Improved Transplanted Stem Cell Survival in a Polymer Gel Supplemented with Tenascin C Accelerates Healing and Reduces Scarring of Murine Skin Wounds

Abstract

Mesenchymal stem cells (MSCs) remain of great interest in regenerative medicine because of their ability to home to sites of injury, differentiate into a variety of relevant lineages, and modulate inflammation and angiogenesis through paracrine activity. Many studies have found that despite the promise of MSC therapy, cell survival upon implant is highly limited and greatly reduces the therapeutic utility of MSCs. The matrikine tenascin C, a protein expressed often at the edges of a healing wound, contains unique EGF-like repeats that are able to bind EGFR at low affinities and induce downstream prosurvival signaling without inducing receptor internalization. In this study, we utilized tenascin C in a collagen/GAG-based polymer (TPolymer) that has been shown to be beneficial for skin wound healing, incorporating human MSCs into the polymer prior to application to mouse punch biopsy wound beds. We found that the TPolymer was able to promote MSC survival for 21 days in vivo, leading to associated improvements in wound healing such as dermal maturation and collagen content. This was most marked in a model of hypertrophic scarring, in which the scar formation was limited. This approach also reduced the inflammatory response in the wound bed, limiting CD3e⁺ cell invasion by approximately 50% in the early wound-healing process, while increasing the numbers of endothelial cells during the first week of wound healing as well. Ultimately, this matrikine-based approach to improving MSC survival may be of great use across a variety of cell therapies utilizing matrices as delivery vehicles for cells.

Introduction

Mesenchymal stem cells/multipotent stem cells (MSCs) possess strong potential to positively influence the wound microenvironment and therefore are considered attractive candidates for regenerative therapy. MSCs have shown beneficial effects in preclinical models of wound healing by accelerating wound closure, reducing scar formation, and even improving neovascularization ^{1 3} . However, the exact mechanism for MSC-mediated improvements in these models of tissue repair is still poorly understood, as the cells do not contribute to the repaired tissue, due to limited survival upon implant, typically less than a week after application ⁴ . Thus, the beneficial effects are deemed to be due to paracrine signaling through soluble and matrix factors ^{5 7} . Unfortunately, countering this beneficial aspect, the rapid death of the implanted MSC creates a nonspecific inflammation response dampening the success of these preclinical studies.

Efforts to optimize the therapeutic effects of MSCs now focus on the secretion of proregenerative factors to stimulate host recovery and regeneration ⁸ . Several well-documented studies suggest that the multiple anti-inflammatory, angiogenic, neurotrophic, immunomodulatory, and antifibrotic factors are derived from the transplanted MSCs, suggesting that the release of these factors may depend on the specific contextual demands of the type and stage of injury ⁹ . This allows for the consideration of MSCs to be viewed as a “smart,” site-specific multidrug release system.

Although MSCs are shown to demonstrate clear therapeutic potential, current paradigms for their use are clearly not optimal ¹⁰ . Among these obstacles are delivery, safety, survival, dosing, immune response, and costly infusion procedures. Current injury models have shown that MSCs are cleared within days ¹¹ in the heart, kidney, and liver with the delivery of MSCs to cutaneous wound sites demonstrating even lower persistence ¹² . Engineering-based strategies have suggested that the use of biomaterial-based therapeutics for cell delivery may not only enhance cell viability but also contribute to MSC fate determination ¹³ . We have found earlier that cell surface-restricted activation of epidermal growth factor (EGF) receptor (EGFR), accomplished by tethering of the EFGR ligands themselves, promotes MSC survival ¹² while allowing for expansion of the MSCs and maintaining differentiation capacity ² . One way to prolong survival of MSCs is using the matrikine tenascin C (TNC); this matricellular protein contains cryptic low-affinity ligands for the EGFR ¹⁴ . Incorporation of TNC into a matrix protects MSCs from killing by death cytokines^15,16.

We have proposed that to improve pathologic healing, the cellular delivery system must promote survival and allow the cells to be responsive to environmental cues ⁶ . Delivery systems such as hydrogels and acellular dermal matrices lack the beneficial structural, mechanical, and chemical properties needed to provide microenvironment-responsive regeneration. As such, we have developed a gel that shapes the space to be replaced by tissue containing matrikines to enhance survival ¹⁶ . Our biodegradable 3,4-dihydroxyphenylalanine (DOPA)-cross-linked gel is built from polyethylene glycol (PEG) and lysine diisocyanate (LDI). Incorporated into this gel are bioactive agents collagen type I to provide hemostasis and TNC to promote engraftment efficiency and survival^17,18. The antimicrobial and hemostatic agents incorporated into this polymer-based wound gel bode well as a cell delivery system, as these components would reduce innate inflammatory responses and reduce subsequent scarring. Herein we examined the beneficial effect of this polymer for xenogeneic MSC delivery using our recently developed wound-scarring model based on the absence of wound resolution through the chemokine receptor CXCR3 signaling axis. This model yields mice that develop sterile chronic inflammation and hypertrophic scars^19,20.

Materials and Methods

Reagents

For all cell culture, α-minimum essential medium (MEM) (15-012-CV) was obtained from Mediatech (Manassas, VA, USA), fetal bovine serum (FBS) (S11550H) was obtained from Atlanta Biologicals (Flowery Branch, GA, USA), and L-glutamine (25030-081) was obtained from Thermo Fisher Scientific (Gibco, Grand Island, NY, USA). CM-DiI cell tracker (C7000) used in culture was obtained from Life Technologies (Grand Island, NY, USA), and an infrared cell tracker was also obtained from Life Technologies (D12731). For plate coating, rat tail collagen I (354236) was obtained from BD Biosciences (San Jose, CA, USA) at 4.41 mg/ml concentration, and TNC (CC065) was obtained from EMD Millipore (Billerica, MA, USA) at 100 μg/ml. Na₂B₄O₇ (S9640-25G) for prepolymer solvent was obtained from Sigma-Aldrich (St. Louis, MO, USA), and AgNO₃ and (NH₄)₂S₂O₈ for prepolymer catalyst solutions were obtained directly from the polymer manufacturer CM-Tec (Newark, DE, USA). For plug digestion and cell processing for flow, Liberase TL (0540102001) was obtained from Roche (Indianapolis, IN, USA), and 40-μm filters (352340) for cell filtration were obtained from BD Biosciences.

Synthesis of Polymers

The PEG-(HDI-DOPA)₂ (DHP) prepolymer-based gel was generated by CM-Tec in the manner described previously ¹⁷ . In brief, a PEG-DOPA/Col I prepolymer was dissolved in 0.1 M Na₂B₄O₇ aqueous solution (0.3 g prepolymer/ml solution). This underwent polymerization initiated by the redox initiation system AgNO₃/(NH₄)₂S₂O₈ (7 μl/ml gel), yielding 20 μg of Ag/gm polymer gel wet weight. This formulated the PEG-polymer. Specifics of polymer stability and silver diffusion have been published previously ¹⁷ . TNC incorporation into the PEG-polymer was accomplished by coating tissue culture dishes for 16 h at 37°C with 1 μg/cm ² Col I and 1 μg/cm² TNC diluted in phosphate-buffered saline (PBS). PBS was aspirated, and the coated surfaces were placed under UV light for 10 min before cell seeding. Human bone marrow-derived MSCs were seeded on the dish for 24 h before harvesting the matrix and cells (if present) by rubber policeman and mixing with the PEG-polymer (referred to hereafter as PEG-TPolymer).

Animals

C57BL/6J and FVB wild-type (WT) mice and CXCR3 expression-abrogated mice [obtained from Dr. Wayne Hancock and Dr. Bao Lu (Millennium Pharmaceutical Inc., Cambridge, MA, USA and Perlmutter Laboratory, Children's Hospital, Boston, MA, USA)] were bred inhouse. All of the offspring were genotyped by polymerase chain reaction (PCR) before use. Control, age-matched C57BL/6J mice were obtained from Jackson Laboratories (Bar Harbor, ME, USA). Serological analyses did not detect blood-borne pathogens or evidence of infection. Mice were housed in individual cages after wound induction and maintained under a 12-h light/dark cycle and room temperature (20 ± 2°C). Both male and female mice were used, as earlier studies did not reveal any gender differences in healing among the mice at the ages used. All studies on these animals were performed after approval by, and in compliance with, the Institutional Animal Care and Use Committee (IACUC) of the Veteran's Administration and University of Pittsburgh; these animals were housed in a facility of the Veteran's Affair Medical Center (Pittsburgh, PA, USA) accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care.

Wounding

Male and female mice [7 to 8 weeks (considered as young) weighing approximately 25 g] were anesthetized with an intraperitoneal (IP) injection containing ketamine (75 mg/kg; Phoenix Scientific Inc., Solana Beach, CA) and xylazine (5 mg/kg; VEDCO, Saint Joseph, MO, USA). No less than eight mice from each group were used in each experiment. The backs were cleaned, shaved, and sterilized with betadine solution. For full-thickness wounds, a 6-mm-diameter punch biopsy through the epidermis and dermis was made on one side of the dorsal midline (to minimize wound contraction), and the contralateral uninjured skin served as a control. Wounds were either left untreated or treated with TPolymer or TPolymer-MSCs. Thirty minutes after the treatments, we covered the wounds with Tegaderm (3M, St. Paul, MN, USA) in all five groups to maintain uniformity and prevent the mice from removing the treatments. Wounds were examined at 3, 7, 14, 21, 60, or 90 days postwounding. This was done to capture the transitions from inflammatory to regenerative and regenerative to resolving phases of wound healing. Animals were euthanized by CO₂ inhalation, as approved by the American Veterinary Medical Association (AVMA), and the wounds harvested for assessment. Each wound was measured and then removed from the animal, with unwounded skin taken from the contralateral dorsum as a control.

Histological Analysis

Wound bed biopsies surrounded by a margin of non-wounded skin were collected after euthanasia at 3, 7, 14, 21, 60, or 90 days postwounding. Wound biopsies were fixed in 10% buffered formalin, processed, and embedded in paraffin blocks using standard protocols. Tissue sections (5 μm) were stained with hematoxylin and eosin (H&E) (Sigma-Aldrich), Masson's trichrome (Sigma-Aldrich), and PicroSirius Red (Sigma-Aldrich) and analyzed for general tissue and cellular morphology and analyzed using the MetaMorph software (Molecular Devices LLC, Sunnyvale, CA, USA). All images were collected at original magnification 10× and 40×.

Epidermal and Dermal Maturation

Histopathological examination of mouse tissue was performed blinded by a trained histopathologist. Qualitative assessments were made concerning aspects of dermal and epidermal maturation as previously described ²⁰ . In brief, the samples were scored on a scale of 0 to 4 for epidermal healing (0, no migration; 1, partial migration; 2, complete migration with partial keratinization; 3, complete keratinization; and 4, normal epidermis) and dermal healing (0, no healing; 1, inflammatory infiltrate; 2, granulation tissue present-fibroplasias and angiogenesis; 3, collagen deposition replacing granulation tissue >50%; and 4, complete healing).

Flow Cytometry

Twelve-millimeter punch biopsies of the wound and 12-mm punch biopsies of unwounded regions were isolated for quantification of MSC engraftment in wounded skin at days 3, 7, 14, and 21. Tissue was minced and incubated in 0.5 mg/ml Liberase TL (Roche) for 1 h at 37°C. Tissue was filtered; cells were washed and labeled for analysis. Human leukocyte antigen (HLA) markers were used to identify positive human MSCs in the wound bed. Appropriate isotype controls, unstained cells, and untreated wounds were used as controls. Flow cytometry was performed on an LSR II Cytometer (BD Biosciences) and subsequently analyzed using FlowJo digital FACS software (Tree Star, Inc., Ashland, OR, USA).

Cell Culture and MSC Labeling

Primary bone marrow-derived MSCs were obtained from the National Institutes of Health (NIH)-funded MSC repository at Texas A&M under the direction of Dr. Darwin Prockop and primary human mesenchymal stem cells (prhMSCs) immortalized by human telomerase reverse transcriptase (hTERT) (imhMSCs) were obtained from Junya Toguchida's laboratory (Kyoto University, Japan). Both imhMSCs and primary bone marrow-derived cells are considered both established cell lines and commercially available primary cells, and as such are IRB exempted 4e at the University of Pittsburgh (IRB No. PRO13080262). These cells are certified and analyzed for appropriate surface marker expression/differentiation potential by the repository prior to shipment. All primary MSCs were cultured in α-MEM supplemented with 17.5% triple-dialyzed FBS and L-glutamine. Passage 1 (P1) MSCs were expanded at 37°C to approximately 70% confluence, when they were passaged until P3 to allow for sufficient cell numbers for implant.

For cell tracking, CM-DiI (C7000) (Thermo Fisher) was resuspended at 1 mg/ml in dimethyl sulfoxide (DMSO) (Sigma-Aldrich) as a stock solution, and then diluted to 1 μM in warm PBS, where it was left on cells at 37°C for 5 min followed by 15 min at 4°C. The flask was then washed and incubated in proliferation media overnight. Fluorescent signal from MSCs in culture was confirmed prior to use in experiments. For infrared tracking of implanted cells, 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindotricarbocyanine iodide (D12731) (Life Technologies) was added directly to culture media at 1 μM from a stock solution of 1 mg/ml. Labeling solution was incubated on the adherent MSCs for 30 min, followed by replacement with normal proliferation media.

Polymer Characterization Analyses

For in vitro assessment of the polymer, degradation assays and cell incorporation assays were performed. Degradation was allowed to proceed with 0.4 g of formed polymer in 2 ml of PBS at 37°C over 10 days, with a daily scoring of remaining polymer area as a means of analyzing degradation. Cells tracked with CM-DiI label were visualized via fluorescence microscopy within the polymer immediately after incorporation.

Statistical Analyses

For wound-healing outcomes (wound maturation, dermal maturation, and collagen content), a two-way Student's t-test was performed for direct comparisons of treatments, with values of p < 0.05 considered statistically significant. For inflammatory infiltrate, one-way analysis of variance (ANOVA) of each mouse model group was performed for all three conditions, with p < 0.05 considered significant. Posteriori test [Tukey's honestly significant difference (HSD)] was used to confirm where the differences occurred between groups on all groups that showed an overall significant difference. Data are presented as mean and SEM. GraphPad software was used.

Results

MSC-Incorporated PEG-Polymer Is Biodegradable

To assess the polymer as a biodegradable MSC delivery system, we evaluated its degradation and cell survival properties. PEG-TPolymer degradation was tested in vitro by placing 100 mg of polymer in phosphate buffer solution and changing the buffer daily. Approximately 60% of the PEG-TPolymer was still intact 10 days after formulation compared to PEG-Polymer alone being almost absent by this time (Fig. 1A and B). Thus, the addition of TNC promoted gel retention. Imaging of the PEG-TPolymer with and without MSC incorporation indicated that the gel maintained viable MSCs during the studied time period (Fig. 1C). The polymer with TNC and MSCs quickly embedded within a mouse wound (Fig. 1D) and remained intact for days (data not shown).

Figure 1.

The PEG-collagen polymer degrades gradually in vitro and integrates into mouse wounds. The PEG-collagen polymer was assessed in vitro for degradation in PBS over 10 days as a function of preparation state (prepolymer, polymer alone, or polymer + TNC) (A), with percent degradation assessed as a size of the original polymer aggregate (B; day 8 shown). (A) Gel polymerizes with a tissuelike consistency in vitro. For MSC incorporation testing, cell presence was monitored via CM-DiI cell tracker on MSCs prior to addition to the polymer (C; day 8 shown). 10× images, 200 mM. For polymer application, the functional polymer showed adequate integration into wound beds after 40–60 min (D). Shown is one of at least two experiments with similar outcomes. Scale bars: 400 μm.

PEG-TPolymer/MSC Survival and Engraftment In Vivo Capabilities

To determine whether the MSCs survive in the wound bed, we labeled the cells with an infrared tracking dye and imaged them over a weeklong period. Qualitatively, there was a decrease in the xenotransplanted human MSCs (Fig. 2A). To quantitate this effect, we assessed MSC survival after 3, 7, or 21 days, using flow cytometric detection of cells stained for human HLA (Fig. 2B). Three days after treatment with PEG-TPolymer + MSCs, 50% of the transplanted cells were still present in the wound bed. The MSCs survived in the gel over 21 days, but declined in cell number, as expected in immunocompetent animals. This survival was noted in both WT mice and CXCR3^−/- [knockout (KO)] mice. This demonstrated that not only does the polymer have superior biocompositional properties but that it also enhances the persistence of the human MSCs.

Figure 2.

Mesenchymal stem cells survive in the polymer in vivo over 21 days in the presence of TNC. MSCs were tracked prior to polymer application with an infrared membrane label, and wound areas were imaged directly on wild-type mice using a LiCOR Odyssey imaging system to detect the IR label. Shown here are three representative images over 18 days (A). To quantify the MSC survival, cells were extracted from mouse wounds via collagenase digestion and probed with a human-specific HLA antibody to detect human MSCs via flow cytometry (B). MSC survival in the wound bed was assessed for both wild-type and CXCR3^−/- mice. Percent positive for HLA (left) normalized to PEG-TPolymer and absolute numbers (right) were determined via flow cytometry analysis. Background was subtracted. n = 3–5 for flow cytometry analyses; n = 1 for IR tracking confirmation.

PEG-TPolymer/MSC Effect on Improved Scarring

With the enhanced MSC survival, we next determined whether this would improve the healing of wounds and normalize the scar phenotype. Full-thickness wounds were created on the WT and CXCR3^−/- (KO) mice, and followed for up to 90 days to track the full healing or scarring as noted in our earlier publications ¹⁸ . Wound maturation was measured to account for a composite of epidermal thickness, keratinization, chronic inflammation, fibroblast cellularity, vascularity, and collagen modeling. Quantification of these wounds revealed that wound repair and dermal maturation in mice that were treated with PEG-TPolymer + MSCs showed significant improvement in comparison with untreated mice in the first 60 days postwounding (Fig. 3A and B). The PEG-TPolymer itself, without MSCs, does not alter the wound healing significantly.

Figure 3.

MSC-based polymer yields improved wound-healing outcomes. Full-thickness sections of mouse wounds across 60 days were assessed via H&E stain (A; day 7 shown) and quantified for whole-wound maturation and dermal maturation (B; day 7 shown) by a veterinary pathologist in a double-blinded analysis. Both wild-type and CXCR3^−/- mice were assessed. Wound tissue was also sectioned and treated with a trichrome stain to assess matrix content and organization at the same time points (C; day 60 shown). Total percent collagen content was analyzed through a MetaMorph analysis normalized to unwounded tissue (D). n = 3. Scale bars: 200 μm.

More strikingly, wounds treated with PEG-TPolymer + MSCs resulted in significantly more collagen earlier in the healing process than untreated at days 7 and 21 postinjury; this was determined by Masson's trichrome staining of histological sections and subsequent MetaMorph analysis for quantitation (Fig. 3C and D). PicroSirius Red staining, which detects appropriately aligned collagen bundles, shows that the PEG-TPolymer + MSCs promoted the replacement of immature collagen fibers that fail to provide strength in the KO scarring mice to more mature aligned fibers (Fig. 4). Of interest, and more strikingly in KO mice, PEG-TPolymer (even without the added MSCs) improved the collagen alignment. This indicates a dual benefit of the therapy, in part coming from the polymer gel and in part from the cellular products.

Figure 4.

Collagen alignment and maturation improve using the MSC-based polymer. Full-thickness mouse skin sections from wound beds were excised over 60 days and probed with PicroSirius Red (PS red) stain to assess collagen outcomes in the wounds (A; day 21 shown; scale bar: 200 μm) for both wild-type and CXCR3^−/- mice. PS red was scored for collagen alignment in a double-blinded fashion by a veterinary pathologist for wild-type (B) and knockout (C) models across 90 days of wound healing. n = 3. Full-thickness mouse skin sections from wound beds were excised over 60 days and probed with PS red stain to assess collagen outcomes in the wounds (A; day 21 shown) for both wild-type and CXCR3^−/- mice. n = 3. *p < 0.05.

PEG-TPolymer/MSC Effect on Immune Response

To assess potential paracrine activity from MSCs surviving in the wound bed, we assessed CD31⁺ cell infiltration as a metric of angiogenic signaling and CD3e⁺ infiltration as an indicator of inflammatory response in relation to T-cell invasion (Fig. 5). Flow cytometric analysis suggested that the inflammatory infiltrate early in the wound-healing response was blunted with less CD3⁺ cell invasion and at both day 3 and day 7 during the peak inflammatory response (Fig. 5A). Additionally, we assessed CD11B, a leukocyte adhesion and migration marker, to mediate the inflammatory response in general, and results did now show a consistent change as a result of including MSCs in the PEG-TPolymer for CD3⁺ cell results, although in the presence of the MSCs, the CD3⁺ and CD11B⁺ cell infiltration tended to be less.

Figure 5.

MSC-based polymer augments murine cell infiltrate in wound beds. Cell extracts from mouse wounds in both wild-type and CXCR3^−/- mice were probed for the presence of CD3e⁺ and CD11b⁺ (A) and CD31⁺ (B) cells to assess inflammatory and angiogenic infiltrate over the first 21 days of wound repair. Treatment conditions analyzed for murine infiltrate were untreated mice, mice treated with a TNC-based polymer only, and mice treated with the TNC-based polymer combined with primary human MSCs. n = 3–5.

CD31⁺ cell infiltration, as a method for assessing angiogenic response in the wound bed, did show marked changes as a result of PEG-TPolymer treatment (Fig. 5B). Both WT and KO mice demonstrated an approximately twofold increased presence of CD31⁺ cells after 3 days, with relatively comparable levels to the untreated mice at day 21. Day 7 results also showed a trend toward increased CD31⁺ cell invasion in both models, although the KO mice PEG-TPolymer + MSC condition demonstrated statistical significance. However, the WT model reverted to a lower level of invasion at day 7. Again, there did not appear to be a consistent change as a result of MSC presence in the wound bed, suggesting the PEG-TPolymer itself acted as a paracrine attractant for the invading murine cells.

Discussion

In this study, we utilized a PEG-collagen-based polymer to effectively deliver xenogeneic MSCs into the wound and appropriately alter the pathological state of the wound. Herein we have shown that our polymer-based approach enhances the survival of these cells while retaining stem cell markers, resulting in a microenvironment projected toward a more regenerative healing. Compared to hydrogel and injection delivery systems, the PEG-collagen delivers the MSCs in a protective scaffold that biodegrades^17,21,22 over time to allow for a reduction in the otherwise robust and damaging immune influx, and increased angiogenesis with a subsequent improvement in healing.

Matrix-based approaches to wound healing are not a particularly novel innovation, and many commercially available treatments for chronic wound dressings utilize fabricated or existing matrices to that effect. This has previously consisted of collagen-based dressings, including decellularized collagen matrices in concert with glycosaminoglycans ²³ , bilayered matrices incorporating fibroblasts and keratinocytes to jump-start the healing process ²⁴ , and others. While these approaches provide the host tissue with space for native cells to migrate and engraft into the new wound matrix, they ultimately fail to address many of the key pathologies for chronic wound healing, such as extended or persistent inflammation or reduced angiogenesis ^{25 28} . Thus, novel treatments to address these factors have become necessary in the next generation of wound-healing strategies. We have investigated MSCs for this purpose because of their widely acknowledged role in normal wound healing, as well as their capacity to modulate the chronic inflammation and avascularity that plague such wounds^1,7,29–32.

While MSCs undoubtedly hold great promise for improving on existing matrix-based approaches to wound healing, a fundamental constraint in all MSC therapies has been limited cell survival upon implant due to the harsh nature of the wound environments. Because of the matrix-based character of the wound-healing polymer (and other cited treatments for wound healing), a logical additive to our treatment were proteins that might assist with improving survival of implanted cells. We utilized the EGF-like repeats of TNC here for this purpose, which are able to bind EGFR at low affinity but high avidity to induce pro-survival signaling constitutively. While we showed here that this approach can in fact promote weeks-long survival of implanted cells, this treatment strategy also affords any migrating resident cells the opportunity to survive much longer than expected in the wound bed. Indeed, a major factor contributing to the chronic nature of diabetic ulcers and other comparable conditions is the lack of healthy fibroblasts and/or keratinocytes lining the wound bed, which are subjected to the same conditions that implanted MSCs face in vivo ^{33 35} . By incorporating an abundance of TNC, it is probable that these cells (fibroblasts, keratinocytes, CD31⁺ cells, and others) will have the ability to invade and engraft in the same matrix, ideally surviving on the same time scale and thus allowing a more complete healing response. By the same token, invading inflammatory cells might also be able to exploit the TNC infrastructure to resist anti-inflammatory strategies, reinforcing the importance of maintaining the immunomodulatory MSC population at high quality in the long term.

A limitation of this study is the well-acknowledged drawbacks of murine skin wound healing. As can be seen even in our data in this study, mouse wounds rarely fail to heal and, in fact, mostly heal at a rapid pace, entering the remodeling phase within a few weeks at most. Long-term outcomes that are analyzed such as our day 60 or day 90 time points here provide little insight beyond potential scarring implications, as seen in the CXCR3^−/- model as characterized previously^18,20. As such, while we do show an improved rate of healing and collagen maturation in mouse wounds treated with our polymer, we have not here delineated effects in a truly human relevant “chronic wound” model or one with a delayed healing response in general. We have also here only examined bone marrow-derived MSCs, whereas many groups have proposed adipose-derived MSCs for skin wound therapy in order to ease the burden of translation to humans and potential differences in paracrine profiles of the two subpopulations. Future studies examining the optimal conditions for MSC success in this wound-healing polymer must include an exploration of the MSC subtype most suited for the TNC approach to extending survival while also leading to the most desirable wound-healing outcomes.

In addition to the functional importance of MSCs in our treatment strategy, the delivery polymer must also encapsulate the other necessary aspects of complete wound healing. Previously, we have shown that this PEG polymer is able to promote wound healing in comparable mouse models ²¹ , also leveraging an AgNO₃/(NH₄)₂S₂O₈ catalyst system for polymerization for this purpose. Silver has a well-documented antimicrobial role^36,37, and as such we originally applied this as the rationale for a silver nitrate-based redox catalysis method. Silver ions in solution as part of the catalysis process are ultimately able to act in a dual fashion as both a catalyst and an antimicrobial agent, which we showed was functional in previous studies with this polymer^17,21. In concert with the wound-healing matrix and MSC-based therapeutic for reducing inflammation and inducing a more robust wound-healing response, this polymer as a whole appears to encapsulate the entire wound-healing process and holds great promise for scaling to chronic wound therapies in the future.