. Collagen-fibrin gels exhibited both morphologies forming an interpenetrating polymer network.

Figure 2

Effects of matrix composition and geometry on osteopontin (A), MMP-2 (B), proliferation (C), and ERK phosphorylation (D). The matrices used were tissue culture plastic (TCP), collagen (C), collagen-fibrin (CF), and fibrin (F). The three matrix proteins were used in 2D (white bars) and 3D (gray bars) cultures. Osteopontin and pERK were measured using ELISA, MMP-2 activity was measured using zymography, cell number was assessed using a Hoechst-dye based DNA quantification assay (1-fold indicates there was no change in population). Osteopontin, MMP-2, and pERK were higher for 3D samples relative to 2D. Conversely, proliferation was markedly lower in 3D than in 2D. Panel E shows a representative zymogram, loaded based on cell number, for the aggregate data shown in Panel B. Data in bar charts are expressed as mean ± SE for four experiments. *P<0.1 compared to 2D ECM control (student t-test).

Figure 3

Effects of matrix geometry over time on osteopontin secretion (A), MMP-2 activation (B), cell proliferation (C), and ERK phosphorylation (D) on TCP, 2D collagen, and in 3D collagen gels. Samples were taken after 1 (black), 3 (gray), and 7 (white) days in culture. Both osteopontin and MMP-2 decreased with time in monolayer cultures. Osteopontin in 3D cultures increased between days 1 and 3, and did not change at day 7. MMP-2 decreased sharply after day 3. Proliferation increased in monolayers, but decreased in 3D. pERK did not change in TCP, decreased with time in 2D ECM, and peaked on day 3 in 3D.

Figure 4

Correlations between phenotypic markers and pERK with varying geometry. Osteopontin secretion (A) and MMP-2 activation (B) demonstrated a strong positive correlation with pERK, whereas cell number and pERK were negatively correlated when fit with a logarithmic curve (C).

Figure 5

Correlations between phenotypic markers and pERK with varying 3D ECM composition. Osteopontin secretion (A), MMP-2 activation (B), and cell number (C) demonstrated a positive correlation with pERK.

Figure 6

Effect of ERK pathway inhibition using the ERK pathway inhibitor PD98059 on osteopontin, MMP-2, and proliferation on TCP, collagen (C), collagen-fibrin (CF), and fibrin (F) both on 2D (white) and in 3D (gray) matrices. Effects of inhibition are shown in % change, where 0 indicates that there was no change relative to control in the measured parameter (i.e. - 100 indicates a parameter was completely suppressed relative to control, and 100 indicates a parameter is twice the control value). Osteopontin secretion (A) and MMP-2 activation (B) decreased more in 3D ECM relative to 2D ECM, whereas cell number (C) decreased in monolayer cultures, but increased in 3D cultures. Upon inhibition of the ERK pathway, there was no detectable amount of pERK. Data are expressed as mean ± SE for four experiments. *P<0.1 compared to 2D ECM control (student t-test).