All BALB/c mice (six-week-old) were purchased from the Laboratory Animal Center of Southern Medical University and were housed in the SPF facility at Southern Medical University. The animal experiments were carried out in the light of the approval and guidelines of Animal Care and Use Committee of Southern Medical University. All procedures in this study strictly complied with the Animal Welfare Act and principles stated in the Guide for the Care and Use of Laboratory Animals, National Research Council, 1996.

Both sexes of adult F. multistriata (n = 3; weighing about 40 g) with no specific permissions in need for sampling were captured from Guangzhou city, Guangdong Province, China (23.12°N, 113.28°E) and were euthanized by CO₂ before their skins were subsequently sheared and preserved in liquid nitrogen. Total RNA of the F. multistriata frog skin was extracted with Trizol (Life Technologies, CA, United States) according to the manufacturer’s protocols and prepared as temples for PCR amplifications as previously reported by us (Chai et al., 2021). The sense primer (5′-GGA​TGA​AGG​TCT​GGC​AGT​GTG​TGC-3′) was used for the 5′ sequence PCR amplifications of cDNA as described in our previous study (Chai et al., 2021). Prediction of physical and chemical parameters of FM-CATH was done by the ExPASy-ProtParam tool (http://web.expasy.org/protparam/).

The peptide ordered from GL Biochem Ltd. (Shanghai, China) was purified to 95% with an Inertsil ODS-SP (C18) reverse-phase HPLC column (SHIMAZU, osumi, Japan) before being lyophilized and further confirmed by MALDI-TOF Mass Spectrometry.

Circular dichroism (CD) was conducted to identify the secondary structure and the stability of FM-CATH in solutions using Jasco-810 spectropolarimeter (Jasco, Tokyo, Japan) as described in our previous study (Zeng et al., 2018). For secondary structure investigation, peptide at the final concentration of 50 µM were dissolved in 0, 30, 60, 90, and 120 mM SDS solutions. For stability evaluation, FM-CATH (50 µM) dissolved in 60 mM SDS was incubated in 25, 37, 50, 70, and 90°C or 0, 100, 200, and 400 mM NaCl for 1 h before CD spectra measurement. To characterize the binding of peptide to bacteria polysaccharides, 0.2 mg/ml of LPS (L2880, Escherichia coli O55:B5, Sigma-Aldrich, St. Louis, Missouri, MO) and LTA (L2512, Staphylococcus aureus, Sigma-Aldrich, St. Louis, MO) were dissolved in H₂O or 30 mM SDS solution, respectively. FM-CATH was added to the sugar suspension at the final concentration of 50 μM for 1 h at room temperature. Binding of the peptide to LPS and LTA aggregates was studied by monitoring the change in the secondary structure of peptide (Nankar and Pande, 2013). CD data were expressed as the mean residue ellipticity (θ) of three consecutive scans per sample in deg cm²·dmol⁻¹.

The antimicrobial activity of FM-CATH was measured using two-fold dilution method as reported previously by us (Ye et al., 2020). Microorganisms bought from Guangdong Institute of Microbiology were cultured in Muller-Hinton (MH) broth at 37°C to exponential phase and subsequently diluted with fresh MH broth to reach 10⁶ CFU/ml of density. An equal volume of microbial inoculums in 96-well plates was incubated with different concentrations of FM-CATH at 37°C for 14 h. The absorbance at 600 nm was measured by a microplate spectrophotometer (Infinite M1000 Pro, Tecan Company, Switzerland) to calculate the minimal inhibitory concentration (MIC) values. To measure the inhibitory effect of LPS and LTA on the antimicrobial activity of FM-CATH, LPS or LTA at final concentration of 0.2 mg/ ml was pre-incubation with FM-CATH for 1 h at room temperature before MIC values were measured. Ampicillin and polymyxin B were used as positive controls.

The salt, thermal and serum stabilities of FM-CATH were investigated with antimicrobial assays as described previously by us (Zeng et al., 2018). Briefly, the activities of FM-CATH against E. coli ATCC 25922 were measured after peptide incubated with 0, 50, 100, 150, 200, and 400 mM NaCl at room temperature or at 25, 37, 50, 70 and 90°C for 1 h. For serum stability, FM-CATH solubilized in saline was incubated with human serum in a volume ratio of 1:4 for 0–6 h at 37°C before their MICs were determined at 0, 1, 2, 4, and 6 h. All experiments were repeated at least three times.

The agglutination assay was performed with S. aureus ATCC 25923 and E. coli ATCC 25922. Bacteria at exponential phase were harvested, washed twice and diluted to 10⁹ CFU/ml of density with fresh MH broth. The microorganisms were treated with BSA in TBS, FM-CATH (25 μM), FM-CATH (25 μM) plus equal volume of 0.2 mg/ ml LPS or 0.2 mg/ ml LTA for 30 min at room temperature. The incubation solution was dropped on a glass slide and dyed with a Gram staining kit (Solarbio Technology Co., Ltd., Beijing), and the result was observed under an oil microscope (Nikon Corporation, Japan).

The cytotoxicity of FM-CATH on different mammal cells was detected by MTT. Three tumor cell lines (A549, HepG2, and MDA-MB-231) and two normal mammalian cell lines (MH-S and MDCK) were seeded in 96-well plates at a density of 5,000 cells per well. Cells were grown in DMEM or RPMI 1640 medium, or medium containing continuous concentrations of FM-CATH (2.5, 5, 10, 20, 40 μM) at 37°C for 24 h before MTT was added in the dark and the culture was continued for 4 h. The supernatant was removed and DMSO was added before the absorbance at 490 nm was measured. Hemolytic assay was carried out with fresh blood from mouse heart as reported in our previous article with minor modification (Chai et al., 2021). In short, 2% of mouse erythrocyte suspension in TBS solution (v/v) was treated with 2.5, 5, 10, 20, and 40 μM FM-CATH in a 96-V-well plate at room temperature for 2 h. 1% Triton X-100 and PBS were applied as the positive and negative control, respectively. The absorbance of the supernatant at 540 nm was measured with a microplate spectrophotometer. The hemolysis rate was calculated using the following formula: percentage hemolysis = (OD_sample—OD_PBS)/(OD_Triton—OD_PBS) × 100%.

ITC assay was carried out to investigate the thermodynamics of interactions between peptides and LPS or LTA using MicroCal PEAQ-ITC (Malvern; United Kingdom). 50 mM of PBS (pH 7.2) was used to prepare stock solutions of peptide, LPS, and LTA. With a constant stirring of 250 rpm/s at 25°C, 1.5 μL aliquots of FM-CATH or polymyxin B at concentration of 1 mM were titrated into the sample cell filled with 280 μL of 50 μM LPS. In LTA binding experiment, 100 μM LTA in the syringe was titrated into 280 μL of 50 μM FM-CATH in the cell with the condition described above. Instrument was operated in high feedback mode. After the heats derived from dilution were subtracted, the equilibrium disassociation constant (K _D ) and the enthalpy change (ΔH) were analyzed by fitting to a single-site binding model using the MicroCal Origin software. The Gibb’s free energy change (ΔG) and entropy change (ΔS) were obtained from the basic thermodynamic equations, respectively.

PlexArrayTM HT A100 system (Plexera LLC, Bothell, Washington, United States) and bare gold SPRi chip (Nanocapture gold chip, with a gold layer or 47.5 nm thickness) was used to explore the real-time binding reaction of FM-CATH with LPS, LTA, and proteases as described in our previous method (Chai et al., 2021). In detail, FM-CATH (2 mM) and BSA dissolved in PBS were spotted in multiplex onto the gold chip surface and then stored at 4°C for 14 h in a humid box according to the instruction of the manufacturer. The SPRi chip was washed with PBS and blocked with 1 M ethanolamine/H₂O solution (pH 8.5) for 30 min, and then mounted in the instrument. Different concentrations of proteinase factors were flowed over the chip at the speed of 2 μL/ s 0.5% (v/v) H₃PO₄ in H₂O was added to regenerate the chip surface. Data were analyzed with the PLEXEA data analysis module and ORIGINLab software (OriginLab).

The chromogenic substrate assay was performed to identify effects of FM-CATH on blood clotting factors. In brief, FM-CATH at final concentrations of 12.5, 25, and 50 μM was mixed with thrombin, plasmin, β-tryptase, tPA for 10 min at room temperature before chromogenic substrates (S2238 for thrombin and β-tryptase; S2302 for plasmin; S2288 for tPA) were added into the mixtures. Substrate hydrolysis at 37°C was measure by reading the absorbance 405 nm at 1-min time intervals with microplate spectrophotometer.

The anticoagulant activity of FM-CATH was examined by the plasma recalcification time (PRT) and activated partial thromboplastin time (APTT) assays of normal human platelet-free plasma. For PRT measurement, 50 μL of the platelet-poor plasma was mixed with 50 μL of FM-CATH in a 96-well microplate and incubated at 37°C for 10 min before addition of 50 μL of 25 mM CaCl₂. Absorbance was measured at 405 nm at intervals of 1 min using microplate spectrophotometer. For APTT assay, 90 µL of the platelet-poor plasma was incubated with 10 µL of FM-CATH (final concentration 12.5, 25, and 50 µM) at 37°C for 10 min before 100 µL of pre-warmed APTT assay reagent (Shanghai yuanyeBio-Technology Co., Ltd, China) was added. The 200 µL of mixture was continued to incubate at 37°C for another 5 min before 100 µL of pre-warmed 25 mM CaCl₂ was added and the clotting time was recorded.

The bleeding time of the transected-tail mice was measured as reported in our previous paper (Ma et al., 2011). Briefly, mice were injected intravenously via the tail vein with saline or FM-CATH dissolved in saline (10 mg/ kg). After 1 h, the distal 2 mm segment of the tail was transected and vertically immersed into saline at 37°C. Bleeding time was calculated from bleeding to termination and the end point was the arrest of bleeding lasting for more than 30 s.

CLP was carried out as described by us with minor modifications (Chai et al., 2021). In brief, six-week-old BALB/c mice of either gender were randomly grouped to sham, CLP, and FM-CATH-treated groups (n = 6). Each group was weighed and anesthetized with ketamine and xylazine. The cecum was isolated under a sterile environment and ligated at 1.0 cm from its tip before a single puncture was performed in the middle of the ligated cecum with a 20-gauge needle. Subsequently, small amount of feces was extruded and the cecum was returned to the abdomen before the abdomen was sewn up. Sham controls underwent the same surgical procedure but CLP was not performed. Immediately after surgery, percutaneous injection of 1  ml of pre-warmed saline was done into the recovering mice. 2 h after the sham or CLP procedure, mice were intraperitoneally administrated with FM-CATH (10 mg/ kg) or saline, and then the survival rate was recorded every 12 h for up to 3 days. Another set of mice were prepared as described above and the serum, bronchoalveolar lavage fluid (BALF), peritoneal lavage fluid and lung tissues were collected 24 h after the CLP operation for histopathological analysis, pathological scoring, wet/dry weight ratio, bacterial colony forming unit (CFU) counts, cytokine and signaling pathway assays.

All data were presented as mean ± SEM. Data were analyzed using one-way ANOVA. ^* p < 0.05, ^** p < 0.01, and ^*** p < 0.001 were considered statistically significant as compared to control.

The cDNA encoding one novel antimicrobial peptide named FM-CATH was obtained by PCR-based cDNA cloning method from the cDNA library of F. multistriata. The complete nucleotide sequence and translated amino acid sequence of FM-CATH precursor were shown in Figure 1. The cDNA sequence encoding FM-CATH was composed of 557 bp and the deduced amino acid precursor contained 148 amino acids. The sequence of mature FM-CATH was LKTKALNKLKQKLQAVGNLIGSVIKG which displayed high sequence similarities with other representative cathelicidins. FM-CATH had a theoretical PI of 10.70 with + 6 net charge and GRAVY index was -0.004. Its relative mass was measured to 2762.37.

The secondary structure of FM-CATH as well as its stability in different solution environments were investigated through CD. As presented in Figure 2A, a small negative peak at 198 nm was observed in the CD spectra of H₂O-dissolved peptide, suggesting that FM-CATH might contain random coil construction. Meanwhile, while dissolved in membrane-like SDS solution, the CD spectra presented a big positive peak at 192 nm and two small negative peaks at 208 and 222 nm, suggesting the presence of α-helix in the secondary structure of FM-CATH. In addition, some slight changes in the secondary structure components of FM-CATH were also observed in different concentration of SDS solution (Table 1). Furthermore, the CD spectra of FM-CATH treated with different temperatures also displayed highly identical peaks (Figure 2B). However, though it showed some changes in various NaCl concentrations (100, 200 and 400 mM), the α-helix structure of FM-CATH was preserved (Figure 2C). Peptides binding to LPS/LTA are known to adopt significant α-helical structures (Nankar and Pande, 2013). To determinate the LPS- and LTA-binding ability of FM-CATH, peptide was individually incubated with the LPS or LTA dissolved in H₂O or 30 mM SDS solution before the CD spectra were determined, respectively. As presented in Figure 2D, in the presence of LPS or LTA, FM-CATH showed obviously different CD spectra which are features of α-helical structure including two negative bands at 222 and 208 nm and a positive band at 192 nm, suggesting FM-CATH binds to LPS and LTA.

The MICs of FM-CATH against Gram-positive bacteria, Gram-negative bacteria and fungi were presented in Table 2. Under our condition, FM-CATH could suppress the growth of E. coli ATCC 25922, P. acnes ATCC 6919, B. subtilis CMCC 63501, C. albicans ATCC 10231, and P. aeruginosa ATCC 27853 within a range of 6.25–50 μM, which were more potent or equal to those of polymyxin B. However, its antimicrobial activity against S. aureus ATCC 25923 was more than 100 µM. Moreover, the antimicrobial activity of FM-CATH against E. coli ATCC 25922 was abrogated after its incubation with LPS and LTA. However, our result showed that the antibacterial activity of polymyxin B at 100 µM was not inhibited in the presence of LPS (0.2 mg/ ml), which is line with the report by Krishnakumari et al. (Krishnakumari et al., 2020).

The antimicrobial ability of many AMPs is usually affected by temperature condition, salt concentration, and serum components like proteases (Travis et al., 2000). To further confirm the results from CD experiment, the antimicrobial activity of FM-CATH under different NaCl concentrations and temperature conditions was investigated. As shown in Figure 3, FM-CATH maintained the antibacterial activity against E. coli ATCC 25922 at the tested temperatures, NaCl solutions and serum. Although the antibacterial activity decreased after incubation with 400 mM NaCl, solubilization in saline at 90°C for 1 h or human serum (1:4 (v/v)) at 37°C for 6 h, FM-CATH still possessed antibacterial activity. The results indicated that FM-CATH has good stability.

The bacterial agglutination of FM-CATH against S. aureus ATCC 25923 and E. coli ATCC 25922 was tested. As illustrated in Figure 4A, agglutination occurred in the presence of FM-CATH (25 μM) after 30 min of incubation. However, it was totally abolished by LPS and LTA. ITC experiment was next performed to investigate the binding of FM-CATH to LPS and LTA, respectively (Figures 4B,D). Binding saturation occurred at 12 min while the molar ration of peptide to LPS reach to 20 while the molar ration of LTA to peptide reach to 2at 40 min, respectively. Furtherly, LPS and LTA bound to FM-CATH with K _D values of about 4.90 μM and 0.714 nM, respectively (Table 3). However, as shown in Figure 4C and Table 3, the K _D value for polymyxin B binding to LPS was about 0.414 μM, indicating that the binding ability of polymyxin B to LPS was stronger than that of FM-CATH to LPS. SPRi is an optical method monitoring and quantifying biomolecular interactions (Homola, 2008). In agreement, there was strong SPRi signal when LPS and LTA were flowed through the gold chip containing FM-CATH and BSA (Figures 4E,F). Resonance units increased in a concentration-dependent manner and the K _D values of FM-CATH binding to LPS and LTA were 2.57 μM and 6.21 nM, which was relatively consistent with the ITC results. FM-CATH could thus directly bind to both LPS and LTA.

The application of AMPs is usually limited due to their cytotoxicity to erythrocytes and mammalian cells (Gajski et al., 2016). The cytotoxicity of FM-CATH against different cells were measured by MTT. As illustrated in Figures 5A–E, FM-CATH concentration-dependently inhibited the proliferation of A549, HepG2 and MDCK cells and the IC₅₀ values were 11.74, 33.40 and 336.8 μM, respectively. However, it showed very low cytotoxicity against MH-S and MDA-MB-231. Accordingly, we speculated that FM-CATH exerts strong killing activity against some of cancer cells and a low toxicity to normal mammalian cells in vitro. Additionally, the hemolytic activity of FM-CATH was measured using mouse heart blood. The hemolysis rates of mouse heart blood at different concentration of FM-CATH were presented in Figure 5F and at the highest concentration, 40 μM, FM-CATH had almost no hemolytic activity.

To explore whether FM-CATH affected the activities of proteases, its binding reactions were measured by SPRi. As illustrated in Figure 6A, FM-CATH could bind to thrombin, plasmin, β-tryptase, and tPA in a dose-dependent manner but not BSA. In addition, chromogenic substrate assay revealed that FM-CATH increased the activity of tPA while suppressed the one of thrombin, plasmin, β-tryptase (Figure 6B). Given that FM-CATH reacted with serine proteases related to the coagulation cascade, the activity of FM-CATH on plasma was further investigated by PRT and APTT measurements. FM-CATH had ability to prolonged PRT and APTT in a dose-dependent manner, suggesting that coagulation could be inhibited by FM-CATH (Figures 6C,D). To clarify whether FM-CATH had anticoagulant effect in vivo, tail bleeding was observed in mice. As shown in Figure 6E, the duration of tail bleeding in the saline-treated group was 3.03 ± 0.52 min. However, the tail bleeding time in the group treated by 10 mg/ kg FM-CATH at a dose of was markedly prolonged to 14.29 ± 4.12 min. The positive control, Tablysin-15, which was described in our previous study (Ma et al., 2011), also significantly extended the tail bleeding time to 10.57 ± 2.67 min at a dose of 2.5 mg/ kg.

To assess the effect of FM-CATH on polymicrobial sepsis, CLP-induced sepsis mice model was established. Interestingly, FM-CATH obviously improved the prognosis of the septic mice. The survival rate at the assay end point had increased to 43% in the group intraperitoneally injected with FM-CATH while 0% in the control CLP group (Figure 7A). Histological staining of the lung tissue with hematoxylin and eosin in the CLP group showed a severe organ damage including the increase of alveolar wall thickness and inflammatory cell infiltration accompanied by destruction of the alveolar structures and reductions in the alveolar spaces (Figure 7B , panel b). It was found that administration of FM-CATH protected against CLP-induced organ damage in the lung (Figure 7B , panel c) and lung damage scores were obviously lower than those in the CLP model group (Figure 7C). In line with the above data, the lung wet/dry ratio and BALF total protein level were clearly increased after CLP induction and FM-CATH effectively decreased their levels (Figures 7D,E). Furthermore, MPO activity of lung was evidently abated after treatment with FM-CATH (Figure 7F). Consistently, the FM-CATH treatment also definitely cut down the contents of the inflammatory cytokines IL-1β, IL-6 and TNF-α in the lung tissue (Figures 7G–I) and serum (Figures 7J–L). It was also confirmed that the FM-CATH treatment significantly decreased the contents of some blood biochemical factors like aspartate aminotransferase (AST), urea nitrogen (BUN) as well as alanine aminotransferase (ALT) (Figures 7M–O). Additionally, the FM-CATH treatment reduced the bacterial colony forming unit (CFU) counts in peritoneal fluid (Figure 7P). To further explore the protective mechanism of FM-CATH in CLP-induced sepsis, the expression of MAPK signaling in lung was investigated by western blotting. The activation of ERK, JNK, and p38 were markedly enhanced after CLP administration, which was significantly alleviated by post-treatment with FM-CATH (Figures 7Q,R). Altogether, these data demonstrated that the FM-CATH treatment can significantly increase survival rate, and decrease disease severity in CLP-induced septic mice.