This led us to investigate the therapeutic effectiveness of iMSCs derived from cord tissue-iPSCs (CT-iPSCs) in promoting wound healing and skin regeneration in our established porcine thermal injury model. An MSC differentiation protocol was conducted to generate CT-iMSCs from a pre-established, validated CT-iPSC line. CT-iMSCs were applied to a well-recognized porcine thermal injury model, where they displayed superior healing outcomes compared to the control groups. These data could transform the burn care practice and provide hope for millions of burn patients.
Figure 1 represents the schematic of the study flow, outlining the main steps and methods used throughout the research. Mohamed et al. previously demonstrated that human cord-tissue stem cells can be reprogrammed into induced pluripotent stem cells (CT-MSC-iPSCs). These cells were fully characterized through standard assays showing the expression of pluripotency markers Nanog, Oct4, and Sox2. In addition, they showed the ability to differentiate into the three germ layers (ectoderm, mesoderm, and endoderm) in vitro, and finally, they successfully formed tumors when injected subcutaneously in nude mice. CT-MSC-iPSCs, here termed CT-iPSCs, were generously donated to us by Dr. Ian Rogers, University of Toronto, Canada. To confirm that cells retained their pluripotency, immunofluorescence and flow cytometry analyses were conducted, showing that the cells positively expressed pluripotency markers Nanog, Sox2, and TRA-1-60, respectively (Supplementary Fig. 1A, B). Next, a previously published protocol was utilized to induce the differentiation of CT-iPSCs into iPSCs-induced MSCs or iMSCs. Cells were kept in a mesodermal induction medium for three days, before switching to MSCs induction medium for another ten days. As shown in supplementary Fig. 1, the expression of mesodermal markers Eomesodermin (Eomes) and MIXL1 was markedly enhanced by day 7, suggesting early differentiation of CT-iPSCs towards mesodermal derivatives (Supplementary Fig. 1C, D). Cells further differentiated into CD105 + , CD73 + , CD90 + , CD45- MSCs (Fig. 2A). Differentiated cells showed apparent morphological changes adopting the characteristic plastic adherent spindle shape (Fig. 2B). Additionally, the multi-differentiation capacity into adipocytes, osteoblasts, and chondrocytes was confirmed through staining analysis with Oil-Red-O, Alcian Blue, and Alizarin Red staining (Fig. 2C). Collectively, these results suggested that the CT-iPSCs were efficiently induced to mesodermal-lineage cells that could be furtherly differentiated into functional MSCs in vitro.
After positively characterizing the nature of iMSC, A well-established porcine thermal injury model was used to assess their in vivo therapeutic effectiveness. Previously, it was shown that MSCs could be effectively incorporated into Integra®, an FDA-approved, synthetic meshed bilayer wound coverage material scaffold for burn wound treatment. Therefore, a similar protocol to incorporate iMSCs into Integra® was applied. The successful seeding of iMSCs onto Integra® was visually assessed by the absence of floating cells, confirming full incorporation and attachment to the scaffold before application to the experimental animal. Next, scaffolds containing iMSCs were placed on the 5 × 5 cm full-thickness excisional wounds as previously described. Notably, no adverse events related to the treatments, such as local or systemic signs of inflammation or infection, were observed throughout the experiment.
Interestingly, gross wound analysis showed that all iMSC-treated wounds exhibited an accelerated wound closure as early as 32 days post-burn compared to burn alone and acellular Integra groups (Fig. 3A, and Supplementary Fig. 2A, B). However, all wounds showed a complete closure by day 40 (sacrifice day). Interestingly, analysis of the wound closure rate over the 40 days treatment period showed that the control groups showed an initial accelerated closure rate up to 12 days post burn, however, iMSCs-treated groups displayed a notable accelerated closure rate during the re-epithelization period (approximately between days 12-25 post-burn) suggesting a direct effect of the seeded stem cells on the epithelization process (Supplementary Fig. 2A, B). Indeed, analysis of the rate of epithelization significantly increased in 5K-iMSC and 10K-iMSC groups compared to burn and acellular Integra groups (Fig. 3B). Further, the 10K-iMSCs exhibited the largest re-epithelialized area, significantly higher compared to the acellular Integra (Fig. 3C). Of note, the extent of re-epithelialization was assessed after removal of the silicone (bi-) layer of Integra® between days 10 and 14. Scarring susceptibility is another key factor in wound healing assessment. Scar quality on day 40 was assessed by evaluating the rate of wound contracture and the Vancouver Scar Scale (VSS) score. The analysis of macroscopic images suggested a manifested decrease in wound contraction in iMSCs-treated wounds relative to the control wounds (burn and acellular Integra) (Fig. 3 A and D). Furthermore, quantitative analysis revealed that both the 5K-iMSC and 10K-iMSC groups exhibited the lowest contracture rates, although the differences were not statistically significant (Fig. 3E). Nevertheless, no statistically significant differences in VSS scores were observed between groups; scar scores were higher in burn (6.67 ± 1.0) and acellular Integra (6.00 ± 0.57) compared to 5K-iMSCs (5.58 ± 0.91), 10K-iMSCs (5.33 ± 1.47), and 20K- iMSCs (6.00 ± 0.54) (Fig. 3F).
On the other hand, granulation tissue formation is an important step in the wound healing process. Wounds were scored for the presence, feature, and amount of granulation tissue throughout healing using the Bates-Jensen Wound Assessment tool. Scores range were as follows 1 = Skin intact or partial thickness wound, 2 = Bright, beefy red; 75% to 100% of wound filled and/or tissue overgrowth, 3 = Bright, beefy red; < 75% and > 25% of wound filled 4 = Pink, and/or dull, dusky red and/or fills < 25% of wound, 5 = No granulation tissue present. By postburn day 4, no apparent granulation tissue deposition was observed in all groups. However, the collagen layer of the Integra templates was seen to integrate into wounds with no significant differences in iMCS-seeded groups compared to controls (burn and acellular Integra). Between postburn days 9-12, all wounds showed the presence of granulation tissue covering between 75 and 100% of the total wound area (Supplementary Fig. 2C). Upon entering into the reepithelialization phase, granulation tissue in iMCS-seeded groups started to be replaced by intact new epithelial tissue as reflected in granulation tissue scores (Supplementary Fig. 2C). In particular, 10 k/cm iMSCs showed a steady increase in the score, indicating an improved wound-healing process that matched the wound closure course (Supplementary Fig. 2C).
Next, we sought to evaluate the wound tissue morphology. Wound tissue sections were stained for hematoxylin and eosin. A distinct skin morphology consisting of epidermis and dermis layers could be observed in wounds of all groups (Fig. 4A). Interestingly, apart from 10 k/cm iMSCs, all groups showed a significant increase in epidermis thickness relative to the non-burned porcine skin (normal skin) (Fig. 4C). All groups demonstrated visibly apparent layers of epidermis; stratum corneum, stratum lucidum, stratum granulosum, and stratum basale being distinctive (Fig. 4A- lower magnified panel). The stratum corneum was noticeable as the uppermost flattened keratinocyte layer containing a dense network of keratin that varied in thickness between different groups. Acellular Integra and iMSCs 20 K/cm showed a thicker keratin layer than the 5 and 10 K iMSCs groups. An underlying transparent stratum lucidum layer could be observed especially in the Integra group followed by the Stratum Basale as the dense deepest layer of the epidermis (Fig. 4A-middle panel). Close examination showed that wounds treated with 10 k/cm iMSCs demonstrated a histological morphology of the epidermis most comparable to non-burned skin (Fig. 4A-lower magnified panel). Subsequently, the formation of rete ridges, the epithelial protrusion at the junction between epidermis and dermis, was quantified in wound histological sections. Control groups (burn and acellular Integra) followed by 20K-iMSCs had the lowest number of rete ridges, a statistically significant difference compared to 5K-iMSCs, 10K-iMSCs, and non-burned skin. The average number of ridges was 3.8 ± 1 and 4.3 ± 0.83 for burn and acellular Integra groups, respectively (Fig. 4D). While the average number of ridges in iMSCs wounds at 5 k/cm = 7.6 ± 1, 10 k/cm = 10.2 ± 0.83, and 20 k/cm = 4 ± 1 per 1 mm (Fig. 4D). Notably, wounds treated with 10K-iMSCs exhibited rete ridges formation most comparable to non-burned skin (10.2 ± 0.83 and 12.8 ± 1.2, respectively). Dermis layer regeneration was then analyzed. Histologically, the dermis is a mainly fibrous connective tissue layer with a predominant collagen component and an embedded cellular population of fibroblasts, macrophages, and mast cells. Therefore, we analyzed wound samples with Mason-trichrome staining. Interestingly, acellular Integra wounds exhibited apparent hyperplasia in the dermis (Supplementary Fig. 3A, B) with an average of 53 × 10 ± 6700 cells per mm. On the other hand, 20k followed by 10 K iMSCs groups showed the most comparable cell number compared to the non-burned skin (44 × 10 ± 2500, 40.8 × 10 ± 6613, and 45.5 × 10 ± 7770 cells per mm respectively (Supplementary Fig. 3B). Additionally, 10k iMSCs-treated wounds had the highest collagen deposition, followed by 5 K iMSCs wounds (66 ± 15 and 56 ± 12% of total area), significantly higher than the burn and acellular Integra (22 ± 12 and 33 ± 8%), respectively; however, lower than the non-burned skin 78 ± 5.3 (Fig. 4E). Collectively, these histological assessments demonstrated that iMSCs treatment could improve the healing skin tendency for restoration of the natural skin features including both dermis and epidermis layers. In particular, a 10 K iMSCs dose tends to produce better wound healing outcomes and restore the skin histological features (Supplementary Fig. 3C).
To investigate iMSC-mediated mechanisms in wound repair, we analyzed the expression of key wound-healing-associated markers starting with major pro-inflammatory cytokines; IL1β, IL4, IL6, TNF-α, and growth factor; TGFβ3. Interestingly, all iMSC-treated wounds exhibited reduced IL-1β expression compared to the burn control, with the 10 K iMSC group showing a statistically significant decrease (Fig. 5A). Similar patterns were observed for IL-4 and TNF-α, with the 10 K iMSC group again displaying the lowest, though not statistically significant, expression levels (Fig. 5B and Supplementary Fig. 4A). Likewise, for IL-6 and TGF-β3, the 10 K iMSC group showed the lowest expression among all treated groups, albeit without statistical significance, whereas the other iMSC-treated and acellular Integra groups exhibited expression levels comparable to the burn control (Fig. 5C, D). Further, we analyzed the expression of the two major dermal collagens: Collagen 1 and 3. Notably, all iMSCs-treated wounds exhibited a comparable non-significant 2.5-fold increase in the expression of Collagen 1α1 compared to the burn control Fig. 5E). However, Collagen 3A1 expression was similar across groups, with 5 K and 10 K iMSCs groups exhibiting slightly higher and lower expression, respectively (Fig. 5F). Replacement of collagen 3 with collagen 1 is particularly important for sound wound healing. Therefore, our evaluation showed that the 10 K iMSCs group had the highest Collagen1/3 ratio followed by acellular Integra and 5 K iMSCs groups (Supplementary Fig. 4B). Since collagen deposition is regulated by the action of the matrix metalloproteinases (MMPs) and their tissue inhibitors of metalloproteinases (TIMPS), we evaluated how major MMPs might be regulated with iMSCs treatment. Interestingly, wounds treated with iMSCs (5 and 20 K) showed a significantly higher MMP2 expression than burn and acellular Integra controls (Fig. 5G). On the contrary, they showed a significant decrease in MMP9 expression (Fig. 5H). Interestingly, plasminogen activator inhibitor-1 (PAI-1; SERPINE1), expression was increased in 5K-iMSC and 10K-iMSCs-treated wounds; significantly increased in 5 and 10 K while nonsignificant in the 20 K group (Supplementary Fig. 4C). Finally, to analyze the angiogenic potential of iMSCs, we analyzed the expression of the extracellular matrix (ECM) glycoprotein Tenascin-C (TNC) and vascular endothelial growth factor (VEGF), essential for angiogenesis during wound healing. All iMSCs groups showed a 2-fold increase of TNC compared to the burn group (Supplementary Fig. 4D) and a slight increase of VEGF in 5 and 10 K iMSCs groups, though all changes were statistically non-significant (Supplementary Fig. 4E).
Together, these data showed that treatment with iMSCs enhanced burn wound healing outcomes that could be attributed to a general improvement in the expression of major molecules associated with different stages of wound healing. Particularly, a dose of 10 K iMSCs/cm tends to produce lower inflammatory, higher collagen, and more angiogenic markers gene expressions which could be translated into better overall healing features.