Matrix Stiffness and Viscoelasticity Influence Human Mesenchymal Stem Cell Immunomodulation

Introduction: Chronic, non-healing wounds are a growing financial and medical burden, with the global wound care market valued at approximately $20B in 2022.1 Human mesenchymal stem cells (hMSCs) are an attractive therapy for chronic, non-healing wounds due to their immunomodulatory behavior.2 However, hMSCs have low viability, poor engraftment, and a risk of eliciting an immune response, thereby reducing therapeutic potential.3 Additionally, variability in donor, age, and tissue source results in heterogeneity.4 Prior studies have demonstrated that substrate elasticity modulates hMSC differentiation.5,6 In addition to substrate elasticity, researchers have also looked at the effects of viscoelasticity on hMSC fate, as human tissues are naturally viscoelastic.7 However, the effects of stiffness and viscoelasticity on hMSC immunomodulation remain largely unknown, with the literature thus far suggesting that a soft,8 more viscous (ie, higher stress relaxation)8 extracellular matrix maximizes the production of paracrine factors by hMSCs that are indicative of immunomodulation. However, other studies suggest that 3D culture of hMSCs is more important than matrix stiffness and viscoelasticity in expressing key immunomodulatory genes.10 Despite the clear impact that biophysical properties have on hMSC immunomodulatory potential, the mechanisms by which these biophysical cues affect immunomodulation have yet to be elucidated. In addition, a comprehensive study seeking to demonstrate the synergistic relationship between matrix stiffness and viscoelasticity on hMSC immunomodulatory for a range of elastic, storage, and loss moduli is missing. We address these gaps in knowledge by investigating the synergistic effects of substrate stiffness and viscoelasticity on hMSC immunomodulatory behavior using highly tunable polyacrylamide (PAAm) gels.

Materials and

Methods: PAAm gels were fabricated as described in the literature.11 Briefly, three gels were fabricated, a low-stiffness gel, a medium-stiffness gel, and a high-stiffness gel, through the crosslinking of acrylamide and bis-acrylamide monomers. Altering the concentration of acrylamide and bis-acrylamide changes the elasticity of the resulting gel. To impart viscoelastic properties to the gels, linear acrylamide was fabricated and mixed with the acrylamide and bis-acrylamide monomer mixture prior to gelation.6,12 Young’s (elastic) modulus (E) was characterized using dynamic mechanical analysis at 37 °C. Viscoelasticity, defined as shear stress-mediated storage and loss moduli, G’ and G’’, was characterized using a rheometer using a frequency sweep from 0.05 to 50 Hz at 37 °C, a controlled 2% strain rate, and 0.1 N normal force. Values at 0.16 Hz were reported; these values were selected due to their biological relevance.14 The PAAm gels were coated in fibronectin via sulfo-SANPAH to aid cell attachment and bone marrow-derived hMSCs were seeded at 5,000 cells/cm2 on the fibronectin-coated PAAm gels. Viability was assessed using a CCK-8 assay and a live/dead assay, and cellular morphology was characterized by fluorescently staining F-actin using Alexa Fluor 488 Phalloidin and nuclei using DAPI. Immunomodulatory cytokine secretion was characterized by fluorescent immunocytochemistry. All fluorescent samples were imaged and quantified using a Nikon Ti2E inverted epifluorescence microscope. All cell studies were completed in either normal growth media or growth media containing the inflammatory molecules, TNF-alpha and IFN-gamma.

Results, Conclusions, and Discussions: Elastic (E) and storage moduli (G’) values increase for the three gel compositions; loss moduli (G’’) and tan delta values also increase in gels with the dissipative linear acrylamide added (Figure A). Cell-matrix interactions directly influence F-actin distribution and focal adhesion formation, leading to known changes in hMSC behavior. Thus, we expect that culture on gels of varying stiffness with or without a viscoelastic component will change the cell-matrix interactions, and that these differing cell-matrix interactions will affect the immunomodulatory potential of the cells. Viscoelastic gels yielded significantly fewer cells per gel with the same seeding density, and there are clear morphological differences between cells on the different stiffness gels (Figure B). Cells on soft gels exhibited larger numbers of cells with smaller surface areas and more rounded morphologies, whereas cells on medium gels exhibited a greater number of cells with a more spindle-like, fibroblast-like morphology (high aspect ratio). Cells on stiff gels exhibited more cells with a greater surface area (more spreading) and protrusions (lower aspect ratio). These results are also consistent with the proliferation rates (Figure C), showing higher proliferation rates for the stiff gels, as well as higher proliferation rates for gels without a viscoelastic component. Additionally, PGE2 expression was enhanced on high stiffness gels (Figure D). PGE2 is known to support resolution of the inflammation stage and transition into the proliferative stage of wound healing, therefore these results are also consistent with the morphology and proliferation rate data. Current work is investigating the synergistic effects of matrix stiffness and viscoelasticity on immunomodulatory cytokine expression.

Figure. A) Gel compositions and expected and measured elastic modulus (E), storage modulus (G’), loss modulus (G’’), and tan delta. B) hMSC morphology (F-actin (green), vinculin (red) and nuclei (blue) staining) of cells cultured on PAAm gels with (+) or without (-) a viscoelastic element after 21 days, scale = 200 µm. C) Proliferation rate over time of hMSCs cultured on PAAm gels in normal growth media. D) Prostaglandin E2 (PGE2) expression on PAAm gels without a viscoelastic component in normal growth media after 21 days, scale = 200 µm.

Acknowledgements (Optional): We would like to acknowledge funding from Clarkson University.

References:
[1] Sen C. et al., Adv Wound Care, 12(12):657-670, 2023.
[2] Keshavarz R. et al., Bioeng. & Trans. Med., 9(1):e10598, 2023.
[3] Im G. et al., Int J Mol Sci., 20(19):4835, 2019.
[4] Chen S. et al., Colloids Surf B Biointerfaces, 140:574-582, 2016.
[5] Wang S. et al., Advanced Healthcare Materials, 10(3):2001244, 2020.
[6] Engler A. et al., Cell, 126(4): 677-689, 2006.
[7] Chaudhuri et al., Nature Materials 15: 326-334, 2016.
[8] Wong S. et al., Science Advances, 6(15): eaaw0158, 2020.
[9] Phuagkhaopong et al., ACS Appl. Mater. Interfaces, 13(26): 30420-20433, 2021.
[10] Gonzales-Pujana et al., Biomaterials, 257: 120266, 2020.
[11] Tse J. and Engler A., Cur. Prot. in Cell Bio., 47(1):10.16.1-10.16.16; 2010.
[12] Pogoda K. et al., Bio-protocol, 11(16):e4131, 2021.