Stiffening hydrogels for investigating the dynamics of hepatic stellate cell mechanotransduction during myofibroblast activation

Abstract

Tissue fibrosis contributes to nearly half of all deaths in the developed world and is characterized by progressive matrix stiffening. Despite this, nearly all in vitro disease models are mechanically static. Here, we used visible light-mediated stiffening hydrogels to investigate cell mechanotransduction in a disease-relevant system. Primary hepatic stellate cell-seeded hydrogels stiffened in situ at later time points (following a recovery phase post-isolation) displayed accelerated signaling kinetics of both early (Yes-associated protein/Transcriptional coactivator with PDZ-binding motif, YAP/TAZ) and late (alpha-smooth muscle actin, α-SMA) markers of myofibroblast differentiation, resulting in a time course similar to observed in vivo activation dynamics. We further validated this system by showing that α-SMA inhibition following substrate stiffening resulted in attenuated stellate cell activation, with reduced YAP/TAZ nuclear shuttling and traction force generation. Together, these data suggest that stiffening hydrogels may be more faithful models for studying myofibroblast activation than static substrates and could inform the development of disease therapeutics.

Figure 1. Visible-light mediated stiffening of MeHA gels.

(a) Hyaluronic acid was modified with methacrylates to enable both addition and radical crosslinking. (b) RGD-containing peptide was coupled to MeHA followed by an initial Michael-type addition reaction to form a soft gel. Radical polymerization in the presence of visible blue light and the photoinitiator LAP enabled substrate stiffening. (c) Rheology profile showing stiffening of MeHA in response to blue light (3 mW/cm², 15 min, shaded) in the presence of LAP (6.6 mM). Closed circles represent storage modulus (G’), open circles represent loss modulus (G”). (d) AFM measurements of both soft and stiff (blue light: 10 mW/cm², 5 min, LAP: 6.6 mM) hydrogel elastic moduli (n = 3 gels per group, error bars represent s.e.m.). ***P < 0.001.

Figure 2. Hepatic stellate cells spread and assume myofibroblast morphology in response to in situ stiffening.

(a) Schematic of in situ stiffening process. (b) Phase contrast images of stellate cells on soft and stiff static substrates, as well as a dynamically stiffened soft-to-stiff substrate (in situ stiffening performed on day 4). Scale bars: 50 μm. (c) Representative stellate cell spread area quantification over 24 h time lapse following in situ stiffening on day 4 (n = 3 cells per group, each shape represents a single cell whose spread area was monitored every 15 min). Spread areas are relatively constant in soft and stiff static groups, but increase steadily in soft-to-stiff group, especially during the first 12 h. (d) To ensure that stellate cells were spreading in response to stiffness and not free radical generation, soft MeHA gels were fabricated where the remaining methacrylates were capped with a thiol, so that in the presence of light and initiator free radicals would be generated but no secondary crosslinking (stiffening) would occur. Representative phase contrast images showed no differences in cell spread areas between groups. Scale bars: 50 μm. (e) Quantification of cell area confirmed no differences in spread area between groups (n > 22 cells per group per time point, error bars represent s.e.m.).

Figure 3. Delayed stiffening promotes more rapid YAP/TAZ nuclear translocation and α-SMA stress fiber assembly.

(a) Schematic of experimental design. (b) Quantification of stellate cell spread area (n > 12 cells per group), (c) YAP/TAZ nuclear to cytoplasmic intensity ratio (n > 29 cells per group), (d) cell fraction displaying organized α-SMA stress fibers (n > 20 cells per group), (e) and cell fraction displaying organized F-actin stress fibers (n > 20 cells per group) 1, 6, 24, or 72 h following either early (Day 1) or later (Day 6) stiffening (error bars represent s.e.m.). (f) Representative phase contrast images of cells 1 or 72 h following stiffening at either day 1 or 6. (g) Representative immunostaining showing more rapid YAP/TAZ nuclear translocation and α-SMA stress fiber assembly at later stiffening time point (day 6). While YAP/TAZ nuclear localization was more evident 72 h following earlier stiffening (day 1), α-SMA staining was still diffuse and did not co-localize with F-actin stress fibers. In contrast, within 1 h following later stiffening (day 6) YAP/TAZ nuclear translocation was present and 72 h following stiffening the majority of stellate cells displayed α-SMA stress fibers. Representative immunostaining for all experimental groups is presented in the Supplementary Information. Dashed lines represent levels on soft gels at Day 1 (light blue) or Day 6 (dark blue). *P < 0.05, **P < 0.01, ***P < 0.001 where comparisons shown in figure are between day 1 and day 6 results for each post-stiffening time point (e.g., in panel (c) the day 6 YAP/TAZ nuclear intensity ratio 1 h post-stiffening is significantly greater than day 1 YAP/TAZ intensity 1 h post-stiffening, P < 0.05). Scale bars: 50 μm.

Figure 4. Blocking α-SMA polymerization results in reduced YAP/TAZ nuclear shuttling.

(a) Stellate cells were seeded on soft gels that were stiffened after 3 days and cultured in normal media (BP(−)) or media supplemented with α-SMA blocking peptide (25 μg/mL, BP(+)) for an additional 4 days. The blocking peptide almost completely abrogated α-SMA expression without impacting other filamentous actin, but reduced nuclear YAP/TAZ accumulation. Scale bars: 50 μm. (b) Immunostaining quantification indicates that α-SMA blocking does not affect the fraction of cells displaying F-actin stress fibers (n > 30 cells per group, error bars represent s.e.m.). (c) Tukey box plot of YAP/TAZ nuclear to cytoplasmic intensity ratio demonstrated significant reduction in YAP/TAZ nuclear localization in BP(+) group (n > 30 cells per group). ***P < 0.001.

Figure 5. Inhibiting α-SMA does not alter stellate cell spread area but significantly reduces traction force generation.

(a) Representative traction force vector maps and phase contrast images illustrating similar spreading but stronger tractions in BP(−) group. Scale bar: 50 μm. (b) Tukey box plot quantification of cell spread area (n > 12 cells per group). (c) Tukey box plot quantification of total force generated per cell (n > 12 cells per group). (d) Tukey box plot quantification of average traction stress (n > 12 cells per group). ***P < 0.001.