Impaired wound healing and ulcer complications are major causes of morbidity in patients with diabetes. Impaired wound healing is associated with increased inflammation and poor angiogenesis in diabetes patients. Here, we demonstrate that topical administration of a secreted recombinant protein (Meteorin-like [Metrnl]) accelerates wound epithelialization and angiogenesis in mice. We observed a significant increase in Metrnl expression during physiological wound healing; however, its expression remained low during diabetic wound healing. Functionally, the recombinant protein Metrnl significantly accelerated wound closure in normal and diabetic mice models including db/db, high-fat diet/streptozotocin (HFD/STZ), and STZ mice. Mechanistically, keratinocytes secrete quantities of Metrnl to promote angiogenesis; increase endothelial cell proliferation, migration, and tube formation; and enhance macrophage polarization to the M2 type. Meanwhile, M2 macrophages secrete Metrnl to further stimulate angiogenesis. Moreover, the keratinocyte- and macrophage-produced cytokine Metrnl drives postinjury angiogenesis and reepithelialization through activation of AKT phosphorylation (S473) in a KIT receptor tyrosine kinase (c-Kit)–dependent manner. In conclusion, our study suggests that Metrnl has a biological effect in accelerating wound closure through c-Kit–dependent angiogenesis and epithelialization.

Article Highlights
  • The topical application of Metrnl hydrogel accelerates wound healing in both normal and diabetic mice.

  • Metrnl has multiple roles in cell proliferation, migration, tube formation, and macrophage M2 polarization, and thus promotes angiogenesis and reepithelialization in mouse models.

  • Metrnl drives postinjury angiogenesis and reepithelialization in a c-Kit/AKTS473–dependent manner.

Impaired wound healing, especially in patients with diabetes, is an important cause of morbidity and mortality and can have serious socioeconomic consequences (13). In addition, poor wound healing in patients with diabetes often lead to infection, causing chronic inflammation, sepsis, wound dehiscence, and even death (4). Despite the major impact of this chronic trauma, effective treatments are lacking. Current treatments, including dressing modifications and administration of growth factors and cytokines (5,6), are still inadequate.

Diabetic wounds exhibit hyperinfiltration and activation of neutrophils, impaired angiogenesis, and defects in epithelial cell migration and proliferation (7). These defects contribute to a persistent inflammatory phase in wounds, which leads to further tissue damage through increased production of inflammatory cytokines, reactive oxygen species, and destructive enzymes (8,9). Angiogenesis, the growth of new blood vessels to nourish damaged tissue, is a critical molecular event in wound healing (10). Furthermore, reepithelialization involves proliferation and migration of keratinocytes, and plays a crucial role in wound healing (11). Therefore, strategies targeting inflammation, angiogenesis, and reepithelialization are considered promising for the initiation of the proliferative phase of wound healing. There is an urgent need to investigate multifunctional factors that can simultaneously impact these processes.

Meteorin-like hormone (Metrnl) is widely expressed in adipose, skeletal muscle, skin, and cardiac tissues (12,13). Specifically, Metrnl functions as a myokine or adipokine that regulates metabolism (12,1416). Metrnl was recently identified as a cardiac factor that protects the heart from cardiac dysfunction (1719). Metrnl also supports skeletal muscle and cardiac repair and regulates the activity of multiple cell types, including alternately activated macrophages, osteocytes, and endothelial cells (13,20,21). Furthermore, Metrnl induction during tissue injury and its role in immune regulation are both required for efficient wound closure, suggesting a potential wider function of Metrnl in tissue repair. The high expression of Metrnl observed in the skin prompted us to investigate the role of Metrnl as a secreted protein in wound healing.

In this study, we investigated the role of Metrnl in physiological and pathological wound healing of the skin. We describe the multifunctional role of Metrnl in diabetic wound healing by promoting postinjury angiogenesis and epithelialization of endothelial cells and keratinocytes. The results indicate that Metrnl secretion mainly occurs in macrophages and keratinocytes in response to local injury. Furthermore, Metrnl acts directly on keratinocytes and endothelial cells through a KIT receptor tyrosine kinase (c-Kit)–dependent mechanism to promote proliferation, migration, and angiogenesis. Importantly, topical application of the secreted recombinant protein Metrnl (rMet) effectively promoted wound healing in both normal and diabetic mice, suggesting that Metrnl may be an attractive biological application for healing diabetic wounds.

Antibody and Reagents

We used the following antibodies: Ki67 (Proteintech; #27309-1), CK14 (Proteintech; #10143-1), CD31 (Invitrogen; #TLD3A12), VEGF (Proteintech; #19003-1-AP), α-tubulin (Proteintech; #11224-1-AP), GAPDH (Proteintech; #60004-1-Ig), c-Kit (Proteintech; #18696-1), p-c-Kit (CST; #48347), AKT (Proteintech; #10176-2), p-AktSer473 (CST; #4060), and CD206 (R&D Systems; #AF2535). All secondary Alexa Fluor (488 or 555) antibodies were from Invitrogen. GTVisionTM III Detection System/Mo&Rb (including DAB) was from Gene Tech (GK500705). Kit inhibitor PLX-3397 was obtained from MCE (HY-16749). Endotoxin-free recombinant mouse Metrnl was produced by AtaGenix (Wuhan, China).

Preparation of Hydrogels Containing rMet

rMet solution was added to Pluronic F-127 (PF-127, σ) aqueous solutions and stirred at 4°C for 2 h. The final concentration of rMet was 2.0 mg/mL, and the hydrogel concentration was 20% (w/v).

Animals and Wounding Experiments

After a 2-week acclimatization period, 8-week-old male C57BL/6 mice, purchased from GuangDong Medical Laboratory Animal Center (Guangzhou, China), were administered a 100 mg/kg streptozotocin (STZ) intraperitoneal injection for two consecutive days. A fasting blood glucose (FBG) level greater than 13.88 mmol/L was used as the diagnostic standard for diabetes. To establish type 2 diabetes mouse models, 8-week-old C57BL/6 mice received a 50 mg/kg STZ injection for 3 days and were subsequentially fed with high-fat diets (HFDs) (Dyets) for 16 weeks. Additionally, 10-week-old homozygote BKS db/db mice obtained from GemPharmatech (Nanjing, China) were used as a mouse model of type 2 diabetes, with age-matched C57BL/6 mice serving as controls.

For the full-thickness cutaneous wounds, mice were first shaved and wiped. They were then anesthetized with isoflurane, and two symmetrical full-thickness wounds were made on both sides of the spine using a round sterile 6-mm biopsy punch. The cutaneous wounds were topically treated twice daily with rMet hydrogels or vehicle (30 μL). Digital images of the wounds were taken at 0, 1, 4, 7, 11 and 13 days, and the wound area was measured using ImageJ software. The changes in the wound areas were expressed as a proportion of the original wound areas. At the specified time intervals after the surgery, skin tissues with a 2-mm margin around the wound were collected for further analysis.

All animal experiments were conducted in accordance with the guidelines for the care of laboratory animals and in strict compliance with the regulations of the Guizhou Medical University Institutional Animal Ethics Committee (Guiyang, China).

Human Samples

Human skin tissue samples and serum samples were obtained from Chen Qi at the affiliated hospital of Southwest Medical University. The collection of samples was approved by an ethical committee review board (reference no. KY2021286 [Luzhou, China] and Chinese Clinical Trial Registry [ChiCTR2100054579]) (Chengdu, China). Fasting blood and debrided tissues (within 1 cm of the wounds) were collected from patients with type 2 diabetes with ulcers during their podiatry assessment clinic. The detailed data of patients with type 2 diabetes were collected as described in Supplementary Tables 1 and 2. Fasting blood samples were centrifuged within 1 h of collection and stored at 4°C. Devitalized tissue was obtained from desloughing and debridement procedures. These samples were stored in liquid nitrogen until further analysis. Control samples were obtained from patients without type 2 diabetes who had undergone amputation due to car accidents at the same clinic.

Cell Culture and Treatment

Human keratinocyte line HaCat and mouse RAW264.7 macrophages were obtained from ATCC and cultured in DMEM (4.5 g/L glucose) supplemented with 10% FBS and penicillin/streptomycin. Human umbilical vein endothelial cells (HUVEC) obtained from ATCC were cultured in RPMI 1640 supplemented with 10% FBS. Human primary cells including keratinocytes and vascular endothelial cells with eligible appraisal reports were purchased from Pricella Biotechnology Co., Ltd. and Shanghai Zhong Qiao Xin Zhou Biotechnology Co., Ltd., respectively. Human primary keratinocytes were cultured using the primary keratinocyte culture system purchased from iCell (PriMed-iCELL-010). Human primary vascular endothelial cells were cultured using Endothelial Cell Medium purchased from SclenCell (#1001) All cells were cultured at 37°C with 5% CO2.

Adenovirus-Mediated Gene Expression and Knockdown

Recombinant adenovirus was generated using the AdEasy Adenovirus System (22). The Metrnl gene was cloned into the pAdTrack-CMV vector. The linearized plasmid was transformed into AdEasier cells to generate recombinant adenoviral plasmids. These plasmids were then transfected into 293A cells to obtain recombinant adenovirus (referred to as Ad-Metrnl below). For adenovirus-mediated gene knockdown, shRNA-Metrnl was cloned into the pAdTrack-1H-U6 vector. The recombinant virus was prepared following the same procedure mentioned above. The shRNA-Metrnl sequence used was CAAGGACTTCCAGAGGATGT.

Quantitative Real-Time PCR

Total RNA was extracted from tissue or cultured cell samples using TRIzol reagent (Invitrogen) according to the manufacturer’s instructions. The RNA concentration was determined using a NanoDrop One spectrophotometer (ThermoFisher Scientific). Subsequently, RNA was converted to cDNA using a PrimeScript Reverse Transcription kit (Takara). Quantitative PCR analysis was performed using quantitative real-time PCR (qPCR) SYBR Green Master Mix (Yeasen). The relative expression of the target genes was calculated using the 2-ΔΔCT method. The primers used in the study are listed in Supplementary Table 3.

Immunohistochemistry and Immunofluorescence

Paraffin-embedded skin tissues were sliced into sections with a thickness of 3 μm. The sections were then deparaffinized and rehydrated, followed by antigen retrieval using sodium citrate solution. To quench endogenous peroxidase enzymes, the samples were incubated with 3% H2O2 for 20 min, and nonspecific binding was reduced by incubating the samples with a blocking buffer containing 1% BSA for 30 min. Next, the samples were incubated with the indicated primary antibody overnight at 4°C, followed by staining with an immunohistochemical kit (Gene Tech) or Alexa 488 or Alexa 594 secondary antibodies. Images were captured using an Olympus BX53 microscope (Olympus) or FV3000 laser scanning confocal microscope system (Olympus). The staining intensity was quantified using ImageJ Pro Plus 6.0 software.

Histological Analysis of Skin Tissue

Patients’ and mice skin tissues were embedded in paraffin and cross-sectioned (3 μm) for histological examination. Hematoxylin-eosin staining (H&E) was performed following routine procedures. Masson staining was performed using staining kits (Solarbio) as per the manufacturers’ instructions. Slices were photographed using an Olympus BX53 microscope (Olympus).

Metrnl Content in Serum and Medium Supernatant

The level of Metrnl in serum collected from patients with diabetic feet was measured using the Metrnl ELISA Kit (R&D), according to the manufacturer’s instructions. After applying various treatments, the supernatant medium of cells was replaced with serum-free medium for 24 h and then collected to measure the concentration of Metrnl.

Western Blotting

Cultured cells were lysed with radioimmunoprecipitation assay buffer containing protease inhibitors, phenylmethanesulfonyl fluoride (PMSF), and phosphatase inhibitors. The resulting soluble supernatant was carefully transferred to fresh eppendorf (EP) tubes for protein assay using the bicinchoninic acid (BCA) protein assay. Proteins were separated by 10–15% SDS-PAGE and transferred onto polyvinylidene fluoride membranes. The membranes were then blocked with 5% milk/Tris-buffered saline with Tween for 1 h at room temperature. The selected proteins were probed with the primary antibody mentioned above, incubated overnight at 4°C. Prior to incubation with horseradish peroxidase-conjugated secondary antibodies for 1 h at room temperature, the membranes were washed. The Western blotting (WB) signals were visualized using enhanced chemiluminescence detection reagents.

Colony Formation

Eight hundred cells were seeded into six-well plates and incubated for 1 week in an atmosphere with 5% CO2 at 37°C. The clones were fixed with 4% paraformaldehyde and then stained with 0.1% crystal violet (Solarbio) for 30 min. The clones were then washed using double distilled water, and the images were captured using a digital camera.

Cholecystokinin-8 Assay

A cholecystokinin-8 (CCK-8) assay was conducted to observe cell growth. Briefly, cells were inoculated in 96-well plates at a density of 1 × 103 cells per well and incubated overnight. After that, the cells were supplemented with the indicated treatment for 72 h. Following the incubation, a CCK-8 solution was added to the cells and incubated for 1 h at 37°C. The absorbance of the cells was then detected using a spectrophotometer at a wavelength of 450 nm.

5-ethynyl-2′-deoxyuridine Incorporation Assay

Cell proliferation was assessed using a 5-ethynyl-2-deoxyuridine (EdU) incorporation assay kit (Beyotime Biotechnology). Following treatment, cells were washed twice with PBS and stained using the EdU incorporation assay as per the manufacturer’s instructions. Subsequently, the stained cells were imaged using the FV3000 laser scanning confocal microscope system (Olympus).

Scratch Wound Healing Assay

HUVECs or HaCat cells were seeded in six-well plates and allowed to grow until they reached 100% confluence. Subsequently, the cells were scratched using pipette tips. Images of the scratch wounds were captured at 0, 24, and 48 h using an Olympus BX53 microscope. The measurement of the wound areas was done using ImageJ software. The initial wound area at 0 h was considered as the baseline (100%).

Transwell Migration Assay

HUVECs (2 × 104 cells per well) or HaCat (1 × 105 cells per well) were suspended in 200 μL serum-free DMEM and added to the upper chamber of transwell. The upper chamber was supplemented with or without rMet (1 μg/mL) or Ad-Metrnl, while the lower chamber was filled with 600 μL DMEM. After incubation for 24 or 48 h, the upper chamber was fixed with 4% paraformaldehyde for 30 min. The cells were then stained with 0.1% crystal violet. Nonmigrated cells on the inner side of the upper chamber were removed using a cotton swab. The cells on the lower surface were photographed using an Olympus BX53 upright microscope.

Tube Formation Assay

After precooling, 96-well culture plates were coated with 50 μL ABW Matrigel (Nova Medical Science) and solidified for 1 h at 37°C. HUVEC cells (2 × 104 cells per well) suspended in 50 μL RPMI 1640 with or without rMet (1 μg/mL) were then seeded in the 96-well plates. For the coculture system, HUVECs were suspended with medium collected from HaCat or Raw246.7 cells and added to each well. After incubation at 37°C for 8 h, capillary-like tubes formed by HUVECs were photographed under an Olympus IX70 microscope. The length of the tubes was calculated using ImageJ software.

Statistical Analysis

The experimental data were presented as mean ± SD using GraphPad Prism 8.0. Statistical significance between two groups was assessed using Student t tests. For comparisons at multiple time points, a two-way ANOVA followed by a Dunnett test was used. A value of P < 0.05 was considered statistically significant.

Data and Resource Availability

Data and resource are available from the corresponding authors.

Characterization of Metrnl Expression During Skin Wound Healing

Metrnl is ubiquitously expressed in various tissues, including the heart, liver, kidney, and skin (https://www.proteinatlas.org). Given the high expression of Metrnl in the skin, we investigated the role of Metrnl in wound healing. First, we examined the temporal and spatial expression profiles of Metrnl mRNA and protein during full-thickness splinted wound healing in C57BL/6 mice. qPCR showed that Metrnl mRNA levels in skin tissues of C57BL/6 mice were significantly increased during wound healing; however, increased Metrnl expression was inhibited in the skin tissues of STZ, HFD/STZ, and db/db diabetic mice models after injury (Fig.1A). Immunohistochemical (IHC) staining and WB showed that the Metrnl protein levels were significantly lower in skin biopsies of diabetic mice (HFD/STZ and db/db mice) compared with C57BL/6 mice (Fig.1B). Likewise, decreased Metrnl expression level was demonstrated in diabetic foot ulcers (DFU) compared with that in normal human skin tissues (Control) (Fig.1C).

Figure 1

Characterization of Metrnl expression during skin wound healing. (A) qPCR analysis of Metrnl mRNA levels in skin tissues from C57BL/6, STZ, HFD/STZ, and db/db mice on days 0, 1, 4, 7, and 11 after injury (n = 3 to 4 mice per group). **P < 0.01, ***P < 0.001, ****P < 0.0001, compared with 0 days in C57BL/6 mice; &&P < 0.01, compared with 0 days in db/db mice; #P < 0.05, compared with 0 days in HFD/STZ mice. (B) Representative IHC and WB images of Metrnl expression in skin biopsies of C57BL/6, HFD/STZ, and db/db mice. Scale bar: 100 μm. (C) Representative IHC images of Metrnl expression in human skin tissues from healthy (Control) (n = 7) and DFU (n = 11) individuals. Scale bar: 200 μm (top) and 100 μm (bottom). (D) Immunofluorescence staining of Metrnl and CK14 in skin biopsies of normal individuals and C57BL/6 mice. Scale bar: 20 μm. (E) WB of Metrnl expression in HaCat cells, which were starved overnight in medium containing 5 mmol/L glucose, then exposed to high concentration of glucose (HG15 and 30 mmol/L) for 48 and 72 h. *P < 0.05, **P < 0.01, compared with NG. HG, high glucose; NG, normal glucose. Data are mean ± SD (n = 3 independent experiments).

Figure 1

Characterization of Metrnl expression during skin wound healing. (A) qPCR analysis of Metrnl mRNA levels in skin tissues from C57BL/6, STZ, HFD/STZ, and db/db mice on days 0, 1, 4, 7, and 11 after injury (n = 3 to 4 mice per group). **P < 0.01, ***P < 0.001, ****P < 0.0001, compared with 0 days in C57BL/6 mice; &&P < 0.01, compared with 0 days in db/db mice; #P < 0.05, compared with 0 days in HFD/STZ mice. (B) Representative IHC and WB images of Metrnl expression in skin biopsies of C57BL/6, HFD/STZ, and db/db mice. Scale bar: 100 μm. (C) Representative IHC images of Metrnl expression in human skin tissues from healthy (Control) (n = 7) and DFU (n = 11) individuals. Scale bar: 200 μm (top) and 100 μm (bottom). (D) Immunofluorescence staining of Metrnl and CK14 in skin biopsies of normal individuals and C57BL/6 mice. Scale bar: 20 μm. (E) WB of Metrnl expression in HaCat cells, which were starved overnight in medium containing 5 mmol/L glucose, then exposed to high concentration of glucose (HG15 and 30 mmol/L) for 48 and 72 h. *P < 0.05, **P < 0.01, compared with NG. HG, high glucose; NG, normal glucose. Data are mean ± SD (n = 3 independent experiments).

Close modal

Given that Metrnl is a secreted protein, we also detected the plasma Metrnl levels in C57BL/6 mice and db/db mice with or without wounding. It was found that the plasma Metrnl levels in db/db mice were lower than that in C57BL/6 mice, but there were no significant changes in the plasma Metrnl levels between wounded and nonwounded C57BL/6 and db/db mice (Supplementary Fig. 1A). At the same time, we collected human serum samples and found that the plasma Metrnl levels in type 2 diabetes mellitus (DM) patients showed a decreasing trend compared with the healthy control, although there were no statistical differences, and there were no significant changes among DM patients with DFU (Supplementary Fig. 1B). This may also be due to the small sample size in human, individual variance, different application of hypoglycemic agents, and multiple factors that affecting blood Metrnl levels (such as temperature, exercise, etc.).

Furthermore, Metrnl was mainly expressed in keratinocytes in the epidermis of skin tissues of normal individuals and C56BL/6 mice, as demonstrated by immunofluorescence for colocalization of Metrnl and CK14, a specific marker for keratinocytes (Fig.1D). Having observed differential expression of Metrnl in the process of wound healing in C57BL/6 and diabetic mice and mainly located in keratinocytes, we further tested the expression of Metrnl under diabetic conditions in vitro. The Metrnl protein levels were significantly decreased in high glucose–treated keratinocytes (HaCat cells) (Fig.1E). These results suggest that Metrnl may play a role in the process of wound healing, particularly in diabetic wound healing.

Metrnl Regulates Various Physiological Functions in Keratinocytes

Since Metrnl is mainly localized in keratinocytes and may be related to wound healing, the function of Metrnl in keratinocytes was further studied. To test the potential relevance of Metrnl-induced proliferation and migration, adenovirus-mediated gene overexpression and knockdown were performed in HaCat cells, the effects of Metrnl overexpression (Ad-Met) and knockdown (sh-Met) induced by adenovirus were confirmed by WB, and the levels of Metrnl secreted by HaCat cells were detected in cultured medium by ELISA (Fig.2A and F). The results showed that CCK-8 and colony formation assays showed significantly enhanced cell proliferation in HaCat cells after Ad-Met infection (Fig.2B and C). Similarly, Metrnl overexpression further resulted in an increased number of EdU+ HaCat cells (Fig.2C). In wound healing experiments, Metrnl overexpression resulted in significantly higher wound closure rates at 24 and 48 h postinjury (Fig.2D). To evaluate the effect of Metrnl on the migration of HaCat cells, we performed a transwell assay (Fig.2E). The results showed that Metrnl overexpression increased HaCat cell migration. In contrast, Metrnl deficiency significantly inhibited the physiological functions of HaCaT cells, including proliferation and migration (Fig.2G–J).

Figure 2

Metrnl regulates various physiologic functions in keratinocytes. (A) Metrnl expression in HaCat cell and its CM was detected using WB and ELISA, respectively, after Metrnl adenovirus (Ad-Vector or Ad-Metrnl) overexpression for 48 h. (B and C) The growth or the viability of cells was measured by CCK-8 assay (B) and clonogenicity assay and EdU staining (C) in HaCat cells. (D) Representative images and quantification of HaCat cells scratch wound healing after transfection with adenovirus. Scale bar: 1 mm. (E) Transwell assay for HaCat cells transfected with adenovirus for 48 h. (F) Metrnl expression in HaCat cell and its CM was detected using WB and ELISA, respectively, after Metrnl knockdown adenovirus (sh-Vector or sh-Metrnl) for 72 h. (G and H) The growth or viability of cells was measured by CCK-8 assay (G) and clonogenicity assay and EdU staining (H) in HaCat cells after infection with sh-Vector or sh-Metrnl adenovirus. (I and J) The migration was measured by scratch wound healing (I) and transwell assay (J) in HaCat cells for 48 h after infection with sh-Vector or sh-Metrnl adenovirus. (K and L) The growth or viability of cells was measured by CCK-8 assay (K) and clonogenicity assay and EdU staining (L) in HaCat cells after treatment with control (Ctrl) or rMet (1 μg/mL). (M and N) The migration of cells was measured by scratch wound healing (M) and transwell assay (N) in HaCat cells after treatment with Ctrl or rMet for 48 h. rMet, recombinant protein Metrnl. Data are mean ± SD (n = 3 independent experiments). Compared with respective controls, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 2

Metrnl regulates various physiologic functions in keratinocytes. (A) Metrnl expression in HaCat cell and its CM was detected using WB and ELISA, respectively, after Metrnl adenovirus (Ad-Vector or Ad-Metrnl) overexpression for 48 h. (B and C) The growth or the viability of cells was measured by CCK-8 assay (B) and clonogenicity assay and EdU staining (C) in HaCat cells. (D) Representative images and quantification of HaCat cells scratch wound healing after transfection with adenovirus. Scale bar: 1 mm. (E) Transwell assay for HaCat cells transfected with adenovirus for 48 h. (F) Metrnl expression in HaCat cell and its CM was detected using WB and ELISA, respectively, after Metrnl knockdown adenovirus (sh-Vector or sh-Metrnl) for 72 h. (G and H) The growth or viability of cells was measured by CCK-8 assay (G) and clonogenicity assay and EdU staining (H) in HaCat cells after infection with sh-Vector or sh-Metrnl adenovirus. (I and J) The migration was measured by scratch wound healing (I) and transwell assay (J) in HaCat cells for 48 h after infection with sh-Vector or sh-Metrnl adenovirus. (K and L) The growth or viability of cells was measured by CCK-8 assay (K) and clonogenicity assay and EdU staining (L) in HaCat cells after treatment with control (Ctrl) or rMet (1 μg/mL). (M and N) The migration of cells was measured by scratch wound healing (M) and transwell assay (N) in HaCat cells after treatment with Ctrl or rMet for 48 h. rMet, recombinant protein Metrnl. Data are mean ± SD (n = 3 independent experiments). Compared with respective controls, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Close modal

Similar results such as increased proliferation and migration, were also obtained in rMet-treated HaCat cells (Fig.2K–N) and human primary keratinocytes (Supplementary Fig. 2A and B). These results indicate that Metrnl positively regulates the proliferation and migration of keratinocytes in vitro.

Metrnl Promotes Wound Healing in C57BL/6 Mice

Based on the in vitro results, we examined the effect of Metrnl on wound healing in C57BL/6 mice using a Ctrl (Pluronic F-127) and 2.0 mg/mL rMet (mixed in Pluronic F-127) application. The treatments were topically applied daily for 11 days. Wound healing rates were recorded on days 0, 1, 4, 7, and 11 after injury (Fig.3A). The results showed that, on the 4th, 7th, and 11th days, the wound area of the rMet treatment group was significantly smaller than that of the control group, as shown in Fig.3B.

Figure 3

Metrnl promotes wound healing in C57BL/6 mice. (A) Timeline for in vivo experiments (n = 24 mice per group). (B) Representative images and quantitative analysis of the wounds treated with rMet or control (Ctrl) for 11 days. Wound size at each time point was normalized to the 0th day. (C) H&E staining of dorsal skin section in mice 4, 7, and 11 days postinjury. Independent images with the same magnifications were used to show the unhealed length and the granulation area of the wounds, and respective statistic data were provided. Red dashed lines represent unhealed wound space. Orange dashed lines mark the newly formed epidermis. (D) Masson staining of dorsal skin section in mice treated 11 days postinjury with or without rMet hydrogel. Scale bar: 100 and 200 μmol/L. (E) Immunofluorescence staining of Ki67 (red) in the wounded region at the 7th day. Scale bars: 50 μmol/L. Data are mean ± SD; n = 5 to 6 mice per group at different points in time. ReT, reepithelialized tongue. Compared with Ctrl, *P < 0.05, **P < 0.01.

Figure 3

Metrnl promotes wound healing in C57BL/6 mice. (A) Timeline for in vivo experiments (n = 24 mice per group). (B) Representative images and quantitative analysis of the wounds treated with rMet or control (Ctrl) for 11 days. Wound size at each time point was normalized to the 0th day. (C) H&E staining of dorsal skin section in mice 4, 7, and 11 days postinjury. Independent images with the same magnifications were used to show the unhealed length and the granulation area of the wounds, and respective statistic data were provided. Red dashed lines represent unhealed wound space. Orange dashed lines mark the newly formed epidermis. (D) Masson staining of dorsal skin section in mice treated 11 days postinjury with or without rMet hydrogel. Scale bar: 100 and 200 μmol/L. (E) Immunofluorescence staining of Ki67 (red) in the wounded region at the 7th day. Scale bars: 50 μmol/L. Data are mean ± SD; n = 5 to 6 mice per group at different points in time. ReT, reepithelialized tongue. Compared with Ctrl, *P < 0.05, **P < 0.01.

Close modal

To further confirm the wound re-epithelization and granulation tissue formation, we observed the wounds in each group on days 4, 7, and 11 after wounding. As shown in Fig.3C, the reepithelialized tongue can be visualized in the images on both ends, and the unhealed wound length was shorter and the granulation tissue area was greater in the rMet-treated group compared with the control group at 7 days (Fig.3C), which showed that Metrnl accelerates the process of wound healing in C57BL/6 mice. The rMet-treated mice developed epithelial spines on the wounds 4 days after wounding, and inflammatory cell infiltration was significantly reduced compared with that in the control group (Fig.3C). At 7 days after injury, the wound epidermis in the rMet treatment group significantly grew, and the wound was covered with a continuous epidermal layer. Moreover, the mice treated with rMet showed excessive collagen deposition in the wound area (Fig.3D). On the 11th day, the epidermis and dermis of the mice in the rMet group were basically formed, and the skin tags were clear and complete compared with those in the control group (Fig.3C). One of the key processes in wound healing is the formation of granulation tissue, which indicates the proliferation and angiogenesis of the cell types involved in wound healing (endothelial cells and keratinocytes). Notably, rMet treatment significantly increased granulation tissue proliferation and increased immunostaining for Ki67, a ubiquitous marker of cell proliferation expressed during the active cell cycle (Fig.3E). Taken together, these results demonstrate that Metrnl can effectively accelerate the rate of wound healing in normal mice during skin repair.

Metrnl Treatment Reverses Impaired Wound Healing in Diabetic Mice

Prolonged wound healing time in patients with diabetes is an insurmountable clinical problem. We therefore investigated whether Metrnl also effectively promotes wound healing in diabetic mice. Diabetic mouse models (diabetic db/db, HFD/STZ, and STZ mice) with FBG greater than 13.8 mmol/L were considered diabetic mice for the wound healing experiments (Fig. 4A, Supplementary Fig. 3B, and Supplementary Fig. 4B).

Figure 4

Metrnl treatment accelerates wound healing in diabetic db/db mice. (A) FBG of db/db mice before wound creation. (B) Representative images and wound size change of the wounds treated with rMet hydrogel for 13 days in db/db mice. (C and D) H&E staining and respective statistic data of unhealed length and granulation area are presented (C), and Masson staining of dorsal skin section in db/db mice treated with or without rMet hydrogel, 7 and 11 days postinjury (D). Scale bar: 100 and 200 μmol/L. Red dashed lines present unhealed wound space. Orange dashed lines mark the newly formed epidermis. (E) Immunofluorescence staining of Ki67 (red) in the wounded region at day 7 in db/db mice treated with or without rMet hydrogel. Scale bars: 50 μm. Data are mean ± SD; n = 5 to 6 mice per group at different points in time. Compared with control (Ctrl), *P < 0.05, **P < 0.01, ****P < 0.0001.

Figure 4

Metrnl treatment accelerates wound healing in diabetic db/db mice. (A) FBG of db/db mice before wound creation. (B) Representative images and wound size change of the wounds treated with rMet hydrogel for 13 days in db/db mice. (C and D) H&E staining and respective statistic data of unhealed length and granulation area are presented (C), and Masson staining of dorsal skin section in db/db mice treated with or without rMet hydrogel, 7 and 11 days postinjury (D). Scale bar: 100 and 200 μmol/L. Red dashed lines present unhealed wound space. Orange dashed lines mark the newly formed epidermis. (E) Immunofluorescence staining of Ki67 (red) in the wounded region at day 7 in db/db mice treated with or without rMet hydrogel. Scale bars: 50 μm. Data are mean ± SD; n = 5 to 6 mice per group at different points in time. Compared with control (Ctrl), *P < 0.05, **P < 0.01, ****P < 0.0001.

Close modal

We also performed macroscopic assessments of dorsal dermal defects and full-thickness wounds. Wound closure in the rMet-treated db/db mice was approximately 30% at 7 days, 58% at 11 days, and 80% at 13 days (compared with the Ctrl-treated db/db mice with 18%, 33%, and 62%, respectively) (Fig.4B). H&E staining showed a shorter unhealing epidermis and more abundant granulation tissues in the wound of db/db mice after rMet treatment at 7 days (Fig.4C). By day 7 postwounding, the wounds in the rMet-treated group were nearly covered with new epidermis. By the 11th day, the regenerated epidermis of the rMet-treated wounds had completely recovered, and the dermal arrangement was close to that of normal skin; however, the control group showed a disordered arrangement of fibroblasts and a small number of inflammatory cells in the dermis (Fig.4C). Masson staining showed that collagen deposition was more extensive in the rMet-treated group (Fig.4D). Furthermore, rMet treatment significantly promoted granulation tissue proliferation, and Ki67 expression was increased in the db/db diabetic mice (Fig.4E). rMet also had a similar effect on HFD/STZ-induced wound healing in diabetic mice (Supplementary Fig. 3AD), suggesting that rMet promotes skin wound repair and regeneration in type 2 diabetic mouse models.

Additionally, to further confirm our results, we assessed the effect of rMet on wound healing in the STZ-induced type 1 diabetic mice (Supplementary Fig. 4A). From day 7 postinjury, the wound area of the rMet-treated group was significantly smaller than that of the control group (Supplementary Fig. 4C). H&E staining results showed that the epidermis of newborns in the rMet treatment group was significantly thickened on day 7, a large number of cells appeared in the superficial dermis, and granulation tissue was well formed (Supplementary Fig. 4D). Masson staining revealed increased extracellular collagen matrix proteins (Supplementary Fig. 4E). In particular, Ki67 staining revealed substantial cell proliferation, including granulation tissue neovascularization, in the rMet-treated group (Supplementary Fig. 4F). Taken together, these observations suggest that Metrnl accelerates the speed of wound healing in diabetic mice (db/db, HFD/STZ, and STZ diabetic mice). These results suggest that rMet may be a potentially biological activity in the treatment of diabetic wound healing.

Safety and toxicity are both critical in pharmaceutical research. Therefore, we further assessed the safety and toxicity of rMet in vitro and in vivo. Firstly, we detected the cell viability of HaCat and HUVECs cells using CCK-8 assay at the different doses of rMet, ranging from 0 to 5,000 ng/mL, which were applied in other research (14,23). We did not find significant changes in cell viability after rMet treatment even at high rMet concentrations (Supplementary Fig. 5A and B). To test the safety and toxicity of rMet in vivo, histological detection and blood biochemistry analysis were used to evaluate the kidney and liver injury in C57BL/6 and db/db mice at the end of the experiments. The results showed that no obvious changes were found in H&E staining of liver and kidney tissues (Supplementary Fig. 5C), serum alanine transaminase, and creatinine between rMet group and the respective control group (Supplementary Fig. 5D and E). These findings suggest that rMet hydrogel is a relatively nontoxic remedy for wound healing through local application.

Moreover, impaired glucose tolerance and insulin sensitivity are both important factors in delayed wound healing (24). Therefore, the plasma Metrnl content was determined in the STZ, STZ+HFD, and db/db mouse models; however, the local application of Metrnl hydrogel did not change the plasma Metrnl content (Supplementary Fig. 6A). We further detected the glucose tolerance and insulin sensitivity by intraperitoneal glucose tolerance test and insulin tolerance test experiments in STZ, STZ+HFD, and db/db mice, and found that the glucose tolerance and insulin sensitivity in respective mouse models were not altered by local Metrnl hydrogel application (Supplementary Fig. 6BG). These results indicate that the accelerating effect of Metrnl in wound healing is not dependent on glucose tolerance and increased insulin sensitivity.

Metrnl Promotes Angiogenesis In Vitro and In Vivo

Vascular endothelial cells play an important role in angiogenesis, and insufficient local angiogenesis is considered an important cause of poor chronic wound healing (25). To determine whether Metrnl enhances wound angiogenesis, we detected the mRNA levels of CD31 and VEGF (biomarkers of angiogenesis) in the wound tissues of db/db mice, and found CD31 and VEGF mRNA expression were upregulated by rMet treatment, especially at 7 and 11 days (Fig.5A). Similar results were also confirmed by IHC assay (anti-VEGF) (Fig.5B) and by immunofluorescence staining of CD31 in wound biopsies on days 7 and 11 (Fig.5C).

Figure 5

Metrnl promotes angiogenesis in vitro and in vivo. (A) mRNA expression levels of CD31 and VEGF (the biomarkers of angiogenesis) were detected using qPCR in the wound tissues of db/db mice. (B) VEGF protein level was measured by IHC in the wound tissues of db/db mice, n = 6. (C) Immunofluorescence staining of CD31 (green) in the wound region at days 7 and 11 in db/db mice after rMet hydrogel treatment (n = 6). Scale bars: 20 μm. (D) The growth or the viability of cells was measured by clonogenicity assay, CCK-8, and Edu assay in HUVECs cells after rMet treatment. (E) Representative images and quantification of scratch wound healing assay in HUVECs cells after rMet treatment at 24 and 48 h. (F) Transwell assay in HUVECs cells after rMet treatment for 24 h. (G) Formation of capillary tube structures and tube length of HUVECs following treatment with rMet for 8 h. Data are mean ± SD (n = 3 independent experiments). Compared with control (Ctrl), *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 5

Metrnl promotes angiogenesis in vitro and in vivo. (A) mRNA expression levels of CD31 and VEGF (the biomarkers of angiogenesis) were detected using qPCR in the wound tissues of db/db mice. (B) VEGF protein level was measured by IHC in the wound tissues of db/db mice, n = 6. (C) Immunofluorescence staining of CD31 (green) in the wound region at days 7 and 11 in db/db mice after rMet hydrogel treatment (n = 6). Scale bars: 20 μm. (D) The growth or the viability of cells was measured by clonogenicity assay, CCK-8, and Edu assay in HUVECs cells after rMet treatment. (E) Representative images and quantification of scratch wound healing assay in HUVECs cells after rMet treatment at 24 and 48 h. (F) Transwell assay in HUVECs cells after rMet treatment for 24 h. (G) Formation of capillary tube structures and tube length of HUVECs following treatment with rMet for 8 h. Data are mean ± SD (n = 3 independent experiments). Compared with control (Ctrl), *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Close modal

Angiogenesis is highly dependent on the proliferation, migration, and invasion of endothelial cells. Therefore, we examined the activity of Metrnl in HUVECs. Monoclonal formation, CCK-8 detection, and Edu assay demonstrated that Metrnl plays a key role in the proliferation of HUVECs (Fig.5D). Next, we examined the effect of Metrnl on the horizontal and vertical migration of HUVECs in vitro. In wound healing and transwell migration experiments, Metrnl significantly reduced wound width and increased horizontal motility of HUVECs compared with the controls (Fig.5E and F). This finding indicated that Metrnl significantly promoted the migration of HUVECs. Furthermore, to test the effect of Metrnl on angiogenesis, we assessed capillary network formation in HUVECs, using tube formation assays. The results showed that Metrnl significantly promoted the tube formation of HUVECs, and the capillary plexus in the Metrnl-treated group was more complex and stable than that in the control group (Fig.5G). These effects were also found in human primary vascular endothelial cells (Supplementary Fig. 2CE). Thus, based on these observations, Metrnl promotes angiogenesis in vitro and in vivo.

Keratinocyte- and M2 Macrophage-Derived Metrnl Promotes Angiogenesis

Keratinocytes produce various cytokines, promote angiogenesis, and play an important role in skin wound healing. Since Metrnl is a secreted protein mainly expressed in keratinocytes, we investigated whether keratinocyte-derived Metrnl affects angiogenesis. We found increased proliferation of HUVECs cultured in supernatant media from the Metrnl-overexpressing HaCat cells, as shown by clone formation and CCK-8 assays (Fig.6A). In wound healing experiments, we found that HUVECs migratory activity was significantly increased after treatment with the Metrnl-overexpressing HaCat cell conditional medium (CM), as shown by transwell assays (Fig.6B). Furthermore, in the tube formation assay, Metrnl significantly promoted tube formation in the HUVECs treated with CM from the Metrnl-overexpressing HaCat cells (Fig.6C). Thus, keratinocyte-derived Metrnl can enhance endothelial cell proliferative activity, migration, and tube formation.

Figure 6

Keratinocyte- and M2 macrophage–derived Metrnl promotes angiogenesis. (AC) The growth or the viability of cells, migration, and tube formation were measured by clonogenicity assay and CCK-8 assay, transwell assay, and tube formation assay in HUVECs cells cultured with CM from Ad-Vector– or Ad-Metrnl–treated HaCaT. (D) The mRNA expression level of Cd206 and Arg1, the makers of M2 macrophages, in Raw264.7 cells treated with IL4 (20 ng/mL for 24 h). (E) The expression level of Cd206, in the skin tissues of db/db mice treated with rMet hydrogel or without, during wound healing at days 1, 4, and 7 postinjury was detected by qPCR. (F) The expression level of CD206 in the skin tissues of db/db mice at day 3 was detected by immunofluorescence. (G) Metrnl content in culture supernatant of M2 Raw264.7 cells (induced by IL4) detected by ELISA after adenovirus infection (Ad-Vector or Ad-Metrnl) for 48 h. (HJ) The growth or the viability of cells, migration, and tube formation were measured by clonogenicity assay and CCK-8 assay, transwell assay, and tube formation assay in HUVECs cells cultured with CM from Ad-Vector or Ad-Metrnl treated-M2 Raw264.7 cells. (KM) CCK-8, transwell, and tube formation assays were conducted after incubating CM from HaCat cell challenged with Ad-Met and treated with or without Metrnl neutralizing antibody. (NP) CCK-8, transwell, and tube formation assays were conducted after incubating CM from Raw264.7 cell challenged with Ad-Met and treated with or without Metrnl neutralizing antibody. Data are mean ± SD (n = 3 independent experiments). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, compared with respective control (Ctrl) group.

Figure 6

Keratinocyte- and M2 macrophage–derived Metrnl promotes angiogenesis. (AC) The growth or the viability of cells, migration, and tube formation were measured by clonogenicity assay and CCK-8 assay, transwell assay, and tube formation assay in HUVECs cells cultured with CM from Ad-Vector– or Ad-Metrnl–treated HaCaT. (D) The mRNA expression level of Cd206 and Arg1, the makers of M2 macrophages, in Raw264.7 cells treated with IL4 (20 ng/mL for 24 h). (E) The expression level of Cd206, in the skin tissues of db/db mice treated with rMet hydrogel or without, during wound healing at days 1, 4, and 7 postinjury was detected by qPCR. (F) The expression level of CD206 in the skin tissues of db/db mice at day 3 was detected by immunofluorescence. (G) Metrnl content in culture supernatant of M2 Raw264.7 cells (induced by IL4) detected by ELISA after adenovirus infection (Ad-Vector or Ad-Metrnl) for 48 h. (HJ) The growth or the viability of cells, migration, and tube formation were measured by clonogenicity assay and CCK-8 assay, transwell assay, and tube formation assay in HUVECs cells cultured with CM from Ad-Vector or Ad-Metrnl treated-M2 Raw264.7 cells. (KM) CCK-8, transwell, and tube formation assays were conducted after incubating CM from HaCat cell challenged with Ad-Met and treated with or without Metrnl neutralizing antibody. (NP) CCK-8, transwell, and tube formation assays were conducted after incubating CM from Raw264.7 cell challenged with Ad-Met and treated with or without Metrnl neutralizing antibody. Data are mean ± SD (n = 3 independent experiments). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, compared with respective control (Ctrl) group.

Close modal

M2 polarization of macrophages around diabetic wounds favors angiogenesis (26,27). Previous studies have shown that Metrnl induces macrophages toward M2 polarization (20). In M2 macrophages induced by IL4, the mRNA expression levels of CD206 and Arg1 were obviously upregulated by rMet treatment (Fig.6D). Similar results showed that the mRNA and protein expression of CD206 (an M2 macrophage marker) was significantly increased in wound biopsies of the rMet-treated db/db mice (Fig.6E and F). In addition, we further found that M2 macrophages secreted Metrnl, as detected by ELISA, and the level of Metrnl in the culture supernatant of the M2 macrophages overexpressing Metrnl was significantly increased (Fig.6G). Similarly, culture supernatants from the M2 macrophages overexpressing Metrnl promoted the proliferation, migration, and tube formation of HUVECs (Fig.6H–J). Further, to evaluate whether Metrnl secreted from HaCat and M2 Raw264.7 cells promotes angiogenesis, Metrnl neutralizing antibody was used to block secreted Metrnl from cells overexpressing Metrnl. The results showed that the increased proliferation, migration, and tube formation of HUVECs induced by the CM from HaCat cells overexpressing Metrnl were significantly blocked by specific Metrnl neutralizing antibody (Fig.6K–M). Moreover, the increased proliferation and tube formation of HUVECs cultured with the CM from M2 Raw264.7 cells overexpressing Metrnl were inhibited by specific Metrnl neutralizing antibody (Fig.6N–P).

These results suggest that both keratinocytes and M2 macrophages can release Metrnl to promote angiogenesis, thereby accelerating wound healing in diabetic-stressed skin by increasing endothelial cell proliferation, migration, and tubule formation.

Metrnl Promotes Keratinocyte and Endothelial Cell Activity in a Manner Dependent on the c-Kit/Akt Manner

To further elucidate the mechanism by which Metrnl accelerates wound healing by promoting the proliferation, migration, and angiogenesis of keratinocytes and endothelial cells, we focused on c-Kit, which has been described as a receptor for Metrnl (23). Notably, the PI3K/Akt pathway, a main downstream pathway of c-Kit, is actively involved in angiogenesis and cell proliferation (28,29). As shown in Fig.7A, increased c-Kit and Akt phosphorylation was observed 15 min after rMet stimulation. Importantly, the KIT inhibitor abrogated rMet treatment-induced Akt activation in HUVECs and HaCat cells for 24 h (Fig.7A), suggesting that Metrnl activates Akt signaling through c-Kit activation.

Figure 7

Metrnl promotes keratinocyte and endotheliocyte functions during wound healing dependent on c-Kit/Akt manner. (A) Representative Western blot of the protein levels of p-c-Kit, c-Kit, AKT, and p-AKTSer473 in HaCaT and HUVECs cells treated with or without rMet and PLX-3397, a c-Kit inhibitor, for 24 h (n = 3 blots). (B) The growth or the viability of cells was measured by CCK-8 assay in HaCaT and HUVECs cells cultured with or without rMet and PLX-3397 for 24, 48, and 72 h. (C) Migration capacity of HaCaT for 48 h and HUVECs cells for 24 h was detected by transwell assay with or without rMet and PLX-3397. (D) Formation of capillary tube structures of HUVECs treated with or without rMet and PLX-3397 for 8 h. Data are mean ± SD (n = 3 independent experiments). Compared with control (Ctrl), *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. ns, no significance.

Figure 7

Metrnl promotes keratinocyte and endotheliocyte functions during wound healing dependent on c-Kit/Akt manner. (A) Representative Western blot of the protein levels of p-c-Kit, c-Kit, AKT, and p-AKTSer473 in HaCaT and HUVECs cells treated with or without rMet and PLX-3397, a c-Kit inhibitor, for 24 h (n = 3 blots). (B) The growth or the viability of cells was measured by CCK-8 assay in HaCaT and HUVECs cells cultured with or without rMet and PLX-3397 for 24, 48, and 72 h. (C) Migration capacity of HaCaT for 48 h and HUVECs cells for 24 h was detected by transwell assay with or without rMet and PLX-3397. (D) Formation of capillary tube structures of HUVECs treated with or without rMet and PLX-3397 for 8 h. Data are mean ± SD (n = 3 independent experiments). Compared with control (Ctrl), *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. ns, no significance.

Close modal

Next, we investigated the effect of rMet on keratinocytes and endothelial cells by c-Kit. Inhibition of c-Kit expression with inhibitors abrogated rMet-mediated proliferation, as demonstrated by CCK-8 assay of HUVECs and HaCat cells (Fig.7B). Similarly, inhibition of c-Kit expression significantly abrogated the rMet-mediated increase in horizontal migration of HUVECs and HaCat cells (Fig.7C) and interfered with the rMet-mediated promotion of HUVECs tube formation in the tube formation assay (Fig.7D), indicating that c-Kit inhibitors counteract the protective effect of Metrnl on keratinocytes and endothelial cells.

These results suggest that Metrnl improves chronic wound healing in patients with diabetes who are under metabolic stress, by activating the c-Kit–mediated Akt signaling pathway to promote keratinocyte and endothelial cell proliferation, migration, and angiogenesis.

Complications of wound healing can lead to foot ulcers and even amputation and are a leading cause of death in DM (30,31). The process of wound healing can be divided into four steps: hemostasis, inflammation, granulation, and remodeling. Excessive inflammation and vasculopathy induced by hyperglycemia around diabetic wounds prolong the inflammatory phase and delay the wound healing process (32,33). Delayed wound healing can significantly increase the risk of wound infection, further disrupting periwound angiogenesis and aggravating the healing process, causing a vicious cycle (34). Therefore, anti-inflammatory and proliferative therapeutic strategies may be important and promising approaches for diabetic skin wound healing.

In this study, we assessed Metrnl as a novel potential cytokine for wound healing. Hydrogels containing recombinant Metrnl protein promote wound closure in C57BL/6 and diabetic mice (db/db, HFD/STZ, and STZ mice). Secretory Metrnl expression significantly increased in the skin tissues during physiological wound healing, while Metrnl expression remained low throughout diabetic wound healing. Local application of rMet modulates multiple regulatory networks involved in cell migration, angiogenesis, and inflammation. Furthermore, Metrnl mainly expressed in keratinocytes and was significantly reduced at high glucose levels, suggesting that decreased Metrnl expression is pathologically associated with diabetic wound healing. The mechanism of Metrnl in skin wound healing is summarized in Fig. 8.

Figure 8

A graphic model depicting the roles of Metrnl in promoting wound healing. Keratinocyte-derived Metrnl promotes the proliferation, migration, and tube formation of keratinocytes and/or endotheliocytes, and, meanwhile, increases M2 phenotypic macrophages in dermis. M2 macrophages secretes Metrnl to accelerate angiogenesis. These effects of reepithelialization and angiogenesis are dependent on c-Kit/Akt axis. Hydrogel containing Metrnl fuels the processes mentioned above.

Figure 8

A graphic model depicting the roles of Metrnl in promoting wound healing. Keratinocyte-derived Metrnl promotes the proliferation, migration, and tube formation of keratinocytes and/or endotheliocytes, and, meanwhile, increases M2 phenotypic macrophages in dermis. M2 macrophages secretes Metrnl to accelerate angiogenesis. These effects of reepithelialization and angiogenesis are dependent on c-Kit/Akt axis. Hydrogel containing Metrnl fuels the processes mentioned above.

Close modal

Patients with diabetes, particularly those with type 2 diabetes, often experience hyperglycemia and hyperlipidemia. We found that Metrnl content in the serum of patients with diabetes and HaCat cells treated with high glucose was reduced. Moreover, reduced Metrnl expression levels were also found in high glucose– and fatty acid–treated renal tubular epithelial cells (35). Thus, the high glucose– and high fat–mediated lower Metrnl expression in local skin tissues may be the critical factor underlying impaired wound healing in diabetes. Metrnl expression was significantly increased during normal wound healing, but not in diabetic mice. Therefore, our therapeutic approach aims to restore physiological levels of Metrnl in diabetic wounds using exogenous Metrnl recombinant protein. Indeed, topical application of rMet restored wound reepithelialization, angiogenesis, and anti-inflammatory effects. Angiogenesis in diabetic wounds is critical for skin regeneration (36). We found that Metrnl can increase the density and tube formation of endothelial cells in vitro and significantly promote wound microvessel density in vivo. In addition, the proliferation and migration of keratinocytes is critical for epithelialization, and we found that Metrnl promotes these processes. Thus, Metrnl can not only promote reepithelialization but also enhance angiogenesis by effectively stimulating the migration and proliferation of keratinocytes and endothelial cells. Inflammation plays both positive and negative roles in skin repair (37,38); the level and duration of inflammation can determine healing time and the quality of wound healing. Inhibiting the inflammatory response has been shown to accelerate wound healing and reduce scarring in chronic wounds (32,39). Increased expression of phenotypic M2/anti-inflammatory macrophage markers (CD206 and Arg1) was observed after treatment with Metrnl, suggesting alternate activation/differentiation of M2 macrophages in wound-related conditions. One possible mechanism has been suggested to reduce inflammation: Metrnl activates Stat3 in macrophages, causing them to develop an anti-inflammatory phenotype (20,40).

Mechanistically, during the process of physiological wound healing, keratinocyte-secreted Metrnl is increased, thus promoting angiogenesis and macrophage polarization to the M2 phenotype and increasing endothelial cell proliferation and migration. Therefore, Metrnl helps to shorten the long-term inflammatory period of chronic wounds, and facilitates diabetic wound healing. M2 macrophages can also promote angiogenesis by releasing proangiogenic mediators such as VEGF, FGF, and EGF (41). In addition, activated M2 macrophages can secrete Metrnl to promote angiogenesis. Both Metrnl overexpression and exogenous Metrnl treatment were found to promote keratinocyte reepithelialization and endothelial cell angiogenesis, while activating AKT signaling through increased c-Kit receptor-mediated proliferation and migration, consistent with the finding that Metrnl-mediated c-Kit is a high-affinity ligand for c-Kit in cardiac repair (27,42). However, further research is still needed to identify the precise acting subunits of PI3K that are regulated by rMet. We found that Metrnl mainly expressed in the epidermis of skin, and other research has reported that M2 macrophage is an important source of Metrnl, so keratinocyte and M2 macrophage are considered to be the two important sources of Metrnl to promote angiogenesis in the process of wound healing in skin tissue. Nevertheless, other types of cells, such as neutrophils (14) and skin fibroblasts, are suggested as the potential source of Metrnl in wound healing, which also should be investigated in future studies. In addition, it also should be noticed that the db/db, HFD/STZ, and STZ models used in this study have their own limitations. The duration of the study was short, lasting less than 14 days. Both treated and untreated wounds showed a trend toward wound closure, but at different speeds, which does not accurately replicate the complexity of diabetic wounds, as diabetic wounds typically do not close in clinical cases. Based on animal models, our current data demonstrate that Metrnl or Metrnl hydrogel can enhance the healing process of normal and diabetic skin. This finding may have potential applications in minor or nonsevere cases of diabetic wounds. However, it is important to note that the effectiveness of Metrnl treatment for infected DFU with exposed bone and osteomyelitis, which are severe cases of DFU, remains uncertain. Further research is necessary to address this issue. Moreover, the clinical application of Metrnl or Metrnl hydrogel may be limited because of various challenges such as regulatory requirements, manufacturing complexities, and cost considerations associated with biologic therapies.

Taken together, our results reveal a novel biological role of the secreted Metrnl protein in the process of skin wound healing. Specifically, we discovered that the cytokine Metrnl, produced by keratinocytes and macrophages, plays a crucial role in promoting posttraumatic angiogenesis and re-epithelialization. This effect is mediated through a c-Kit–dependent signaling pathway. Additionally, our results provide further evidence that Metrnl is a versatile cytokine that not only promotes angiogenesis but also exerts anti-inflammatory effects. These findings suggest that Metrnl influences the interactions between keratinocytes, endothelial cells, and macrophages, ultimately enhancing the repair of skin injuries, including diabetes-associated ulcers.

Acknowledgments. The authors express our gratitude to the patients who generously donated their skin tissue for the purpose of this study. The authors would also like to acknowledge the Department of Endocrinology and Metabolism at the Affiliated Hospital of Southwest Medical University for providing the cutaneous slice samples.

Funding. This study was supported by National Natural Science Foundation of China (82000741, 32160207, 82170743, and 82060111), China Postdoctoral Science Foundation (2020M683374), Guizhou Provincial Science and Technology Projects (ZK [2021]402), Universities Young Science and Technology Talent Growth Project in Guizhou Province (KY [2021]170), Excellent Young Talents Plan of Guizhou Medical University (2021105), and Guizhou Provincial Natural Science Foundation ([2021]4029 and [2022]4017).

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

Author Contributions. B.G. and Y.Z. conceived the idea for the project, designed experiments, and wrote the paper. L.S. performed data analysis, WB, and cell imaging experiments. X.C. performed IHC, H&E, immunofluorescence, and Masson staining experiments. B.J., M.S., and Q.C. were involved in the discussion of ideas and paper writing. L.X. performed pathological analysis. La.H., Lu. L., G.W., and Y.H. performed animal experiments. J.S. and Li.H. collected the clinical samples. T.Z., Y.W., Y.X., F.Z., and Li.L. edited the manuscript. Y.Z. is the guarantor of this work and, as such, had full access to all the data in the study and takes full responsibility for the integrity of the data and the accuracy of the data analysis.

This article contains supplementary material online at https://doi.org/10.2337/figshare.24018906.

L.S., X.C., and La.H. contributed equally to this article.

1.
Gottrup
F
,
Apelqvist
J.
Present and new techniques and devices in the treatment of DFU: a critical review of evidence
.
Diabetes Metab Res Rev
2012
;
28
(
Suppl. 1
):
64
71
2.
Lipsky
BA
,
Senneville
É
,
Abbas
ZG
, et al;
International Working Group on the Diabetic Foot (IWGDF)
.
Guidelines on the diagnosis and treatment of foot infection in persons with diabetes (IWGDF 2019 update)
.
Diabetes Metab Res Rev
2020
;
36
(
Suppl. 1
):
e3280
3.
McDermott
K
,
Fang
M
,
Boulton
AJM
,
Selvin
E
,
Hicks
CW.
Etiology, epidemiology, and disparities in the burden of diabetic foot ulcers
.
Diabetes Care
2023
;
46
:
209
221
4.
Alavi
A
,
Sibbald
RG
,
Mayer
D
, et al
.
Diabetic foot ulcers: part I. pathophysiology and prevention
.
J Am Acad Dermatol
2014
;
70
:
1.e1
1.e18
; quiz 19–20
5.
Nowak
NC
,
Menichella
DM
,
Miller
R
,
Paller
AS.
Cutaneous innervation in impaired diabetic wound healing
.
Transl Res
2021
;
236
:
87
108
6.
Niu
Y
,
Li
Q
,
Ding
Y
,
Dong
L
,
Wang
C.
Engineered delivery strategies for enhanced control of growth factor activities in wound healing
.
Adv Drug Deliv Rev
2019
;
146
:
190
208
7.
Yang
S
,
Gu
Z
,
Can Lu
M
, et al
.
Neutrophil extracellular traps are markers of wound healing impairment in patients with diabetic foot ulcers treated in a multidisciplinary setting
.
Adv Wound Care (New Rochelle)
2020
;
9
:
16
27
8.
Falanga
V.
Wound healing and its impairment in the diabetic foot
.
Lancet
2005
;
366
:
1736
1743
9.
Mosser
DM
,
Hamidzadeh
K
,
Goncalves
R.
Macrophages and the maintenance of homeostasis
.
Cell Mol Immunol
2021
;
18
:
579
587
10.
Veith
AP
,
Henderson
K
,
Spencer
A
,
Sligar
AD
,
Baker
AB.
Therapeutic strategies for enhancing angiogenesis in wound healing
.
Adv Drug Deliv Rev
2019
;
146
:
97
125
11.
Pastar
I
,
Stojadinovic
O
,
Yin
NC
, et al
.
Epithelialization in wound healing: a comprehensive review
.
Adv Wound Care (New Rochelle)
2014
;
3
:
445
464
12.
Huang
S
,
Cao
L
,
Cheng
H
,
Li
D
,
Li
Y
,
Wu
Z.
The blooming intersection of subfatin and metabolic syndrome
.
Rev Cardiovasc Med
2021
;
22
:
799
805
13.
Miao
ZW
,
Hu
WJ
,
Li
ZY
,
Miao
CY.
Involvement of the secreted protein Metrnl in human diseases
.
Acta Pharmacol Sin
2020
;
41
:
1525
1530
14.
Rao
RR
,
Long
JZ
,
White
JP
, et al
.
Meteorin-like is a hormone that regulates immune-adipose interactions to increase beige fat thermogenesis
.
Cell
2014
;
157
:
1279
1291
15.
Lee
JO
,
Byun
WS
,
Kang
MJ
, et al
.
The myokine meteorin-like (metrnl) improves glucose tolerance in both skeletal muscle cells and mice by targeting AMPKα2
.
FEBS J
2020
;
287
:
2087
2104
16.
Jung
TW
,
Lee
SH
,
Kim
H-C
, et al
.
METRNL attenuates lipid-induced inflammation and insulin resistance via AMPK or PPARδ-dependent pathways in skeletal muscle of mice
.
Exp Mol Med
2018
;
50
:
122
17.
Rupérez
C
,
Ferrer-Curriu
G
,
Cervera-Barea
A
, et al
.
Meteorin-like/Meteorin-β protects heart against cardiac dysfunction
.
J Exp Med
2021
;
218
:
e20201206
18.
Xu
L
,
Cai
Y
,
Wang
Y
,
Xu
C.
Meteorin-Like (METRNL) attenuates myocardial ischemia/reperfusion injury-induced cardiomyocytes apoptosis by alleviating endoplasmic reticulum stress via activation of AMPK-PAK2 signaling in H9C2 cells
.
Med Sci Monit
2020
;
26
:
e924564
19.
Lu
Q-B
,
Ding
Y
,
Liu
Y
, et al
.
Metrnl ameliorates diabetic cardiomyopathy via inactivation of cgas/sting signaling dependent on LKB1/AMPK/ULK1-mediated autophagy
.
J Adv Res
2023
;
51
:
161
–1
79
20.
Baht
GS
,
Bareja
A
,
Lee
DE
, et al
.
Meteorin-like facilitates skeletal muscle repair through a Stat3/IGF-1 mechanism
.
Nat Metab
2020
;
2
:
278
289
21.
Zheng
SL
,
Li
ZY
,
Song
J
,
Liu
JM
,
Miao
CY.
Metrnl: a secreted protein with new emerging functions
.
Acta Pharmacol Sin
2016
;
37
:
571
579
42.
Zhou
Y
,
Liu
Z
,
Zhang
S
, et al
.
RILP restricts insulin secretion through mediating lysosomal degradation of proinsulin
.
Diabetes
2020
;
69
:
67
82
23.
Reboll
MR
,
Klede
S
,
Taft
MH
, et al
.
Meteorin-like promotes heart repair through endothelial KIT receptor tyrosine kinase
.
Science
2022
;
376
:
1343
1347
24.
Otranto
M
,
Nascimento
AP
,
Monte-Alto-Costa
A.
Insulin resistance impairs cutaneous wound healing in mice
.
Wound Repair Regen
2013
;
21
:
464
472
25.
Shaabani
E
,
Sharifiaghdam
M
,
Faridi-Majidi
R
,
De Smedt
SC
,
Braeckmans
K
,
Fraire
JC.
Gene therapy to enhance angiogenesis in chronic wounds
.
Mol Ther Nucleic Acids
2022
;
29
:
871
899
26.
Louiselle
AE
,
Niemiec
SM
,
Zgheib
C
,
Liechty
KW.
Macrophage polarization and diabetic wound healing
.
Transl Res
2021
;
236
:
109
116
27.
Boniakowski
AE
,
Kimball
AS
,
Jacobs
BN
,
Kunkel
SL
,
Gallagher
KA.
Macrophage-mediated inflammation in normal and diabetic wound healing
.
J Immunol
2017
;
199
:
17
24
28.
Tsai
CL
,
Changchien
CY
,
Chen
Y
, et al
.
Accelerated wound healing and keratinocyte proliferation through PI3K/Akt/pS6 and VEGFR2 signaling by topical use of pleural fluid
.
Cells
2022
;
11
:
817
29.
Cheng
HW
,
Chen
YF
,
Wong
JM
, et al
.
Cancer cells increase endothelial cell tube formation and survival by activating the PI3K/Akt signalling pathway
.
J Exp Clin Cancer Res
2017
;
36
:
27
30.
Armstrong
DG
,
Boulton
AJM
,
Bus
SA.
Diabetic foot ulcers and their recurrence
.
N Engl J Med
2017
;
376
:
2367
2375
31.
Bus
SA
,
Armstrong
DG
,
van Deursen
RW
,
Lewis
JE
,
Caravaggi
CF
;
International Working Group on the Diabetic Foot
.
IWGDF guidance on footwear and offloading interventions to prevent and heal foot ulcers in patients with diabetes
.
Diabetes Metab Res Rev
2016
;
32
(
Suppl. 1
):
25
36
32.
Raziyeva
K
,
Kim
Y
,
Zharkinbekov
Z
,
Kassymbek
K
,
Jimi
S
,
Saparov
A.
Immunology of acute and chronic wound healing
.
Biomolecules
2021
;
11
:
700
33.
Sim
SL
,
Kumari
S
,
Kaur
S
,
Khosrotehrani
K.
Macrophages in skin wounds: functions and therapeutic potential
.
Biomolecules
2022
;
12
:
1659
34.
Chang
M
,
Nguyen
TT.
Strategy for treatment of infected diabetic foot ulcers
.
Acc Chem Res
2021
;
54
:
1080
1093
35.
Zhou
Y
,
Liu
L
,
Jin
B
, et al
.
Metrnl alleviates lipid accumulation by modulating mitochondrial homeostasis in diabetic nephropathy
.
Diabetes
2023
;
72
:
611
626
36.
Shao
Z
,
Yin
T
,
Jiang
J
,
He
Y
,
Xiang
T
,
Zhou
S.
Wound microenvironment self-adaptive hydrogel with efficient angiogenesis for promoting diabetic wound healing
.
Bioact Mater
2022
;
20
:
561
573
37.
Eming
SA
,
Wynn
TA
,
Martin
P.
Inflammation and metabolism in tissue repair and regeneration
.
Science
2017
;
356
:
1026
1030
38.
Hassanshahi
A
,
Moradzad
M
,
Ghalamkari
S
,
Fadaei
M
,
Cowin
AJ
,
Hassanshahi
M.
Macrophage-mediated inflammation in skin wound healing
.
Cells
2022
;
11
:
2953
39.
Kharaziha
M
,
Baidya
A
,
Annabi
N.
Rational design of immunomodulatory hydrogels for chronic wound healing
.
Adv Mater
2021
;
33
:
e2100176
40.
Ushach
I
,
Arrevillaga-Boni
G
,
Heller
GN
, et al
.
Meteorin-like/Meteorin-β is a novel immunoregulatory cytokine associated with inflammation
.
J Immunol
2018
;
201
:
3669
3676
41.
Lin
CW
,
Hung
CM
,
Chen
WJ
, et al
.
New horizons of macrophage immunomodulation in the healing of diabetic foot ulcers
.
Pharmaceutics
2022
;
14
:
2065
42.
Stender
S
,
Kozlitina
J
,
Nordestgaard
BG
,
Tybjærg-Hansen
A
,
Hobbs
HH
,
Cohen
JC.
Adiposity amplifies the genetic risk of fatty liver disease conferred by multiple loci
.
Nat Genet
2017
;
49
:
842
847
Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at https://www.diabetesjournals.org/journals/pages/license.