Synergistic effect of Hypoxic Conditioning and Cell-Tethering Colloidal Gels enhanced Productivity of MSC Paracrine Factors and Accelerated Vessel Regeneration

Abstract

Microporous hydrogels have been widely used for delivering therapeutic cells. However, several critical issues, such as the lack of control over the harsh environment they are subjected to under pathological conditions and rapid egression of cells from the hydrogels, have produced limited therapeutic outcomes. To address these critical challenges, cell-tethering and hypoxic conditioning colloidal hydrogels containing mesenchymal stem cells (MSCs) are introduced to increase the productivity of paracrine factors locally and in a long-term manner. Cell-tethering colloidal hydrogels that are composed of tyramine-conjugated gelatin prevent cells from egressing through on-cell oxidative phenolic crosslinks while providing mechanical stimulation and interconnected microporous networks to allow for host-implant interactions. Oxygenating microparticles encapsulated in tyramine-conjugated colloidal microgels continuously generated oxygen for 2 weeks with rapid diffusion, resulting in maintaining a mild hypoxic condition while MSCs consumed oxygen under severe hypoxia. Synergistically, local retention of MSCs within the mild hypoxic-conditioned and mechanically robust colloidal hydrogels significantly increased the secretion of various angiogenic cytokines and chemokines. The oxygenating colloidal hydrogels induced anti-inflammatory responses, reduced cellular apoptosis, and promoted numerous large blood vessels in vivo. Finally, mice injected with the MSC-tethered oxygenating colloidal hydrogels significantly improved blood flow restoration and muscle regeneration in a hindlimb ischemia (HLI) model.

Keywords: colloidal gel; hMSC; hypoxic conditioning; ischemic disease; mechanical stimulation; oxygenating microparticles; paracrine effect; vessel regeneration.

Figure 1:. Preparation and characterization of OMP-encapsulated GelTA/GelMA colloidal hydrogel.

(a) Schematic representation of TA moiety and OMP incorporation effect on the crosslinking networks of inter- and intra-colloidal particles after an annealing process that formed the polypeptide helix structures in the order for: Pristine Gelatin/GelMA colloidal hydrogel (GC), Cell-tethering GelTA/GelMA colloidal hydrogel (TC), and Cell-tethering self-oxygenating OMP-laden GelTA/GelMA colloidal hydrogel (TOC). (b) The size distribution of OMPs obtained by DLS analysis and SEM images show the morphology of OMPs. (c) The size distribution of TC, GC, and TOC colloidal particles obtained by DLS analysis. (d) Fluorescent microscope images to confirm porous networks of TC, GC, and TOC hydrogels indicated by infiltration of FITC labeled with dextran (FITC-dextran). Black color indicates the presence of a colloidal particle. (e) Surface topology of TC, GC, and TOC hydrogels analyzed using the in-fluid imaging set up of the AFM. (f) Z-stack confocal microscopy images of TOC hydrogels at regular intervals to confirm the degree of the OMP encapsulation inside TC colloidal particles. Rhodamine-conjugated OMPs encapsulated in TC hydrogels that were infiltrated by FITC-dextran (porous space) are observable. (g) Alizarin red and H&E staining images of TOC hydrogels to indicate calcium peroxide (CaO₂) in the OMPs and corresponding distribution of the distances between OMPs present in the hydrogels. (h) Swelling test of TOB, GC, TC, GOC and TOC hydrogels in DPBS at room temperature after 1 day. Swelling ratio for the group of the colloidal gels including GC, GOC, TC, and TOC were significantly lowered compared to the TOB group. (‘****’: p<0.0001, ‘***’: p<0.001, ‘**’: p<0.01; unpaired t test; n=3). (i) Porosity of TOB, TC, GOC, and TOC hydrogels calculated from the void area in H&E staining images of colloidal hydrogels (‘****’: p<0.0001, ‘***’: p<0.001, ‘**’: p<0.01, ‘*’: p<0.05; One-way ANOVA with Tukey’s multiple comparisons test; n=5). (j) Distribution of pore size diameters of TOB, TC, GC, GOC, and TOC hydrogels measured by mercury porosimeter and schematic illustration of estimated porous structures in TC and TOC hydrogels. (k) Oxygen release profiles of OMPs and TOC hydrogels in oxygen-depleted PBS solution (n=3). (l) Cryo-SEM images of TOC particles show nano-porous morphology. (m) Simulation data show oxygen diffusion in the TOC hydrogels. (n) Young’s moduli of TOB, GC, TC, GOC, and TOC colloidal particles indicating microscale stiffness determined by AFM-assisted nano-indentation (‘**’: p<0.01, ‘*’: p<0.05; unpaired t test; n=3). (o) Compressive modulus of TOB,GC, TC, GOC, and TOC hydrogels indicating macroscale stiffness obtained by using Instron mechanical tests (‘****’: p<0.0001, ‘***’: p<0.001, ‘*’: p<0.05; One-way ANOVA with Tukey’s multiple comparisons test; n=5).

Figure 2:. Cell-tethering effect on hMSCs behavior in TC and GC colloidal hydrogels.

(a) Brightfield and fluorescent images of GC and TC hydrogels that were enzymatically crosslinked with AF647-labeled TA (AF647-TA). (b) A schematic diagram illustrating the enzymatically-crosslinked hMSCs on the surface of TC hydrogel via the TA moiety with HRP and H₂O₂ after a vigorous washing step. Fluorescent images showed a higher number of green fluorescence (cell tracker)-labeled hMSCs on the TC hydrogels that were labeled by AF647-TA. (c) Live and dead images of hMSCs encapsulated in GC and TC hydrogels on Day 1 and (d) its quantification data (ns=0.94; One-way ANOVA with Tukey’s multiple comparisons test; n=10). (e) Proliferation of hMSCs in GC and TC hydrogels for 7 days using 3D MTS assay (‘***’: p<0.001; unpaired t test; n=3). (f) F-actin/DAPI stained images to confirm hMSC morphology in GC and TC hydrogels for 7 days and (g) its quantification data showing total cell number and (h) stretched cell number in the colloidal hydrogels for 7 days (‘****’: p<0.0001, ‘***’: p<0.001, ‘**’: p<0.01, ‘*’: p<0.05; unpaired t test; n=3). (i) A schematic diagram of increased cell binding sites of hMSCs in the TC hydrogels through both chemical and biological cell binding compared with GC hydrogels shows the cell tethering effect of the TA moiety. All experiments in Figure 2 were performed under a normoxic condition.

Figure 3:. Cooperative stimulation with oxygenation and mechanical cues on cellular morphology and the paracrine effect of hMSCs-laden colloidal hydrogels cultured in a severe hypoxic condition.

(a) Oxygen consumption of hMSC-laden TC, GOC and TOC hydrogels compared with acellular colloidal hydrogels on Day 1. (‘****’: p<0.0001, ‘***’: p<0.001, ‘**’: p<0.01, ‘*’: p<0.05; unpaired t test; n=3). (b) Viability of hMSC-laden TC, TOC, and GOC hydrogels using 3D MTS assay (‘*’: p<0.05; unpaired t test; n=3). (c) Schematic description of increased secreted paracrine factors from hMSCs in TOC hydrogel; on the other side, TC hydrogel induces a pro-apoptosis effect in a severe hypoxic condition. (d) F-actin/DAPI stained fluorescein images of hMSC-laden TC, TOC, and GOC hydrogels for 14 days, and (e) its quantification data showed total cell numbers (‘*’: p<0.05; Two-way ANOVA with Sidak’s multiple comparisons test; n=6). (f) Quantified wound closure and (g) morphology of cell HUVEC migration treated by conditioned media collected from hMSC-laden TC, TOC, and GOC hydrogels on Day 1. Control group treated by EGM-2 endothelial cell culture media (‘****’: p<0.0001, ‘***’: p<0.001, ‘**’: p<0.005; Two-way ANOVA with Sidak’s multiple comparisons test; n=10). (h) Accumulated VEGF secretion from hMSC-laden TC, TOC, and GOC hydrogel for 14 days (‘***’: p<0.001; Two-way ANOVA with Tukey’s multiple comparisons test; n=3) (i, j) Cytokine array of conditioned media collected from hMSC-laden TC, TOC, and GOC hydrogels on Day 3 (indicating explanations of abbreviations in Table S1). (k-q) Gene expression according to angiogenic factors (VEGF and PIGF), representative hypoxic marker (HIF1α), mechanical characteristics and cell binding (ITGβ1, ITGα5 and YAP2), and SDF-1α expression of hMSC-laden TOC, TC and GOC hydrogels on Day 3 (‘***’: p<0.001, ‘**’: p<0.01, ‘*’: p<0.05; unpaired t test; n=3). All experiments in Figure 3 were performed under 0.1% O₂ severe hypoxic conditions.

Figure 4:. Effect of OMPs, TA moiety, and interconnected porous networks in colloidal hydrogels on vascularization in subcutaneous implantation.

(a) Schematic description of subcutaneous implantation of hMSC-laden colloidal hydrogels. Photographs show well-integrated TOC hydrogels into the host tissue at 2-weeks post-implantation. (b,c) Representative low and high magnified (b) H&E and (c) M&T staining images of colloidal hydrogels at 2-weeks post-implantation. (d) Quantification of total number of cells inside colloidal hydrogels at 1- and 2-weeks post-implantation (‘****’: p<0.0001, ‘**’: p<0.01; One-way ANOVA Tukey’s multiple comparisons test; n=10, 3 gels were implanted per each group). (e) Representative images for TUNEL assay of colloidal hydrogels implantation at 2-weeks post-implantation. (f) Quantification of TUNEL-positive cells among implanted hMSCs and invaded host cells into the colloidal hydrogels at 2-weeks post-implantation (‘****’: p<0.0001, ‘**’: p<0.01; One-way ANOVA Tukey’s with multiple comparisons test; n=20). (g) Representative immunofluorescence images stained with CD31 (orange) on colloidal hydrogels at 2-weeks post-implantation. (h) Quantification of vessel diameter and (i) number of vessels at 2-weeks post-implantation (‘****’: p<0.0001, ‘***’: p<0.001, ‘**’: p<0.01, ‘*’: p<0.05; One-way ANOVA with Tukey’s multiple comparisons test; n=30). (i-m) Representative immunofluorescence images of cells stained with (i) hCD44 and (l) HIF1α for each gel on colloidal hydrogels at 1-weeks post-implantation and quantitative analysis of (k) CD44 and (m) HIF1α-positive cells at 1- and 2-weeks post-implantation (‘****’: p<0.0001, ‘***’: p<0.001, ‘**’: p<0.01, ‘*’: p<0.05; One-way ANOVA with Tukey’s multiple comparisons test; n=15).

Figure 5:. In vivo delivery of injectable hMSC-laden TOC hydrogels to improve blood flow restoration in nude mice experiencing hindlimb ischemia

(a) Schematics show experimental setup and intramuscular (I.M.) injection regimen of hMSC-laden TOC hydrogel. (b) Rheological behavior of TC and TOC hydrogels that show deformation and recovery of strain rate. Colloidal hydrogels evolved from repeated cycles of 3-min low strain (1% strain) and 2-min high strain (100% strain) oscillations over time. (c) The images after injection of injectable TOC gel with fluorescence dye ex vivo. (d) Representative LDI images of hindlimbs immediately after FAL and at 3, 7, and 14 days later. (e) Quantification of blood flow (surgical limb/contralateral limb) by LDI images, normalized to measurements immediately after surgery (‘****’: p<0.0001, ‘***’: p=0.0002, ‘**’: p=0.0016; Two-way ANOVA with Sidak’s multiple comparisons test; n=4–6). (f) Necrosis score of ischemic foot 2 weeks after FAL.

Figure 6:. Evaluation of vessel and muscle regeneration after injection of hMSC-laden TOC hydrogel.

(a) Representative Masson’s trichrome staining images of no treatment (control), TC (no hMSCs) and hMSC-laden TOC hydrogels at 2-weeks post-implantation and (b) quantified total tissue area on sectioned samples. (c) Representative immunofluorescence images of cells stained with IB4 (green) and α-SMA (red) for all groups at 2-weeks post-implantation: (i) middle area of TOC hydrogel, (ii) interface area between TOC hydrogel and host tissue, (iii) area of TOC hydrogel in close proximity to the host tissue. (d-f) Quantification of the (d) isolectin IB4-positive capillaries, and (e) diameter and (f) number of α-SMA-positive arterioles found in the harvested samples at 2-weeks post-implantation (‘****’: p<0.0001, ‘**’: p=0.0072; One-way ANOVA with Tukey’s multiple comparisons test; n=20). (g) Representative immunofluorescence images of cells stained with hCD44 (green) and MyHC (red) on samples from all conditions were harvested at 2-weeks post-implantation, and (h) the counted number of myofibers containing centrally located nuclei was determined per high-power field (HFP) (‘****’: p<0.0001, ‘*’: p<0.005; One-way ANOVA with Tukey’s multiple comparisons test; n=20). (i-k) Representative immunofluorescence images with (i) VEGF, (j) PIGF and (k) IGFBP3 of harvested samples in TC and TOC groups as well as the untreated injury site at 2-weeks post-implantation and their quantification data with (l) VEGF, (m) PIGF and (n) IGFBP3 (‘****’: p<0.0001, ‘***’: p<0.001, ‘**’: p<0.01, ‘*’: p<0.05; One-way ANOVA with Tukey’s multiple comparisons test; n=20). (o-q) Representative immunofluorescence images with (o) C-kit, (p) CD34 and (q) SDF-1α of harvested samples in TC and TOC groups as well as the untreated injury site at 2-weeks post-implantation and their quantification data with (r) C-kit, (s) CD34 and (t) SDF-1α (‘****’: p<0.0001, ‘***’: p<0.001, ‘**’: p<0.01, ‘*’: p<0.05; One-way ANOVA with Tukey’s multiple comparisons test; n=20).

Schematic 1.. The schematic illustration of oxygenating and cell-tethering colloidal hydrogels increasing the productivity of paracrine factors locally and in a long-term manner for hindlimb ischemia (HLI) treatment.

GelTA/GelMA colloidal particles containing OMPs release oxygen constantly and provide both irreversible chemical (tyramine) and reversible biological (RGD) cell binding sites. A mild hypoxic environment is maintained inside the colloidal hydrogel despite the oxygen consumption by encapsulated cells under anoxic conditions. Additionally, the presence of the TA group on the colloidal particles prevents natural cell egression through oxidative phenolic crosslinking with hMSCs. The oxygenating colloidal hydrogels could also provide favorable mechanical cues to stimulate mechanosensing signaling of hMSCs. These key factors significantly increase the productivity of paracrine factors from hMSCs that can promote blood flow restoration and muscle regeneration in an HLI model.