Combination product of dermal matrix, human mesenchymal stem cells, and timolol promotes diabetic wound healing in mice

Abstract

Diabetic foot ulcers are a major health care concern with limited effective therapies. Mesenchymal stem cell (MSC)-based therapies are promising treatment options due to their beneficial effects of immunomodulation, angiogenesis, and other paracrine effects. We investigated whether a bioengineered scaffold device containing hypoxia-preconditioned, allogeneic human MSCs combined with the beta-adrenergic antagonist timolol could improve impaired wound healing in diabetic mice. Different iterations were tested to optimize the primary wound outcome, which was percent of wound epithelialization. MSC preconditioned in 1 μM timolol at 1% oxygen (hypoxia) seeded at a density of 2.5 × 10⁵ cells/cm² on Integra Matrix Wound Scaffold (MSC/T/H/S) applied to wounds and combined with daily topical timolol applications at 2.9 mM resulted in optimal wound epithelialization 65.6% (24.9% ± 13.0% with MSC/T/H/S vs 41.2% ± 20.1%, in control). Systemic absorption of timolol was below the HPLC limit of quantification, suggesting that with the 7-day treatment, accumulative steady-state timolol concentration is minimal. In the early inflammation stage of healing, the MSC/T/H/S treatment increased CCL2 expression, lowered the pro-inflammatory cytokines IL-1B and IL6 levels, decreased neutrophils by 44.8%, and shifted the macrophage ratio of M2/M1 to 1.9 in the wound, demonstrating an anti-inflammatory benefit. Importantly, expression of the endothelial marker CD31 was increased by 2.5-fold with this treatment. Overall, the combination device successfully improved wound healing and reduced the wound inflammatory response in the diabetic mouse model, suggesting that it could be translated to a therapy for patients with diabetic chronic wounds.

Keywords: animal models; diabetes; hypoxia; mesenchymal stem/stromal cells (MSCs).

FIGURE 1

The combined treatment of MSC and timolol improves wound healing in diabetic mice. A, in vivo wounding assay. Bilateral 6 mm skin excisional wounds were made with splints to reduce skin contraction (left panel, day 0) and the different treatment matrices were inserted to the wound beds (right panel, day 7). B, Quantitation of wound healing. A representative image of H&E staining of the day 7 wound is shown. The original wound edge (yellow dashed lines) on each side is determined by the absence of subdermal adipose tissue. Re‐epithelialization (red arrows) is defined by epithelial cell growth above the Integra matrix (scale bar = 1000 μm). C, MSC embedded in Integra improve wound healing. The mouse wounds were treated with different concentrations of MSCs seeded on Integra and collected on day 7. All the MSC‐Integra scaffolds significantly improved wound epithelialization by more than 70% relative to the Integra alone group (*P < .05 as compared to the Integra control). D, Timolol increases wound healing. Treatment of wounds post‐implantation of MSC‐Integra scaffolds with either 1 μM or 2.9 mM timolol increased wound re‐epithelialization as compared to nontimolol treated controls (mean ± SEM, n = 12 wounds from six mice in each group, *P < .05)

FIGURE 2

Quantitation of systemic absorption of timolol. A, Timolol standard curve by UHPLC. A standard curve for UHPLC was prepared with 3 to 300 ng/mL. The lowest value 3 ng/mL of the standard curve is the limits of quantitation (LOQ). B, Timolol and metoprolol peaks in an UHPLC chromatograph. Timolol (100 ng/mL) and metoprolol (1000 ng/mL, control) were added to the untreated mouse plasma, extracted and detected by UHPLC. The retention time of timolol is 6.7 and 8.3 minutes for Metoprolol. One ng/mL is the limits of detection (LOD) or the minimal visible peak in the chromatograph. Any values between 1 and 3 ng/mL (between LOD and LOQ) are considered detectable, but not quantifiable. C, Plasma timolol concentrations. The mouse blood samples were collected on day 7 and examined by UHPLC. Metoprolol (1000 ng/mL) was added as internal standard. The daily accumulative timolol dose is 0.092 mg/mouse, but the detected timolol concentrations are extremely low (mean ± STDEV, n = 6 mice in each group)

FIGURE 3

The MSC/T/H/S treatment decreases expression of inflammatory cytokines in mouse wounds. The relative amounts of cytokine mRNA on the day 3 wounds were measured by quantitative PCR and normalized to unwounded mouse skin. A‐C, CCL‐2, IL‐1β, and IL‐6 decreased significantly with the MSC/T/H/S treatment when compared with the Integra control (*P < .05). D,E, CXCL‐1 and CXCL‐2 show a trend of decreasing, but the fold changes are not statistically significant (mean ± SEM, n = 12 wounds from six mice)

FIGURE 4

The MSC/T/H/S treatment decreases neutrophil numbers and shifts macrophages to M2. A‐C,G, Gating strategy to identify neutrophils, monocytes, and M1 and M2 macrophages. D‐F, Quantification of cell populations. H, Shift of macrophages from M1 to M2 phenotype. Representative dot plots to show the gating strategy in flow cytometry in A. Live cells were selected with Zombie red and the immune cells were identified by the CD45+ population. Neutrophils were identified as Ly6G+ CD11b+ and monocytes/macrophages were identified as Ly6G− CD11b+ in B,C. The CD45+ immune cells were decreased with the MSC/T/H/S treatment in D (*P < .05). Panels E and F were gated in Ly6G− CD11b+ neutrophil cells. Decreased numbers of neutrophils are found in the wounds treated with MSC/T/H/S compared to Integra controls on day 4 post injury in F (*P < .05). M1 and M2 monocytes/macrophages were further subclassified as M1 cells (Ly6C+) and M2 cells (CD206+) in G. On day 4, the percentages of pro‐inflammatory M1 macrophages (Ly6C+) are decreased and the M2/M1 ratio increased with the MSC/T/H/S treatment when compared to controls (Integra alone) in (H) (*P < .05, mean ± STDEV, n = 10 wounds from five mice)

FIGURE 5

Quantitation of angiogenesis. Anti‐CD31 staining across the entire wound section of day 7 mouse wounds with, A, Integra control and, B, MSC/T/H/S treatments (scale bar = 500 μm). CD31 signals are in red, nuclei in blue, and red blood cells in green in the higher magnification (×60) images with, C, Integra control and, D, MSC/T/H/S treatments (scale bar = 25 μm). E, Quantification of the CD31 signals (CD31 positive objects/ wound area) in each treatment groups. The MSC/T/H/S treatment significantly increased the CD31 staining at the wound bed (*P < .05, mean ± STDEV, n = 12 wounds from six mice.)