Background

Ray-finned teleost fishes occupy the vast majority of the more than 36,000 currently described fish species [1]. Over their 400 million years of evolutionary history [2, 3], teleosts have experienced three rounds of whole-genome duplications (WGD) [4, 5], one more than other vertebrates [6]. Many duplicated genes may lose one of the duplicates via pseudogenization by the accumulation of deleterious mutations during or after duplication events [7,8,9]. Populations of a species that lose different copies of a duplicated gene may become genetically isolated, according to the divergent resolution model of speciation [10, 11]. Duplicated genes might also be retained, rather than lost, and acquire novel functions [12]: this has been suggested as a driving force for major evolutionary transitions, although the evidence for this is mixed [11, 13].

The adaptive immune system is found in all jawed vertebrates [14, 15], but has undergone fundamental modifications of the immune gene repertoire [16,17,18]. Through the loss or doubling of key immune genes during or after WGD events, the immune response system of vertebrate species may have evolved compensatory mechanisms in both adaptive and innate immunity, especially for teleost fishes that contain a number of specific genes because of their unique living environments.

The functional roles of NOD-like receptors (NLRs), key components of the innate and adaptive immune system in invertebrate and vertebrate species, have attracted much research attention. In mammals, there are 20–30 NLR family members [19, 20], while larger numbers of NLR member repertoires have been identified in fish species and other early-diverging metazoans [21, 22]. Based on different types of structural domains, NLRs are usually divided into four subfamilies; NLRA, NLRB, NLRC and NLRP [23]. NLRA subfamilies have been well characterized in teleost fishes including grass carp (Ctenopharyngodon idella) [24], channel catfish (Ictalurus punctatus) [25], and miiuy croaker (Miichthys miiuy) [26]. NLRB and NLRC subfamilies have been also identified preliminarily in several fish species, such as turbot (Scophthalmus maximus L.) [27], miiuy croaker [28] and black rockfish (S. schlegelii) [29]. Meanwhile, the important roles of NLRC genes were also mentioned in several fish species. It is well known that CD4+ T cells were the key component in the immune system, which played as the center in orchestrating adaptive immune responses against pathogenic infections [30, 31]. In Nile tilapia (Oreochromis niloticus), NLRC genes were concentratedly detected in T cells, especially for NLRC3 gene that was mainly observed a high expression level in CD4+ T Cell during LPS/LTA stimulation [32], which might suggest the potential functional roles of NLRC3 gene in teleost adaptive immune response.

According to their physiological functions, NLRs can also be classified into three subgroups; inflammasome-forming, reproductive and regulatory NLRs [33]. Several members of the inflammasome-forming NLRs have been well studied and found to cooperate with the maturation of IL-1β and IL-18 to process pyroptosis [34]. Moreover, regulatory NLRs may perform important functions by acting as either positive or negative regulators on several immune signaling pathways, including the NF-kB and mitogen-activated protein kinase (MAPK) signaling pathway, the type I IFN response and the NOD1-RIPK2 antibacterial pathway [33, 35,36,37]. In fish, regulatory NLRs, including the piscine NLRC3 subfamily proteins, have been also identified and their immune functions investigated [38, 7).

Discussion

While it is known that the small number of NLR proteins in mammals are involved in immune defense and recognize PAMPs [45,46,47], there have been few studies of the larger number of NLR proteins found in taxa other than mammals [28, http://://ftp.ncbi.nlm.nih.gov/blast/db/FASTA/) and conserved domain database (CDD) in the NCBI website [101]. The nucleotide and protein sequences of black rockfish NLRC3 subfamily homologues were extracted from the CDS and protein databases of black rockfish and confirmed by referencing to its genomic database. Python scripts and codes were edited to remove the reduplicated sequences.

To assess the functions of NLRC3 genes, the immune gene repertoires of the teleost genomes, including several conserved immune genes and several genes interacting with NLRC3 genes, were investigated through a comparative gene mining pipeline comprising BLAST searches, prediction of ORFs and annotation. AP1M2 (adaptor related protein complex 1 subunit mu 2), AP2M1 (adaptor related protein complex 2 subunit mu 1), AP3M2 (adaptor related protein complex 3 subunit mu 2), RIPK2 (receptor interacting serine threonine kinase 2), TLR4 (toll-like receptor 4), TLR9 (toll-like receptor 9), TRAF6 (TNF receptor associated factor 6) and NLRP3 (NACHT, LRR and PYD domains-containing protein 3) were confirmed. Moreover, two highly conserved genes (RAG1: recombination activating gene 1, and RAG2: recombination activating gene 2) were included as controls. Then, the orthologue sequences of these selected genes for nine species, including chicken, Chinese softshell turtle, African clawed frog, lumpfish, yellow perch, nine-spined stickleback, pike, zebrafish and honeycomb rockfish, were collected as queries to determine the copy numbers of these genes in all 17 species using the above methodology.

Phylogenetic analyses of NLRC3 subfamily

To evaluate the phylogenetic orthology of the NLRC3 subfamily, NLRC3 protein sequences of the 14 selected fish species and other high vertebrates (chicken, turtle and frog) were aligned using the MAFFT v.7.505 program. The best-fit substitution model (JTT + F + I + G4) was calculated by IQ-TREE v.2.2.0.3 and a maximum-likelihood phylogenetic tree was constructed on RAxML-NG v.1.1.0. The NLRC3 subfamily sequences of black rockfish were defined and annotated by the phylogenetic relationships between black rockfish and other species and the protein domain architecture on CDD and SMART databases [102]. Two additional phylogenetic analyses were run, using the sequences of NLRC3 genes of black rockfish, honeycomb rockfish and high vertebrates, and only the FISNA domains of the black rockfish to compare the NLRC3 orthology inference. The iTOL online tool was employed to modify the final phylogenetic trees [103].

Functional enrichment of differentially expressed NLRC3 genes

The data for differentially expressed genes of the black rockfish spleen challenged by A. salmonicida, intestine infected by E. tarda and liver with A. salmonicida, at different time points post infection (2 h, 12 and 24 h) were obtained from previous studies [73, 84, 85]. Differential expression analysis of the transcripts was performed using StringTie software and the DESeq R package (3.0.3) [104, 105]. Reads per kilobase of transcript per million mapped reads (RPKM) were obtained. The Benjamini-Hochberg correction procedure was used to adjust the resulting P-value for false discovery rate [106]. The expressed level of transcripts with | log2(Fold Change) | > 0 and P-value < 0.05 were assigned as differentially expressed. Subsequently, based on NLRC3 target sequences of the black rockfish determined above, differential expression patterns for NLRC3 genes were obtained and displayed using Heatmap and volcano plots on the web server of ImageGP [107]. In addition, the transcriptomic data sets of different organs of black rockfish, such as testis and ovary sexual organs in different develo** stages (PRJNA573572), were downloaded and analyzed in the same methods descripted above. The Heatmap plots of NLRC3 genes in these organs were drawn to support the relevant findings in this study.

In order to verify the functional processes in which differentially expressed NLRC3 genes in the black rockfish participate, Gene Ontology (GO) analyses were performed using Blast2GO v.6.0.3 software via a series of analysis schemes [108]. In detail, the query sequences of differentially expressed NLRC3 genes in the three black rockfish transcriptomic databases were matched against the GO annotation database with GO map** after running BLAST at the NCBI. Next, functional annotation was conducted to select GO terms from the GO pool obtained by the map** step and assign them to the query sequences. Then, GO-slim was carried out to functionally summarize a sequence dataset in a uniform and species-specific way. Finally, GO graph visualization was used to generate combined gene ontology annotation graphs in three GO functional categories (Biological Process, Molecular Function, Cellular Component).