Heme oxygenase-1 accelerates cutaneous wound healing in mice

Abstract

Heme oxygenase-1 (HO-1), a cytoprotective, pro-angiogenic and anti-inflammatory enzyme, is strongly induced in injured tissues. Our aim was to clarify its role in cutaneous wound healing. In wild type mice, maximal expression of HO-1 in the skin was observed on the 2(nd) and 3(rd) days after wounding. Inhibition of HO-1 by tin protoporphyrin-IX resulted in retardation of wound closure. Healing was also delayed in HO-1 deficient mice, where lack of HO-1 could lead to complete suppression of reepithelialization and to formation of extensive skin lesions, accompanied by impaired neovascularization. Experiments performed in transgenic mice bearing HO-1 under control of keratin 14 promoter showed that increased level of HO-1 in keratinocytes is enough to improve the neovascularization and hasten the closure of wounds. Importantly, induction of HO-1 in wounded skin was relatively weak and delayed in diabetic (db/db) mice, in which also angiogenesis and wound closure were impaired. In such animals local delivery of HO-1 transgene using adenoviral vectors accelerated the wound healing and increased the vascularization. In summary, induction of HO-1 is necessary for efficient wound closure and neovascularization. Impaired wound healing in diabetic mice may be associated with delayed HO-1 upregulation and can be improved by HO-1 gene transfer.

Figure 1. Effect of SnPPIX (10 µmol/L) on time of gap closure by HaCaT cells cultured in vitro in the presence of hydroxyurea (10 mmol/L).

Scratch assay. A – Quantitative data. Each point represents mean±SD of 3 experiments done in duplicates. * P<0.05 in comparison to control. B – representative pictures. Scale bar = 200 µm.

Figure 2. A - Expression of HO-1 protein in healthy and wounded skin.

H - healthy skin, PC - positive control (HaCaT cells stimulated with 10 µmol/L of hemin for 24 h). Western blot analysis. Tubulin was used as a housekeeping gene to control the protein loading. One of 3 similar blots. B - Effect of SnPPIX (45 µmol/kg of body weight) injected subcutaneously (once a day for 10 days) on wound closure in C57BL mice. C – Effect of SnPPIX (45 µmol/kg of body weight), injected intraperitoneally (once a day for 10 days) on wound closure in C57BL mice. Each bar represents mean+SD; N = 10 animals per group. * P<0.05, ** P<0.01 in comparison to control, vehicle injected animals. D – representative pictures taken immediately after wounding, and on the 3^rd and 7^th days. Scale bar = 1 mm.

Figure 3. Closure of cuteneous wounds in the HO-1^+/+ (WT), HO-1^+/− (HT), or HO-1^−/− (KO) C57BLxFVB mice.

A – 3-month old animals. B – 6-month old animals. Each point represents individual animal (N = 4–5), lines connect the median values. Crossed points represent animals subjected to euthanasia. * P<0.05, ** P<0.01 in comparison to WT. C – representative pictures showing the wounds in 6-month old animals immediately after wounding and on day 10^th. Scale bar = 5 mm.

Figure 4. HO-1 overexpression in murine keratinocytes.

A – Expression of human HO-1 mRNA in the skin of HO-1^+/+ and transgenic HO-1^Tg mice C57BL mice. Electrophoresis of RT-PCR products (2% agarose gel). EF2 was used as a housekeeping gene. One of 4 similar analyses. B – Representative pictures of immunocytofluorescent staining for HO-1 in primary murine keratinocytes isolated from newborns and cultured in vitro. Scale bar = 100 µm. C – Concentration of HO-1 in lysates of primary murine keratinocytes isolated from newborns and cultured in vitro. ELISA. Each bar represents mean+SD of 6 measurements. *** P<0.001 in comparison to HO-1^+/+.

Figure 5. Activity of primary murine keratinocytes isolated from HO-1^+/+ and HO-1^Tg newborns and cultured in vitro.

A – Migration of cells measured by time of gap closure in the presence of hydroxyurea (10 mmol/L). Scratch assay. B – Spontaneous proliferation of cells cultured for 48 h. BrdU incorporation assay. C – Viability of cells cultured in hypoxia (1% O₂) for 24 h. MTT reduction assay. D – Concentration of VEGF in media harvested from cell cultures after a 24 h incubation. Each bar represents mean+SD of 3–5 experiments. * P<0.05, ** P<0.01 in comparison to HO-1^+/+.

Figure 6. A – Closure of cutaneous wounds in the HO-1^+/+ wild type and HO-1^Tg mice.

Each bar represents mean+SD. N = 10 animals per group. * P<0.05, ** P<0.01, *** P<0.001 in comparison to HO-1^+/+ mice. B – Representative pictures demonstrating CD31 staining of endothelial cells in the wounded skin (3 days after wounding) in the 3-month old mice of different genotypes. Scale bar = 100 µm. C – Number of vessels in wounded skin (3 days after wounding, CD31 staining) in the 3-month old mice of different genotypes. Each bar represents analysis of samples from 5–8 animals. Data are presented as mean+SD. * P<0.05 in comparison to HO-1^+/+ animals.

Figure 7. A – Closure of cutaneous wounds in the wild type (WT) and db/db diabetic C57BL mice.

Each bar represents mean+SD; N = 10 animals per group. ** P<0.01, *** P<0.001 in comparison to WT animals. B – Number of vessels in healthy and wounded skin in the WT and db/db mice. Each bar represents analysis of samples from 10 animals. Data are presented as mean+SD. * P<0.05 in comparison to healthy skin (day 0); # P<0.05 in comparison to WT animals. C – Representative pictures showing blood vessels in healthy and wounded skin of WT and db/db mice. Immunohistochemical staining for CD31. Scale bar = 100 µm. D – Western blot analysis of HO-1 protein expression in healthy and wounded skin of the WT and db/db mice. One of 5 similar blots. E – Western blot analysis of HO-1 protein expression in healthy skin of the wild type and db/db mice 24 h after intradermal injection with hemin (10 mg/kg of body weight). One of 2 similar blots. Tubulin was used as a housekeeping gene to control the protein loading.

Figure 8. Expression of GFP and HO-1 transgenes in skin.

A – Representative pictures showing the expression of GFP in the wounded skin of db/db diabetic mice, 3 days after local injection with AdGFP adenoviral vectors (2.3×10⁷ IU). Bright field and fluorescence microscopy. Scale bar = 100 µm. B – Concentration of HO-1 protein in tissue lysates of healthy and wounded skin of db/db mice, measure on the 3^rd and 14^th days after AdGFP and AdHO-1 vector delivery. Each bar represent mean+SD for 5–8 animals. # P<0.05 in comparison to healthy skin; * P<0.05 in comparison to AdGFP injected animals. C – RT-PCR analysis of rat HO-1 mRNA and total (rat/murine) mRNA in wounded skin, measured on the 1^st and 3^rd days after AdGFP and AdHO-1 vector delivery. Electrophoresis of RT-PCR products in 2% agarose gel. NC-negative control, HS-healthy skin.

Figure 9. Effect of HO-1 transgene delivery on wounds.

A – Effect of HO-1 transgene delivery on wound closure in the db/db diabetic mice. Adenoviral vectors (2.3×10⁷ IU in 100 µL of PBS) were injected subcutaneously near the wound immediately after injury. Control animal were injected with the same amount of AdGFP carriers. Each bar represents mean+SD; N = 5–8 animals per group. * P<0.05 in comparison to control, AdGFP treated mice. B – representative pictures showing blood vessels in the wounded skin of db/db mice injected with AdHO-1 or AdGFP vectors. CD31 staining of the skin cross-section. Scale bar = 100 µm. C – Number of vessels in wounded skin in the db/db mice injected with AdHO-1 or AdGFP, on the 3^rd and 14^th days after wounding. Analysis of specimens stained for CD31 to visualize endothelial cells. Each bar represents mean+SD values for 5–8 animals. * P<0.05 in comparison to control, AdGFP injected animals.