Ligand presentation controls collective MSC response to matrix stress relaxation in hybrid PEG-HA hydrogels

Abstract

Cell interactions with the extracellular matrix (ECM) influence intracellular signaling pathways related to proliferation, differentiation, and secretion, amongst other functions. Herein, bone-marrow derived mesenchymal stromal cells (MSCs) are encapsulated in a hydrazone crosslinked hyaluronic acid (HA) hydrogel, and the extent of stress relaxation is controlled by systemic introduction of irreversible triazole crosslinks. MSCs form elongated multicellular structures within hydrogels containing RGD peptide and formulated with elastic composition slightly higher than the hydrogel percolation threshold (12 % triazole, 88 % hydrazone). A scaling analysis is presented ( ${<{R_{g\text{Structure}}}^2>}^{\frac{1}{2}}$ ∼N^α) to quantify cell-material interactions within these structures with the scaling exponent (α) describing either elongated (0.66) or globular (0.33) structures. Cellular interactions with the material were controlled through peptides to present integrin binding ECM cues (RGD) or cadherin binding cell-cell cues (HAVDI) and MSCs were observed to form highly elongated structures in RGD containing hydrogels ( $\alpha =0.56\pm 0.05$ ), whereases collapsed structures were observed within HAVDI containing hydrogels ( $\alpha =0.39\pm 0.04$ ). Finally, cytokine secretion was investigated, and a global increase in secreted cytokines was observed for collapsed structures compared to elongated. Taken together, this study presents a novel method to characterize cellular interactions within a stress relaxing hydrogel where altered cluster morphology imparts changes to cluster secretory profiles.

Keywords: Hydrogels; Ligand interactions; Mesenchymal stromal cells; Stress relaxation.

Graphical abstract

Fig. 1

Network structure of hybrid, stress relaxing hydrogel containing mimetic peptides. A) Schematic representation of the dual-network containing both the dynamic hydrazone bond, and the elastic triazole bond and the mimetic peptide's, RGD and HAVDI, attached to the HA-Hyd or PEG-BCN, respectively. B) The schematics of the macromers present within the hydrogel including the hydrazide functionalized HA, aldehyde functionalized HA and BCN functionalized PEG groups. The reactant groups forming the various bonds throughout the hydrogel, including the dynamic hydrazone bonds (either alkyl- or benzyl based), and the triazole bonds. C) The small molecule and peptide epitopes present throughout the hydrogel, enabling control over the formation of the SPAAC network and composition of the hydrogel, respectively. D) Tables containing the formulations of the differing stress relaxing conditions, as well as the concentrations of the varied RGD: HAVDI peptide ratios utilized throughout this paper. E) Hydrogel in situ formation, reaching a final modulus ranging from 1300 Pa to 1600 Pa. F) The stress-relaxation of the hydrogels, over 6 h, after in situ formation with the 94 % and 88 % alkyl hydrazone bonds conditions relaxing the most stress over the timescale. G) The percentage of stress relaxed by the differing alkyl hydrazone compositions. ∗p < 0.05, ∗∗p < 0.01, ns: no significance.

Fig. 2

Multi-cellular structure formation dependence on stress relaxing hydrogels. A-C) Representative images of the multi-cellular structures observed in the 82 % (A), 88 % (B) and 94 % (C) alkyl hydrazone bonds conditions, scale bar = 50 μm. D) The total number of nuclei/cluster within the multi-cellular clusters for the 82 %, 88 % and 94 % alkyl hydrazone bond conditions. ∗p < 0.05, ns: no significance.

Fig. 3

Scaling analysis model to determine cell-cell and cell-ECM interactions. A) Skeletonized image (white), overlaid with AF-647 phalloidin (red) for a representative image within the 88 % alkyl hydrazone bonds condition, including the radius of gyration (magenta circle, with orange radius), yielding an α value of 0.57 for this multi-cellular structure, scale bar = 50 μm. B) The radius of gyration, <R_{g, structure}²>^1/2 for the 82 %, 88 % and 94 % conditions, yielding the largest sized multi-cellular structures in the 88 % alkyl hydrazone bonds condition. C) The scaling exponents for the 82 %, 88 % and 94 % alkyl hydrazone bonds conditions, with the 88 % condition yielding the highest α value, near 0.6, and the 82 % yielding the lowest, significantly different, α-value near 0.5. ∗p < 0.05, ∗∗p < 0.01 ns: no significance.

Fig. 4

Influence of RGD and HAVDI integrin-binding on multi-cellular structures. A) Representative image of a multi-cellular structure within the 88 % alkyl hydrazone bonds condition containing 1 mM HAVDI, scale bar = 50 μm. B) Representative image of the 88 % alkyl hydrazone bond condition cell-ECM interactions via visualization of mature focal adhesions, scale bar = 50 μm. C) Representative image of the 88 % alkyl hydrazone bond condition cell-cell interactions via N-cadherin, scale bar = 50 μm. D) The intensity expression of a mature focal adhesion marker, Paxillin, indicating similar expression regardless of presented peptide epitope. E) The intensity expression of a cell-cell junction marker, N-Cadherin, indicating similar expression regardless of presented peptide epitope. F) The total nuclei/structure, again, showing regardless of presented peptide-epitope the number of cell nuclei within each structure is similar. G) The average distance of all the pixels within the structure to the center of mass, or the radius of gyration, for the 1 mM RGD and 1 mM HAVDI conditions within the 88 % alkyl hydrazone bonds condition, with the 1 mM RGD condition leading to a greater radius of gyration, indicating an increase in cell-ECM interactions. H) The derived scaling exponent compared between the 1 mM RGD condition and the 1 mM HAVDI condition within the 88 % alkyl hydrazone bonds condition. The 1 mM RGD condition led to an increased scaling exponent, averaging around 0.56, in comparison to the 1 mM HAVDI condition, averaging around 0.39, indicating increased cell-ECM interactions as compared to cell-cell. ∗∗∗∗p < 0.0001, ns: no significance.

Fig. 5

Influence of RGD: HAVDI dual incorporation on MSCs cell-ECM interactions. A-E) Representative images of a multi-cellular structure within the 88 % alkyl hydrazone bonds condition containing 1 mM RGD (A), 3:1 RGD: HAVDI (B), 1:1 RGD: HAVDI (C), 1:3 RGD:HAVDI (D) and 1 mM HAVDI (E) conditions, scale bar = 50 μm. F). The PitStop2, endocytosis inhibitor control representative image, nuclei (blue), and F-actin (red), scale bar = 50 μm. G) The derived radius of gyration compared between PitStop2, 1 mM RGD, 3:1 RGD: HAVDI, 1:1 RGD: HAVDI, 1:3 RGD:HAVDI and the 1 mM HAVDI condition within the 88 % alkyl hydrazone bonds condition. H) The derived scaling exponent compared between PitStop2, 1 mM RGD, 3:1 RGD: HAVDI, 1:1 RGD: HAVDI, 1:3 RGD:HAVDI and the 1 mM HAVDI condition within the 88 % alkyl hydrazone bonds condition. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗∗p < 0.0001, ns: no significance.

Fig. 6

Integrin presentation influences MSC secretion independent of stable cell-ECM interactions. A) Heatmap of cytokine expression of MSCs encapsulated in the 88 % alkyl hydrazone condition with either 1 mM RGD or 1 mM HAVDI, normalized to total DNA concentration/hydrogel. Blue intensities represent high expression values, and white intensities represent low expression values.

Fig. 7

Representative schematic of the influence of varying mimetic peptides and stress relaxation on MSC mechanosensing, specifically the cell-ECM interactions and the ensuing formation of multi-cellular structures.