Relationship between amino acid sequence diversity and antimicrobial activity of Japanese quail cathelicidins

Background Japanese quail is a low-fat, meat-bird species exhibiting high disease resistance. Cathelicidins (CATHs) are host defense peptides conserved across numerous vertebrate species that play an important role in innate immunity. The activity of host defense peptides can be affected by amino acid substitutions. However, no polymorphisms in avian CATH genes have been reported to date. The aim of this study was to clarify the effect on antimicrobial activity of polymorphisms in CATHs. Results DNA for genomic analyses was extracted from the peripheral blood of 99 randomly selected quail from 6 inbred lines. A total of six, four, four, and six CjCATH3 , CjCATH2 , CjCATHB1 , and CjCATH1 alleles were identified, respectively. Nine haplotypes, including four that were strain specific, were identified in alleles from the CjCATH3 , CjCATH2 , CjCATHB1 , and CjCATH1 loci. In addition, two and one amino acid substitutions (I145F, Q148H, and P245H) predicted by PROVEAN and PolyPhen-2 to have deleterious effects were detected at CjCATH2 and B1, respectively. Synthetic CjCATH2 and B1 peptides exhibited greater antibacterial activity against Escherichia coli than chicken CATH2 and B1, respectively. Furthermore, CjCATHB1*04 peptide exhibited less-potent antimicrobial activity than other CjCATHB1 peptides examined. Conclusions This is the first report of amino acid substitutions accompanied by changes in antibacterial activity in avian CATHs. These findings could be employed as indicators of improvements in the innate immune response in poultry. potent than suggesting that charge is important to antimicrobial activity. The difference in antimicrobial activity of the two CjCATHB1 peptides was associated with substitution of proline with a more-hydrophilic histidine residue as the first amino acid. The amino acid substitutions identified in our study affect the antimicrobial activity of CATHs, which could in turn affect the binding affinity for LPS and ultimately the effectiveness of innate immune responses in this commercially important species. Overall, these results suggest that genetic diversity may affect resistance to infection by pathogens in Japanese quail.

In the genome of Japanese quail, we identified a cluster of four chicken orthologous CATH genes (CjCATH3, CjCATH2, CjCATHB1, and CjCATH1) located within approximately 13 kb of one another [10]. This cluster was highly conserved in synteny with chickens. Therefore, investigation of polymorphisms and the function of quail CATHs may contribute to improving the anti-disease properties of chickens and other pheasants.
The identification of amino acid substitutions in mediators of innate immune responses can help to elucidate the relationship between host genomic variations and protection against invading pathogens [20,21,22,23]. Single nucleotide polymorphisms (SNPs) and other mutations in HDP genes can affect the function of HDPs and cause differential susceptibility to infections [23,24,25,26].
Multiple CATHs have been identified in a variety of vertebrate species but not in euarchontoglires (e.g., primates, rabbits, and rodents) and carnivores (e.g., cats and dogs) [4]. CATHs are polymorphic in nature, with several known sequence variations in frogs [27], buffalos [28], and humans [29]. Furthermore, copy number variations have been reported in CATH genes in cattle [30]. However, no relationships between the identified genetic variations and function of CATH peptides have been reported to date. In 2017, we provided the first report of amino acid substitutions in avian CATHs from Japanese quail [10]. Analyses using two different software programs predicted that amino acid substitutions in the Japanese quail variants CATH2 and B1 (CjCATH2 and B1) would have potentially significant deleterious effects on peptide function [10]. Although sequence variations have been reported in CATH peptides of other species, the effects of 5 polymorphisms in Japanese quail CATH genes on peptide function have not been reported to date. Thus, the aim of this study was to characterize CATH diversity in Japanese quail and determine the effects of amino acid substitutions on peptide function. We identified the haplotypes and amino acid substitutions in the CATH gene cluster of Japanese quail (CjCATH) and characterized the effects of amino acid substitutions on antimicrobial activity.

Animal care
The management of Japanese quail and all procedures involving animals in the present study were performed in accordance with the Animal Experimental Guidelines of Tokyo University of Agriculture.

Nucleic acid isolation
Peripheral blood samples were collected from the jugular vein. Genomic DNA was isolated from peripheral red blood cells using a DNeasy Blood and Tissue kit (QIAGEN) according to the manufacturer's protocol.

Amplification of genomic fragments of four CjCATH genes
Four fragments (approximately 325 to 517 bp) were amplified using four primer pairs (Additional file 6) designed based on each CjCATH gene (GenBank accession no. LC136907). Each 50-μL PCR reaction mixture contained 50 ng of quail genomic DNA, 1.25 U of Prime STAR ® GXL DNA polymerase (Takara), 1× PrimeSTAR ® GXL buffer (Mg 2+ concentration 1 mmol/L), 0.2 mmol/L of each dNTP, and 0.2 μmol/L of each primer. Cycling parameters were as follows: initial denaturation at 98°C for 1 min, followed by 35 cycles of denaturation at 98°C for 10 s, annealing at 58°C for 15 s, and extension at 68°C for 30 s. A 2-μL aliquot of each PCR reaction was then analyzed by electrophoresis on a 1.0% agarose gel. PCR products were purified using a QIAquick PCR Purification kit (QIAGEN) and then subjected to direct sequencing using an ABI Capillary System (Macrogen Research).

Amplicon sequencing using MiSeq
The partial nucleotide sequences of CjCATH2 and B1 and CATH3 , including insertions or deletions, were determined using MiSeq (Illumina). The forward and reverse primers included the Illumina adapter overhang nucleotide sequences (Additional file 2). Each 50-μL PCR reaction mixture contained 50 ng of quail genomic DNA, 25 μL of KAPA HIFI HotStart Ready Mix (KAPA), and 1 μM forward and reverse primers. PCR products were purified using an AMPURE XP (Beckman Coulter). The index PCR was conducted using a 50-µL mixture consisting of 25 µL of KAPA HIFI HotStart Ready Mix, 5 µL of Nextera XT Index Primer 1 (Illumina), 5 µL of Nextera XT Index Primer 2 (Illumina), 10 µL of sterile H 2 O, and 5 µL of first-PCR product. PCR products were purified using AMPURE XP (Beckman Coulter).
Library quality was assessed on an Agilent 2200 Tapestation (Agilent Technologies). The libraries were sequenced as paired-end, 300-bp reads using a MiSeq (Illumina) platform according to the manufacturer's instructions. All sequence data associated with this 7 project were submitted to the DNA Data Bank of Japan Sequence Reads Archive (DRA) (DRA006654).

Amplicon sequencing analysis
The overall quality of the MiSeq reads was evaluated using FastQC. The reads were trimmed 1 bp from both the site and low-quality ends (phred score <30) using the FASTX tool kit (http://hannonlab.cshl.edu/fastx_toolkit/), and unpaired reads were removed. The MiSeq reads were joined using fastq-join [46] with the option -p 5 -m 40, and adapter trimming was performed using cutadapt 1. containing complete primer sequences were recovered from among the un-mapped reads, and primer sequences were trimmed. Trimmed primer reads with an average quality score >30 and without unknown nucleotides were retained for further analysis. High-quality reads were converted to a fasta file, and identical sequences were collapsed using the FASTX tool kit. The BLAST database was generated from each CATH using the makeblastdb program included with the BLAST 2.5.0+ package. BLAST searches (blastn algorithm) were used to identify the most-similar CATH sequences. In the analysis of CATH3, identification as a heterozygote or homozygote was based on whether the second highest number of reads was more than one-third of the highest read number.

Comparison of nucleotide sequences
PCR product sequences were aligned using ClustalW, and alleles were then identified.

Comparison of nucleotide and amino acid sequences
Net charges were calculated using Innovagen's Peptide Property Calculator (https://pepcalc.com/). Hydropathy indexes used the value reported by Kyte & Doolittle (1982) [50]. The impact of all non-synonymous SNPs on the function of each CATH peptide was assessed using the PROVEAN (http://provean.jcvi.org/index.php) [32] and PolyPhen-2 [33] software packages. A resulting PROVEAN score of less than −2.500 indicated a variant was deleterious, and a score of greater than −2.500 indicated a neutral variant [32].

Antimicrobial region peptide synthesis
GgCATH peptides were designed based on the mature peptide region (Figure 1). CjCATH peptides were designed from six peptides isolated from the mature peptide region of CjCATHs included in this experiment based on amino acid substitutions ( Figure 1). All 10 peptides were synthesized by the Funakoshi Corporation, and all were >95% pure.
Lyophilized peptides were stored in desiccant at −20°C and dissolved in dimethyl sulfoxide and diluted in 10 mmol/L phosphate buffer (pH 7.2) before use.

Antimicrobial activity assay
Escherichia coli (NRIC 1023) was provided by the Nodai Culture Collection Center and used to assess the antimicrobial activity of the 10 synthesized peptides (Figure 1). Following overnight incubation at 37°C, E. coli cells were subculture for an additional 2 h at 37°C to the mid-logarithmic phase. The cells were then washed once with 10 mmol/L phosphate buffer (pH 7.2) and suspended to a concentration of 1 × 10 4 colony-forming units/mL in the same buffer. Ninety microliters of the E. coli suspension were placed in 0.2-mL tubes, 9 followed by the addition of 10 μL of serially diluted peptide in triplicate. After a 3-h incubation at 37°C, the cultures were plated onto LENNOX L BROTH agar, incubated overnight at 37°C, and then the number of surviving bacterial colonies was determined.

Statistical analyses
Antimicrobial activity was analyzed using a one-way analysis of variance, followed by the Holm-Bonferroni test for unpaired data. Differences with a P value of <0.05 were considered statistically significant.

Sequence diversity
To assess allelic diversity in the CjCATH mature peptide-encoding regions, PCR products were amplified from 99 quail specimens representing six strains. A total of 10 SNPs were detected, 5 each at intron3 and exon4 between partial exon3 and partial exon4 of the CjCATH3 locus (325 bp) (Additional file 1). However, any SNPs did not exist in the mature peptide region. Six alleles (CjCATH3*01~*06: GenBank accession nos. LC426739-LC426744) were identified based on these 10 SNPs of CjCATH3 (Table 1) A total of 11 SNPs were detected, 1 at intron3 and 10 at exon4 (mature peptide region) between partial exon3 and partial exon4 of the CjCATH2 locus (464 bp) (Additional file 2).
A total of five SNPs were detected, three at intron3 and two at exon4 (mature peptideencoding region) between partial exon3 and partial exon4 of the CjCATHB1 locus (509 bp) (Additional file 3). Four alleles (CjCATHB1*01~*04: GenBank accession nos. LC426745-LC426748) were identified based on these five SNPs of CjCATHB1 (Table 1). A total of seven SNPs were detected, five at intron3 and two at exon4 between partial exon3 and partial exon4 of the CjCATH1 locus (517 bp) (Additional file 4). One SNP with synonymous substitution was detected in the mature peptide-encoding region. Six alleles (CjCATH1*01~*06: GenBank accession nos. LC426729-LC426734) were identified based on these seven SNPs of CjCATH1 (Table 1) A comparison of the CATH3, CATH2, CATHB1, and CATH1 genes in the 99 quail specimens revealed six, four, four, and six alleles, respectively. We identified nine haplotypes (HT1-HT9) based on combinations of the CATH3, CATH2, CATHB1, and CATH1 alleles (Table 1).
HT1 was the primary haplotype in strains ND and P, whereas HT2 was the primary haplotype in strain B. Haplotypes HT3 and HT4 were detected only in strain ND, and HT9 was found only in strain B. HT7 was a major haplotype specific to strain K. Among the 26 AA of the sequence of the mature CjCATH1 peptide, two differed from the sequence of GgCATH1 (L128W and R147K). However, no significant differences in net charge or hydropathy index were observed between CjCATH1 and GgCATH1. (Table 2) Synthetic peptides representing all quail and chicken CATHs exhibited concentrationdependent antibacterial activity against E. coli. The amino acid sequence of the antimicrobial region of both chicken and Japanese quail CATH3 was the same. However, the amino acid sequences of CATH2, CATHB1 and CATH1 in the quail and chicken differed from each other.

Antimicrobial activity
At a concentration of 0.1 to 1 µM, the antimicrobial activity of CjCATH2s against E. coli tended to be higher than that of GgCATH2. The antimicrobial activity of CjCATH2*01*03 and CjCATH2*02 in particular was significantly more potent than that of GgCATH2.
However, there were minimal differences in the antimicrobial activity of CjCATH2 and GgCATH2 against E. coli at concentrations of 10 and 100 µM.
At concentrations of 0.1 to 1 µM, the antimicrobial activity of CjCATHB1s against E. coli tended to be higher than that of GgCATHB1. The antimicrobial activity of CjCATHB1*01*02*03 at a concentration of 0.5 µM was significantly more potent than that of GgCATHB1. However, there were minimal differences in the antimicrobial activity of CjCATHB1 and GgCATHB1 against E. coli at concentrations of 10 and 100 µM. In addition, there were minimal differences in the antimicrobial activity of CjCATH1 and GgCATH1 against E. coli at all concentrations examined.
At a peptide concentration of 0.1 µM, CjCATH3 (=GgCATH3), CjCATH1, and GgCATH1 exhibited significantly more-potent antimicrobial activity against E. coli than GgCATH2, CjCATHB1, and GgCATHB1. At a peptide concentration of 0.5 μM, CjCATH3, CjCATH2s, CjCATHB1*01*02*03, CjCATH1, and GgCATH1 exhibited significantly more-potent antimicrobial activity against E. coli than GgCATH2 and GgCATHB1. At a peptide concentration of 1 μM, CjCATH2*01*03 and CjCATH1 exhibited significantly more-potent antimicrobial activity against E. coli than GgCATH2 and GgCATHB1. In contrast, at a peptide concentration of 10 μM, no differences between CATH-derived peptides were observed in terms of antimicrobial activity against E. coli. At a CATH-derived peptide concentration of 100 µM, no survival of E. coli was observed.

Discussion
In nature, CATHs are polymorphic, exhibiting several known sequence variants in frogs [27], buffalos [28], and humans [29]. In addition, copy number variations have been reported in CATH genes in cattle [30]. No polymorphisms were detected in the amino acid sequence of the mature antimicrobial peptide from Japanese quail liver-expressed antimicrobial peptide 2 (CjLEAP-2), however, and the sequence was found to be the same as that of chicken GgLEAP-2. This sequence similarity suggests that LEAP-2 is highly conserved in birds and may therefore play an as yet unidentified key role in mediating innate immunity [31]. The amino acid sequences of mature CjCATH3 and GgCATH3 peptides are identical, and the similarity between CjCATH1 and GgCATH1 is >90%.
Furthermore, CjCATH2/GgCATH2 and CjCATHB1/GgCATHB1 exhibit 83.1% and 68.7% similarity, respectively [10]. These data thus suggest that, similar to CjLEAP-2, CATH1 and CATH3 exhibit high sensitivity in terms of recognition of universally present antigens. The amino acid substitutions identified in CATH2 and CATHB1 may be associated with enhanced antigen sensitivity. The observed less-potent antimicrobial activity of CATH2 and CATHB1 against E. coli compared with CATH3 and CATH1 supports this hypothesis.
14 Moreover, amino acid substitutions in both CjCATH2 and CjCATHB1 were shown to affect these peptides' antimicrobial activity against E. coli. These results were supported by analyses using two different software programs that predicted the amino acid substitutions in CjCATH2 and CjCATHB1 would adversely affect peptide function (PROVEAN:32 and PolyPhen-2 33).
Previous studies focusing on ostrich defensins and avian NK-lysin identified an association between membrane leakage/microbicidal activity and peptide charge [34,35]. However, this association has not been verified with quail and chicken CATH2. CjCATH2 has a lower net charge than GgCATH2 but exhibited more-potent antimicrobial activity at concentrations of 0.1, 0.5, and 1 µM. Hydrophobicity is thought to be an important mediator of the interaction of the CATHs with bacterial membranes [36,37]. Loss of the first tryptophan residue in GgCATH1 diminishes the peptide's antimicrobial activity, indicating the importance of this hydrophobic residue for the activity of the peptide [18].
Loss of the more-hydrophobic C-terminal residue in GgCATH2 analogues also results in diminished bactericidal activity [38], although C-terminal truncation of a GgCATH2 analogue (C1-15 mature peptide) enhanced the antibacterial activity [39]. Interestingly, Nterminal truncation leaving only the hydrophobic C-terminal alpha-helix resulted in almost complete loss of antibacterial activity, probably because the first interaction with bacterial cells involves the polar portions of the N-and C-termini [38,39]. Substitution of phenylalanine with the more-hydrophobic tryptophan in the C1-15 mature peptides resulted in enhanced bactericidal activity and better stability in the presence of high salt concentrations [40]. Additionally, substitution of tyrosine with an alanine in a cecropin Amagainin-2 fusion peptide resulted in markedly reduced antibacterial activity [41,42].
In the present study, the antimicrobial activity of the three CjCATH2 peptides against E.
coli was higher than that of the GgCATH2 peptide, presumably because the C-terminal hydrophobicity of CjCATH2 was higher than that of GgCATH2. However, differences in the C-terminal hydrophobicity among the three CjCATH2 peptides had no apparent effect on the peptides' antimicrobial activity. Substitution of isoleucine with phenylalanine in phenol soluble modulin α (PSMα) greatly reduced antimicrobial activity against Legionella pneumophila [43], whereas the antimicrobial activity of warnericine RK against L. pneumophila was not affected by this substitution [43]. These data thus suggest that the antibacterial activity of CATH2 is mediated by some factor other than hydrophobicity.
The antimicrobial activity of CjCATHB1 peptides against E. coli was more potent than that of GgCATHB1 peptides. The hydrophobicity of the mature CjCATHB1 peptides was lower than that of GgCATHB1. Conversely, the net charge of the CjCATHB1*01*02*03 and CjCATHBB1*04 peptides was higher than that of the GgCATHB1 peptide. These observations suggest that the antibacterial activity of CATHB1 is affected by net charge.
In contrast, the antimicrobial activity of CjCATHB1 peptides differed at a concentration of 10 nM. Substitution of proline with a more-hydrophilic histidine as the first amino acid of mature CjCATHB1 enhanced the peptide's antimicrobial activity (Table 2). These results are in agreement with those of previous reports indicating that the antimicrobial activity of avian CATHs is affected by hydrophobicity [18,38,39,40]. Therefore, the antimicrobial activity of CjCATHB1 may be affected by hydrophobicity in a manner similar to CjCATH1 and CjCATH2.

Conclusions
In this study, several SNPs in the mature peptide-encoding regions of CjCATH genes were identified. CjCATH2 contains six non-synonymous SNPs, whereas CjCATHB1 contains one; three and two different peptides, respectively, were synthesized based on each allele. The antimicrobial activity of the three CjCATH2 peptides against E. coli determined in vitro was more potent than that of GgCATH2 peptide, suggesting that hydrophobicity is important to 16 antimicrobial activity. The antimicrobial activity of the two CjCATHB1 peptides was more potent than that of GgCATHB1 peptide, suggesting that net charge is important to antimicrobial activity. The difference in antimicrobial activity of the two CjCATHB1 peptides was associated with substitution of proline with a more-hydrophilic histidine residue as the first amino acid. The amino acid substitutions identified in our study affect the antimicrobial activity of CATHs, which could in turn affect the binding affinity for LPS and ultimately the effectiveness of innate immune responses in this commercially important species. Overall, these results suggest that genetic diversity may affect resistance to infection by pathogens in Japanese quail.

Ethics approval
The management of Japanese quail and all experimental procedures undertaken in the present study were performed according to the recommendations of the Guidelines by the ethics committee for animal experiments of the Tokyo University of Agriculture (Tokyo, Japan). All Japanese quails were maintained in order to preserve the strain after the study.

Consent for publication
Not applicable.

Availability of data and material
The data sets supporting the results of this article are included within the article and its additional files. All Next Generation Sequencer of sequences data associated with this project were submitted to the DNA Data Bank of Japan Sequence Reads Archive (DRA) (DRA006654).

23
The authors declare that they have no competing interests.