Anti-infective Effects of a Fish-Derived Antimicrobial Peptide Against Drug-Resistant Bacteria and Its Synergistic Effects With Antibiotic

Antimicrobial peptides (AMPs) play pivotal roles in protecting against microbial infection in fish. However, AMPs from topmouth culter (Erythroculter ilishaeformis) are rarely known. In our study, we isolated an AMP from the head kidney of topmouth culter, which belonged to liver-expressed antimicrobial peptide 2 (LEAP-2) family. Topmouth culter LEAP-2 showed inhibitory effects on aquatic bacterial growth, including antibiotic-resistant bacteria, with minimal inhibitory concentration values ranging from 18.75 to 150 μg/ml. It was lethal for Aeromonas hydrophila (resistant to ampicillin), and took less than 60 min to kill A. hydrophila at a concentration of 5 × MIC. Scanning electron microscope (SEM) and SYTOX Green uptake assay indicated that it impaired the integrity of bacterial membrane by eliciting pore formation, thereby increasing the permeabilization of bacterial membrane. In addition, it showed none inducible drug resistance to aquatic bacteria. Interestingly, it efficiently delayed ampicillin-induced drug resistance in Vibrio parahaemolyticus (sensitive to ampicillin) and sensitized ampicillin-resistant bacteria to ampicillin. The chequerboard assay indicated that topmouth culter LEAP-2 generated synergistic effects with ampicillin, indicating the combinational usage potential of topmouth culter LEAP-2 with antibiotics. As expected, topmouth culter LEAP-2 significantly alleviated ampicillin-resistant A. hydrophila infection in vivo, and enhanced the therapeutic efficacy of ampicillin against A. hydrophila in vivo. Our findings provide a fish innate immune system-derived peptide candidate for the substitute of antibiotics and highlight its potential for application in antibiotic-resistant bacterial infection in aquaculture industry.


INTRODUCTION
To improve the dietary protein ingestion for the increasing world population, the world aquaculture is experiencing a rapid expansion, and fish farming comprises an important source of protein demands for people (Yousefi et al., 2018). In order to achieve a better profitability, fish culture systems have been changed into more complicated systems with higher density that corresponds to a higher production level (Yousefi et al., 2018). However, the intensive culture of fish simultaneously brings more and more fish diseases, which cause severe economic losses (Xiong et al., 2019). Especially, bacterial infection comprises one of the major fish diseases . Up to now, the usage of antibiotics is still the most favorable strategy to control bacterial infection in fish . The wide usage of antibiotics has resulted in the increasing spread of antibiotic-resistant bacteria in farmed fish and antibiotic residues in farmed fish food. Considering the strict regulations for administration of antibiotic for fish disease treatment, we are obligated to develop novel strategies to control the emergence of antibiotic-resistant bacteria in farmed fish and antibiotic residues in farmed fish food. Therefore, many researchers have explored the potential application of various types of substitutes for antibiotics over the past decades. Among them, antimicrobial peptides (AMPs) gained more and more research interests because of their broad antimicrobial activities, effective immunomodulatory activities, and none induced drug resistance risk.
AMPs, also known as host defense peptides (HDPs), are pivotal molecules of innate immunity that provide the first line of host defense against microbial invasion (Zasloff, 2002(Zasloff, , 2019. AMPs are widely expressed in vertebrates, including fishes, amphibians, reptiles, birds and mammals . As a lower vertebrate, the innate immune system is a fundamental defense mechanism of fish (Shabir et al., 2018), and it serves as a bridge between innate and adaptive immunity, which is considered as an excellent model for innate immunity and comparative immunology investigations (Liang et al., 2013;Zhu et al., 2013). So far, several families of AMPs have been characterized from fishes, including piscidins, cathelicidins, defensins, and liver-expressed antimicrobial peptide 2 (LEAP-2, also known as hepcidins; Shabir et al., 2018). Among these AMPs, piscidins, and cathelicidins are linear peptides, while defensins and LEAP-2 contain intramolecular disulfide bonds. They were found to disrupt pathogenic bacterial membranes . In addition, they were also observed to regulate innate immune response . These properties of AMPs effectively reduce the possibility of development of resistance against bacteria.
Topmouth culter (Erythroculter ilishaeformis) is a kind of freshwater fish belonging to the genera of Culter fishes (Dong et al., 2017). It is one of the important economical fishes in Taihu Lake, which is the third largest freshwater lake in China (Dong et al., 2018). Topmouth culter, whitebait (Neosalanx taihuensis), and white shrimp (Fenneropenaeus chinensis) are well-known as "three whites" of Taihu Lake that provide a lot of dietary protein for people (Dong et al., 2017). Due to overfishing, the wild topmouth culter is almost exhausted. Recently, topmouth culter exhibits great cultural potential with rapid growth rate. However, research on the innate immunity of topmouth culter is rarely conducted. In this study, we isolated and cloned an AMP (LEAP-2) from topmouth culter. The effects of topmouth culter LEAP-2 on aquatic pathogenic bacteria were tested by minimal inhibitory concentration (MIC) assay, bacterial killing kinetics assay, minimal inhibitory concentration (SEM) observation, and SYTOX Green uptake assay. The possibility of inducible drug resistance of topmouth culter LEAP-2 against aquatic pathogenic bacteria was evaluated. The combinational usage of topmouth culter LEAP-2 and traditional antibiotic was investigated both in vitro and in vivo. Our findings provide a topmouth culter-derived AMP for controlling antibiotic-resistant bacterial infection-induced diseases in cultured fishes.

Animals and Ethics Approval
Healthy topmouth culter (E. ilishaeformis; body weight 350-450 g) were purchased from a commercial farm in Suzhou, China. Topmouth culter were reared in circulating filtered water tanks at the temperature of at 20-25°C. All surgery of topmouth culter were performed after they were anesthetized in 0.2% tricaine and euthanized by incubating in ice water for 15 min. C57BL/6 mice (female, 18-20 g) were purchased from Shanghai Slac Animal Co. Inc. and housed in a pathogen-free facility. All surgery of mice were performed under sodium pentobarbital anesthesia with minimum fear, anxiety, and pain. Animal experiments were approved by the Animal Care and Use Committee and the Ethical Committee of Soochow University (SYXK2017-0043).

Peptide Purification
After healthy topmouth culter (E. ilishaeformis) were anesthetized in 0.2% tricaine and euthanized via incubation in ice water for 15 min, head kidneys were collected and homogenized in 0.1 M phosphate buffer (PBS; pH 6.0, 1 g tissue/ml) containing 1% (v/v) protease inhibitor cocktail (Sigma-Aldrich). The homogenates were centrifuged at 12,000 × g for 10 min, and the supernatants were harvested and lyophilized. Lyophilized samples were dissolved in 9 ml phosphate buffer (0.1 M, pH 6.0, OD280 = 56.3). The dissolved samples were subjected to a Sephadex G-50 (Superfine, Amersham Biosciences, 2.6 cm × 100 cm) gel filtration column. phosphate buffer (0.1 M, pH 6.0) was used as eluted buffer at a flow rate of 3.0 ml/10 min. The eluted fractions were measured at 280 nm, and the fractions with antimicrobial activity were pooled. The fractions with antimicrobial activity were then subjected to a C18 reversed-phase high-performance liquid chromatography column (RP-HPLC, 5 μm particle size, 110 Å pore size, 250 mm × 4.6 mm, Gemini, CA, United States) using a linear gradient of 0-60% acetonitrile supplemented with 0.1% (v/v) trifluoroacetic acid/water. The eluted peaks were collected for antimicrobial assay, and the eluted peak with antimicrobial activity was spotted onto a matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) plate for purity assay. The purified peptide was applied to a protein sequencer (PPSQ-31A; Shimadzu, Kyoto, Japan) following manufacturer's instruction.

cDNA Cloning
Total RNA from the head kidney of topmouth culter was extracted using Trizol reagent (Life Tech, United States). Head kidney cDNA library was constructed using a SMART ™ PCR cDNA synthesis kit (Clontech, CA). Two primers, a sense primer (5' PCR primer, 5'-AAGCAGTGGTATCAACGCAGAGT-3', provided by the cDNA library construction kit) and an antisense primer (S1(5'-A(A/G)CAT(G/A/T)ATIC(G/T)CCA(A/G/C/T)A(A/G) (A/G/C/T)GG-3'), designed from the amino acid sequence of topmouth culter LEAP-2 obtained by Edman degradation), were used to screen the 5' fragment of cDNA encoding topmouth culter LEAP-2. Then, a sense primer (5'-ATGCAGACCCAC CCCAACAG-3', designed according to the 5' fragment of cDNA) and an antisense primer (3' PCR primer, 5'-ATTCTAGAG GCCGAGGCGGCCGACATG-3', provided by the cDNA library construction kit) were used to screen the full-length cDNA that encodes topmouth culter LEAP-2. The PCR conditions were 2 min at 95°C, and 28 cycles of 10 s at 92°C, 30 s at 56°C, and 30 s at 72°C, and followed by 10 min extension at 72°C. The PCR products were cloned into pGEM-T Easy vector (Promega, Madison, WI, United states), and positive clones were selected for DNA sequencing performed by Genewiz Co. Ltd. (Suzhou, China).

Sequence Analysis
The deduced amino acid sequence of topmouth culter LEAP-2 precursor was translated from the cDNA sequence using the ExPASy Translate Tool. 1 Blast search was performed with Blastx 1 http://web.expasy.org/translate/ provided by NCBI. 2 Multiple alignment of amino acid sequence of topmouth culter LEAP-2 with its homologues was performed with ClustalW. 3 Theoretical isoelectric point (pI) and molecular weight (MW) were calculated by an online tool 4 (Wei et al., 2015).

Antimicrobial Assay
A standard 2-fold broth microdilution method was used to evaluate the antimicrobial activity of topmouth culter LEAP-2 against aquatic pathogenic bacteria. Aquatic pathogenic bacteria were diluted with fresh nutrient broth (Oxoid, UK) to 10 5 CFU/ ml after bacteria were cultured to exponential phase in nutrient broth. A series of topmouth culter LEAP-2 dilutions were prepared in 96-well plates (50 μl/well), and an equal volume of bacterial dilution was added. The plates were cultured at 37°C for 18 h. Ampicillin was used as positive control in antimicrobial assay. The MIC at which no visible bacterial growth occurred was recorded as MIC value .

Scanning Electron Microscope Assay
To examine if topmouth culter LEAP-2 impaired the bacterial surface morphology, A. hydrophila were cultured in nutrient broth to exponential phase, washed with PBS, and suspended in PBS (about 5 × 10 6 CFU/ml). Topmouth culter LEAP-2 (5 × MIC) was added into the bacterial suspension and incubated at 37°C for 30 min. After centrifuged at 1,000 × g for 10 min, A. hydrophila were fixed for SEM assay following standard operating conditions. The surface morphology was observed with a Hitachi SU8010 SEM (Japan) following the manufacturer's instruction.

Drug Resistance Test
Vibrio parahaemolyticus were sensitive to ampicillin, and were selected for drug resistance test against topmouth culter LEAP-2 and ampicillin. Briefly, drug resistance was induced in V. parahaemolyticus by repeated treatment with ampicillin or topmouth culter LEAP-2 for 10 passages and tested via MIC measurement. The same volume of vehicle (PBS) was used as control. For each passage, V. parahaemolyticus were exposed to a sub-MIC concentration of ampicillin or topmouth culter LEAP-2 (1/8 MIC at that particular passage) until growing to the log phase, and the MIC values of topmouth culter LEAP-2 and ampicillin against V. parahaemolyticus were determined as mentioned above.

Combinational Usage of Topmouth Culter LEAP-2 With Ampicillin
To explore the potential of the combinational usage of topmouth culter LEAP-2 with antibiotic, we tested the ampicillin-induced drug resistance in V. parahaemolyticus as mentioned above in the presence of topmouth culter LEAP-2. Briefly, V. parahaemolyticus were passaged for 10 times in the presence of ampicillin and topmouth culter LEAP-2. For each passage, V. parahaemolyticus was exposed to a sub-MIC concentration of ampicillin (1/8 MIC at that particular passage) and a constant concentration of topmouth culter LEAP-2 (18.75 μg/ml, 1/4 MIC of LEAP-2 against V. parahaemolyticus) until growing to the log phase, and the MIC values of ampicillin against V. parahaemolyticus were determined via MIC assay.
In addition, we also tested the MIC values of ampicillin against aquatic pathogenic bacteria (including ampicillin-sensitive and ampicillin-resistant strains) in the presence of topmouth culter LEAP-2 at a concentration of 1/4 MIC of LEAP-2 against respective bacterial strain.

Chequerboard Assay
Interactive interactions between LEAP-2 and ampicillin were tested by a chequerboard assay according to the method described previously . In brief, a series of 2-fold LEAP-2 or ampicillin dilutions were prepared, and equal volume of LEAP-2 (50 μl/well) and LEAP-2 (50 μl/well) were mixed in a well of 96-well plate. Then bacteria (10 5 CFU/ml, 100 μl/well) was added and cultured at 37°C for 18 h bacterial growth was recorded by a microplate reader at 600 nm.

In vivo Antimicrobial Assay
The anti-infective effect against antibiotic-resistant bacteria of topmouth culter LEAP-2 in fish was tested according to method described previously with slight modification . Topmouth culter were randomly divided into four groups, and were intraperitoneally challenged with A. hydrophila (100 μl, 2 × 10 7 CFU/ml). After bacterial challenge, topmouth culter in each group were immediately intraperitoneally administrated with LEAP-2 (10 mg/kg), ampicillin (10 mg/kg), LEAP-2 and ampicillin used in combination (10 mg/kg each), or an equal volume of PBS (vehicle) respectively. At 6 h post infection, peritoneal lavage was collected by intraperitoneal injection of 5 ml PBS. Its anti-infective effect against antibiotic-resistant bacteria was also evaluated in mice as described previously . C57BL/6 mice (female, 18-20 g, n = 6) were intraperitoneally challenged with A. hydrophila (resistant to ampicillin, 2 × 10 7 CFUs/mouse). Topmouth culter LEAP-2 (10 mg/kg) or an equal volume of PBS (vehicle) was intraperitoneally administered into mice post bacterial infection. At 18 h post infection, peritoneal lavage was collected by intraperitoneal injection of 2 ml PBS. A series of 10-fold dilutions of fish and mouse peritoneal lavage were prepared, and 50 μl of different dilutions was plated on nutrient broth agar plates. The CFUs were counted after the agar plates were cultured at 37°C for 18 h.

Synthetic Peptides
Synthetic peptide was purchased from Synpeptide Co. Ltd. (Shanghai, China). The purity of the synthetic peptide was analyzed by RP-HPLC and MALDI-TOF MS.

Statistical Analysis
Statistical analysis was performed using Student's t-tests or one-way ANOVA provided by GraphPad Prism software (GraphPad Software Inc., La Jolla, CA, United States). Data were presented as mean ± standard deviation from three independent experiments. p < 0.05 was considered as statistically significant of difference between the groups.

A Novel AMP, LEAP-2, Was Isolated From the Head Kidney of Topmouth Culter
To isolate topmouth culter (E. ilishaeformis)-derived AMPs, head kidney, an important immune organ of topmouth culter, was collected and homogenized. The condensed supernatants of head kidney homogenates were first separated by Sephadex G-50 gel filtration. As shown in Figure 1A, the condensed supernatants were separated into five fractions, and the fraction containing antimicrobial activity was indicated by an arrow. The antimicrobial fraction from Sephadex G-50 gel filtration was then subjected to a C18 RP-HPLC column ( Figure 1B). The eluted peak containing antimicrobial activity was marked with an arrow. The purified AMP was subjected to mass spectrometry analysis and had an observed MW of 4652.17 Da ( Table 1). The N-terminal amino acid sequence of the purified peptide was determined as MTPLWRIMLLFKPHALCQNNY by Edman degradation.

The Purified AMP From Topmouth Culter Belongs to LEAP-2 Family
According to N-terminal amino acid sequence of the purified peptide by Edman degradation, we designed primer and tried to clone the cDNA sequence from the head kidney of topmouth culter, and we successfully obtained the open reading frame of the purified AMP (Figure 2). The cDNA sequence that encoded the precursor of the purified AMP contained 522 base pairs (Figure 2; GenBank accession number, MW199736). The precursor of topmouth culter LEAP-2 was composed of 92 amino acid residues. Blast search indicated that the precursor is a member of LEAP-2 AMP family that contained a signal peptide (italic), a prodomain (bold), and a mature peptide (underlined; Figure 2). The deduced mature peptide of topmouth culter LEAP-2 is composed of 41 amino acids (Figure 2, Table 1), and the N-terminal amino acid sequence of the deduced mature LEAP-2 is consistent with that (MTPLWRIMLLFKPHALCQNNY) obtained by Edman degradation. The mature peptide of topmouth culter LEAP-2 has a theoretical MW of 4652.56 Da, net charges of +3, and a theoretical isoelectric point of 8.91 (Table 1). The theoretical MW of mature LEAP-2 matched well with the observed MW. The alignment of topmouth culter LEAP-2 precursor with selected cyprinid fish LEAP-2 precursors indicated that there are a conserved "RXXR" amino acid sequence before the cleavage site separating the prodomain and the mature LEAP-2 peptide, a conserved "MTPLWR" amino acid sequence at the N-terminal of mature LEAP-2 peptide, and a conserved "RXGH" amino acid sequence at the C-terminal of mature LEAP-2 peptide as described previously (Figure 3; . In addition, the mature LEAP-2 peptide of topmouth culter contains four conserved cysteine residues that form two intramolecular disulfide bonds, and the four cysteine residues are probably linked in Cys1-Cys2 and Cys2-Cys4 patterns as described previously (Figure 3; .

Topmouth Culter LEAP-2 Showed Antimicrobial Activity Against Aquatic Pathogenic Bacteria
To investigate the antimicrobial effects of LEAP-2, we detected the MIC values of topmouth culter LEAP-2 against aquatic pathogenic bacteria. As shown in Table 2, all the tested aquatic bacterial strains were sensitive to topmouth culter LEAP-2, and the MIC values ranged from 18.75 to 150 μg/ml. Among these tested aquatic bacterial strains, A. sobria, A. hydrophila, V. anguillarum, V. vulnificus, V. splendidus, and V. cholerae were resistant to ampicillin with MIC values higher than 200 μg/ml, but these ampicillin-resistant aquatic bacterial strains were all sensitive to topmouth culter LEAP-2 with MIC values ranging from 18.75 to 150 μg/ml. These results revealed that topmouth culter LEAP-2 showed antimicrobial activity against aquatic pathogenic bacteria, including antibioticresistant bacterial strains.

Topmouth Culter LEAP-2 Had a Rapid Bactericidal Speed
To evaluate the antimicrobial efficacy of topmouth culter LEAP-2, its bacterial killing kinetics against A. hydrophila (an ampicillin resistant strain, MIC > 200 μg/ml) was evaluated. As listed in Table 3, topmouth culter LEAP-2 rapidly inhibited the bacterial growth as compared to PBS. Topmouth culter LEAP-2 (5 × MIC, 93.75 μg/ml, equal to 20.2 μM) took less than 60 min to kill all the bacteria, and it was proved to be lethal for A. hydrophila since A. hydrophila were not capable of resuming growth on agar plates. However, ampicillin (1 mg/ml, equal to 2862.0 μM) could not completely kill A. hydrophila until incubation for 180 min. On contrary, A. hydrophila dramatically growed after the addition of ampicillin, and the CFUs increased from 6.7 × 10 4 to 9.6 × 10 5 after the bacteria were incubated with ampicillin (1 mg/ml) for 180 min, which again indicated that this strain of A. hydrophila is resistant to ampicillin. These results suggested that topmouth culter LEAP-2 had a rapid bactericidal speed against antibiotic-resistant aquatic pathogenic bacteria.  Antimicrobial peptides were usually membrane-active agents, which lead to the alteration of bacterial membrane morphology (Khara et al., 2014). To see if topmouth culter LEAP-2 altered the membrane morphology of aquatic pathogenic bacteria, we observed the surface morphology of A. hydrophila by SEM after bacteria were exposed to PBS (vehicle, control) or topmouth culter LEAP-2. As shown in Figure 4, PBS-exposed A. hydrophila exhibited a regular, smooth, and intact surface ( Figure 4A). Whereas LEAP-2-exposed A. hydrophila showed a significant change of surface morphology with raised vesicle and pore formation in the surface, and the bacterial surface was covered with irregular debris (Figure 4B). The results indicated that topmouth culter LEAP-2 exhibited direct antimicrobial activity by membrane-disruption mechanism. FIGURE 3 | Multiple alignment of the amino acid sequences of topmouth culter LEAP-2 and its homologues. The identical sites ( * ), conserved sites (:), and less conserved sites (.) were indicated in the multiple alignment of topmouth culter LEAP-2 with its homologues, the accession numbers of the selected homologues were listed following each sequence. The predicted cleavage site for the signal peptide and the cleavage site for mature peptide were marked with an arrow. The boxes indicated "RXXR," "MTPLWR," and "RXGH" motifs respectively. The four conserved cysteine residues in the mature peptide are bold. Two cysteine residues joined by a solid line represented a disulfide bond.

Topmouth Culter LEAP-2 Induced Bacterial Membrane Permeabilization
To further confirm the membrane-disruption mechanism of topmouth culter LEAP-2, A. hydrophila were incubated with LEAP-2 (1 × MIC, 18.75 μg/ml, equal to 4.0 μM) in the presence of SYTOX Green and the fluorescence intensity was monitored. Compared to PBS-treated bacteria, the fluorescence intensity dramatically increased in 16 min after the addition of topmouth culter LEAP-2, while no significant increment of fluorescence intensity was observed in ampicillin-treated (200 μg/ ml, equal to 572.4 μM) bacteria ( Figure 5). The results indicated that topmouth culter LEAP-2 treatment increased the uptake of fluorescent dye by A. hydrophila, and the bacterial nucleic acids were subsequently stained with SYTOX Green, resulting in an increment of fluorescence intensity. Combined with results observed by SEM assay, LEAP-2 was demonstrated to evidently cause pore formation on the bacterial membrane.
Topmouth Culter LEAP-2 Showed None Inducible Drug Resistance to Bacteria Antimicrobial peptides are less likely to induce bacterial drug resistance and were usually selected as promising candidates for peptide antibiotic development (Mwangi et al., 2019). We herein detected whether topmouth culter LEAP-2 induce drug resistance against aquatic pathogenic bacteria. An ampicillin sensitive strain, V. parahaemolyticus, was exposed to sub-therapeutic doses of topmouth culter LEAP-2 or ampicillin (control) over 10 passages to simulate drug resistance . As shown in Table 4, PBS (vehicle) did not affect the drug resistance of V. parahaemolyticus against topmouth culter LEAP-2 or ampicillin, and resistance of V. parahaemolyticus against topmouth culter LEAP-2 did not readily develop with sub-therapeutic treatment

Topmouth Culter LEAP-2 Delayed Ampicillin-Induced Bacterial Drug Resistance
The combinational usage of AMPs and traditional antibiotics might be a promising way to overcome the emergence of bacterial drug-resistance and generate a greater antimicrobial effect . Therefore, we detected ampicillin-induced drug resistance against V. parahaemolyticus as mentioned above in the presence or absence of a constant concentration of topmouth culter LEAP-2 (18.75 μg/ml, 1/4 MIC of LEAP-2 against V. parahaemolyticus). As shown in     Topmouth Culter LEAP-2 Showed Synergistic Effects With Ampicillin As mentioned above, topmouth culter LEAP-2 could delay ampicillin-induced drug resistance and sensitize drug-resistant aquatic pathogenic bacteria to ampicillin, indicating that topmouth culter LEAP-2 might have an additive effect or synergistic effect with ampicillin. To investigate which effect was involved in the interaction of topmouth culter LEAP-2 with ampicillin, a chequerboard assay was performed as shown in Table 7. The combinational usage of topmouth culter LEAP-2 and ampicillin generated an FICI value of 0.375 against V. harveyi, and an FICI value of 0.5 against V. parahaemolyticus, indicating that topmouth culter LEAP-2 exhibited synergistic effects with ampicillin. Of note, peptide-ampicillin used in combination produced a lower minimum effective concentration for each agent. Combined with the membrane-disrupted action of topmouth culter LEAP-2 mentioned above, the synergistic effects between LEAP-2 and ampicillin is likely attributed to the LEAP-2-mediated permeation of ampicillin from outer membrane to cytoplasmic targets.

Topmouth Culter LEAP-2 Alleviated Ampicillin-Resistant Bacterial Infection and Enhanced the Therapeutic Efficacy of Ampicillin Against Antibiotic-Resistant Bacteria in vivo
The in vitro experiments indicated that topmouth culter LEAP-2 had an excellent combinational usage potential in treating drug-resistant bacterial infection. We next evaluated its antiinfective effects against A. hydrophila (resistant to ampicillin) combined with ampicillin in vivo. Compared with PBS-treated topmouth culter, the intraperitoneal bacterial loads of topmouth culter were significantly reduced after the administration of LEAP-2 (10 mg/kg), and the CFUs in the peritoneal lavage of topmouth culter were reduced by 47.8% ( Figure 6A).
While ampicillin (10 mg/kg) did not significantly reduce the intraperitoneal bacterial loads of topmouth culter ( Figure 6A). As expected, the combinational usage of topmouth culter LEAP-2 (10 mg/kg) with ampicillin (10 mg/kg) showed the best efficacy against ampicillin-resistant A. hydrophila infection, and the CFUs in the peritoneal lavage of topmouth culter were reduced by 67.2% (Figure 6A). A similar result was observed in mice. Topmouth culter LEAP-2 (10 mg/kg) significantly reduced the loads of ampicillin-resistant A. hydrophila in mouse peritoneal lavage, reducing about 46.9% CFUs (Figure 6B). But ampicillin (10 mg/kg) did not markedly affect the bacterial loads in mice ( Figure 6B). Whereas the combinational injection of topmouth culter LEAP-2 (10 mg/kg) and ampicillin (10 mg/kg) exhibited the best therapeutic efficacy among these groups, reducing about 72.1% CFUs (Figure 6B). These results suggested that topmouth culter LEAP-2 markedly alleviated ampicillin-resistant bacterial infection in vivo and enhanced the therapeutic efficacy of antibiotic against antibiotic-resistant bacterial infection in vivo.

DISCUSSION
In recent years, the research activities about AMPs have gained many achievements throughout the world. Due to the small MW and simple molecular structure of these peptides, it is easy to produce these peptides with low production cost. Therefore, more and more researchers are trying their best to develop peptide drugs derived from AMPs. In mammals, AMPs showed diverse biologic activities such as enodotoxin neutralization, anti-inflammation, chemotaxis of leukocytes, promotion of wound healing, and induction of angiogenesis (Hancock et al., 2016). They have been regarded as very attractive therapeutic molecules. Especially, AMPs have been considered as most promising agents for production of new generation antibiotics. Many peptide antibiotics derived from AMPs are undergoing clinical trials. For example, four cathelicidin analogs, omiganan (bovine indolicidin analogue, MBI-226, CPI226), MX-594AN (bovine indolicidin analogue), and iseganan (porcine protegrin-1 analogue and IB-367) are undergoing clinical trials as novel anti-infective agents .
In case of fishes, AMPs are crucial members of innate immunity that provide first line of host defense. AMPs from fishes was initially characterized from the secretion of Moses sole fish Pardachirus marmoratus in 1980, which is called pardaxin (Primor and Tu, 1980). Pardaxin was first described as a toxic peptide and was used as a shark repellent (Primor and Tu, 1980). Later, pardaxin was demonstrated to possess a potent pore-formation property (Lazarovici et al., 1986) and a high antibacterial activity (Oren and Shai, 1996). Over the past decades, several AMP families, like piscidins, cathelicidins, defensins, and hepcidins (also known as liver expressed APs, LEAPs), have been identified from fishes (Shabir et al., 2018). In general, these AMPs from fishes showed direct antimicrobial potency against bacteria, fungi, viruses and parasites. Almost all AMPs from fishes showed direct antibacterial or bacteriostatic activities against several Gramnegative and Gram-positive bacterial strains (Rajanbabu and Chen, 2011). Some AMPs from fishes displayed anti-fungal activities. Piscidin 2 secreted by striped bass could act as a fungicide by disrupting fungal membranes (Sung et al., 2008). A few AMPs from fishes exhibited antiviral activities. Rainbow trout β-defensin 1 showed antiviral activity against viral hemorrhagic septicemia virus (VHSV) infection (Falco et al., 2008). AMPs from fishes also showed anti-parasitic functions. Pisicidin 2 isolated from hybrid striped bass showed potent anti-parasitic effects against three protistan ectoparasites of marine fish and one ciliate ectoparasite of freshwater fish (Colorni et al., 2008). In addition, AMPs from fishes also acted as immunostimulants that modulated the anti-bacterial immune response and basal immune response of fish. Cathelicidins from Brachymystax lenok and Gadus morhua exhibited potent immunoregulatory activities by inhibition of pro-inflammatory cytokine (tumor necrosis factor-α, interleukin-1β, and interleukin-6) expression in bacteria-infected zebrafish and induction of chemokine (interleukin-8) expression in healthy zebrafish. In this study, topmouth culter LEAP-2 exhibited direct anti-bacterial activities against aquatic pathogenic bacteria via a membrane-disruption mechanism. The direct anti-bacterial properties of topmouth culter LEAP-2 are consistent with those observed from other fish-derived AMPs. In addition, it also showed potent anti-bacterial activities in different temperatures (Supplementary Table S1) and different buffers (Supplementary Table S2) that mimicked the fish environment. In the present study, we did not test the anti-fungal, antiviral, anti-parasitic, and immunomodulatory functions of topmouth culter LEAP-2, which need to be further elucidated in future.
Because of the widespread of antibiotics, the increasing antibiotic-resistant bacterial strains have become a severe threat to fish farming. In the past decades, many researchers tried to develop the substitutes of antibiotics in aquaculture industry aquaculture, which will control the diseases in cultured aquatic animals induced by the spread of antibiotic-resistant bacterial strains. AMPs are largely membrane-active agents that alter bacterial membrane integrity by inducing pore formation (Khara et al., 2014). The membrane-disruption mechanism of AMPs against bacteria are different from the mechanism of antibiotics against bacteria. Compared to the mechanism of action of antibiotics, the membrane-disrupted mechanism of AMPs against bacteria is non-specific, which makes host AMPs have broad anti-bacterial spectrum, including antibiotic-sensitive and antibiotic-resistant bacterial strains. Besides, this non-specific anti-bacterial mechanism is difficult to induce the bacterial resistance against AMPs. In our study, topmouth culter LEAP-2 significantly disrupted membrane integrity of aquatic pathogenic bacterial membrane and showed none inducible drug-resistance to aquatic pathogenic bacteria. Interestingly, topmouth culter LEAP-2 delayed ampicillin-induced drug resistance against V. parahaemolyticus, sensitized ampicillin-resistant aquatic pathogenic bacteria to ampicillin, and generated synergistic effects with ampicillin. It is more likely that the induction of pore formation on bacterial membrane mediates the permeation of ampicillin from outer membrane to cytoplasmic targets. In vivo test proved that topmouth culter LEAP-2 did inhibit ampicillin-resistant A. hydrophila infection and did enhance the therapeutic efficacy of ampicillin against antibiotic-resistant A. hydrophila infection. Since the membrane-disrupted mechanism is a non-specific way, topmouth culter LEAP-2induced pore formation on bacterial membrane possibly also promotes the permeation of other antibiotics from bacterial outer membrane to cytoplasmic targets, thereby generating interactive effects with other antibiotics. These findings made topmouth culter LEAP-2 an excellent candidate for the combinational usage with antibiotic, and the combinational usage of topmouth culter LEAP-2 and antibiotic may slow the progression of antibiotic-resistant aquatic pathogenic bacteria and even prevent the emergence of antibiotic-resistant aquatic pathogenic bacteria.
In conclusion, we identified a novel AMP from topmouth culter that belonged to LEAP-2 family. Topmouth culter LEAP-2 showed antimicrobial activities against aquatic pathogenic bacteria with rapid bactericidal speed. It dramatically induced bacterial membrane permeabilization by eliciting pore formation on bacterial membrane. It showed none inducible drug resistance to aquatic pathogenic bacteria and efficiently delayed ampicillin-induced drug resistance against aquatic pathogenic bacteria. It could sensitize ampicillin-resistant aquatic pathogenic bacterial strains to ampicillin and could generate synergistic effect with ampicillin. In vivo, topmouth culter LEAP-2 markedly alleviated ampicillin-resistant bacterial infection and enhanced the therapeutic efficacy of antibiotic against antibiotic-resistant bacterial infection. Our findings highlighted its development potential for antibiotic-resistant bacterial infection in aquaculture industry.

DATA AVAILABILITY STATEMENT
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.