Critical in vivo roles of WNT10A in wound healing by regulating collagen expression/synthesis in WNT10A-deficient mice

Abstract

Background: We have reported that WNT10A plays a critical role in the growth of fibroblasts/myofibroblasts and microvascular endothelial cells, i.e.; wound healing/scarring. To ascertain the in vivo regulatory, central functions of WNT10A, we examined the net effects of WNT10A depletion using WNT10A-deficient mice (WNT10A-/-).

Methods and results: We generated WNT10A-/-mice, displaying a range of unique phenotypes of morpho/organogenetic failure, such as growth retardation, alopecia, kyphosis and infertility, and then focused on the functions of WNT10A in wound healing. We subjected C57BL/6J wild-type (WT) or WNT10A-/-mice to skin ulcer formation. The WNT10A-/-mice had significantly larger injured areas and delayed wound healing, which were associated with (a) a smaller number of fibroblasts/myofibroblasts and microvessels; and (b) more reduced expression and synthesis of collagen, compared with WT mice with intact WNT10A expression, especially in those with activated myofibroblasts.

Conclusions: These observations indicate that WNT10A signaling can play a pivotal in vivo role in wound healing by regulating the expression and synthesis of collagen, as one of fibrogenic factors, at least in part, and critical in vivo roles of WNT10A-mediated effective wound healing are extremely closely associated with collagen expression.

Fig 1. Our generated WNT10A^–/–mice (KO) showing a range of unique phenotypes of morpho/organogenetic failure.

A) Under basal conditions, morphological examinations between the 12-week-old wild-type (WT) and KO mice (n = 10 mice per group) show that the KO mice are grossly smaller in size and have marked alopecia and kyphosis compared to WT. B) Representative whole-body X-ray confirms significant growth retardation in the KO mice, accompanied by marked kyphosis and many bony hyperlucent areas (arrows), in contrast to WT mice (n = 10 mice per group). C) From 1 to 12 months of age (whole one year), both male and female KO display significantly smaller body weights, as reflected by more essential growth retardation, than WT (n = 10 mice per group). D) The bone length and bone mineral density (BMD) in both male and female 12-week-old KO mice are significantly lower than those of WT mice (n = 10 mice per group), corresponding to the gross and imaging findings. E) Female KO mice with a phenotype of infertility demonstrate that the tissue of the grossly smaller KO ovary contains markedly fewer (or no) follicles than the ovary of WT mice on representative H&E sections. F) The number of follicles is significantly smaller in the KO ovary than in the WT ovary (n = 10 mice per group). Scale bars = 100 μm. Values are means ± SE. *P < 0.05, **P < 0.001, ***P < 0.0001.

Fig 2. WNT10A^–/–KO mice showing the delayed skin ulcer/wound healing associated with suppressed stromagenesis, especially in fibroblasts/myofibroblasts.

A) Grossly, KO mice display significantly larger injured areas of skin along with more delayed wound healing than WT mice (n = 8 mice per group) within 10 days of observation of this murine in vivo model. Interestingly, WT mice at day 10 post-injury showed new hair growth, whereas KO mice did not. B) Accordingly, the diameters of the KO skin wound were significantly longer than those of WT mice (n = 8 mice per group) from days 2 to 10 post-injury. C) The representative histopathological pictures reveal that the inflammatory/edematous granulation tissue in KO skin ulcer (bottom) is significantly larger and less cellular (H&E), accompanied by more decreased number of α-SMA-positive spindle fibroblasts/myofibroblasts, than that in WT mice (upper) (n = 8 mice per group), only in which not only WNT10A but β-catenin expression is immunohistochemically overt especially in those activated (myo)fibroblasts. In addition, the incurable KO skin wound contains significantly more repressed and fewer CD31-positive microvessels than that of WT mice. Scale bars = 100 μm. D) Correspondingly, there are significantly fewer α-SMA-positive fibroblasts/myofibroblasts and CD31-positive microvessels in KO mice than in WT mice (n = 8 mice per group). Values are means ± SE. *P < 0.05, **P < 0.001, ***P < 0.0001.

Fig 3. Specific WNT10A expression in the WNT10A^+/+ WT mice activated fibroblasts/myofibroblasts in the murine wound healing model.

Double-immunofluorescence staining show that a large number of α-SMA-positive fibroblasts/myofibroblasts (red-stained) in the WT skin wounds (upper) had specific expression of WNT10A (green-stained), whereas those of KO mice did not (bottom) (n = 8 mice per group). Scale bars = 20 μm.

Fig 4. WNT10A^–/–KO skin wound showing reduced fibrosis/fibrogenesis associated with the decreased expression of stromagenesis-related genes.

A) Representative pictures of Masson’s trichrome staining (Scale bars = 100 μm) show that collagen deposits (blue-stained) were markedly smaller with more reduced fibrosis in the injured incurable skin lesions of KO mice than in those of WT mice (n = 8 mice per group). In addition, an immunofluorescence study (Scale bars = 20 μm) reveals that a large number of WT spindle (myo)fibroblasts (blue-stained in nuclei) have significant expression of Type I/III collagen at day 10 post-skin injury, which is very rarely or not observed in KO (myo)fibroblasts. B) On a quantitative analysis, the blue-stained collagen content of KO injured skin was found to be markedly lower than in WT skin. C) Correspondingly, real-time RT-PCR showed that the mRNA expression of Type I and Type III collagen was significantly lower in the KO skin of the day-3 wound healing model than in the WT skin. Values are means ± SE and were normalized for 18s rRNA expression (real-time RT-PCR). *P < 0.05, ***P < 0.0001.

Fig 5. Immunofluorescence study of in vitro murine dermal fibroblasts showing a unique phenotype of little stromagenesis on WNT10A^–/–KO mice.

Immunofluorescence staining shows that a significantly larger number of cultured dermal fibroblasts (blue-stained in nuclei) contain elevated expression of Type I/III collagen (green-stained) in WT mice than in KO mice (n = 3 per group). The basal condition of fibroblastic proliferation is overtly better in WT mice than that in KO mice, and the morphology of fibroblasts from KO skin is relatively poorly preserved. Scale bars = 50 μm.

Fig 6. The deficiency of WNT10A suppressed cell proliferation activity in wound healing mice model.

Immunofluorescence staining showed that a significantly larger number of histone H3-positive cells (green-stained) were observed in WT mice wound skin, compared to WNT10A^–/–mice (n = 3 per group). The fluorescence of histone H3 was located in nuclei (inset) (blue-stained in nuclei). Scale bars = 100 μm.

Fig 7. A schematic presentation of the critical in vivo roles of WNT10A in ‘wound’ healing.

This diagram depicts the central, key roles of WNT10A in the current murine model of wound healing. Our obtained data confirm that WNT10A expression, especially in fibroblasts/myofibroblasts, can be crucially, responsible for various potentially beneficial effects against wound healing.