A frog cathelicidin peptide effectively promotes cutaneous wound healing in mice

The skin serves as a protective barrier against the outside world and any break in it must be rapidly and efficiently mended. Cutaneous wound healing is a highly orchestrated biological process, which is essential to restore the integrity of the skin barrier and is critical to the survival of the organism after injury. Among vertebrates, skins of amphibians are exposed to more dangers of biological or non-biological injuries than others [1]. Amphibians have evolved the capacity to respond to environmental challenges. Their skins, limbs or tails possess the excellent ability to heal and regenerate as needed for normal physiology and pathophysiology [2–4].

Cathelicidins are a family of antimicrobial and immune-stimulatory peptides found exclusively in vertebrates thus far. Two cathelicidins (Bac5 and Bac7) were first isolated from bovine neutrophils in 1989 [5], and then many cathelicidins have been identified from different species of mammals [6], birds [7], reptiles [8], amphibians [9] and fishes [10]. Mammalian cathelicidins are one of the best characterized classes of host defense peptides [11] and they have multifaceted roles in immune modulation, inflammation, wound healing [12], angiogenesis [13], cell death [14] and autoimmune disorders [15]. However, relatively limited research has been done about cathelicidins in amphibians (Anura, Caudata and Gymnophiona). As of April 2018, ∼7858 amphibian species have been described in the AmphibiaWeb database (http://www.amphibiaweb.org/). The first amphibian cathelicidin (cathelicidin-AL) possessing antimicrobial activity was characterized in 2012 [16], and then eight additional cathelicidins (cathelicidin-RC1 and -RC2, Lf-CATH1 and -2, BG-CATH, cathelicidin-Bg, cathelicidin-PY and cathelicidin-PP) with antimicrobial and/or anti-inflammatory activities have been identified [9]. Currently, only two cathelicidins with wound-healing ability have been characterized in amphibians. One is tylotoin, which was isolated from the salamander Tylototriton verrucosus (Caudata: Salamandridae) [17]; the other is cathelicidin-OA1, which was identified from the Chinese odorous frog Odorrana andersonii (Anura: Ranidae) [18]. Additionally, two non-cathelicidin peptides (CW49 and AH90) from frog skin secretions and one designed peptide (tiger17) based on frog antimicrobial peptides, also showed wound-healing ability [19–21].

The plateau frog Nanorana ventripunctata is endemic to northwestern Yunnan province in China. It lives at high altitudes (3120–4100 m) with extremely cold weather. Thus, to survive, N. ventripunctata must respond rapidly and effectively to various biotic and abiotic damage, including cutaneous damage caused by ultraviolet radiation and cold. It would be useful to explore whether there are some molecules involved in wound healing in the skins of N. ventripunctata.

In the current study, a novel cathelicidin peptide named cathelicidin-NV was identified from the skin of the plateau frog N. ventripunctata. It showed potential wound healing-promoting activity in a murine model with a full-thickness dermal wound. The cytokines, chemokines and mitogen-activated protein kinases (MAPKs) (ERK, JNK and p38) signaling pathways that are involved in the wound healing-promoting effect were also studied. Our results suggest that cathelicidin-NV might be an excellent candidate for the development of a novel wound-healing agent.

Adult N. ventripunctata (n = 30; weight range 20–25 g) was collected from Shangri-La, Yunnan province of China. Skin secretions were collected as previously reported [9]. Frogs were stimulated with volatilized anhydrous ether immersed in absorbent cotton, and their skin surface was seen to exude secretions. Skin secretions were washed with 0.1 M phosphate buffer, pH 6.0 (PBS) [containing 1% (v/v) protease inhibitor cocktail, Sigma, P8340-5]. The collected solutions containing skin secretions were quickly centrifuged (10 000 g for 10 min) and the supernatants were lyophilized. The study was approved by the Animal Care and Use Ethics Committee of Kunming Medical University.

The peptide purification procedures were performed according to the method described in our previous work [22]. An aliquot (1 g) of lyophilized skin secretion was dissolved in 10 ml of PBS and centrifuged at 5000 g for 10 min. The supernatant was applied to a Sephadex G-50 (Superfine, Amersham Biosciences) gel filtration column (2.6 cm diameter, 100 cm length) equilibrated with 0.1 M PBS for preliminary separation. Elution was performed with the same buffer, collecting fractions of 3.0 ml. The absorbance of the elute fractions were monitored at 280 nm. The cell proliferation activity was determined as described below. The fraction containing cell proliferation activity was further purified by a C₁₈ reversed-phase high-performance liquid chromatography (RP-HPLC, Gemini C₁₈ column, 5 μm particle size, 110 Å pore size, 250 mm length, 4.6 mm diameter) column. The elution is performed using a linear gradient of 0–80% acetonitrile containing 0.1% (v/v) trifluoroacetic acid in 0.1% (v/v) trifluoroacetic acid/water over 60 min, as illustrated in Figure 1B. UV-absorbing peaks were collected, lyophilized and assayed for cell proliferation activity. Peaks with cell proliferation activity were collected and lyophilized for a second HPLC purification procedure using the same conditions as illustrated in Figure 1C.

The N-terminal sequence of the purified peptide was determined by Edman degradation on an Applied Biosystems pulsed liquid-phase sequencer (model ABI 491). Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) was used to identify the purified peptide. The AXIMA CFR mass spectrometer (Kratos Analytical) was analyzed in the linear and the positive ion mode using an acceleration voltage of 20 kV and an accumulating time of single scanning of 50 s.

The experiment was performed according to the method described in our previous work [22]. Total RNA was extracted from the skin of N. ventripunctata using the RNeasy Protect Mini Kit (QIAGEN, Germany) according to the manufacturer's instructions. An In-Fusion SMARTerTM Directional cDNA Library Construction Kit was used for cDNA synthesis. The synthesized cDNA was used as the template for PCR to screen the cDNAs encoding the peptide (cathelicidin-NV). According to the sequence determined by Edman degradation, an antisense degenerate primer (cathelicidin-NV-R1) was designed and coupled with a 5′ PCR primer (the adaptor sequence of the 3′ In-Fusion SMARTer CDS Primer provided in the kit) to screen the 5′ fragment of cDNA encoding cathelicidin-NV. Then, a sense primer (cathelicidin-NV-F1) was designed according to the 5′ coding region and coupled with the 3′ PCR primer from the library kit to screen the full-length cDNAs. The PCR conditions were 2 min at 95°C, and 30 cycles of 10 s at 92°C, 30 s at 50°C and 40 s at 72°C, followed by a 10 min extension at 72°C. The PCR products were cloned into pGEM®-T Easy vector (Promega, Madison, WI). DNA sequencing was performed on an Applied Biosystems DNA sequencer, model ABI PRISM 377. Primers used in this research are listed in the Supplementary Table S1.

Antimicrobial activity of cathelicidin-NV was assayed according to our previous methods [9]. Microorganisms were obtained from the First Affiliated Hospital of Kunming Medical University. The growth of the microbe was determined by monitoring the absorbance at 600 nm. The minimal concentrations at which no microbial growth occurred were recorded as MIC values.

Cytotoxicity and hemolytic assay were performed according to the method described in our previous work [9]. Cytotoxicity against human skin fibroblasts (HSFs), human HaCat keratinocyte cells and murine macrophage cell line RAW264.7 was determined by the MTT assay. Cathelicidin-NV dissolved in serum-free RPMI 1640 medium was added to cells in 96-well plates (2 × 10⁴ cells/well), and the serum-free RPMI 1640 medium without cathelicidin-NV was used as control. After incubation for 24 h, 20 µl of MTT solution (5 mg/ml) was added to each well, and the cells were further incubated for 4 h. Finally, cells were dissolved in 200 µl of Me₂SO₄, and the absorbance at 570 nm was measured. Rabbit erythrocyte suspensions were incubated with cathelicidin-NV and then the absorbance of the supernatant was measured at 540 nm. 1% (v/v) Triton X-100 and PBS were used as the positive and the negative control, respectively.

The proliferation of HSFs and immortalized human HaCat keratinocyte cells was measured using a colorimetric assay according to the method described in our previous work [17]. Cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Invitrogen) supplemented with 10% fetal bovine serum (FBS) and penicillin (100 u/ml)–streptomycin (100 mg/ml) at 37°C in a humidified 5% CO₂ atmosphere, respectively. HaCat and HSF cells were plated into 96-well plates (2 × 10⁴ cells/well), respectively. After adhering to the plate, cells were incubated with vehicle (DMEM) or cathelicidin-NV with different concentrations for 24 h. Based on previous reports, the amphibian cathelicidin (tylotoin) [17] and three non-cathelicidin peptides (CW49, AH90 and tiger17) from amphibians [19–21], as well as the sole human cathelicidin LL-37 [23], exhibited wound-healing ability with concentrations ranging from 100 ng/ml to 250 µg/ml. We optimized the concentration (2, 5, 10 and 20 µg/ml) of cathelicidin-NV to measure its activities. Then 20 µl of 3-(4,5-dimethylthiazol-2-yl)-2,5- diphenyl-2H-tetrazolium bromide (MTT) solution (R&D Systems Inc., Minneapolis, MN) was added to each well for a further 4 h incubation at 37°C. After the cells were washed three times with PBS (pH 7.4), the insoluble formazan product was dissolved by incubation with 150 µl of DMSO. The absorbance of each well was measured on an enzyme-linked immunosorbent assay (ELISA) micro-plate reader at 570 nm. The optical density reflects the level of cell metabolic activity. Each experiment was performed in triplicate.

For wound assays in the presence of cathelicidin-NV, HaCat Cells were seeded into 6-well plates (1 × 10⁶ cells/well) to confluence in DMEM supplemented with 10% FBS. After 12 h of serum starvation (DMEM supplemented with 1% FBS) for increasing the peptide sensitivity, the cell monolayer was then subjected to a mechanical scratch wound using a sterile pipette tip. After washing twice with PBS to remove detached cells, cells were then cultured for additional periods (from 0 to 48 h) in a serum-free basal medium in the continued presence of vehicle or cathelicidin-NV (20 µg/ml) and mitomycin C (5 mg/ml) to prevent cell proliferation.

The same fields of the wound margin were photographed using a Primovert microscope (Zeiss, Germany) at different time points (at 0, 24 and 48 h after scratch wounding). Cell migration activity was expressed as the percentage of the gap relative to the total area of the cell-free region immediately after the scratch, named the repair rate of scarification, using the Image J software (National Institutes of Health, Bethesda, MD, U.S.A.). For each plate, six randomly selected images were acquired. All experiments were carried out in triplicate.

Raw 264.7 murine macrophage cells were seeded and adhered to 96-well culture plates (2 × 10⁴ cells/well). The cells were treated with cathelicidin-NV (2, 5, 10 and 20 µg/ml) or vehicle for 48 h; then culture supernatants were collected and assessed for MCP-1 (monocyte chemoattractant protein-1), TNF-α (tumor necrosis factor-α), VEGF (vascular endothelial growth factor) and TGF-β1 (transforming growth factor-β1) by using ELISA kits (DAKAWE, Beijing, China). ELISA was performed according to the manufacturer's instructions. Experiments were independently carried out in triplicate.

Raw 264.7 murine macrophage cells were cultured in 6-well culture plates (2 × 10⁶ cells/well) and transferred to serum-free DMEM for a 16-h incubation. The cells were incubated with cathelicidin-NV (2, 5, 10 and 20 µg/ml) or vehicle for 3 h. Then, the cells were washed twice with ice-cold phosphate-buffered saline (PBS) and lysed with RIPA lysis buffer (Beyotime, China). The proteins were extracted for western blot analysis according to our previously described method [15]. The concentration of protein was determined by the Bradford protein assay. Then the cellular proteins were separated on a 12% SDS–PAGE gel and electroblotted onto a polyvinylidene difluoride membrane. Primary antibodies against β-actin (1:5000, Santa Cruz Biotechnology, U.S.A.) and Erk1/2, SAPK/JNK and p38 MAPK (1:2000; Cell Signaling Technology, Beverly, MA, U.S.A.) were used in western blot analysis.

It would be valuable to test the role of cathelicidin-NV in a frog model. However, the captured frogs in the wild are different in age, sex, size and body weight, which may lead to inaccurate results in the frog model of wound healing. Furthermore, commercial antibodies in frogs are not available for our further studies, including immunohistochemistry, ELISA and western blot analysis. A mouse model of wound healing was used to determine the activity of cathelicidin-NV. Kunming mice (6–7-week-old male) were anesthetized using 1% pentobarbital sodium (0.1 ml/20 g body weight). After the dorsal hairs were shaved by an electric clipper and the dorsal skin was cleansed with Betadine, two full-thickness skin wounds of 6-mm diameter were created independently on the back of each mouse using a biopsy punch, as previously reported [24]. After wounding, mice were caged individually until termination of the experiment. Mice were treated with vehicle, cathelicidin-NV (20 μl, 200 µg/ml) or EGF (20 μl, 100 µg/ml) applied directly to the wound site twice daily. To observe the natural course of wound healing in mice, the wounds were photographed on days 0 to 10 after the skin injury. Time-dependent alterations of the wound area are shown in Figure 4A. The wound areas were calculated from the photographs using PhotoShop (Adobe Photoshop Element 2.0, Adobe Systems) (n = 10 per group). The researchers performing the experiment and analysis were both blinded. The study was approved by the Animal Care and Use Ethics Committee of Kunming Medical University.

The biopsy specimens involving the central part of the wounds (days 0, 1, 2, 4 and 6) were obtained from mice for tissue ELISA. Skin specimens were homogenized in 4 ml 0.1 M PBS [containing 1% (v/v) protease inhibitor cocktail, Sigma, P8340-5]; then the homogenates were transferred to Eppendorf tubes and centrifuged at 13 000 g for 30 min at 4°C. The total protein concentrations were adjusted using the Bio-Rad protein assay dye (Bio-Rad Laboratories, Hercules, CA), and MCP-1, TNF-α, VEGF and TGF-β1 protein levels were determined using ELISA kits (DAKAWE, Beijing, China).

At days 4 and 10 after wounding, mice were killed (five animals in each group, one paired wounds per animal), and the biopsy specimens involving the central part of the wounds were obtained from mice for light microscopy. Skin specimens were fixed in 4% paraformaldehyde for 24 h and then were dehydrated by washing with a series of ethanol solutions. Subsequently, skin specimens were cleared in xylene, and embedded in paraffin wax. Thick sections (5 µm) were prepared and stained with Hematoxylin and Eosin (H&E) for histological analysis. Thick sections (3 µm) were stained with the Masson collagen staining method to observe the collagen fibers. The IPLab imaging software (BD Biosciences, Bedford, MA) was used to measure the changes of the wound. Width of the wound and distance of the neo-epithelium were measured on H&E-stained sections, and percent re-epithelialization was calculated according to the following formula: (distance covered by neo-epithelium)/(distance between wound bed) × 100 (n = 5 per group).

Immunohistochemistry was performed following standard procedures. A total of 3 µm sections were reacted with anti-α SMA primary antibody (1:1000 dilution, ab32575, abcam) after blocking endogenous peroxidase and nonspecific binding. Then, they were incubated with biotinylated goat anti-rabbit IgG (1:200 dilution, ab150077, abcam) for 1 h at room temperature. As a control, sections were treated with the same dilution buffer without the primary antibodies.

Cathelicidin-NV (ARGKKECKDDRCRLLMKRGSFSYV) and the scrambled version of cathelicidin-NV called sNV (KLSRGVCDKRDCASMRYFLKGREK), which are stabilized by one intramolecular disulfide bond between Cys7-Cys12, were synthesized by the peptide synthesizer GL Biochem Ltd (Shanghai, China). The synthetic peptides were purified and then analyzed by HPLC and MALDI-TOF MS to confirm that the purity was higher than 98%.

Statistical differences were determined using the one-way analysis of variance test or Student's t-test. P < 0.05 and P < 0.01 were considered statistically significant. Results are shown as mean ± SEM.

As shown in Figure 1A, the skin secretions of N. ventripunctata were divided into five fractions after Sephadex G-50 gel filtration. Then the fraction containing activity to promote cell proliferation was pooled and subjected to a C₁₈ RP-HPLC column for further purification (Figure 1B,C). The purified peptide with cell proliferation activity was named cathelicidin-NV (Figure 1C) (marked by an arrow). Unless indicated otherwise, the purified cathelicidin-NV was used in the experiments which are involved in the wound healing-promoting activities.

After Edman degradation, the complete amino acid sequence of purified cathelicidin-NV was identified with the following sequence: ARGKKECKDDRCRLLMKRGSFSYV. Cathelicidin-NV is composed of 24 amino acid residues including two cysteines, which are possible to form an intramolecular disulfide bridge. MALDI-TOF MS analysis (Supplementary Figure S1) indicated that the molecular mass of cathelicidin-NV was 2845.9 Da. It is 1.47 Da lower than the calculated mass (2847.37 Da), suggesting that one disulfide bridge was formed by two cysteines in the mature peptide. The complete nucleotide sequence encoding the cathelicidin-NV precursor (GenBank accession number MH010580) and the encoded amino acid sequence are shown in Figure 2; the cathelicidin-NV precursor is composed of 146 amino acid residues, including a predicted 20 amino acid signal peptide and a conserved 102 amino acid cathelin domain, followed by a 24 amino acid mature peptide.

BLAST search indicated that the precursor is a member of the cathelicidin family containing a conserved cathelin domain and a variable C-terminus (Figure 2). Cathelicidin-NV shared the highest identity of 75% with the predicted Cath-2-like-isoform X1 from the frog Nanorana parkeri. Multisequence alignment of the cathelicidin-NV precursor with other amphibian cathelicidins (Supplementary Figure S2) indicated that the cathelicidin-NV precursor exhibits a high degree of similarity with the other cathelicidins in the cathelin regions, but they are highly variable in the C-terminal mature region. Additionally, a characteristic feature of all these mature peptides is the presence of one intramolecular disulfide bond.

Among the 15 strains of the microorganisms tested (Supplementary Table S2), cathelicidin-NV did not show antimicrobial activity against Gram-positive bacteria (7 strains), Gram-negative bacteria (6 strains) and fungi (2 strains) at the concentration of 200 µg/ml (70.28 µM). Cathelicidin-NV did not exhibit cytotoxicity toward HSFs, HaCat cells and RAW264.7 at the concentration of 200 µg/ml. In addition, it did not have hemolytic activity against erythrocytes at the same concentration.

The proliferation and motility of keratinocytes and fibroblasts are the most critical factors in wound healing, which promote the re-epithelialization process and accelerate wound closure. As illustrated in Figure 3A–C, cathelicidin-NV obviously enhanced the proliferation of HaCat and HSF cells in a concentration-dependent manner. At the concentration of 20 μg/ml, the rates of proliferation were increased by 43% and 63% for HaCat and HSF cells, respectively. The scrambled version of cathelicidin-NV, called sNV, did not show any activities on the proliferation of keratinocytes and fibroblasts (Figure 3C,D and Supplementary Figure S3).

To observe the effect of cathelicidin-NV on keratinocyte migration, an in vitro cells scratch assay was performed. Cathelicidin-NV significantly promoted keratinocyte migration; as illustrated in Figure 3C,D, the migration of HaCaT cells across the wound chasm was significantly enhanced in cells treated with cathelicidin-NV compared with those treated with vehicle. The repair rate of scarification of cathelicidin-NV reached 51 or 74% at 24 or 48 h, respectively, but that of vehicle only reached 32 or 56% at 24 or 48 h, respectively (Figure 3C).

Considering that cathelicidin-NV has strong ability to promote proliferation and motility of keratinocytes and fibroblasts in vitro, we performed an excisional wound-healing mouse model to address whether topical application of cathelicidin-NV would modify the healing of full-thickness dermal wounds in vivo. The time course observations indicated that topical application of cathelicidin-NV and sCathelicidin-NV (synthetic cathelicidin-NV) (200 μg/ml, 20 μl) on mice significantly accelerated wound closure, compared with mice treated with vehicle and sNV control (Figure 4A,B). The sNV did not show any accelerated activities on the healing of full-thickness dermal wounds (Figure 4A,B). Cathelicidin-NV also showed effects on wound healing comparable to the positive control epidermal growth factor (EGF) (Figure 4A,B). Furthermore, there were obviously contractions of wounds in cathelicidin-NV-treated mice, compared with wounds treated with vehicle control or EGF (Figure 4A). On post-injury day 2, the wound area in cathelicidin-NV-treated mice was ∼22% smaller than that in control mice (Figure 4B). Subsequently, there was a rapid acceleration of wound healing in cathelicidin-NV- and EGF-treated mice. On day 8, the wound treated with cathelicidin-NV or EGF was ∼88 and 90% smaller than controls, respectively. On day 10, the wounds of cathelicidin-NV- and EGF-treated mice were almost closed completely, whereas the wounds of control mice remained ∼6% (Figure 4A,B). No adverse effects on the body weight, general health or behavior of the mice were observed for the topical cathelicidin-NV treatment.

In a parallel experiment, we killed mice in cathelicidin-NV-treated and control groups at post-injury days 4 and 10, and then performed a histological evaluation of skin tissue sections stained with H&E. Cathelicidin-NV-treated mice exhibited enhanced restoration of dermal and epidermal and formation of granulation tissue in the wound (Figure 4C–E). The measurement of the distance between epithelial tongues revealed a clear enhancement of the re-epithelialization process in most of the cathelicidin-NV-treated wounds at days 4 and 10 post wounding (Figure 4C,D). The percentage of re-epithelialization was improved by 35% after 4 days, and by 26% after 10 days of treatment with cathelicidin-NV. At day 4, an increase in the lengths of the tongues of new migrating epithelium was clearly observed in cathelicidin-NV-treated wounds (Figure 4C,D). At day 10, a similar increase was observed, as in most cases the epithelial tongues had converged to completely cover the wound when cathelicidin-NV was applied, whereas in the corresponding vehicle wounds the epithelial gap was still evident (Figure 4C,D). Granulation tissue developed more slowly in control wounds, as the cell density in the central area of the 4,10-day cathelicidin-NV-treated wounds was higher than that of control wounds (Figure 4E).

Given the above observation that cathelicidin-NV obviously promoted the proliferation of HSFs in vitro and enhanced mice wound healing by contractions, we further explored the potential effect of cathelicidin-NV on fibroblasts in a skin wound. Fibroblast-to-myofibroblast transition plays an important role in cutaneous wound healing [25]. In wound tissues at post-injury day 4, 10 myofibroblasts were stained brown with an antibody against alpha-smooth muscle actin (α-SMA), a differentiation marker of smooth muscle cells. As illustrated in Figure 5A,B, there is strong expression of α-SMA in myofibroblasts of the granulation tissue below the wound surface at days 4 and 10 in cathelicidin-NV-treated mice, but only very weak signals in vehicle-treated mice. There was ∼43 and 20% α-SMA positive area in the wound tissue of mice treated with cathelicidin-NV and vehicle, respectively, on day 4; then the proportion of the α-SMA positive area reached ∼76 and 47%, respectively, on day 10 (Figure 5B). The result indicates that cathelicidin-NV obviously induces the fibroblast-to-myofibroblast transition.

To further examine collagen deposition from fibroblasts in the wounds, wound tissues at post-injury days 4 and 10 were stained with Masson trichrome. The wounds in the cathelicidin-NV-treated mice contain higher levels of collagen, both on day 4 and day 10, compared with the vehicle-treated wounds, respectively (Figure 5C,D); the percentage of collagen positive area was improved by 65% after 4 days, and by 77% after 10 days of treatment with cathelicidin-NV (Figure 5C).

Wound healing is a complex biological process that is regulated by numerous growth factors, cytokines and chemokines. To determine the effect of cathelicidin-NV on cytokine and chemokine secretion in vitro, we used ELISA to measure the MCP-1,TNF-α, VEGF and TGF-β1 level in murine macrophage cell line RAW264.7. As illustrated in Figure 6A–D, cathelicidin-NV significantly increased MCP-1,TNF-α, VEGF and TGF-β1 secretion in a dose-dependent manner in the culture supernatants of RAW264.7.

Furthermore, to determine cathelicidin-NV on cytokine and chemokine secretion at the wound site, wounds were harvested 0, 1, 2, 4 and 6 days after wounding, homogenized and analyzed using ELISA kits. All of the tested proteins in the pointed day were expressed at higher levels in cathelicidin-NV-treated wounds than in vehicle-treated wounds (Figure 6E–H). Overall, these results show that cathelicidin-NV significantly induced expression of several cytokines and chemokines in vitro and in vivo, which are related with wound healing.

Western immunoblot analysis was performed to further explore the effect of cathelicidin-NV on the MAPK signaling pathway in RAW 264.7 macrophage cells. As illustrated in Figure 7A–D cathelicidin-NV significantly increased ERK and p38 phosphorylation in a concentration-dependent manner, but showed little effect on JNK phosphorylation. After a 1 h treatment with 2, 5, 10 and 20 μg/ml tylotoin, ERK phosphorylation was increased by 0.1, 0.6, 1.3 and 1.8 times, while the corresponding p38 up-regulation was 1.0, 2.4, 3.2 and 3.4 times, respectively.

The classical model of wound healing is a highly complex, orchestrated and overlapping process that can be divided into four main phases: hemostasis, inflammation, proliferation and tissue remodeling [26–28]. Chronic wounds, defined as a barrier defect that has not healed in an orderly set of phases or in a timely fashion, have become a major therapeutic challenge [29]. Thus, identifying more effective wound therapies (medicines) of treating chronic wounds is urgently necessary. Amphibian skins display excellent wound-healing ability and represent a resource for prospective wound healing-promoting peptides. As a step toward understanding amphibians’ wound healing and identifying novel wound healing peptides, we address this issue and have characterized a potential wound healing-promoting peptide (cathelicidin-NV) from N. ventripunctata in this work.

Cathelicidin-NV is a short peptide (ARGKKECKDDRCRLLMKRGSFSYV) and is composed of just 24 amino acid residues. The precursor of cathelicidin-NV belongs to the cathelicidin family, which are a unique and diverse group of effector molecules involved in humoral immunity among vertebrates. Cathelicidin-NV is a highly basic peptide with a net charge of +5, which is an important parameter contributing to the ability of cationic peptides to bind to and disrupt microbial membranes. However, cathelicidin-NV did not have any direct effect on fifteen strains of microorganisms at concentrations up to 200 µg/ml, including 7 strains of Gram-positive bacteria, 6 strains of Gram-negative bacteria and 2 strains of fungi (Supplementary Table S2). In addition, it showed no cytotoxicity and hemolytic activity toward mammalian cells at the same concentration of 200 µg/ml (Data not shown). The synthetic cathelicidin-NV (sCathelicidin-NV) with one intramolecular disulfide bond exhibited the same bioactivity in vitro and in vivo with the purified cathelicidin-NV from the skin secretions (Figure 4A,B and Supplementary Figure S3), which further confirmed the structure of cathelicidin-NV. Moreover, the scrambled version of cathelicidin-NV (sNV) did not show any activities on the motility and proliferation of keratinocytes and fibroblasts (Figure 3C,D and Supplementary Figure S3). Cathelicidin-NV seems to be a specific wound healing-promoting peptide.

Wound healing after damage to the skin involves a complex interplay of primarily keratinocytes, fibroblasts, endothelial cells of vessels, recruited immune cells and their associated extracellular matrix [29]. The proliferation, migration and differentiation of these cells resulted in the formation of new tissue and ultimately wound closure [26,30]. Re-epithelialization is a crucial process, which entails re-covering the surface of the wound with new epithelium and requires proper migration and proliferation of keratinocytes at the periphery of the wound [31]. Cathelicidin-NV treatment markedly enhanced the proliferation and motility of keratinocytes (Figure 3A–D) in vitro. Further in vivo investigations demonstrated that cathelicidin-NV can effectively promote cutaneous wound healing (Figure 4A–E). Topical application of cathelicidin-NV greatly accelerated full-thickness skin wound healing in a mouse model of a full-thickness skin wound (Figure 4A,B). Accelerated wound closure was observed in the mice treated with cathelicidin-NV compared with vehicle control. As illustrated in Figure 4C,D, cathelicidin-NV directly accelerated re-epithelialization of the wound site. In addition to re-epithelialization resulting from the proliferation and motility of keratinocytes, fibroblasts are also important repairing cells that dominantly take part in the proliferation phase of wound healing. Some of fibroblasts differentiate into myofibroblasts and these contractile cells will help bridge the gap between the wound edges [32]. As illustrated in Figure 5A,B, cathelicidin-NV significantly promoted the expression of α-SMA in vivo, a differentiation marker of smooth muscle cells, suggesting that cathelicidin-NV accelerated the fibroblast-to-myofibroblast transition to accelerate wound closure. Additionally, cathelicidin-NV was found to significantly induce collagen production from fibroblasts in the skin wound as illustrated in Figure 5C,D, which is essential for the formation of granulation tissue. Taken together, cathelicidin-NV obviously promotes the formation of granulation tissue by inducing proliferation of fibroblasts (Figure 3B), fibroblast-to-myofibroblast transition (Figure 5A,B) and collagen production from fibroblasts (Figure 5C,D). Meanwhile, cathelicidin-NV directly enhances the proliferation of keratinocyte cells, resulting in accelerated re-epithelialization of the wound site. The results suggest that a more efficient wound-healing process in mice treated with cathelicidin-NV might result from enhanced proliferation, differentiation and secretion of fibroblasts in the injured area, which facilitate the formation of granulation tissue.

Macrophages almost take part in every process during wound healing and they provide a continual source of growth factors. They can produce many cytokines associated with wound healing such as VEGF, MCP-1,TNF-α and TGF-β1 [32], which are important producers of inflammatory mediators. VEGF and MCP-1 are known to promote angiogenesis in wounds, and this vascularization in the wound-bed area is important for the proper supply of nutrients and oxygen to allow rapid growth of the skin [33–35]. TNF-α is also able to influence the rate of wound healing through regulation of a smooth muscle actin expression in dermal fibroblasts [36]. TGF-β1 is a multifunctional cytokine that plays important roles in cell proliferation, differentiation and especially in formation of extracellular matrix (ECM) [37]. Our results indicated that cathelicidin-NV has an effect on cytokine and chemokine secretion in vitro and in vivo. As illustrated in Figure 6A,D, cathelicidin-NV significantly increased MCP-1, TNF-α, VEGF and TGF-β1 secretion in a dose-dependent manner in the culture supernatants of RAW264.7 in vitro. Furthermore, cathelicidin-NV also significantly up-regulated MCP-1, TNF-α, VEGF and TGF-β1 levels in wounds in vivo (Figure 6E–H). MCP plays an important role in wound-healing processes, which chemotactically attracts neutrophils and monocytes into the wound [11]. The monocytes differentiate into macrophages, which initiate the development of granulation tissue and release a variety of pro-inflammatory cytokines and growth factors contributing to wound healing [38]. Additionally, cathelicidin-NV was found to induce VEGF secretion as illustrated in Figure 6C,G. VEGF can enhance endothelial cell migration and proliferation to stimulate angiogenesis, which plays a critical role in wound healing [39], because it provides nutrients, oxygen and wound-healing factors to enter the wound area.

Our results are partially different from the other two amphibian cathelicidins (tylotoin and cathelicidin-OA1) that showed similar wound-healing effects. Tylotoin increased TGF-β1 and IL-6 secretion in a dose-dependent manner in the culture supernatants of RAW264.7 and it had no significant effect on IL-1β, EGF and TNF-α secretion [17]. Cathelicidin-OA1 promoted the secretion of TNF-α and TGF-β in the human monocyte-like cell line THP-1 [18], and the authors did not detect the effect of cathelicidin-OA1 on other cytokines and chemokines. Additionally, the human cathelicidin LL-37, which was the best characterized member of the cathelicidin family, has been shown to exhibit wound healing-promoting activities in vitro and in vivo. LL-37 increased the production of cytokines and chemokines such as IL-6, IL-8, TNF-α, IL-10, MCP-1, interferon-inducible protein-10 (IP-10) and macrophage inflammatory protein-3 alpha (MIP3a) in keratinocytes [40,41]. Moreover, the wound-healing activity of LL-37 could be mediated not only through the transactivation of the epidermal growth factor receptor (EGFR) but also through the induction of formyl peptide receptor-like 1 (FPRL-1) expression in human keratinocyte cell line HaCaT [23]. Cathelicidin-NV belongs to the cathelicidin family and its functions seem to overlap with those of the mammalian counterparts. The structure and activity relationship for cathelicidin-NV needs to be further investigated.

The MAPK signaling system has been demonstrated to take an important role in wound healing [42–44]. Western blot analysis showed that cathelicidin-NV significantly increased the activation of the ERK and p38 subgroups of the MAPK signaling pathway in a concentration-dependent manner (Figure 7A–C). This demonstrated that both the ERK and p38 signaling pathways are involved in cathelicidin-NV-stimulated cytokines related with wound healing release, and they may have a cross-talk and orchestrate in regulating these cytokines’ expression and wound-healing process.

In conclusion, cathelicidin-NV identified from N. ventripunctata skin is a bioactive/effector compound with potential wound-healing ability. It promotes the proliferation of fibroblasts, the differentiation of fibroblasts to myofibroblasts and collagen production in fibroblasts. Furthermore, it also promotes the secretion of cytokines and chemokines in vivo and in vitro, including MCP-1, TNF-α, VEGF and TGF-β1. It may facilitate the understanding of frog wound healing and skin regeneration. In addition, these properties make cathelicidin-NV a potent candidate for skin wound therapeutics.

DMEM

Dulbecco's modified Eagle's medium

EGF

epidermal growth factor

ELISA

enzyme-linked immunosorbent assay

ERK

extracellular regulated protein kinases

FBS

fetal bovine serum

H&E

Hematoxylin and Eosin

HSFs

human skin fibroblasts

JNK

c-Jun NH2-terminal kinase

MALDI-TOF

matrix-assisted laser desorption ionization time-of-flight

MAPK

mitogen-activated protein kinases

MCP-1

monocyte chemoattractant protein-1

MTT

3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide

p38

p38 mitogen-activated protein kinases

PBS

phosphate-buffered saline

RP-HPLC

reversed-phase high-performance liquid chromatography

TGF-β1

transforming growth factor-β1

TNF-α

tumor necrosis factor-α

VEGF

vascular endothelial growth factor

α-SMA

α-smooth muscle actin

J.W., L.M. and H.Y. conceived and designed the experiments and analyzed the data. J.W., L.M., H.Y., J.Y., L.W., X.W., K.M., Y.S., and T.L. performed the experiments. L.M. and H.Y. contributed reagents/materials/analysis tools and wrote the paper. All the authors read and approved the final version of the manuscript.

This work was supported by Chinese National Natural Science Foundation [81560581, 81673401, 81373380, 81402830], Natural Science Foundation of Yunnan Province [2015FB018, 2017FB136, 2014Y160], Foundation of Kunming Medical University [60117190405] and Natural Science Foundation of Jiangsu Province [BK20140362].

The Authors declare that there are no competing interests associated with the manuscript.