Primitive Teleost Fish Cathelicidin Antimicrobial Peptides in A

Home , Antimicrobial peptides

Distinctive Structural Hallmarks and Biological Activities of the Multiple Cathelicidin Antimicrobial Peptides in a Primitive Teleost Fish This information is current as of September 30, 2021. Xu-Jie Zhang, Xiang-Yang Zhang, Nu Zhang, Xia Guo, Kai-Song Peng, Han Wu, Long-Feng Lu, Nan Wu, Dan-Dan Chen, Shun Li, Pin Nie and Yong-An Zhang J Immunol published online 15 April 2015 http://www.jimmunol.org/content/early/2015/04/15/jimmun Downloaded from ol.1500182

Supplementary http://www.jimmunol.org/content/suppl/2015/04/15/jimmunol.150018 http://www.jimmunol.org/ Material 2.DCSupplemental

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists by guest on September 30, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2015 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published April 15, 2015, doi:10.4049/jimmunol.1500182 The Journal of Immunology

Distinctive Structural Hallmarks and Biological Activities of the Multiple Cathelicidin Antimicrobial Peptides in a Primitive Teleost Fish

Xu-Jie Zhang,*,† Xiang-Yang Zhang,*,† Nu Zhang,*,† Xia Guo,*,† Kai-Song Peng,*,‡ Han Wu,* Long-Feng Lu,*,† Nan Wu,* Dan-Dan Chen,*,† Shun Li,* Pin Nie,* and Yong-An Zhang*

Cathelicidin antimicrobial peptides (CAMPs) represent a crucial component of the innate immune system in vertebrates. Although widely studied in mammals, little is known about the structure and function of fish CAMPs. Further to the previous findings, two more cathelicidin genes and multiple transcripts from rainbow trout were identified in the present study. Interestingly, we found that trout have evolved energy-saving forms of cathelicidins with the total deletion of the characteristic cathelin-like domain. Sequence Downloaded from analysis revealed that salmonid CAMPs have formed a special class of antimicrobial peptides in vertebrates with three distinctive hallmarks: the N terminus is intensified by positive charges, the central region consists of repetitive motifs based on RPGGGS, and the C terminus is lowly charged. Immunofluorescence localization of trout CAMPs demonstrated that these peptides expressed mainly at the mucosal layer of gut. Meanwhile, signals around sinusoids were also detected in head kidney. Moreover, the biological activities of trout CAMPs were proved to be mediated by the N terminus. Additionally, the repetitive motifs characteristically existing in Salmonidae increased the structural flexibilities of peptides and further increased the antibacterial and IL-8–stimu- http://www.jimmunol.org/ lating activities. Unlike most a helical and cytotoxic mammalian CAMPs, trout CAMPs, mainly consisting of b-sheet and random coil, exhibited no cytotoxic activities. The distinctive structural features of trout CAMPs provide new insights into the understanding of the evolution of CAMPs in vertebrates. Moreover, the high bacterial membrane selectivity of trout CAMPs will help to design excellent peptide antibiotics. The Journal of Immunology, 2015, 194: 000–000.

ntimicrobial peptides represent an evolutionarily old Cathelicidins are likely present in all vertebrates (2, 6, 7). The component of the innate immune system in all organisms discovery of three cathelicidin-like members in the ancient ver-

A (1, 2). They form the first line of host defense against tebrate hagfish (Myxine glutinosa) implies that cathelicidins al- by guest on September 30, 2021 infectious microorganisms prior to stimulating the adaptive im- ready existed ∼450 million years ago (8). Until now, only one mune system of animals (1–3). The increasing resistance of bac- cathelicidin antimicrobial peptide (CAMP) has been found in teria to conventional antibiotics stimulates the exploitation of primates (2). Nevertheless, at least six CAMPs have been dis- antimicrobial peptides as candidates for new antibiotics (3, 4). coveredinothermammalsbelongingtoEuungulata,suchas Among them, cathelicidins and defensins represent the major and cattle, goat, sheep, and pig, whereas four have been discovered in well-studied antimicrobial peptide families in animals (2, 5). birds, three in reptiles, and two in amphibians (2, 9, 10). As in fish, three CAMPs have been identified in Atlantic cod (Gadus morhua), whereas two or one has been discovered in several other

*Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan 430072, China; species (5, 7, 11). In vivo studies conducted in mice have sug- †University of Chinese Academy of Sciences, Beijing 100049, China; and gested that the duplication and divergence of cathelicidins are ‡ College of Animal Science and Technology, Anhui Agricultural University, Hefei beneficial to host defense against infections (12, 13). 230036, China In mammals, CAMPs are produced by epithelial cells and Received for publication January 23, 2015. Accepted for publication March 17, 2015. lymphocytes (14, 15) and are mainly expressed at bone marrow This work was supported by National Basic Research Program of China Grant 2014CB138601 and by National Natural Science Foundation of China Grant and the epithelial surfaces of skin, gastrointestinal tract, 31172431. and lung (15–17). CAMPs are first produced as inactive pre- The sequences presented in this article have been submitted to GenBank (http:// cursors, which consist of a signal peptide followed by a cathelin- www.ncbi.nlm.nih.gov/genbank/) under accession numbers KP347620–KP347627. like domain (CLD), as well as a C-terminal mature peptide The structures of trout CAMPs have been submitted to the Protein Model DataBase (http://bioinformatics.cineca.it/PMDB/) under identification numbers PM0079827 containing 12–80 residues (2, 6, 18). The CLD shows high and PM0079829–PM0079833. interspecies sequence homology, whereas the mature peptide Address correspondence and reprint requests to Prof. Yong-An Zhang, Institute shows high diversity both inter- and intraspecies (2, 6, 19). The of Hydrobiology, Chinese Academy of Sciences, 7 Donghu South Road, Wuhan active mature peptide is released via proteolytic processing with 430072, China. E-mail address: [email protected] specific proteases (2, 18). The released CAMPs are small, cat- The online version of this article contains supplemental material. ionic, and amphipathic peptides (2, 5, 6). Although varying in Abbreviations used in this article: CAMP, cathelicidin antimicrobial peptide; CD, circular dichroism; CLD, cathelin-like domain; EF-1a, elongation factor 1a; MHB, amino acid sequence and size, most mammalian CAMPs are Mueller–Hinton broth; pAb, polyclonal Ab; PBL, peripheral blood leukocyte; PI, linear peptides consisting of 23–37 residues and adopting am- propidium iodide; qPCR, quantitative real-time PCR; RTG, rainbow trout gonadal; phipathic a helical structures in the environments mimicking SEM, scanning electronic microscopy; TFE, 2,2,2-trifluoroethanol. biological membranes (6). These CAMPs exhibit a broad spec- Copyright Ó 2015 by The American Association of Immunologists, Inc. 0022-1767/15/$25.00 trum of antimicrobial activities against bacteria, viruses, and

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1500182 2 STRUCTURES AND ACTIVITIES OF TROUT CAMPs fungi (2, 6, 18). In addition to their direct antimicrobial effects, ExPASy proteomics server (http://www.expasy.org/). Multiple sequence mammalian CAMPs possess other modulatory activities, such as alignment was generated using ClustalW2 at the European Molecular LPS neutralizing (20, 21) and cytokine stimulating (22, 23). Biology Laboratory–European Bioinformatics Institute Web site (http:// www.ebi.ac.uk/). Although CAMPs have been widely studied in mammals, research on fish CAMPs has been predominantly limited to gene Expression patterns of trout cathelicidin genes identification and antibacterial activity test. No information is Three healthy trout (50–70 g) were euthanized and total RNA was isolated available about the structure/activity relationships of fish CAMPs. from the blood, head kidney, spleen, thymus, gut, skin, gill, liver, heart, However, the special living environment and evolutionary status muscle, and brain. Meanwhile, two groups of healthy trout were injected make fish an interesting host to study the functions of CAMPs. i.p. with LPS (from Escherichia coli 0111:B4; Sigma-Aldrich; 1 mg/ml in PBS) or PBS (Invitrogen), 100 ml per fish, and total RNA was isolated from Thus, the aim of the present study was two-fold: first, to unearth the head kidney and gut of three fish at 0 h, 12 h, 24 h, 3 d, and 7 d more cathelicidin genes and transcripts in rainbow trout (Onco- postinjection. The first-strand cDNA was synthesized as described above. rhynchus mykiss), a commercially farmed fish throughout the The transcripts of rtCATH-2a and rtCATH-2b were amplified with the world; and second, to further investigate the structure and multi- primer set rtCATH-2rtF/rtCATH-2rtR (Table I) by 38 cycles and then function of these CAMPs. We found that trout possess at least cloned and sequenced as described above to detect the coexpression of both genes in healthy trout tissues. The expression levels of trout cath- four genes and seven transcripts of CAMPs. Sequence analysis elicidin genes in the tissues under normal and challenged situations were revealed that trout CAMPs, together with other salmonid CAMPs, further analyzed by quantitative real-time PCR (qPCR) using primer sets form a special class of antimicrobial peptides with distinctive specific for rtCATH-1a/1b/1c (qCATH1aF/qCATH1aR), rtCATH-2a/2b hallmarks in vertebrates. Moreover, the structure/activity relation- (qCATH2F/qCATH2R), and the reference gene elongation factor 1a (EF- 1a; (qEF-1aF/qEF-1aR) (Table I). The reactions were conducted in du- ships of trout CAMPs have been established. plicate and each 10-ml reaction volume contained 1 ml cDNA template, Downloaded from 5 ml SYBR Green PCR master mix (23; Bio-Rad), and 1 ml of each Materials and Methods forward and reverse primer (10 mM). The amplification profile consisted of Fish an initial denaturation step at 95˚C for 3 min, and then 45 cycles of 95˚C for 10 s, 60˚C for 20 s, and 72˚C for 20 s, followed by melting (dissoci- Rainbow trout (50–70 g) obtained from Zhanghe Reservoir Rainbow Trout ation stage) from 65˚C to 95˚C in a CFX real-time PCR detection system Farm (Jingmen, China) were maintained and acclimated to the laboratory (Bio-Rad). Amplification efficiencies of the primer sets were 96.2% for 2 conditions as previously described (24). All animal experiments were qCATH1aF/qCATH1aR (R = 0.995), 100% for qCATH2F/qCATH2R http://www.jimmunol.org/ approved by the Committee on the Ethics of Animal Experiments of the (R2 = 0.999), and 96.7% for qEF-1aF/qEF-1aR (R2 = 0.995), calculated by Institute of Hydrobiology, Chinese Academy of Sciences. using 2-fold series dilution of cDNA in qPCR. Specificities of the primer sets were verified by the dissociation curves and sequencing the qPCR products. Isolation of cDNAs for trout cathelicidins The relative expression levels of trout cathelicidin genes were normalized by using the expression of EF-1a. Healthy trout were euthanized and total RNA was isolated from the head kidney by using the TRIzol reagent (Invitrogen). First-strand cDNA was Peptide synthesis synthesized from the total RNA by using a reverse transcription system (Promega) with oligo(dT)18 primer and stored at 220˚C for use. Trout CAMPs and the truncated variants were synthesized using a peptide Using the previously reported cDNA sequence of rtCATH-1 (accession synthesizer from Aitepeptide (Nanjing, China) and analyzed by HPLC and . no. AY382478, termed rtCATH-1a in this study) to search trout in the MALDI-TOF mass spectroscopy to confirm that the purity was 95%. by guest on September 30, 2021 Expressed Sequence Tags database (National Center for Biotechnology Human CAMPs LL-37 was synthesized as a reference. All peptides were Information; http://www.ncbi.nlm.nih.gov), we obtained a putative splic- lyophilized and stored at 280˚C for use. ing isoform (GE834401, termed rtCATH-1b). Primer sets rtCATH1-F1/ rtCATH1-R1 and rtCATH1-F1/rtCATH-1rtR (Table I) matching the 59 Expression and purification of GST-tagged trout CAMPs in and 39 ends of the cDNA and open reading frame sequences, respectively, E. coli were applied to amplify other potential splicing isoforms of rtCATH-1a. The DNA fragments encoding trout CAMPs 1a and 2a were amplified using Alternatively, we have previously sequenced a genomic clone carrying the primer sets Ex-Cath1a-F/Ex-Cath1a-R and Ex-Cath2a-F/Ex-Cath2a-R a homolog gene (AY542962, termed rtCATH-2b) of rtCATH-2 (AY542963, (Table I), respectively, and then digested with BamHI and XhoI (New termed rtCATH-2a). Primer set rtCATH2-F1/rtCATH2-R1 (Table I) match- England Biolabs) and ligated to the pGEX-4T-1 expression vector (GE ing the 59 and 39 ends of the genomic sequences was applied to amplify the Healthcare). The expression plasmids were subsequently transformed into transcripts from trout cDNA. E. coli BL21 (DE3)–CodonPlus-RIL competent cells, which were cultured All PCRs were performed in 20-ml reaction volumes containing 14 ml in Luria–Bertani broth and induced with 0.5 mM isopropyl b-D-thio- water, 2 ml103 buffer, 2 ml29-deoxynucleoside 59-triphosphate (2.5 mM galactoside at 12˚C for 12 h. Then cells were harvested and resuspended in each), 0.25 ml Ex Taq HS DNA Polymerase (TaKaRa), 1 ml primer set 25 ml lysis buffer (500 mM NaCl, 20 mM Tris, 1 mM DTT, 1 mM EDTA, (10 mM each), and 0.75 ml cDNA. The amplification program was as 0.1% Triton X-100, 1 mM PMSF, 1 mM aprotinin [pH 7.5]). Cell sus- follows: 95˚C for 5 min, followed by 40 cycles of 95˚C for 20 s, 60˚C for pensions were mechanically disrupted using a JN-02C high-pressure cell 20 s, and 72˚C for 30 s, and then 72˚C for 5 min. Amplified DNA was press (JNBIO, Guangzhou, China) and then centrifuged at 16,000 3 g for analyzed using 1.8% (w/v) agarose gel electrophoresis stained with ethid- 12 min to collect the supernatant, which was subjected to a glutathione- ium bromide and visualized with a UV transilluminator (Bio-Rad). Sepharose 4 Fast Flow resin (GE Healthcare) equilibrated with lysis buffer. Isolation of genomic sequences of trout cathelicidins Recombinant proteins were batch bound to resin by rotating at 4˚C for 2 h. The resin was in turn washed with 30 ml buffers, including lysis buffer, Genomic DNA was extracted from the liver of trout with a genomic DNA wash buffer 1 (500 mM NaCl, 20 mM MES, 1 mM DTT, 1 mM EDTA, extraction kit (Tiangen Biotech) and was used as a PCR template. The 0.1% Triton X-100, 1 mM PMSF, 1 mM aprotinin [pH 5.0]), and wash homolog genes of trout rtCATH-1a and rtCATH-2a were PCR amplified buffer 2 (500 mM NaCl, 20 mM Tris, 1 mM DTT, 1 mM EDTA, 0.1% with the primer sets and program as described above. Amplified DNA was Triton X-100, 1 mM PMSF, 1 mM aprotinin [pH 9.0]). Then the resin was analyzed on a 1% agarose gel. Owing to the high sequence similarity, the packed into an empty PD-10 column (GE Healthcare) and the binding gene fragments of rtCATH-2a and rtCATH-2b were amplified with the protein was eluted with elution buffer (150 mM NaCl, 20 mM Tris, 20 mM same primer set rtCATH-2rtF/rtCATH-2rtR (Table I) and further cloned reduced glutathione [pH 7.2]). Tris-SDS-PAGE (12%) was carried out to and sequenced to detect the coexistence of both genes in individual trout. assess the purity of recombinant proteins. DNA cloning and sequence analysis Production, purification, and detection of rabbit anti-trout CAMP polyclonal Abs All PCR products were cloned into pMD18-T vector (TaKaRa). Plasmid DNA was isolated from the positive colonies and sequenced with an ABI GST-tagged trout CAMPs 1a and 2a were used to raise polyclonal Abs 3730XL sequencer (Applied Biosystems) by Tsingke Company (Wuhan, (pAbs) in rabbits (Yingji Biotech, Shanghai, China). The pAbs in rabbit China). Predictions of the open reading frame and signal peptide were antiserum were purified by affinity chromatography using the GST-tagged performed with the Translate and SignalP programs, respectively, at the trout CAMPs coupled to SulfoLink coupling resin (Thermo Scientific). The Journal of Immunology 3

Thereafter, the GST-specific pAbs were removed by affinity chromatog- (33). The accuracy of the simulated structures was validated by the CD raphy using the GST-coupled SulfoLink coupling resin. Thus, the remaining spectra of the peptides. The structure was visualized by RasMol (http:// pAbs were trout CAMP specific. Western blot was performed to analyze the rasmol.org/) and represented in the form of cartoons. reactivity of the pAbs with trout CAMPs as previously described (25). Hemolysis and cytotoxicity assays Immunofluorescence localization of trout CAMPs Hemolytic activities of trout CAMPs, at the concentration of 60 mM, to Immunofluorescence was performed as previously described (26) to ana- trout and human erythrocytes were evaluated as previously described (27, lyze the tissue localization of trout CAMPs in head kidney and gut. Briefly, 28). The absorbance was measured at 405 nm using an ELx800 micro- cryosections of tissue samples were incubated with rabbit anti-trout CAMP plate reader (BioTek). Percentage hemolysis of peptides was determined pAbs (0.5 mg/ml), followed by FITC-conjugated goat anti-rabbit IgG Ab by comparison with 0 and 100% lysis using the following formula: (2.5 mg/ml, Thermo Scientific). Rabbit anti-GST pAb (0.5 mg/ml) was [(Acathelicidin 2 A0% lysis)/(A100% lysis 2 A0% lysis)] 3 100, where A is the used as the negative control. After staining with DAPI (Beyotime), the absorbance at 405 nm. images were acquired using a confocal microscope (Zeiss). Cytotoxic activities of trout CAMPs were evaluated as previously described (34, 35) with minor modifications. Briefly, rainbow trout gonadal Antibacterial activity assay of trout CAMPs and their (RTG)-2 cells (2 3 104/well) were seeded onto 96-well plates in L-15 truncated variants medium containing L-glutamine (Sigma-Aldrich) supplemented with 10% FBS, 100 U/ml penicillin, and 100 mg/ml streptomycin. Cells were Five species of Gram-negative bacteria, including E. coli ATCC 25922, grown at 20˚C for 48 h, and then growth medium was removed and cells Edwardsiella ictaluri, Vibrio fluvialis, Aeromonas hydrophila,andYersinia were incubated in CAMPs in duplicate at a concentration of 60 mMin ruckeri, and one species of Gram-positive bacteria, Streptococcus dysga- L-15 medium with supplements. As 0 or 100% controls for cell death, L-15 lactiae, were used to test the antibacterial activities of trout CAMPs and or 2% Triton X-100 in L-15 was added in duplicate to the control wells. their truncated variants. All of these bacteria, except E. coli ATCC 25922, After incubated at 20˚C for 48 h, to each well 20 ml MTS/PMS solution are fish pathogenic bacteria. Human CAMP LL-37 was used as a reference. (Promega) was added and incubated at 20˚C for an additional 4 h. The Downloaded from Antibacterial activity assay was performed by 2-fold microtiter broth dilu- absorbance was subsequently measured at 490 nm. The relative cell via- tion method as previously described with minor modifications (27, 28). bility was determined by comparison with 0 and 100% lysis using the Briefly, 80 ml serial 2-fold dilutions of peptides prepared in water were following formula: [(Acathelicidin – A100% lysis)/(A0% lysis 2 A100% lysis)] 3 added to each corresponding well of a 96-well microtiter plate (Thermo 100, where A is the absorbance at 490 nm. Scientific). Then, 20 ml bacteria diluted in Mueller–Hinton broth (MHB; Oxoid) or brain heart infusion broth (BD Biosciences) (for S. dysgalactiae) LPS-binding activities to a concentration of 5 3 105 CFU/ml were added. The wells without The LPS-binding activities of trout CAMPs were measured through ELISA http://www.jimmunol.org/ peptide were used as controls. Initial OD595 was measured using an ELx800 microplate reader (BioTek). Then these plates were incubated at 20˚C, ex- as previously described (20, 36). All reactions were done in duplicate and the experiment was repeated. Binding of peptides to LPS was expressed as cept S. dysgalactiae (at 28˚C), for 18 h. OD595 was measured and corrected a percentage of absorbance developed by 0.1 mg peptide. Additionally, the by initial OD595 values. Bacterial grow rates were calculated as the bacterial densities in the presence of peptides to the bacterial densities of controls. LPS-binding isotherms of trout CAMPs and their truncated variants were quantified by the kinetic chromogenic Limulus amebocyte lysate assay Bacterial membrane permeability analysis of trout CAMPs and (QCL-1000 kit; Lonza) as previously described (21, 37). their truncated variants Induction of chemokine IL-8 expression in trout peripheral The bacterial membrane permeability of trout CAMPs and their truncated blood leukocytes

variants was assessed by using flow cytometry as previously described (29) by guest on September 30, 2021 with minor modifications. Briefly, to 20 ml E. ictaluri cells (106 CFU/ml in Trout peripheral blood leukocytes (PBLs) were isolated and stimulated with MHB), 80 ml peptides diluted in water was added to give a final con- CAMPs at the concentration of 10 mM as previously described (28). Total centration of 8 mM. Meanwhile, a bacterial suspension without peptides RNA was isolated from the PBLs at 0, 4, 8, 16, and 24 h postincubation. was included as a control. After incubated at 20˚C for 1.5 h, propidium Then, cDNA synthesis and qPCR were performed as above described to iodide (PI; Sigma-Aldrich) was added to give a final concentration of analyze the expression changes of IL-8, using primer sets specific for IL-8 9 mM. The influx of PI into bacterial cells was investigated by using an (qrtIL-8F/qrtIL-8R; amplification efficiency 96.6%, R2 = 0.990) and ref- Accuri C6 flow cytometer (BD Biosciences) at 3000 events. The cell- erence genes, including b-actin (qrtb-actinF/qrtb-actinR; 97.0%, R2 = penetrating efficiency was analyzed by using the FlowJo software pack- 0.999) and EF-1a (qEF-1aF/qEF-1aR; 96.7%, R2 = 0.995) (Table I). The age (Tree Star). relative expression levels of trout IL-8 were normalized using the mean expression of the two reference genes. Scanning electronic microscopy imaging Synteny analyses of cathelicidin genes in vertebrates Scanning electronic microscopy (SEM) was performed to observe the bacterial morphological changes after treatment with trout CAMPs as Gene synteny of cathelicidin family members in representative vertebrates, previously described (9) with minor modifications. Briefly, to 20 ml including human, mouse, cattle, goat, chicken, frog, and trout, were ana- E. ictaluri cells (106 CFU/ml in MHB), 80 ml peptides (10 mM in water) or lyzed. Cathelicidin members were located to chromosomes or scaffolds by water was added. After incubated at 20˚C for 4 h (for trout CAMPs) or 2 h using BLAST algorithm. (for LL-37), the bacteria were fixed, dehydrated, vacuum dried, sputter coated with gold, and observed with an S-4800 field emission SEM Statistical analysis (Hitachi) at an accelerating voltage of 3 kV. The statistic p values were calculated by one-way ANOVA with a Dunnett Circular dichroism spectrometry post hoc test (SPSS Statistics, version 19, IBM). A p value ,0.05 was considered significant. Circular dichroism (CD) spectra of trout CAMPs and their truncated variants were recorded at 298 K on a Jasco J-810 spectrometer as previously described (21, 30) with minor modifications. Human LL-37 was used as Results a reference. The spectra of samples contained 20–40 mM peptides in Genomic, cDNA, and protein sequences of trout cathelicidins 20 mM sodium phosphate buffer (pH 6.0) or membrane environments, including 90 mM SDS micelles (in 20 mM sodium phosphate buffer Gene-specific primers (Table I) were used to amplify the tran- [pH 6.0]) and 60% 2,2,2-trifluoroethanol (TFE) (in 20 mM sodium phos- scripts of rtCATH-1. Interestingly, among the PCR products phate buffer [pH 6.0]), were measured between 190 and 250 nm. The amplified by the primer set rtCATH1-F1/rtCATH1-R1, besides secondary structural elements of the peptides, including a helix, b-sheet, rtCATH-1b, an additional splicing isoform (termed as rtCATH-1c) b -turn, and random coil, were calculated using the software supported by the of rtCATH-1a was obtained. While among the PCR products DichroWeb (http://dichroweb.cryst.bbk.ac.uk/html/home.shtml) (31, 32). amplified by the primer set rtCATH1-F1/rtCATH-1rtR, another Simulation of the tertiary structures isoform (termed as rtCATH-1am) of rtCATH-1a and a transcript The tertiary structures of trout CAMPs were predicted using Robetta, an of the second rtCATH-1 gene (termed as rtCATH-1d) were ob- online computational protein design software (http://robetta.bakerlab.org/) tained. Then, the genomic sequences of rtCATH-1a and rtCATH-1d 4 STRUCTURES AND ACTIVITIES OF TROUT CAMPs

Table I. Primers used for gene cloning and expression

Primer Sequence (59-39) Application rtCATH1-F1 AGACATGAAGATGAAGGTCCAG Genomic and cDNA amplification rtCATH1-R1 TTTATTGGCGTAAACATAATCTAGA Genomic and cDNA amplification rtCATH-1rtR TATGCAAAGCGATTTCCATCAC cDNA amplification rtCATH2-F1 ATCAGACATGAAGATGAAGGCTC Genomic and cDNA amplification rtCATH2-R1 TTTATTAGTGCAGTAAACATGAACT Genomic and cDNA amplification rtCATH-2rtF GAAGAGGCAAGGACAGCGGA Genomic and cDNA amplification rtCATH-2rtR AGACAAAAGAGTTACCAGAAGCA Genomic and cDNA amplification Ex-Cath1a-F TATGGATCCAGAAGAAGCAAAGTCAGAATATGCa Protein expression and purification Ex-Cath1a-R ATTCTCGAGCTATGCAAAGCGATTTCCATCACb Protein expression and purification Ex-Cath2a-F TTAGGATCCAGAAGAGGCAAGGACAGCGGAa Protein expression and purification Ex-Cath2a-R ATCCTCGAGTTATTGGGGAGATATCTTGCAGTb Protein expression and purification qCATH1aF AGGCAAGCAACAACCTGAACAC Real-time PCR qCATH1aR CCCCAAGACGAGAGACACAAT Real-time PCR qCATH2F CAACACCCTCAACACTGACCG Real-time PCR qCATH2R GAATCTTTTCTACCCATCTTAGG Real-time PCR qrtIL-8 F ATTGAGACGGAAAGCAGACGA Real-time PCR qrtIL-8 R CCCTCTTCATTTGTTGTTGGC Real-time PCR qrtb-actinF ACAGGTCATCACCATCGGCA Real-time PCR qrtb-actinR GGTCTCGTGGATACCGCAAG Real-time PCR Downloaded from qEF-1aF CAAGGATATCCGTCGTGGCA Real-time PCR qEF-1aR ACAGCGAAACGACCAAGAGG Real-time PCR aThe restriction enzyme BamHI recognition sequence is underlined. bThe restriction enzyme XhoI recognition sequence is underlined. http://www.jimmunol.org/ were amplified with the primer set rtCATH1-F1/rtCATH-1rtR. a mature peptide. The central region of the mature peptide is The deduced amino acid sequences were aligned and schematic characterized by five tandem repeats based on the (R/P)(P/L) diagrams were created to show the organizations of the genes, GGGS motif, and it is flanked by 16 residues at the N-terminal transcripts, and preproproteins (Fig. 1A, 1B). The preproprotein of and C-terminal, respectively. The splicing isoforms rtCATH-1b rtCATH-1a consists of a signal peptide and a CLD followed by and rtCATH-1c are characterized by the loss of small portions by guest on September 30, 2021

FIGURE 1. Schematic diagrams of the gene, mRNA, and protein organizations of trout CAMPs and multiple alignment of the deduced amino acid sequences. (A) Schematic diagrams showing the gene, mRNA, and protein organizations of rtCATH-1a and rtCATH-1d. (B) Multiple alignment of the deduced amino acid sequences of the preproproteins of rtCATH-1a and rtCATH-1d. (C) Schematic diagrams showing the gene, mRNA, and protein organizations of rtCATH-2a and rtCATH-2b. (D) Multiple alignment of the deduced amino acid sequences of the preproproteins rtCATH-2a and rtCATH-2b. The signal peptide, CLD, and mature peptide are denoted above the alignment. The repetitive motifs [(R/P)(P/L)G(G/S/F)GS] in mature peptides are in boldface. The cysteine residues are shaded black. The Journal of Immunology 5 of exon 4, resulting in reduction of the copy numbers of the re- tested tissues, although the expression level of rtCATH-2a was petitive motif in the mature peptide, whereas rtCATH-1am is higher than that of rtCATH-2b (data not shown). Then, the characterized by the loss of a large portion ranging from exon 1 to expression levels of rtCATH-1a/1b/1c and rtCATH-2a/2b were exon 3, resulting in complete deletion of the CLD of the pre- further analyzed by qPCR. As shown in Fig. 2A, in normal con- proprotein. Interestingly, the rtCATH-1d gene has no introns, and ditions, rtCATH-1a/1b/1c and rtCATH-2a/2b are constitutively the preproprotein only consists of a signal peptide and a mature expressed. The expression level of rtCATH-1a/1b/1c was higher peptide, which is similar to that of rtCATH-1am, but without the than that of rtCATH-2a/2b in thymus and gill but lower in heart and repetitive motif. muscle. Interestingly, both types of CAMPs were highly expressed The genomic and cDNA sequences of rtCATH-2a and in thymus and mucosal tissues, including skin and gill. In LPS- rtCATH-2b were amplified. The deduced amino acid sequences stimulated conditions, the expressions of both types of CAMPs were compared and schematic diagrams were created to show in head kidney (Fig. 2B) and gut (Fig. 2C) were increased dra- the organizations of the genes, transcripts, and preproproteins matically. (Fig. 1C, 1D). Similar to rtCATH-1a, the preproprotein of rtCATH- Immunofluorescence localization of trout CAMPs in head 2a and rtCATH-2b also consist of a signal peptide and a CLD kidney and gut followed by a mature peptide. The central regions of the mature peptides possess two tandem repeats of the RPG(S/F)GS motif. Anti-trout CAMP pAbs were used to detect the expression and Interestingly, these two genes share a surprisingly high degree of localization of CAMPs in head kidney and gut of trout. Western sequence identity (99%) at the nucleotide level. However, owing blot showed that anti–rtCATH-1a pAb could specifically recog- to the earlier occurring stop codon, the preproprotein of rtCATH- nize rtCATH-1a, -1b, -1c, and -1d; however, anti–rtCATH-2a Downloaded from 2b is shorter than that of rtCATH-2a, with lack of the C-terminal pAb could only recognize rtCATH-2a, but not rtCATH-2b 16 residues (Fig. 1C, 1D). The present study revealed that both (Supplemental Fig. 1). In head kidney, the immunoreactive sig- genes coexisted in each of the 24 tested trout (data not shown), nals of trout CAMPs were observed in lymphoid cells around the which means that rtCATH-2a and rtCATH-2b are two different sinusoids (Fig. 3A, 3B). However, in gut, the signals were ob- genes and are not segregated in trout. served mainly at the mucosal layer (Fig. 3D, 3E). Thus, trout

CAMPs were mainly expressed by the columnar epithelial cells http://www.jimmunol.org/ Tissue distribution and stimulation of trout cathelicidins of the gut villus. Higher magniﬁcation further revealed that some The expressions of rtCATH-2a and rtCATH-2b in a range of healthy lymphoid cells in the lamina propria of gut also expressed trout tissues were analyzed by RT-PCR. The sequenced PCR CAMPs. Negative controls showed no immunoreactive signal products showed that both genes were coexpressed in all of the (Fig. 3C, 3F). by guest on September 30, 2021

FIGURE 2. Expression patterns of trout cathelicidin genes in healthy and LPS-stimulated conditions. (A) The expression levels of trout cathelicidin genes in healthy conditions analyzed by qPCR and normalized against the expression of EF-1a. (B and C) The expression of trout cathelicidin genes in head kidney (B) and gut (C) stimulated with E. coli LPS. p , 0.01 for both rtCATH-1a/ 1b/1c and rtCATH-2a/2b, ANOVA; *p , 0.05, **p , 0.01 versus PBS control (Dunnett post hoc test). BL, blood; BR, brain; GI, gill; GU, gut; HE, heart; HK, head kidney; LI, liver; MU, muscle; SK, skin; SP, spleen; TH, thymus. 6 STRUCTURES AND ACTIVITIES OF TROUT CAMPs

FIGURE 3. Immunofluorescence detection of trout CAMPs in head kidney and gut. Cryosections were stained with specific pAbs for CAMPs (green) and with DAPI for nuclei (blue). (A and B) Signals of cathelicidins 1 peptides (A) and 2a peptide (B)in head kidney. (D and E) Signals of cathelicidins 1 peptides (D) and 2a peptide (E)in gut. (C and F) Control Ab staining in head kidney (C) and gut (F). The right image of each panel shows the enlarged image of the area outlined in the left image. The white arrowheads indicate lymphoid cells expressing CAMPs around sinusoids (SI) in head kidney or in the lamina propria (LP) of gut. Data are representative of four inde- Downloaded from pendent experiments. Scale bars, 20 mm. http://www.jimmunol.org/ Antibacterial activities of trout CAMPs and their truncated activities of trout cathelicidin 1 peptides and their truncated variants variants was 1a . 1b . 1ΔC . 1c . 1d . 1Nt . 1In . 1DN . 1Ct. Trout CAMPs derived from rtCATH-1a, rtCATH-1d, rtCATH-2a, As for cathelicidin 2 peptides (2a and 2b), with lack of the and rtCATH-2b genes displayed various net charges (Fig. 4) and C-terminal 16 residues, the antibacterial activities of 2b were antibacterial activities (Fig. 5). All of the trout CAMPs, in- slightly lower than those of 2a. Among the series of truncated cluding 1a, 1b, 1c, 1d, 2a, and 2b, were antibacterial to all strains variants of 2a, the lowly cationic central variant 2In and C-terminal tested. variant 2Ct had no antibacterial activities. The deletion of the highly cationic N-terminal of 2a (2ΔN) resulted in a significant As for cathelicidin 1 peptides, with the gradual deletion of the re- by guest on September 30, 2021 petitive motif in the central region, the antibacterial activities of 1a, decrease of the antibacterial activity, whereas the deletion of the Δ Δ 1b, 1c, and 1d were gradually decreased. The antibacterial activities lowly cationic central region (2 In) or C terminus (2 C) almost of 1a and 1b were almost equal, whereas those of 1c and 1d were 3- did not lead to a decrease of the activity. The N-terminal variant to 4-fold lower. Among the series of truncated variants of 1a, only 2Nt independently had antibacterial activity that was slightly Δ Δ the negatively charged C-terminal variant 1Ct had no antibacterial lower than that of 2 In and 2 C. The order of the antibacterial activity. The deletion of the highly cationic N-terminal of 1a (1ΔN) activities of trout cathelicidin 2 peptides and their truncated var- . . Δ . Δ . . D . resulted in a significant decrease of the antibacterial activity, iants was 2a 2b 2 In 2 C 2Nt 2 N 2In = 2Ct. whereas the deletion of the negatively charged C terminus (1ΔC) Bacterial membrane permeability of trout CAMPs and their only resulted in a slight decrease of the activity. The N-terminal variant 1Nt independently had antibacterial activity that was slightly truncated variants lower than that of 1ΔC. Additionally the central variant 1In con- The penetrating efficiencies to bacteria, according to the relative sisting of repetitive motifs exhibited low antibacterial activity, which fluorescence intensities, of trout CAMPs and their truncated variants was almost equal to that of 1ΔN. The order of the antibacterial are shown in Fig. 6. The results indicated that trout CAMPs killed

FIGURE 4. The amino acid sequence, charge distribution, and net charges of trout CAMPs and their truncated variants synthesized in this study. (A) Trout cathelicidin 1 peptides and their truncated variants. (B) Trout cathelicidin 2 peptides and their truncated variants. The repetitive motifs [(R/P)(P/L)G(G/S/F)GS] are in boldface. The Journal of Immunology 7 Downloaded from http://www.jimmunol.org/ by guest on September 30, 2021

FIGURE 5. Antibacterial activities of trout CAMPs and their truncated variants against E. ictaluri, V. fluvialis, E. coli, A. hydrophila, Y. ruckeri, and S. dysgalactiae.(A) Antibacterial activities of cathelicidin 1 peptides and their truncated variants. (B) Antibacterial activities of cathelicidin 2 peptides and their truncated variants. LL-37 was used as a control. The bacteria in mid-logarithmic phase were diluted in medium and incubated with serially diluted 5 peptides at a concentration of 10 CFU/ml for 18 h. OD595 was measured and corrected by initial OD595 values. Bacterial grow rates were calculated as the bacterial density in the presence of peptides to the bacterial density of controls. Data shown are the means of duplicate reactions in a representative of three independent experiments. bacteria through membrane-permeabilizing action. Compared to with complete deletion of the repetitive motif in the central region, LL-37, trout CAMPs showed weaker bacterial membrane perme- the cell wall perturbation ability of 1d was significantly decreased. ability. The penetrating efficiency of 1a was higher than that of 2a. Secondary structural elements of trout CAMPs measured Among the truncated variants of 1a and 2a, 1Nt, 1In, and 2Nt could by CD independently penetrate the bacterial membrane, whereas 1Ct, 2In, and 2Ct could not. The results precisely accorded with the anti- The structural features of trout CAMPs and their truncated vari- bacterial activities of trout CAMPs and their truncated variants. ants in different solvent environments were measured by CD spectroscopy (Fig. 8, Supplemental Fig. 2). In phosphate buffer, SEM observation of the bacteria treated with trout CAMPs the CD spectrum of all peptides, except human LL-37, showed Bacterial morphological changes after treatment with trout CAMPs a strong negative band at 200 nm, indicative of a mostly random- were directly observed by SEM. As shown in Fig. 7, the untreated coiled conformation. Interestingly, in SDS micelles or TFE sol- E. ictaluri cells possessed normal shape and smooth surfaces. In vent, mimic of bacterial cell membrane environment, the CD contrast, E. ictaluri cells treated with trout CAMPs had obvious spectra of cathelicidin 1 peptides (1a, 1b, 1c, and 1d) changed morphological alterations. After treatment with 1a, 1b, 1c, 2a, and dramatically, with double minimal signals at 208 and 222 nm, 2b, the cell walls of bacteria were severely perturbed; however, indicating an amphipathic a helical conformation. Especially in 8 STRUCTURES AND ACTIVITIES OF TROUT CAMPs Downloaded from FIGURE 6. PI labeling of E. ictaluri cells treated with trout CAMPs and their truncated variants. E. ictaluri cells (2 3 104 CFU) were incubated with 8 mM peptides at 20˚C for 1.5 h. Then PI was added and the influx of PI into bacterial cells was investigated by flow cytometry. All bacteria were gated and the cell-penetrating efficiency was shown as percentage. Data are representative of three independent experiments.

TFE, an organic solvent whose dielectric constant resembles that trout CAMPs, the LPS-binding isotherms of 1a and its truncated of biological membrane, the signals were intensified. However, the variants were evaluated by the kinetic chromogenic Limulus http://www.jimmunol.org/ calculated helical content of peptides 1a, 1b, 1c, and 1d in TFE amebocyte lysate assay (Fig. 11B). The EC50 is indicated as was only 10.4, 12.8, 18.1, and 13.3%, respectively (Table II). The a dotted line. The LPS-binding activity of 1ΔC was almost equal results demonstrated that the peptides were mainly composed of to that of 1a. However, the activities of 1ΔN and 1In were sig- b-sheet and random coil. For the truncated variants of 1a, 1Ct nificantly lower. The results suggested that there are at least two showed the most significant signals of a helical conformation, LPS binding sites in 1a, one at the N terminus and the other at the followed by 1ΔN. However, 1ΔC, 1Nt, and 1In showed no signal central region. The 6-fold higher affinity of 1ΔC (EC50 9 mM) of a helical conformation. The results indicated that the a helical conformation locates at the C-terminal of 1a. All of these peptides, b except 1Ct, showed -sheet conformation. Unlike cathelicidin 1 by guest on September 30, 2021 peptides, cathelicidin 2 peptides and the truncated variants showed no obvious signal of a helical conformation, but partial signals of b-sheet conformation. The calculated secondary structural elements of 2a and 2b demonstrated that they mostly consisted of random coil, with 62.0 and 52.4% content, respectively. In contrast with trout CAMPs, human LL-37 showed strong signal of a helical conformation, with 98.3% a helix in TFE. Simulated tertiary structures of trout CAMPs The simulated tertiary structures of trout CAMPs are shown in Fig. 9. As for 1a, the N-terminal 16 aa formed an antiparallel b-sheet, the central repetitive motifs formed another antiparallel b-sheet and a flexible random coiled region, whereas the C-terminal 16 aa formed a one-turn helix flanked by an antiparallel b-sheet on the C-terminal side. From peptide 1a to 1d, with gradual deletion of the repetitive motifs in the central region, the second antiparallel b-sheet and the flexible region were gradually diminished, resulting in a decrease of the structural flexibility of peptides. As for 2a and 2b, except for an obvious antiparallel b-sheet located in the central region, the entire molecule consisted of random coil. Hemolytic and cytotoxic activities of trout CAMPs As shown in Fig. 10, trout CAMPs did not exhibit apparent hemolytic and cytotoxic activities. In contrast, human LL-37 exhibited strong hemolytic and cytotoxic activities against all FIGURE 7. SEM imaging of E. ictaluri cells treated with trout CAMPs. cells tested. E. ictaluri cells (2 3 104 CFU) were incubated with 8 mM peptides at 20˚C LPS-binding activities of trout CAMPs for 4 h (for trout CAMPs) or 2h (for LL-37). Then the bacteria were fixed and collected on polycarbonate filters. After dehydration and vacuum The LPS-binding activities of trout CAMPs were investigated by drying, filters with bacteria were mounted onto aluminum stubs, coated ELISA. As shown in Fig. 11A, trout CAMPs could bind to LPS in with gold, and observed with a SEM. Scale bars, 1 mm. Data are repre- a dose-dependent manner. To investigate the LPS binding sites of sentative of three independent experiments. The Journal of Immunology 9

FIGURE 8. CD spectra of trout CAMPs. The spectra were taken in 20 mM phosphate buffer (pH 6.0), 90 mM SDS (pH 6.0), and 60% TFE (pH 6.0), respectively, and measured between 190 and 250 nm. Three consecutive scans per sample were performed and averaged followed by subtraction of the solvent signals. Downloaded from http://www.jimmunol.org/ by guest on September 30, 2021 for LPS than 1In (EC50 64 mM) suggested that the N-terminal locus has been almost conserved in vertebrates throughout 350 binding site has higher affinity for LPS than does the central one. million y of evolution, although with varied gene numbers. The LPS-binding isotherm of 1a showed a sigmoidal shape, indicating positive cooperativity of the two LPS-binding sites. Discussion Until now, ∼70 cathelicidin family members have been identified Induction of IL-8 expression in PBLs by trout CAMPs from fish to mammals (2, 5, 7). Although only single cathelicidin In human and Atlantic salmon (Salmo salar), CAMPs have been has been found in primates, this family expanded in other verte- associated with the ability to stimulate IL-8 expression (22, 23, 28, brates by gene duplication and divergence (6). All of these cath- 38, 39). As shown in Fig. 11C, trout CAMPs 1a and 2a could also elicidin genes show a conserved four-exon/three-intron organization, significantly stimulate the expression of IL-8 in PBLs at 4 h after except the cathelicidin genes recently found in green lizard (Anolis incubation. Interestingly, with gradual deletion of the repetitive carolinensis) (41), Arctic charr (Salvelinus alpinus), and brook trout motif in the central region, the IL-8 stimulating ability of 1a, 1b, (Salvelinus fontinalis) (7), which miss the third exon, resulting in 1c, and 1d was gradually decreased. Compared to 2a, the IL-8 partial deletion of the CLD. Further to the previous findings of stimulation by 2b was weaker, and the time point with the greatest rtCATH-1a and rtCATH-2a genes from trout (24, 27), two more stimulation effect (16 h after incubation) was much latter than that cathelicidin genes, rtCATH-1d and rtCATH-2b, have been identified by 2a. in this study. Interestingly, rtCATH-1d contains no intron and its preproprotein lacks the whole CLD. Despite absence of the CLD Synteny analyses of cathelicidin genes in vertebrates and (R/P)(P/L)GGGS motif, the protein sequence of rtCATH-1d is As shown in Supplemental Fig. 3, only a single cathelicidin gene exists in humans and mice. However, in cattle, goats, chickens, and frogs, multiple cathelicidin genes form a gene cluster in one chromosome, although the two genes in frogs are separated by Table II. Secondary structural elements of trout CAMPs measured by CD other genes. Unlike higher vertebrates, the cathelicidin genes discovered so far in trout locate in three different chromosomes, Peptide a Helix (%) b-Sheet (%) b-Turn (%) Random Coil (%) which may result from the salmonid-specific fourth whole- genome duplication dated from 25 to 100 million y ago (40). 1a 10.4 37.7 21.0 30.9 Among trout cathelicidin genes, rtCATH-2b shows the highest 1b 12.8 32.5 21.8 33.0 1c 18.1 27.5 22.5 32.0 synteny with other cathelicidin family members in higher verte- 1d 13.3 31.7 22.3 32.6 brates, whereas rtCATH-1d was not located in trout chromosomes, 2a 5.5 21.1 11.4 62.0 restricted to the integrity of the genome. The overall synteny 2b 6.6 27.1 13.9 52.4 analysis revealed that the composition of the cathelicidin gene LL-37 98.3 1.1 0.4 0.2 10 STRUCTURES AND ACTIVITIES OF TROUT CAMPs

FIGURE 9. Tertiary structures of trout CAMPs and representative mammalian CAMPs belonging to different classes. The structures of trout CAMPs were predicted using Robetta (http://robetta.bakerlab.org/) and have been submitted to the Protein Model Data- Base (http://bioinformatics.cineca.it/PMDB/) under identification numbers as follows: 1a, PM0079827; 1b, PM0079829; 1c, PM0079830; 1d, PM0079831; 2a, PM0079832; 2b, PM0079833. The Protein Data Bank entries of LL-37 and PG-1 are 2K6O and 1ZY6, respectively (http://www.rcsb.org/pdb/home/home.do). The identification number of PF-2 in the Protein Model DataBase is PM0075253. Downloaded from http://www.jimmunol.org/ identical to that of rtCATH-1a. Moreover, three splicing isoforms of trout strongly supported the viewpoint that cathelicidin family rtCATH-1a were cloned in this study. When compared with rtCATH- members expanded in vertebrates by gene duplication and diver- 1a, rtCATH-1am lacks the whole CLD, whereas rtCATH-1b and gence (6, 48). The high nucleotide sequence identity of rtCATH- rtCATH-1c lack several of the repetitive motifs in the mature peptide. 2a and rtCATH-2b demonstrated that the duplication occurred Coincidentally, a combination of these splicing sites can almost lead very recently. Based on sequence alignments (data not shown), to the production of rtCATH-1d. These results indicated that a splic- rtCATH-2b is more similar to the CATH-2 in other Salmonidae ing mechanism exists in trout resulting in the production of rtCATH- fish species. Besides, the rtCATH-2b gene shows the highest by guest on September 30, 2021 1d from rtCATH-1a, which might occur very recently, as the se- synteny with other cathelicidin genes in higher vertebrates. The quence variation did not happen in such a short evolutionary time. results indicated that rtCATH-2b is a more conserved gene from To our knowledge, this is the first report of a cathelicidin gene and which rtCATH-2a might be duplicated and diverged. In vivo a splicing isoform with total deletion of the whole characteristic studies conducted in mice have demonstrated that the duplica- CLD in the cathelicidin family. This indicated that trout have tion and divergence of cathelicidins are beneficial to host defense evolved energy-saving forms of cathelicidins. Thus, the discovery against bacterial infections (12, 13). Thus, it is certain that the of more cathelicidins in primitive teleost fish may help to reveal the duplication events that occurred recently in trout can promote host trajectory of the evolution of cathelicidins in vertebrates. disease resistance. Cathelicidins are produced as inactive forms and processed In mammals, CAMPs are produced by epithelial cells and extracellularly to release the active peptides in mammals (42, 43). lymphocytes and are mainly expressed at bone marrow and epi- Most mammalian CAMPs exhibit high cytotoxicity in vitro (44). thelial surfaces of skin, gastrointestinal tract, and lung (2, 15–17). Thus, the function of the CLD has long been considered to neu- In this study, at the mRNA level, trout CAMPs were constitu- tralize the potential intracellular cytotoxicity of CAMPs (45). tively expressed in different tissues, especially in mucosal However, limited studies are concerned with the function of the tissues, including skin and gill. In LPS-stimulated conditions, the CLD, and the biological role of this domain is still a matter for expressions of trout CAMPs were increased dramatically. Fur- debate (46, 47). As shown in this study, trout CAMPs possessed no thermore, specific Abs against trout CAMPs were raised and the obvious hemolytic and cytotoxic activities. Hence, discovery of expression sites of these peptides were detected by immunofluo- the cathelicidin preproproteins with complete (in rainbow trout) or rescence. In head kidney, the equivalent of bone marrow in higher partial (in Arctic charr and brook trout) (7) deletion of the CLD vertebrates (49), the immunoreactive signals of trout CAMPs were provides ideas for revealing the function of the domain. In future observed in lymphoid cells around sinusoids, which would cir- studies, the structure and function of the CLDs in Salmonidae fish culate to the entire body. In gut, trout CAMPs were largely species are worth examination, which will provide implications expressed at the mucosal layer, especially in the columnar epi- for understanding the possible roles of CLDs in vertebrates. thelial cells of the villus, and in some lymphoid cells in the lamina A homologous gene of rtCATH-2a was cloned in this study. propria, which indicated that, similar to their mammalian ortho- Interestingly, both genes almost have identical nucleotide se- logs, CAMPs also form the first line of host defense in fish. quences (99% identity). The present study revealed that they are In higher vertebrates, all discovered CAMPs belong to three indeed two different genes coexisting and coexpressing in trout. structural classes: the a helical class consisting of 23–37 residues, However, owing to the stop codons locating at different positions, the b-hairpin class consisting of 12–18 residues, and linear class the preproprotein of rtCATH-2b lacks the C-terminal 16 residues enriched with repetitive Pro-Arg motifs and consisting of 39–94 of rtCATH-2a. The discovery of rtCATH-2a and rtCATH-2b in residues (6, 50). Among these CAMPs, the a helical class is The Journal of Immunology 11

region. Scocchi et al. (5) proposed that duplications of this region in salmonid CAMPs have occurred very recently, because the other regions of the different orthologs in Salmonidae have almost identical sequences. As shown in Supplemental Table I, CAMPs in some other ﬁsh species also show repetitive motifs in the central regions, but based on different residues. Thus, the emergence and duplication of the repetitive motifs in teleost CAMPs might result from the highly selective pressures produced by the special living environment. What is the signiﬁca- tion of the emergence of the repetitive motifs in teleost CAMPs? The structure and activity analyses of trout CAMPs and their truncated variants in the present study answered this question. Our results showed that trout CAMPs are mainly consisted of b-sheet and random coil in membrane-like environments, the accuracy of which was validated by the CD spectra analysis and tertiary structure simulation. Permeability assay and SEM observation of the bacteria treated with trout CAMPs revealed that trout CAMPs killed bacteria through the disruption of the bacterial

membranes, including the inner and outer membranes. Similar to Downloaded from b-sheet protegrins in pigs stabilized by disulﬁde bonds (2), the N-terminal of trout cathelicidin 1 peptides also contained cysteines and formed antiparallel b-sheets. The antibacterial activities of these peptides were revealed to be mediated by this region. Pre- vious studies have demonstrated that the bacterial membrane is

the key target for CAMPs (6, 19). Thus, the highly cationic N http://www.jimmunol.org/ terminus can facilitate the binding of trout cathelicidin 1 peptides to the negatively charged bacterial membranes. The cell permeability assay revealed that this region can independently disrupt the bacterial membranes. The repetitive (R/P)(P/L)GGGS motifs in trout cathelicidin 1 peptides formed an antiparallel b-sheet and a flexible random coil region. From 1a to 1d, with gradual deletion of the repetitive motifs, the structural flexibilities of these peptides were gradually decreased, accompanied by the reduction of the antibacterial activities. These findings are particularly intriguing by guest on September 30, 2021 in view of the conformational effects of Pro and Gly residues on peptide secondary structures. Many studies have shown that the Pro residue could act as a flexible element in determining peptide structures, and it was a potent helix breaker (51–53). It has been FIGURE 10. The hemolytic and cytotoxic activities of trout CAMPs at reported that a Pro residue insertion in the central region of an- a concentration of 60 mM. (A and B) Hemolytic activities of CAMPs to trout timicrobial peptides could induce a hinge structure and reduce the (A) and human (B) erythrocytes. The CAMPs were added to the erythrocytes hemolytic activities of peptides (51, 52). This is consistent with and incubated at 20˚C for 2 h and then OD405 of the supernatant was mea- the high bacterial membrane selectivity of trout CAMPs, which sured. (C) Cytotoxic activities of CAMPs to trout RTG-2 cells. The CAMPs possessed no hemolytic and cytotoxic activities. Additionally, Gly were added to RTG-2 cells and incubated at 20˚C for 48 h; MTS solution was is frequently found together with Pro in the antiparallel b-sheets then added and incubated at 20˚C for an additional 4 h. The absorbance was or flexible coiled structures of peptides (53, 54). These previous subsequently measured at 490 nm. Data shown are the means 6 SD of findings precisely accord with the simulated tertiary structures of duplicate reactions in a representative of three independent experiments. trout cathelicidin 1 peptides. Furthermore, the Pro-Gly motif was found in the transmembrane domain of several ion channel– present in every vertebrate thus far investigated, even in primates forming peptides (53). Consistent with this, the cell permeability with a sole cathelicidin. Thus, the a helical class is considered to assay revealed that the repetitive motifs in 1a could independently be the ancestral molecule from which this family differentially disrupt the bacterial membranes, although the activity was low. expanded (6). In recent years, the mysterious veil of fish CAMPs Unlike human LL-37, only the C termini of trout cathelicidin 1 has been uncovered. Sequence analysis revealed that teleost peptides formed partial a helical conformations, which retained CAMPs form a novel class of peptides enriched with Arg, Gly, and little effect on the antibacterial activities. In summary, trout Ser (7). In this class, salmonid CAMPs are more special and have cathelicidin 1 peptides possess all the structural elements of three distinctive hallmarks: the N terminus is intensified by pos- mammalian CAMPs, including a helix, b-sheet, and random coil, itive charges, the central region consists of (R/P)(P/L)G(G/S/F/V) which are more likely to represent the ancestral molecules from (G/S/R)S–based repetitive motifs, and the C terminus is lowly which this family differentially expanded. charged (Supplemental Table I). In this study, two isoforms of Unlike trout cathelicidin 1 peptides, the entire molecule of trout rtCATH-1a are characterized by the gradual deletion of the cathelicidin 2 peptides mainly consists of random coil, except (R/P)(P/L)GGGS motif within the mature peptides, from five copies a short piece of antiparallel b-sheet. In mammals, a linear class of to three or one copy, respectively. This repetitive motif is unique CAMPs consisting of random coil also has been discovered in to salmonid CAMPs, with minor sequence variations and variable cattle, sheep, goats, and pigs, and these peptides are enriched with copy numbers, which suggested that it is a genetically unstable residues Pro and Arg (19, 50). In trout CAMPs 2a and 2b, Pro, Arg, 12 STRUCTURES AND ACTIVITIES OF TROUT CAMPs

FIGURE 11. LPS-binding and IL-8– stimulating activities of trout CAMPs. (A) LPS-binding activities evaluated by ELISA. The activities were investigated by incubating peptides in the LPS- immobilized 96-well microtiter plate. The bound peptides were detected by tetramethylbenzidine reaction using rabbit anti-1a or -2a pAbs and HRP-conjugated goat anti-rabbit IgG. Data shown are the means 6 SD of a representative experiment. (B) LPS-binding isotherms of cathelicidin 1a and its truncated variants evaluated by the kinetic chromogenic Limulus amebocyte lysate assay.

The EC50 is indicated as a dotted line. Serially diluted peptides were incubated with LPS in a 96-well microtiter plate, followed by amebocyte lysate reagent. Downloaded from

OD405 was measured at 10 and 16 min after the addition of chromogenic sub- strate. Data shown are the mean values of a representative experiment. (C) IL-8– stimulating activities of trout CAMPs in PBLs. The CAMPs were incubated with PBLs at 18˚C for 24 h, and IL-8 ex- http://www.jimmunol.org/ pression was determined by qPCR at different time points. Data are representative of three independent experiments. *p , 0.05, **p , 0.01 versus PBS control (ANOVA with Dunnett post hoc test).

and Gly residues are distributed throughout the entire molecular, cell membrane to cause conformational changes and further ac- by guest on September 30, 2021 accounting for 40.9 and 46%, respectively. This precisely accords tivate the surface receptors linked to intracellular signaling path- with their simulated tertiary structures. Similar to cathelicidin 1 ways indirectly. In trout CAMPs, from 1a to 1d, with deletion of peptides, cathelicidin 2 peptides also exhibited no cytotoxic and the repetitive motifs, the structural flexibilities were gradually hemolytic activities, and the antibacterial activities were also me- reduced. Thus, the cell membrane conformational changes caused diated by the highly cationic N terminus. In contrast, both of the by these peptides should have been gradually diminished, result- central region (consisting of only two copies of the repetitive ing in the decreased IL-8–stimulating capacities. These results motif) and the C terminus (with low charges) had minimal impact indicated that, besides antibacterial activities, the emergence of on the antibacterial activities of these peptides. Taken together, our the repetitive motifs in salmonid CAMPs has also increased the results suggest that positive charges and structural flexibilities are IL-8–stimulating activities. Compared to 2a, the IL-8–stimulating the key factors determining the antibacterial activities of trout ability of 2b was weaker and the time point with greatest stimu- CAMPs. Substitution studies have proved the critical role of cat- lation effect was significantly latter. As discussed above, 2a may ionic Arg residues for the activity of Bac5 (55). In future studies, be duplicated and diverged from 2b, and this evolution has led to the importance of different residues in trout CAMPs needs to be the acquisition of more potent IL-8–stimulating activities. Thus, in clarified by site-specific amino acid substitutions. fish, CAMPs may recruit immune cells to the site of infection via In addition to the direct antimicrobial effects, mammalian stimulating chemotactic IL-8. CAMPs possess other modulatory activities, such as LPS-neu- Taken together, similar to their mammalian orthologs, trout tralizing (20, 21) and IL-8–stimulating activities (22, 23, 28, CAMPs also formed the first line of host defense. Interestingly, trout 38, 39). Furthermore, the LPS-binding activities are often posi- have evolved energy-saving forms of cathelicidins. The activities of tively correlated with their antibacterial activities (21). Similar to trout CAMPs were mainly mediated by the highly cationic N ter- their mammalian orthologs, trout CAMPs also possessed these minus, whereas the emergence of (R/P)(P/L)G(G/S/F)GS motifs activities. The LPS binding was mainly mediated by the N ter- increased the antibacterial and IL-8–stimulating activities. The minus, whereas the repetitive motifs showed lower affinity, and the distinctive structural hallmarks and biological activities of trout C terminus did not show any. Thus, the LPS-binding activities of CAMPs offered new insights into the understanding of the evolution fish CAMPs are also positively correlated with their antibacterial of this family in vertebrates. Moreover, the high bacterial membrane activities, which demonstrates that the activity is a primordial selectivity of trout CAMPs provided new ideas for the design of function that has been conserved throughout 350 million years of excellent peptide antibiotics based on CAMPs in vertebrates. evolution. Moreover, trout CAMPs exhibited different abilities to stimulate the expression of IL-8 in PBLs. However, the mecha- Acknowledgments nism of IL-8 stimulation by CAMPs is unclear. A study conducted We thank Dr. Dong Han (Institute of Hydrobiology, Chinese Academy of by Braff et al. (39) suggested that LL-37 may directly interact with Sciences, China) for assistance with statistical analyses. The Journal of Immunology 13

Disclosures 27. Chang, C. I., O. Pleguezuelos, Y. A. Zhang, J. Zou, and C. J. Secombes. 2005. Identification of a novel cathelicidin gene in the rainbow trout, Oncorhynchus The authors have no financial conflicts of interest. mykiss. Infect. Immun. 73: 5053–5064. 28. Bridle, A., E. Nosworthy, M. Polinski, and B. Nowak. 2011. Evidence of an antimicrobial-immunomodulatory role of Atlantic salmon cathelicidins during References infection with Yersinia ruckeri. PLoS ONE 6: e23417. Available at: http:// 1. Yeaman, M. R., and N. Y. Yount. 2003. Mechanisms of antimicrobial peptide journals.plos.org/plosone/article?id=10.1371/journal.pone.0023417. action and resistance. Pharmacol. Rev. 55: 27–55. 29. Park, C. B., K. S. Yi, K. Matsuzaki, M. S. Kim, and S. C. Kim. 2000. Structure- 2. Kosciuczuk, E. M., P. Lisowski, J. Jarczak, N. Strzałkowska, A. Jo´źwik, activity analysis of buforin II, a histone H2A-derived antimicrobial peptide: the J. Horbanczuk, J. Krzyżewski, L. Zwierzchowski, and E. Bagnicka. 2012. proline hinge is responsible for the cell-penetrating ability of buforin II. Proc. Cathelicidins: family of antimicrobial peptides. A review. Mol. Biol. Rep. 39: Natl. Acad. Sci. USA 97: 8245–8250. 10957–10970. 30. Wang, Y., J. Hong, X. Liu, H. Yang, R. Liu, J. Wu, A. Wang, D. Lin, and R. Lai. 3. Zasloff, M. 2002. Antimicrobial peptides of multicellular organisms. Nature 2008. Snake cathelicidin from Bungarus fasciatus is a potent peptide antibiotics. 415: 389–395. PLoS ONE 3: e3217. Available at: http://journals.plos.org/plosone/article?id=10. 4. Epand, R. M., and H. J. Vogel. 1999. Diversity of antimicrobial peptides and 1371/journal.pone.0003217. their mechanisms of action. Biochim. Biophys. Acta 1462: 11–28. 31. Whitmore, L., and B. A. Wallace. 2004. DICHROWEB, an online server for 5. Scocchi, M., A. Pallavicini, R. Salgaro, K. Bociek, and R. Gennaro. 2009. The protein secondary structure analyses from circular dichroism spectroscopic data. salmonid cathelicidins: a gene family with highly varied C-terminal antimicro- Nucleic Acids Res. 32: W668–73. bial domains. Comp. Biochem. Physiol. B Biochem. Mol. Biol. 152: 376–381. 32. Whitmore, L., and B. A. Wallace. 2008. Protein secondary structure analyses 6. Zanetti, M. 2004. Cathelicidins, multifunctional peptides of the innate immunity. from circular dichroism spectroscopy: methods and reference databases. Bio- J. Leukoc. Biol. 75: 39–48. polymers 89: 392–400. 7. Maier, V. H., K. V. Dorn, B. K. Gudmundsdottir, and G. H. Gudmundsson. 2008. 33. Liu, Y., and B. Kuhlman. 2006. RosettaDesign server for protein design. Nucleic Characterisation of cathelicidin gene family members in divergent fish species. Acids Res. 34: W235–8. Mol. Immunol. 45: 3723–3730. 34. Broekman, D. C., A. Zenz, B. K. Gudmundsdottir, K. Lohner, V. H. Maier, and G. H. Gudmundsson. 2011. Functional characterization of codCath, the mature

8. Uzzell, T., E. D. Stolzenberg, A. E. Shinnar, and M. Zasloff. 2003. Hagfish in- Downloaded from testinal antimicrobial peptides are ancient cathelicidins. Peptides 24: 1655–1667. cathelicidin antimicrobial peptide from Atlantic cod (Gadus morhua). Peptides 9. Ling, G., J. Gao, S. Zhang, Z. Xie, L. Wei, H. Yu, and Y. Wang. 2014. Cath- 32: 2044–2051. elicidins from the bullfrog Rana catesbeiana provides novel template for peptide 35. Feng, F., C. Chen, W. Zhu, W. He, H. Guang, Z. Li, D. Wang, J. Liu, M. Chen, antibiotic design. PLoS ONE 9: e93216. Available at: http://journals.plos.org/ Y. Wang, and H. Yu. 2011. Gene cloning, expression and characterization of avian plosone/article?id=10.1371/journal.pone.0093216. cathelicidin orthologs, Cc-CATHs, from Coturnix coturnix. FEBS J. 278: 1573–1584. 10. Dalla Valle, L., F. Benato, S. Maistro, S. Quinzani, and L. Alibardi. 2012. Bio- 36. Nagaoka, I., S. Hirota, F. Niyonsaba, M. Hirata, Y. Adachi, H. Tamura, informatic and molecular characterization of beta-defensins-like peptides isolated S. Tanaka, and D. Heumann. 2002. Augmentation of the lipopolysaccharide- from the green lizard Anolis carolinensis. Dev. Comp. Immunol. 36: 222–229. neutralizing activities of human cathelicidin CAP18/LL-37-derived antimicro- 11. Li, Z., S. Zhang, J. Gao, H. Guang, Y. Tian, Z. Zhao, Y. Wang, and H. Yu. 2013. bial peptides by replacement with hydrophobic and cationic amino acid residues. http://www.jimmunol.org/ Structural and functional characterization of CATH_BRALE, the defense molecule Clin. Diagn. Lab. Immunol. 9: 972–982. in the ancient salmonoid, Brachymystax lenok. Fish Shellfish Immunol. 34: 1–7. 37. Xiao, Y., Y. Cai, Y. R. Bommineni, S. C. Fernando, O. Prakash, S. E. Gilliland, and 12. Nizet, V., T. Ohtake, X. Lauth, J. Trowbridge, J. Rudisill, R. A. Dorschner, G. Zhang. 2006. Identification and functional characterization of three chicken V. Pestonjamasp, J. Piraino, K. Huttner, and R. L. Gallo. 2001. Innate antimicrobial cathelicidins with potent antimicrobial activity. J. Biol. Chem. 281: 2858–2867. peptide protects the skin from invasive bacterial infection. Nature 414: 454–457. 38. Murakami, M., B. Lopez-Garcia, M. Braff, R. A. Dorschner, and R. L. Gallo. 13. Lee, P. H., T. Ohtake, M. Zaiou, M. Murakami, J. A. Rudisill, K. H. Lin, and 2004. Postsecretory processing generates multiple cathelicidins for enhanced R. L. Gallo. 2005. Expression of an additional cathelicidin antimicrobial peptide topical antimicrobial defense. J. Immunol. 172: 3070–3077. protects against bacterial skin infection. Proc. Natl. Acad. Sci. USA 102: 3750–3755. 39. Braff, M. H., M. A. Hawkins, A. Di Nardo, B. Lopez-Garcia, M. D. Howell, 14. Agerberth, B., J. Charo, J. Werr, B. Olsson, F. Idali, L. Lindbom, R. Kiessling, C. Wong, K. Lin, J. E. Streib, R. Dorschner, D. Y. M. Leung, and R. L. Gallo. H. Jo¨rnvall, H. Wigzell, and G. H. Gudmundsson. 2000. The human antimi- 2005. Structure-function relationships among human cathelicidin peptides: dis- crobial and chemotactic peptides LL-37 and a-defensins are expressed by sociation of antimicrobial properties from host immunostimulatory activities. J. specific lymphocyte and monocyte populations. Blood 96: 3086–3093. Immunol. 174: 4271–4278. by guest on September 30, 2021 15. Bals, R., X. Wang, M. Zasloff, and J. M. Wilson. 1998. The peptide antibiotic LL- 40. Berthelot, C., F. Brunet, D. Chalopin, A. Juanchich, M. Bernard, B. Noe¨l, 37/hCAP-18 is expressed in epithelia of the human lung where it has broad anti- P. Bento, C. Da Silva, K. Labadie, A. Alberti, et al. 2014. The rainbow trout microbial activity at the airway surface. Proc. Natl. Acad. Sci. USA 95: 9541–9546. genome provides novel insights into evolution after whole-genome duplication 16. Frohm, M., B. Agerberth, G. Ahangari, M. Staˆhle-Ba¨ckdahl, S. Lideń, in vertebrates. Nat. Commun. 5: 3657. H. Wigzell, and G. H. Gudmundsson. 1997. The expression of the gene coding 41. Dalla Valle, L., F. Benato, M. C. Paccanaro, and L. Alibardi. 2013. Bioinformatic for the antibacterial peptide LL-37 is induced in human keratinocytes during and molecular characterization of cathelicidin-like peptides isolated from the inflammatory disorders. J. Biol. Chem. 272: 15258–15263. green lizard Anolis carolinensis (Reptilia: Lepidosauria: Iguanidae). Ital. J. Zool. 17. Agerberth, B., H. Gunne, J. Odeberg, P. Kogner, H. G. Boman, and (Modena) 80: 177–186. G. H. Gudmundsson. 1995. FALL-39, a putative human peptide antibiotic, is 42. Sørensen, O. E., P. Follin, A. H. Johnsen, J. Calafat, G. S. Tjabringa, cysteine-free and expressed in bone marrow and testis. Proc. Natl. Acad. Sci. P. S. Hiemstra, and N. Borregaard. 2001. Human cathelicidin, hCAP-18, is USA 92: 195–199. processed to the antimicrobial peptide LL-37 by extracellular cleavage with 18. Shinnar, A. E., K. L. Butler, and H. J. Park. 2003. Cathelicidin family of anti- proteinase 3. Blood 97: 3951–3959. microbial peptides: proteolytic processing and protease resistance. Bioorg. 43. Panyutich, A., J. Shi, P. L. Boutz, C. Zhao, and T. Ganz. 1997. Porcine poly- Chem. 31: 425–436. morphonuclear leukocytes generate extracellular microbicidal activity by elastase- 19. Zanetti, M. 2005. The role of cathelicidins in the innate host defenses of mediated activation of secreted proprotegrins. Infect. Immun. 65: 978–985. mammals. Curr. Issues Mol. Biol. 7: 179–196. 44. Zelezetsky, I., A. Pontillo, L. Puzzi, N. Antcheva, L. Segat, S. Pacor, S. Crovella, 20. Nagaoka, I., S. Hirota, F. Niyonsaba, M. Hirata, Y. Adachi, H. Tamura, and and A. Tossi. 2006. Evolution of the primate cathelicidin. Correlation between D. Heumann. 2001. Cathelicidin family of antibacterial peptides CAP18 and structural variations and antimicrobial activity. J. Biol. Chem. 281: 19861–19871. CAP11 inhibit the expression of TNF-a by blocking the binding of LPS to 45. Yang, D., A. Biragyn, D. M. Hoover, J. Lubkowski, and J. J. Oppenheim. 2004. CD14+ cells. J. Immunol. 167: 3329–3338. Multiple roles of antimicrobial defensins, cathelicidins, and eosinophil-derived 21. Tack, B. F., M. V. Sawai, W. R. Kearney, A. D. Robertson, M. A. Sherman, neurotoxin in host defense. Annu. Rev. Immunol. 22: 181–215. W. Wang, T. Hong, L. M. Boo, H. Wu, A. J. Waring, and R. I. Lehrer. 2002. SMAP- 46. Zaiou, M., V. Nizet, and R. L. Gallo. 2003. Antimicrobial and protease inhibitory 29 has two LPS-binding sites and a central hinge. Eur. J. Biochem. 269: 1181–1189. functions of the human cathelicidin (hCAP18/LL-37) prosequence. J. Invest. 22. Bowdish, D. M. E., D. J. Davidson, D. P. Speert, and R. E. W. Hancock. 2004. Dermatol. 120: 810–816. The human cationic peptide LL-37 induces activation of the extracellular signal- 47. Pazgier, M., B. Ericksen, M. Ling, E. Toth, J. Shi, X. Li, A. Galliher-Beckley, regulated kinase and p38 kinase pathways in primary human monocytes. J. L. Lan, G. Zou, C. Zhan, et al. 2013. Structural and functional analysis of the Immunol. 172: 3758–3765. pro-domain of human cathelicidin, LL-37. Biochemistry 52: 1547–1558. 23. Zuyderduyn, S., D. K. Ninaber, P. S. Hiemstra, and K. F. Rabe. 2006. The an- 48. Zhu, S. 2008. Did cathelicidins, a family of multifunctional host-defense timicrobial peptide LL-37 enhances IL-8 release by human airway smooth peptides, arise from a cysteine protease inhibitor? Trends Microbiol. 16: muscle cells. J. Allergy Clin. Immunol. 117: 1328–1335. 353–360. 24. Chang, C. I., Y. A. Zhang, J. Zou, P. Nie, and C. J. Secombes. 2006. Two 49. Uribe, C., H. Folch, R. Enriquez, and G. Moran. 2011. Innate and adaptive cathelicidin genes are present in both rainbow trout (Oncorhynchus mykiss) and immunity in teleost fish: a review. Vet. Med. 10: 486–503. atlantic salmon (Salmo salar). Antimicrob. Agents Chemother. 50: 185–195. 50. Scocchi, M., A. Tossi, and R. Gennaro. 2011. Proline-rich antimicrobial pep- 25. Jin, J. Y., L. Zhou, Y. Wang, Z. Li, J. G. Zhao, Q. Y. Zhang, and J. F. Gui. 2010. tides: converging to a non-lytic mechanism of action. Cell. Mol. Life Sci. 68: Antibacterial and antiviral roles of a fish b-defensin expressed both in pituitary 2317–2330. and testis. PLoS ONE 5: e12883. Available at: http://journals.plos.org/plosone/ 51.Bobone,S.,G.Bocchinfuso,Y.Park,A.Palleschi,K.S.Hahm,and article?id=10.1371/journal.pone.0012883. L. Stella. 2013. The importance of being kinked: role of Pro residues in 26. Peng, J. X., J. L. Xie, L. Zhou, Y. H. Hong, and J. F. Gui. 2009. Evolutionary the selectivity of the helical antimicrobial peptide P5. J. Pept. Sci. 19: conservation of Dazl genomic organization and its continuous and dynamic 758–769. distribution throughout germline development in gynogenetic gibel carp. J. Exp. 52. Kim, J. K., S. A. Lee, S. Shin, J. Y. Lee, K. W. Jeong, Y. H. Nan, Y. S. Park, Zoolog. B Mol. Dev. Evol. 312: 855–871. S. Y. Shin, and Y. Kim. 2010. Structural flexibility and the positive charges are the 14 STRUCTURES AND ACTIVITIES OF TROUT CAMPs

key factors in bacterial cell selectivity and membrane penetration of 54. Benincasa, M., M. Scocchi, E. Podda, B. Skerlavaj, L. Dolzani, and R. Gennaro. 2004. peptoid-substituted analog of Piscidin 1. Biochim. Biophys. Acta 1798: Antimicrobial activity of Bac7 fragments against drug-resistant clinical isolates. Peptides 1913–1925. 25: 2055–2061. 53. Jacob, J., H. Duclohier, and D. S. Caﬁso. 1999. The role of proline and glycine in 55. Tokunaga, Y., T. Niidome, T. Hatakeyama, and H. Aoyagi. 2001. Antibacterial ac- determining the backbone ﬂexibility of a channel-forming peptide. Biophys. J. tivity of bactenecin 5 fragments and their interaction with phospholipid membranes. 76: 1367–1376. J. Pept. Sci. 7: 297–304. Downloaded from http://www.jimmunol.org/ by guest on September 30, 2021