Conditioned medium from hypoxic bone marrow-derived mesenchymal stem cells enhances wound healing in mice

Abstract

Growing evidence indicates that bone marrow-derived mesenchymal stem cells (BM-MSCs) enhance wound repair via paracrine. Because the extent of environmental oxygenation affects the innate characteristics of BM-MSCs, including their stemness and migration capacity, the current study set out to elucidate and compare the impact of normoxic and hypoxic cell-culture conditions on the expression and secretion of BM-MSC-derived paracrine molecules (e.g., cytokines, growth factors and chemokines) that hypothetically contribute to cutaneous wound healing in vivo. Semi-quantitative reverse transcriptase polymerase chain reaction (RT-PCR) and enzyme-linked immunosorbent assay (ELISA) analyses of normoxic and hypoxic BM-MSCs and their conditioned medium fractions showed that the stem cells expressed and secreted significantly higher amounts of basic fibroblast growth factor (bFGF),vascular endothelial growth factor A (VEGF-A) interleukin 6 (IL-6) and interleukin 8 (IL-8) under hypoxic conditions. Moreover, hypoxic BM-MSC-derived conditioned medium (hypoCM) vs. normoxic BM-MSC-derived conditioned medium (norCM) or vehicle control medium significantly enhanced the proliferation of keratinocytes, fibroblasts and endothelial cells, the migration of keratinocytes, fibroblasts, endothelial cells and monocytes, and the formation of tubular structures by endothelial cells cultured on Matrigel matrix. Consistent with these in vitro results, skin wound contraction was significantly accelerated in Balb/c nude mice treated with topical hypoCM relative to norCM or the vehicle control. Notably increased in vivo cell proliferation, neovascularization as well as recruitment of inflammatory macrophages and evidently decreased collagen I, and collagen III were also found in the hypoCM-treated group. These findings suggest that BM-MSCs promote murine skin wound healing via hypoxia-enhanced paracrine.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1. Proliferation of BM-MSCs in serum-containing complete culture medium (A) and serum-free α-MEM (B).

Cell proliferation were examined under conditions of hypoxia or normoxia. Data are given as the means± the SEM; *p< 0.05 compared with the norCM group at indicated time points. #p< 0.05 compared with indicated earlier time points.

Figure 2. Relative mRNA and secreted protein expression levels of wound-healing-related growth factors, cytokines and chemokines in normoxic and hypoxic BM-MSCs or conditioned medium.

(A) RT-PCR assays were performed to measure mRNA levels in BM-MSCs. (B, C, D and E) ELISA assays were performed to measure secreted protein levels in BM-MSC norCM and hypoCM. Data are given as the means± the SEM; *p< 0.05 compared with the expression level of each factor under normoxic culture conditions.

Figure 3. Effects of BM-MSC-derived conditioned medium samples on paracrine cell proliferation and migration.

Equal numbers of keratinocytes, fibroblasts and HUVECs were incubated with vehicle control medium, norCM or hypoCM. Cell proliferation was evaluated at indicated time points (A, C and E). Data are given as the means±the SEM; *p< 0.05 compared with the vehicle control or the norCM group. #p< 0.05 the vehicle control compared with the norCM group. Equal numbers of keratinocytes, fibroblasts, HUVECs and CD14+ monocytes were added to the upper chambers of 24-well transwell plates, with the indicated medium added to the lower chambers (n = 4 wells per treatment). Cells that migrated to the bottom of the filter were stained and evaluated (B, D, F and G). Data are given as the means± the SEM;*p< 0.05 compared with the vehicle control or the norCM group. #p< 0.05 the vehicle control compared with the norCM group.

Figure 4. Increased capillary-like tube formation stimulated by BM-MSC hypoCM.

(A) HUVECs were seeded onto a Matrigel matrix and incubated with vehicle control medium, hypoCM or norCM for 12 h. (B)Tube formation was quantified. Scale bar, 400 µm for images in (A) (200×). Results are given as the means ± the SEM; *p< 0.05 compared with the vehicle control or the norCM group.

Figure 5. Effects of BM-MSC hypoCM on murine skin wound healing.

Representative macroscopic views of cutaneous wounds are shown on day 11 after treatment with vehicle control medium (A), norCM (B) or hypoCM (C). (D) The fraction of the wound area at each indicated time point in comparison to the original wound area was quantified as described in Material and methods, and plotted. Values are givenas the means± the SEM;*p<0.01 compared with the vehicle control or the norCM group.

Figure 6. IHC evaluation of wounded mouse skin.

Wound sections were evaluated on day 11 by staining with anti-Ki67 and anti-F4/80 antibodies. The numbers of Ki67+ proliferating cells (A, C) and recruited F4/80+ macrophages (B, D) in each of 4 randomly chosen high-power fields in the dermis were counted. Scale bar, 100 µm (400×). Data are expressed as the mean±the SEM;*p<0.05 compared with the vehicle control or the norCM group.

Figure 7. Angiogenesis in the murine skin after wounding.

Wound sections were evaluated on day 11 by staining with anti-CD31 antibody. Representative CD31+ vessels are shown (A). The extent of vascularization was determined by assessing the number of CD31+ vessels in each of 4 randomly chosen high-power fields within the injury site (B). Scale bar, 100 µm for images in (A) (400×). Results are given as the means ± the SEM; *p< 0.05 compared with the vehicle control or the norCM group.

Figure 8. Collagen I and collagen III synthesis analysis.

(A). Evaluation of relative collagen I, collagen III mRNA expression. (B) Staining of collagen I and collagen III in scars treated with vehicle control, norCM or BM-MSC hypoCM. Scale bar, 100 µm (400×). Data are expressed as the mean±the SEM; *p<0.05 compared with day 20. #p< 0.05 compared with the vehicle control or the norCM group at day 40.