Matrix Adhesiveness Regulates Myofibroblast Differentiation from Vocal Fold Fibroblasts in a Bio-orthogonally Cross-linked Hydrogel

Abstract

Repeated mechanical and chemical insults cause an irreversible alteration of extracellular matrix (ECM) composition and properties, giving rise to vocal fold scarring that is refractory to treatment. Although it is well known that fibroblast activation to myofibroblast is the key to the development of the pathology, the lack of a physiologically relevant in vitro model of vocal folds impedes mechanistic investigations on how ECM cues promote myofibroblast differentiation. Herein, we describe a bio-orthogonally cross-linked hydrogel platform that recapitulates the alteration of matrix adhesiveness due to enhanced fibronectin deposition when vocal fold wound healing is initiated. The synthetic ECM (sECM) was established via the cycloaddition reaction of tetrazine (Tz) with slow (norbornene, Nb)- and fast (trans-cyclooctene, TCO)-reacting dienophiles. The relatively slow Tz-Nb ligation allowed the establishment of the covalent hydrogel network for 3D cell encapsulation, while the rapid and efficient Tz-TCO reaction enabled precise conjugation of the cell-adhesive RGDSP peptide in the hydrogel network. To mimic the dynamic changes of ECM composition during wound healing, RGDSP was conjugated to cell-laden hydrogel constructs via a diffusion-controlled bioorthognal ligation method 3 days post encapsulation. At a low RGDSP concentration (0.2 mM), fibroblasts residing in the hydrogel remained quiescent when maintained in transforming growth factor beta 1 (TGF-β1)-conditioned media. However, at a high concentration (2 mM), RGDSP potentiated TGF-β1-induced myofibroblast differentiation, as evidenced by the formation of an actin cytoskeleton network, including F-actin and alpha-smooth muscle actin. The RGDSP-driven fibroblast activation to myofibroblast was accompanied with an increase in the expression of wound healing-related genes, the secretion of profibrotic cytokines, and matrix contraction required for tissue remodeling. This work represents the first step toward the establishment of a 3D hydrogel-based cellular model for studying myofibroblast differentiation in a defined niche associated with vocal fold scarring.

Keywords: cell-adhesive peptide; myofibroblasts; tetrazine ligation; transforming growth factor beta1; vocal fold fibroblasts.

Figure 1.

Fabrication of MMP-degradable HA hydrogels with varying cell adhesiveness. (A) Mechanisms of tetrazine ligation with TCO and Nb. (B) Hydrogel building blocks include tetrazine modified HA (HA-Tz), TCO-tagged cell adhesive/non-adhesive peptides (RGD/RGE-TCO) and Nb functionalized MMP-degradable crosslinker (SMR-bisNb). (C) Hydrogels with the same stiffness but different adhesiveness were prepared.

Figure 2.

Characterization of hydrogel composition (A), swelling (B) and mechanical properties (C, D). (A) Peptide retention in hydrogels prepared using TCO-tagged RGD or RGE. Hydrogels were equilibrated in HEPES for 24 h before the solution was aspirated for HPLC analysis. Each hydrogel was conjugated with 2 mM of peptide. (B) Equilibrium swelling ratio for gels R_H and R_L. (C, D) Rheological analyses of gels R_H and R_L. Gels R_H and R_L exhibit similar storage and loss moduli. n = 3; ns: nonsignificant, p > 0.05, determined by student’s t-test. Error bars represent the standard error of the mean.

Figure 3.

Characterization of 3D cultures in terms of viability (A, C), morphology (B), and metabolic activities (D). (A) Confocal images of VFFs stained by calcein AM (green) and ethidium homodimer-1 (EthD, red) after 3, 7, and 14 days of culture. Scale bar, 100 μm. (B) Confocal images of VFFs stained by F-actin (red) and DAPI (blue) after 14 days of culture. Scale bar, 20 μm. (C) Quantitative cell viability analysis based on calcein/EthD-stained constructs using ImageJ. ns: nonsignificant, determined by two-way ANOVA. (D) Quantification of cell metabolism by PrestoBlue assay, as fold change relative to the R_LT- level on day 3. *: p < 0.05, ****: p < 0.0001, determined by two-way ANOVA. All error bars represent the standard error of the mean.

Figure 4.

Characterization of αSMA expression in 3D cultures by immunocytochemistry (A) and image analysis (B) and qPCR (C). (A) Representative confocal images of day 14 constructs showing αSMA, F-actin, and nuclei stained green, red, and blue, respectively. Scale bar, 50 μm. (B) Quantification of αSMA expression as the integrated density ratio between αSMA and F-actin signals in R_HT+ and R_LT+ cultures. The sum of fluorescent intensity of αSMA was normalized to that for F-actin. ** p < 0.01, determined by student’s t-test. n = 5. (C) Quantification of Mander’s coefficient (M2) in R_HT+ and R_LT+ cultures. M2 is defined as the fraction of αSMA overlapped with F-actin bundles, with 0 indicating no colocalization and 1 corresponding to complete colocalization. *** p < 0.001, determined by student’s t-test. n = 5. (D) Temporal ACTA2 expression in R_HT+ and R_LT+ cultures by qPCR. Expression was normalized to the R_LT+ level on day 3. *: p < 0.05, **: p < 0.01, ***: p < 0.001 and ****: p < 0.0001, determined by two-way ANOVA. n = 9 (three biological repeats, each with three technical repeats). All error bars represent standard error of the mean.

Figure 5.

Effects of culture conditions on VFF gene expression and cytokine secretion. (A-C) qPCR analysis of day 14 constructs for the expression of proteins involved in cell-ECM adhesion (A), pro-fibrotic growth factors (B) and ECM proteins (C) as a function of RGD concentrations. Fold change normalized to the R_LT+ level on day 14. (D-H) Characterization of day 14 constructs for cellular secretion of IL-6 (D), IL-8 (E), VEGFA (F), pro-collagen 1A1 (G), and TIMP2 (H) by ELISA. *: p < 0.05, **: p < 0.01, and ****: p < 0.0001, determined by student’s t-test or one-way ANOVA where appropriate. ns: nonsignificant. All error bars represent the standard error of the mean.

Figure 6.

Characterization of day 14 constructs for cellular contractility (A, B) and the expression of αSMA and MLC2 (C, D). (A) Digital images of R_HT+ and R_LT+ constructs after 14 days of culture. White and yellow dotted circles outline the border of the day 0 and day 14 constructs, respectively. Scale bar: 2 mm. (B) Quantitative analysis of the degree of contraction. ****: p < 0.0001 determined by student’s t-test. (C) Detection of αSMA and MLC2 by Western blotting. (D) Quantitative analysis of western blot bands by densitometry for αSMA and MLC2. **: p < 0.01, ***: p < 0.001, determined by student’s t-test. All error bars represent standard error of the mean.

Figure 7.

Temporal modification of matrix adhesiveness under 3D cell culture conditions via diffusion controlled bioorthogonal ligation. (A) Schematic illustration of diffusion-controlled reaction in a cell-laden hydrogel. Created with BioRender.com. (B) Four different hydrogel constructs were produced from the original R_L constructs via tetrazine ligation at the gel-liquid interface using RGD / RGE-TCO. (C) Digital picture of the hydrogels showing the gradual disappearance of the pink color of tetrazine functional groups over time.

Figure 8.

Characterization of mechanical properties of hydrogels prepared by diffusion-controlled reaction. (A, B) Storage (A) and loss (B) moduli for the modified hydrogels (3 samples per group). (C) Swelling ratio as a function of hydrogel composition (3 samples per group). ns indicates nonsignificant comparison determined by one-way ANOVA. All error bars represent the standard error of the mean.

Figure 9.

3D cell culture of cell-laden hydrogel constructs with temporal increase of matrix adhesiveness. (A) Experimental timeline. On day 0, VFF-laden constructs with 0.2 mM RGD were prepared. On day 1, cultures were supplemented with 5 ng/mL TGF-β1 supplementation on day 1. On day 3, RGD/RGE-TCO was added to the media to initiate diffusion-controlled reaction. On day 4, TCO media was replaced with TGF-β1 conditioned media. Media was refreshed every other day and cultures were terminated on day 17. (B) Confocal images of day 17 cultures detailing the formation of αSMA stress fiber formation. αSMA, F-actin, and nuclei stained with green, red, and blue, respectively. Scale bar: 20 μm. (C) Quantification of αSMA expression as the integrated density ratio between αSMA and F-actin signals. The sum of fluorescent intensity for αSMA was normalized to that for F-actin. *** p < 0.001, ns: nonsignificant, determined by one-way ANOVA. n=5. All error bars represent the standard error of the mean.