Leptin in Darkbarbel Catfish Pseudobagrus vachellii: Molecular Characterization, Synteny and Phylogeny, Tissue Distribution, and Expression in Response to Different Feeding Status

1Key Laboratory of Sichuan Province for Fishes Conservation and Utilization in the Upper Reaches of the Yangtze River, Neijiang Normal University, Neijiang, 641100, China 2College of Life Science, Neijiang Normal University, Neijiang, 641100, China 3Engineering Research Center of Eco-environment in Three Gorges Reservoir Region, Ministry of Education, China Three Gorges University, Yichang 443002, China Article Information Received 22 July 2021 Revised 25 August 2021 Accepted 06 September 2021 Available online 22 February 2022 (early access)


INTRODUCTION
L eptin is a class-I helical cytokine peptide primarily secreted by adipocytes in mammals, which was first discovered from mouse in 1994 (Zhang et al., 1994). Subsequently, leptin was cloned in human (Cohen et al., 1996) and other mammals (Denver et al., 2011). In nonmammalian animals, leptin was first cloned from puffer fish (Takifugu rubripes) in 2005 (Kurokawa et al., 2005), the delay may be due to the low identity and similarity of Leptin in different lineages (Londraville et al., 2017). Thus far, it has been well demonstrated that Leptin plays vital roles in suppressing food intake, modulating cell morphology and cytokine release, stimulating the reproductive endocrine system, promoting bone formation, and maintaining energy homeostasis (Barash et al., 1996;Steppan et al., 2000;Friedman, 2002;Klok et al., 2006;Lafrance et al., 2010). However, the roles of leptin are variable due to its divergent evolution in vertebrates.
The darkbarbel catfish (P. vachelli) belongs to siluriformes, bagridae, and is an omnivorous freshwater fish native to Asia, which has become an economically important aquaculture species in China due to its fast growth and valuable taste traits (Qin et al., , 2018a. In the present study, a leptin gene was cloned and characterized in darkbarbel catfish (Pvleptin), and its expression level in response to different feeding status was examined. These findings will help us to better understand the evolutionary history and functional roles of Leptin in teleosts, as well as provide potential feeding management measures to improve the production of this species.

Fish sampling
Darkbarbel catfish (bodyweight 4.9 ± 0.3 g) used in this study were purchased from local aquatic market in Neijiang city of China and fishes were transported to the experimental aquarium in the Key Laboratory of Sichuan Province for Fishes Conservation and Utilization in the Upper Reaches of the Yangtze River (Neijiang Normal University). Fishes were cultured in 100 L tanks with a constant flow of filtered water under natural light-dark conditions (12 L/12 D). The aquicultural water was aerated using an air pump, and the water temperature was maintained at 20.0±0.5°C. For cloning and tissue distribution experiments, a total of five fishes were randomly selected and then anesthetized with 10 mg.L -1 MS-222. Subsequently, fishes were sacrificed by decapitation and tissues samples including adipose, brain, gill, heart, intestine, kidney, liver, muscle, spleen and stomach were collected and immersed in liquid nitrogen immediately, then, were kept at -80°C for further utilization.
For fasting and refeeding experiments, fishes were assigned to 3 groups (with triplicate tanks per group; 15 fishes per tank) and then conducted as described in several previous studies (Qin et al., 2018b;Yang et al., 2018;Wen et al., 2020d). Fishes in control tanks were fed once daily at 19:00, while fishes in fasting group were not fed for two weeks. For refeeding group, fishes were fasted for two weeks and then fed at 19:00. Fishes were allowed to feed for 30 minutes then five fishes from each tank (three replicates, a total of 15 fishes for each group) were randomly selected and their livers were collected. All samples were treated as described above and finally were kept at -80°C for further experiments.
The animal experiments were conducted following the approval of the Neijiang Normal University Animal Care and Use Committee and in full compliance with its ethics guidelines.

Molecular cloning of Pvleptin
Total RNA was isolated from liver with the Trizol O n l i n e F i r s t A r t i c l e reagent (Invitrogen, USA) following the manufacturer's instruction, and 1 μg of the RNA from each sample was reversely transcribed to cDNA by using Super Script TM II RT reverse transcriptase (Takara, Japan). Two pairs of primers were designed basing on a transcriptome data established in our previous study , and then they were used to amplify the complete open reading frame (ORF) sequences of Pvleptin by using leptin sequences from yellow catfish and channel catfish as references (Kobayashi et al., 2011;Gong et al., 2013b), and the primer information is shown in Table I. The basic cycling conditions of the PCR were set as follows: a denaturing stage at 94°C for 30 s, an annealing stage at gene-specific temperature for 45 s and an elongation stage at 72°C for 60 s, a total of 34 cycles. The products were purified from agarose gel using the Universal DNA Purification Kit (Tiangen, China), and then cloned into the pMD-19T vector (TaKaRa, Dalian, China) and finally sequenced at BGI-Wuhan (Wuhan, China). Multiple sequences alignment and three-dimension (3D) structure prediction Two cDNA sequences obtained from the sequencing clones were assembled into one complete sequence. Subsequently, the ORF finder (https://www.ncbi.nlm.nih. gov/gorf/gorf.html) and Primer Premier 5.0 software were used to determine the ORF and predict the putative protein sequence of Pvleptin, respectively (Wen et al., 2020a). Meanwhile, signal peptide was predicted using the online tool Signal P 4.1 Server (http://www.cbs.dtu.dk/services/ SignalP/). Furthermore, ClustalX and BioEdit were conducted to perform the multiple alignments as described in our previous studies (Wen et al., 2019(Wen et al., , 2020b. Finally, the SWISS-MODEL (https://swissmodel.expasy.org/) was utilized to predict the Leptin three-dimension (3D) structures of representative species Wen et al., 2020c).

Synteny and phylogenetic analyses
Genetic synteny was conducted to better understand the evolutionary history of leptin genes in representative species. In silico protein similarity-based blast was executed to against the genome datasets of representative species using zebrafish Leptin and its flanking proteins as queries. Genome datasets were downloaded from O n l i n e

F i r s t A r t i c l e
Ensemble (http://asia.ensembl.org/index.html) or NCBI (https://www.ncbi.nlm.nih.gov/) databases. Meanwhile, phylogenetic analysis was performed to declare the relationship of leptin genes in vertebrates. A series of Leptin protein sequences of representative species were also downloaded from NCBI or Ensemble databases, and their accession numbers were listed in Table II. The protein dataset was aligned by using ClustalX software, and then the best-fitting model was evaluated by Mrmodeltest 2.0 and ProtTest 2.4. Subsequently, phylogenetic tree was constructed with neighbor-joining method by using Mega 6.0 software . The robustness of the tree topology was assessed by nonparametric bootstrap analysis with 1,000 resampling replicates. The tree was beautified with FigTree software and spotted gar (Lepisosteus oculatus) was selected as the outgroup species.

Real-time quantitative PCR
Real-time quantitative PCR (qPCR) was used to detect the mRNA expression level of Pvleptin, which could be helpful for better understanding the tissue distribution pattern and nutritional regulation of leptin in the darkbarbel catfish. RNA isolation and first strand cDNA synthesis were conducted as described above. The qPCR reaction system contained 10 μL SYBR Green Master Mix (Thermo Fisher Scientific, Waltham, MA, USA), 8 μL double distilled H 2 O, 0.5 μL forward/reverse specific primer (10 μM), and 1 μL reverse transcribed product, with a final volume of 20 μL. Then qPCR was conducted on a Light Cycler Real-Time system and the running procedure was designed following the manufacturer's instruction. The end products of qPCR were verified with the melting curves that showing a single peak specific for the target gene. Relative leptin mRNA expression was calculated by using method described in previous studies (Pfaffl, 2001;Da et al., 2021;Wen et al., 2021), and β-actin was selected as reference gene after assessing the stability of several potential housekeeping genes. Primers were provided in Table I.

Statistical analysis
Statistical analysis was performed with SPSS 22.0 (IBM, Armonk, NY, USA) and GraphPad Prism (San Diego, CA, USA). All data were shown as mean normalized values ± standard error of the mean. Significant differences were evaluated by using one-way analysis of variance (ANOVA), followed by the post hoc test (least significant difference test and Duncan's multiple range test), after confirming for data normality and homogeneity of variances. Differences were considered to be significant if P < 0.05.

cDNA characterization of the Pvleptin
The characterized cDNA sequence of Pvleptin was 1186 bp long containing a 148 bp 5'-UTR, a 539 bp 3'-UTR, and a 519 bp ORF that predicted to encode a Leptin precursor of 172 amino acids (Fig. 1). A putative signal peptide with 23 amino acids was identified at the N-terminal of the Leptin precursor (Fig. 1). Similar to other teleosts, four conserved a-helix domains were discovered in the mature Leptin of darkbarbel catfish with length ranging from 16-23 amino acids (Fig. 1). The electronic point and molecular weight of the putative PvLeptin were calculated to be 7.96 and 19.89 kDa, respectively. The cDNA sequence of Pvleptin has been deposited into GenBank database with an accession number MW251477.

Multiple sequences alignment and 3D structure prediction
Multiple sequences alignment was performed based on the protein sequences to better understand the structural and functional properties of the vertebrate Leptins. We observed that Leptins commonly contained a signal peptide, four a-helix domains and two conserved cysteine residues were also identified in vertebrates ( Fig. 2A). Meanwhile, sequence identity analysis was also conducted, and results showed that PvLeptin shared low identity with Leptin in tetrapods (human, 19.2%; rat, 19.7%; chicken, 20.1%; frog, 25.9%) and most teleosts (zebrafish LepA, 37.3%; zebrafish LepB, 22.0%; snakehead LepA, 22.1%; snakehead LepB, 15.3%; rainbow trout LepA, 28.4%; rainbow trout LepB, 17.4%), whereas shared high identity with that in siluriformes fish including yellow catfish (99.4%) and channel catfish (90.1%) (Supplemental Table  I). Interestingly, despite PvLeptin shares low identity with that in most teleosts, it seems that PvLeptin is closer to teleost LepA than LepB. Additionally, 3D structures modeling revealed that the 3D structures of Leptin in four representative species (human, rat, frog and darkbarbel catfish) were highly conserved (Fig. 2B).

Genetic synteny and phylogenetic analyses
Genetic synteny and phylogenetic analysis were performed to better understand the evolutionary history and phylogenetic relationship of leptin genes in vertebrates. Synteny analysis showed that only a single copy of LEPTIN gene was found in mammals including human and rat, while two copies of leptin gene were extensively existed in teleosts, such as in zebrafish, pachon cavefish (Astyanax mexicanus) and snakehead fish (Fig. 3)  genomes and teleost genomes respectively, sharing same core cluster of lrrc4-leptin-rbm28 (Fig. 3). Differently, a specific cluster of pax4-lrrc4a-snd1-lepB-impdh1b was found in teleost genomes (Fig. 3). Interestingly, it seems that lepB but not lepA has lost in Siluriformes fishes including yellow catfish and channel catfish (Fig. 3). Phylogenetic analysis showed that the neighbor joining tree was divided into two groups of teleost leptin and tetrapod leptin, and the former was further clustered into two subgroups including teleost lepA and teleost lepB (Fig. 4). The Pvleptin was clustered into lepA clade and shared a close relationship with leptin in yellow catfish and channel catfish (Fig. 4), consistent with the protein identity described above. All clades were supported with high scores, and the spotted gar (Lepisosteus oculatus) was selected as outgroup species due to its special evolution potion in teleost.

Tissue distribution of Pvleptin
Quantitative real-time PCRs were conducted to detect the tissue distribution pattern of Pvleptin. Results showed that the Pvleptin was widely distributed in examined tissues including adipose, brain, gill, heart, intestine, kidney, liver, muscle, spleen, and stomach (Fig. 5). The highest mRNA expression level of Pvleptin was detected in liver, while relative high expression was tested in adipose, heart, intestine, muscle, spleen and stomach (Fig.  5). While, Pvleptin was hardly detectable in brain, gill and kidney (Fig. 5).

Effect of fasting and refeeding on Pvleptin mRNA expression
To investigate the expression patterns of Leptin associated with starvation and feeding schemes, the mRNA expression level of Pvleptin in the liver was detected after food deprivation and refeeding. The mRNA expression level of hepatic Pvleptin was significantly decreased in fishes after a two-week fasting in comparison with that of those in feeding group, while refeeding increased the transcription of hepatic Pvleptin of the fasted fish (Fig.  6). Groups with significant differences were indicated by different letters above the bars. Data were shown as mean ± SEM.

O n l i n e F i r s t A r t i c l e
Leptin in Darkbarbel Catfish Pseudobagrus vachellii Fig. 4. Neighbor joining tree reveals the phylogenetic relationship among vertebrate leptin genes. The tree was constructed by using MEGA 6.0 program based on a protein dataset. Values on the nodes represent bootstrap percentages from 1000 replicates. The aimed species was highlighted by a rhombic diamond and spotted gar (Lepisosteus oculatus) was used as the outgroup species.

DISCUSSION
In the present study, we characterized a leptin gene from the darkbarbel catfish (P. vachelli) for the first time. The cDNA of Pvleptin contained a 519 bp long ORF that predicted to encode a precursor protein of 172 amino acids, which was in line with the findings in yellow catfish (Gong et al., 2013b) and channel catfish (Kobayashi et al., 2011).
PvLeptin was predicted to contain four conserved a-helix domains and two cysteine residues, which was similar to previous studies in pufferfish (Kurokawa et al., 2005), Jian carp (Tang et al., 2013), mandarin fish (He et al., 2013;Yuan et al., 2016), and snakehead fish (Wen et al., 2020d), suggesting these conservative domains or amino acid residues are especially important for maintaining the 3D structures and functions of the vertebrate Leptin. Multiple sequences alignment revealed that Leptins were variable and shared low identity with each other in vertebrates, which was consistent with most studies related to this hormone (Londraville et al., 2017). It is noticed that PvLeptin shared higher identity with teleost LepA in comparison of that with teleost LepB, suggesting the Pvleptin identified in present study may be the ortholog of lepA in teleosts (Kobayashi et al., 2011). Although low sequence identities of Leptins were observed among different animals, 3D structure modeling showed that the 3D structures of Leptins were highly conserved, implying they may experience an independent evolution process while still potentially restrain similar functional roles in vertebrates (Munzberg and Morrison, 2015;Londraville et al., 2017). Genetic synteny analysis showed that yellow catfish and channel catfish possessed a single leptin gene while the other teleosts contained two copies of leptin genes (namely lepA and lepB) in their genomes, and the two single leptin genes shared a consistent genetic synteny with that of other teleost lepA showing a same gene order (cacna2d1-hgfb-lrrc4b-lepA-rbm28) (Fig. 3). These findings were identical to several previous studies (Kobayashi et al., 2011;Gong O n l i n e

F i r s t A r t i c l e
Z-Y. Wen et al. et al., 2013b;Wen et al., 2020d), suggesting that the ortholog of lepA was retained whereas lepB had lost in siluriformes genomes. A recent work has well illustrated the phylogenetic relationship of ray-finned fishes based on big genome and transcriptome datasets (Hughes et al., 2018), which revealed that Cypriniformes, Characiformes, Gymnotiformes and Siluriformes were belonged to Otophysa and the Cypriniformes was located at the root of this lineage. In present study, lepB was identified both in zebrafish and pachon cavefish, a representative species of Cypriniformes and Characiformes, respectively whereas lepB-impdh1b cluster has lost in Siluriformes including yellow catfish and channel catfish, suggesting the lost event of lepB should be specific in Siluriformes and therefore the Pvleptin should be the ortholog of teleost lepA. Phylogenetic analysis showed that the teleost group consisted of two clades of teleost lepA and teleost lepB, which was in line with several previous phylogenetic studies Yuan et al., 2016;Wen et al., 2019), suggesting two leptin genes were widely existed in teleosts and this phenomenon might be caused by a specific whole genome duplication (WGD) event in teleosts (Londraville et al., 2017;Xu et al., 2018;Wen et al., 2020d). Moreover, more copies of leptin were identified in common carp (Huising et al., 2006), Jian carp (Tang et al., 2013), goldfish , Atlantic salmon (Angotzi et al., 2013), and rainbow trout (Murashita et al., 2008;Gong et al., 2013a), which may be due to an additional WGD event was occurred in their genomes (Lien et al., 2016;Xu et al., 2019). In addition, Pvleptin was clustered into the clade of teleost lepA and shared a close relationship with yellow catfish and channel catfish leptin (Fig. 4), which further confirmed our assumption that mentioned above.
Tissue distribution pattern of Pvleptin was detected by using real-time quantitative PCR. Results showed that Pvleptin was widely distributed in various tissues with the highest expression level in liver, which was similar to the pattern of leptin in pufferfish (Kurokawa et al., 2005), channel catfish (Kobayashi et al., 2011), yellow catfish (Gong et al., 2013b), and lepA in mandarin fish (He et al., 2013), Ya-fish (Yuan et al., 2014) and snakehead (Wen et al., 2020d), indicating the Pvleptin may play similar roles in these teleosts and it also may be involved in regulating food intake and energy balance. Differently, lepA was observed to be highly expressed in brain of Atlantic salmon (Ronnestad et al., 2010), cerebellum of orange-spotted grouper (Zhang et al., 2013), and ovary of tongue sole (Xu et al., 2018), implying the distribution characteristic of teleost lepA or its ortholog is species-specific, and its roles may be variable among different teleosts. In addition, lepB was usually found to be highly distributed in central tissues, such as in orange-spotted grouper (Zhang et al., 2013), mandarin fish (Yuan et al., 2016) and snakehead fish (Wen et al., 2020d), suggesting the divergent evolution and function between the two paralogs of leptin in teleosts. However, the extract roles of teleost leptin genes are still not well understood and more studies are required to further clarify.
Previous studies have reported that feeding status can affect the mRNA expression level of leptin in various teleosts. In the present study, we observed that two-week fasting reduced the mRNA expression level of Pvleptin while refeeding improved the corresponding expression level in liver, which was consistent with related researches in Ya-fish (Yuan et al., 2014), seabream (Babaei et al., 2017), Indian major carp (Dar et al., 2018), and Yangtze sturgeon (Chen et al., 2020), implying that the Pvleptin is also involved in the regulation of energy balance in the liver, an important energy metabolic center in teleost. As an anorexigenic factor, Leptin has been reported to suppress the appetite in gold fish (De Pedro et al., 2006) and rainbow trout (Murashita et al., 2008). Therefore, the expression level of Pvleptin was decreased after a two-week food deprivation, suggesting this hormone protein may regulate energy homeostasis by reducing the metabolic energy demand among fasting period in darkbarbel catfish.
In summary, we identified a single leptin gene in darkbarbel catfish for the first time. Multiple sequences alignment and 3D structure modeling revealed low identity but similar 3D structure of Leptins in vertebrates. Genetic synteny and phylogenetic analysis suggested that Pvleptin was the ortholog of teleost lepA and the homolog of lepB had lost in siluriformes teleosts. Similar to other catfishes, Pvleptin was highly expressed in liver of darkbarbel catfish. Finally, fasting and refeeding experiments suggested that PvLeptin was also involved in regulation of food intake and energy homeostasis.

Supplementary material
There is supplementary material associated with this article. Access the material online at: https://dx.doi.