Crosstalk between Growth and Osmoregulation of GHRH-SST-GH-IGF Axis in Triploid Rainbow Trout (Oncorhynchus mykiss)

Smolting is an important development stage of salmonid, and an energy trade-off occurs between osmotic regulation and growth during smolting in rainbow trout (Oncorhynchus mykiss). Growth hormone releasing hormone, somatostatin, growth hormone and insulin-like growth factor (GHRH-SST-GH-IGF) axis exhibit pleiotropic effects in regulating growth and osmotic adaptation. Due to salmonid specific genome duplication, increased paralogs are identified in the ghrh-sst-gh-igf axis, however, their physiology in modulating osmoregulation has yet to be investigated. In this study, seven sst genes (sst1a, sst1b, sst2, sst3a, sst3b, sst5, sst6) were identified in trout. We further investigated the ghrh-sst-gh-igf axis of diploid and triploid trout in response to seawater challenge. Kidney sst (sst1b, sst2, sst5) and sstr (sstr1b1, sstr5a, sstr5b) expressions were changed (more than 2-fold increase (except for sstr5a with 1.99-fold increase) or less than 0.5-fold decrease) due to osmoregulation, suggesting a pleiotropic physiology of SSTs in modulating growth and smoltification. Triploid trout showed significantly down-regulated brain sstr1b1 and igfbp2a1 (p < 0.05), while diploid trout showed up-regulated brain igfbp1a1 (~2.61-fold, p = 0.057) and igfbp2a subtypes (~1.38-fold, p < 0.05), suggesting triploid trout exhibited a better acclimation to the seawater environment. The triploid trout showed up-regulated kidney igfbp5a subtypes (~6.62 and 7.25-fold, p = 0.099 and 0.078) and significantly down-regulated igfbp5b2 (~0.37-fold, p < 0.05), showing a conserved physiology of teleost IGFBP5a in regulating osmoregulation. The IGFBP6 subtypes are involved in energy and nutritional regulation. Distinctive igfbp6 subtypes patterns (p < 0.05) potentially indicated trout triggered energy redistribution in brain and kidney during osmoregulatory regulation. In conclusion, we showed that the GHRH-SST-GH-IGF axis exhibited pleiotropic effects in regulating growth and osmoregulatory regulation during trout smolting, which might provide new insights into seawater aquaculture of salmonid species.


Introduction
Rainbow trout (Oncorhynchus mykiss) are an important salmonid fish cultured all over the world. Rainbow trout and steelhead are the same species (Oncorhynchus mykiss) with different lifestyles (National Wildlife Federation, NWF), rainbow trout are landlocked and spend their life mostly (or entirely) in freshwater; while steelhead is an anadromous species, and the juveniles start out in freshwater and then exhibit migration to marine environments [1,2]. Due to anthropogenic activities (with the resulting global climate change), freshwater is becoming increasingly precious, and a Nature paper reported that "80% of the world's population is exposed to high levels of threat to water security" [3,4]. Global warming results in rising temperatures in ocean and freshwater ecosystems [5,6]. Two recent papers published in Nature series (2017 and 2021) demonstrated the global warming threats to global fishery biodiversity, especially the freshwater faunas [7,8]. Therefore, the

Identification of Rainbow Trout sst Genes
Totally, seven sst genes (sst1a, sst1b, sst2, sst3a, sst3b, sst5, sst6) were identified in rainbow trout (Table 1), and these sst genes were localized in three different chromosomes of 9 (sst2, sst5, sst6), 15 (sst1b, sst3a), and 27 (sst1a, sst3b) ( Figure 1A). The amino acids sequences of SST proteins ranged from 98 to 115 amino acids (Table 1). Multiple sequence alignments of trout SST with vertebrate orthologues are shown in Figure S1. The phylogenetic analysis showed trout sst could divide into six subgroups with two duplicates genes of sst1 (sst1a and sst1b) and sst3 (sst3a and sst3b) ( Figure 1B). The sst4 was absent in rainbow trout ( Figure 1B). Exon-intron structure showed most of the sst genes exerted two exons while sst3a exhibited four exons ( Figure 1C). Compared to two SST1 subtypes, a significant alteration of motif structures was observed between SST3a and SST3b, potentially leading to functional diversity ( Figure 1C). Syntenic analysis showed sst1 and sst3 subtypes exerted conserved alignments with neighboring genes between rainbow trout and Atlantic salmon ( Figure 1D). The sst2, sst5, sst6 exhibited synteny homolog in teleosts ( Figure 1E-G).   The amino acid sequences were analyzed via a neighbor-joining method with 1000 bootstrap replications. Rainbow trout SSTs are highlighted by a red pentacle. (C) Gene structure and protein motif of SST proteins of rainbow trout. (I) Exon and intron structures of SST proteins in rainbow trout. Phylogenetic tree was constructed by a neighbor-joining method with 1000 bootstrap replications.
(II) Protein motif analyses of SST proteins in rainbow trout. Eight specific motifs were found in SST proteins of rainbow trout, different colors represent different motifs. (D-G) Conserved syntenic analyses of sst genes: sst1a/b and sst3a/b (1D); sst2 (1E); sst5 (1F); sst6 (1G). sst genes are shown by the same color. The double slash represents the omission of more than two genes.

The SST System
The SST exhibits a conserved function of suppressing growth in both mammals and teleosts [38][39][40][41]. Triploid trout showed significantly down-regulated brain sstr1b1 and kidney sst2, potentially suggesting triploid trout exerts a better growth after seawater acclimation. A recent human clinical case report showed SST and analogs target the GI tract and pancreas, thus regulating water and salt reabsorption [42]. We observed gene expressions of sst (sst1b, sst2, sst5) and sstr (sstr1b1, sstr5a, sstr5b) were changed in kidney rather than liver after seawater challenge. Considering the pleiotropic physiology of GH in regulating growth and smoltification, we proposed that SST might regulate extracellular osmolarity via the GH-IGF axis in kidney [15,43]. Consistently, recent studies showed SST is involved in physiological regulations of GHRs, IGFs, IGFRs, and IGFBPs [44][45][46].

The IGFBPs System
Smolting is a pan-hyperendocrine state in salmonids, and McCormick indicated that the GH-IGF axis is significantly involved in migratory readiness via regulating salinity tolerance, growth, and metabolism. IGF acts as the primary regulator and target of GH functions, and the local physiology of IGF is co-regulated by IGFRs and IGFBPs. Salmonid specific genome duplication results in expanded igfbp paralogs, however, physiology of these novel igfbp paralogs in response to seawater challenge (migratory readiness) has yet to be determined.

1.
The IGFBPs of triploid trout before and after seawater challenge.
In triploid trout, the upregulated brain igfbp1a2 and kidney igfbp1a1 were observed after seawater challenge, suggesting the central and peripheral endocrine response to salinity challenge is regulated by different igfbp1a subtypes. Previous studies showed the upregulated IGFBP1 could suppress the interaction between IGF and receptors, thus saving the energy from basal metabolism for the processes of acclimation to environmental stress [47,48]. Meanwhile, triploid trout kidney showed an increased trend of igfbp5a subtype expression and significantly downregulated igfbp5b2 and igfbp6b2. The IGFBP5 and IGFBP6 subtypes are involved in regulation of ionic and energy (nutrition) homeostasis. For example, igfbp5a is specifically expressed in zebrafish ionocytes (chloride cell) and dysregulation of IGFBP5a results in abnormal ionocyte proliferation [49]. Consistent to previous studies showing igfbp5a and igfbp5b exert distinctive expressions in zebrafish and grass carp, we showed igfbp5(6)a subtypes exhibited conversed transcriptional signatures with igfbp5(6)b, suggesting an evolutionary divergence in physiology [50,51].

2.
Comparison of IGFBPs in diploid and triploid in response to seawater challenge.
It is necessary to evaluate whether triploid trout exhibit equally biology and physiology performances as diploids in response to environment challenges before mass commercial production [52]. The GH-IGF axis is involved in multiple functions associated with commercially relevant traits, including growth, immunity, and smoltification. IGFBP1 could be induced by stress and acts as a negative regulator of teleost growth and development [24]. Diploid showed up-regulated brain igfbp1a1 (p = 0.057) after seawater challenge. Brain plays a key role in stress reactivity and modulates the physiological and behavioral alterations for stress coping and recovery [53]. Meanwhile, significantly increased brain igfbp2a1 and igfbp2a2 expressions were observed in diploid trout. Previous study showed igfbp2 overexpression results in growth and development reduction [54,55]. Our results suggested triploid were less sensitive to seawater challenge when compared to diploids. The igfbp5 exerted subtype-specific transcription patterns in kidney (down-regulated igfbp5b2) and liver (down-regulated igfbp5a2) in triploid when compared to diploid. Previous studies revealed igfbp5 is involved in both muscle growth and ionic homeostasis in teleost [24,27,49,[56][57][58]. The triploid exhibited potential IGFBP5 subtype-regulated crosstalk between growth (or energy production) and osmoregulation in liver and kidney. The igfbp6, a growth inhibitor of teleost growth [24], showed down-regulation in triploid after seawater challenge, suggesting a better energy crosstalk between seawater acclimation and growth in triploid trout.

Ethics Statement
All experimental protocols were performed in conformity to the Guidelines of Animal Research and Ethics Committees of the Ocean University of China (Permit Number: 20141201), and National Institutes of Health Guide for the Care and Use of Laboratory Animals (NIH Publications, No. 8023, revised 1978). Immature trout were used in this study and sex effects were not considered. No endangered or protected animals were used in this study.

Ploidy Identification
The ploidy identification procedure was combined with CyFlow TM Robby6 (Sysmex, Norderstedt, Germany). Briefly, caudal fin (~0.05 g) was clipped and homogenized with 1×PBS and CyStain TM UV Precise P Nuclei Extraction Buffer (Sysmex, Germany). At least 10,000 cells were measured per sample via flow cytometry. The average relative DNA content of six diploid trout were determined as the diploid control.

Animals Acclimation and Salinity Challenge
Trout juveniles (diploid and triploid,~9.7 g,~10.7 cm) were obtained from Linqu Salmon and Trout Aquatic Breeding LLC (Weifang, Shandong, China). The experimental trout were derived from the same full-sibling family batch with same day age and synchronized development. Trout were acclimated for 10 days before salinity challenge. Fish were acclimated and maintained in a recirculating water system at the Experimental Fish Facility in the Key Laboratory of Mariculture, Ocean University of China, with~16 • C of water temperature,~7 mg/L of dissolved oxygen, saturated feeding (5% of body weight), and natural photoperiod of 12:12 (hours) of light and dark. The acclimation protocols were performed in conformity with the Standards of Linxia Salmon and National Trout Elite Breeding and Protection Farm (Linxia, Gansu, China, Approved by the Department of Agriculture, China, 2009).
After acclimation, trout were divided into four groups (diploid freshwater exposure (DF), diploid seawater exposure (DS), triploid freshwater exposure (TF), and triploid seawater exposure (TS)). Each group contained three biological replications and each replication contained 18 individual trout. The DF and TF groups were cultured in cuboid tanks filled with freshwater, while the DS and TS groups were cultured in cuboid tanks filled with seawater at salinity of 15 (15 ppt) for 7 days. Trout were reared in the same recirculating water system with acclimation protocol, with the same water temperature, dissolved oxygen, feeding regime, and photoperiod. The exposure seawater (15 ppt) was prepared by dilution of the seawater (30 ppt) with aerated freshwater, and the salinity for the exposure media was detected by LS10T (Suwei, Guangdong, China). During exposure, 40% of the exposure media was replaced every two days for the following reasons: (1) to maintain the salinity concentrations, (2) to qualify the water quality, (3) to reduce the animal chemicals. No mortality was observed in DS and TS groups after a 24-h salinity exposure, and the survival rates on day 7 in DF, DS, TF, and TS were 100%, 96.3 ± 3.2%, 98.1 ± 3.2%, and 98.1 ± 3.2%, respectively.

Sampling and RNA-seq
Trout were starved 24 h prior to sacrifice and then anesthetized with tricaine (MS-222, Sigma-Aldrich, Shanghai, China). Tissues of brain, liver, and kidney were quickly removed and then stored in tubes at −80 • C in RNase-free conditions. Total RNA was extracted from brain, liver, and kidney via TRIzol reagent (Vazyme Biotech, Nanjing, China) based on the manufacturer's protocol. The quality and quantity of extracted RNA were evaluated with agarose gel electrophoresis and a biophotometer (OSTC, Shanghai, China).
RNA sequences and data analyses were performed with the general protocol and pipeline for medical and fishery studies of the commercial provider (OEbiotech, Shanghai, China). To reduce the individual differences of RNA-Seq, one RNA-Seq library was constructed via pooling equal quantities of RNA from two individuals within the same replicate (tank). The libraries were sequenced with Illumina HiSeq 4000 platform via a commercial provider (Oebiotech, Shanghai, China). The clean reads of sequences were derived after removing raw reads of low quality and/or poly-N were removed. The latest rainbow trout genome (GCA_013265735.3) was used for mapping of the clean reads via histat2 [59]. The DESeq2 R package was used for gene expression analyses [60], and normalized count (via DESeq2 R package) was used in this study. Details of the data qualities and expression metrics are shown in Tables S1 and S2. Sequence reads are available in the Sequence Read Archive Database (PRJNA844477).

Identification of Rainbow Trout sst Repertoire
Rainbow trout genomic data were used to identify sst repertoire. The SST amino acid sequences of zebrafish (Danio rerio) and goldfish (Carassius auratus) were used for sequence comparisons via TBLASTN. Molecular weight (MW), isoelectric points (pI), and chromosome location of SST genes/proteins were predicted/acquired by online resources of ProtParam tool and NCBI.

Quantitative Polymerase Chain Reaction (qPCR)
Results of RNA-Seq were validated via qPCR ( Figure S2). Total RNA was reversetranscribed into cDNA using PrimeScript RT reagent kit (Takara, Shiga, Japan). The 4× diluted cDNA was served as the template for qPCR. The primers were designed via Primer 5 software (Table S3) and β-actin was selected as the reference gene [63]. The qPCR system was a 10 µL reaction, containing 1 µL cDNA, 5 µL SYBR ® FAST qPCR Master Mix (Monad, Wuhan, China), 0.2 µL of each primer, and 3.6 µL RNAase-free water. The reaction program was 95 • C for 30 s, 40 cycles of 95 • C for 10 s, and Tm for 30 s, followed by 72 • C for 30 s. The qPCR was carried out via StepOnePlus™ Real-time PCR system (Applied Biosystems, Foster City, CA, USA). The results were calculated by the 2 −∆∆Ct method [64].

Data Visualization and Statistical Analysis
According to previous studies in medical and fishery studies [65,66], the count of RNA-Seq data were normalized by DESeq2 [67]. The normalized data were analyzed via the websites of MetaboAnalyst and NetworkAnalyst for multivariate analysis of principal components analysis (PCA), partial least squares discriminant analysis, loading plots, and clustering algorithm (heatmap) [68]. Genes were selected via the screening criteria of Log2|Foldchange| ≥ 1 or p value < 0.05 (Log 10 (count+1) tested by Student's t-test). The correlation analysis of gene expressions was analyzed via the Pearson correlation coefficient by MetaboAnalyst and GraphPad Prism 8.0. Data of basal expression of ghrh-gh-sst-igf axis in diploid trout were retrieved from previously published RNA-Seq data (SRA of NCBI: PRJNA865462 for liver; SRA of NCBI: PRJNA753277 for brain and kidney) [69,70]. The gene count data were analyzed by DESeq2 R package and then the average count was normalized by average β-actin ((average of normalized count × 10,000)/(average of β-actin count)) [60]. Expression matrix are shown in Tables S1 and S2.

Conclusions
In this study, we showed that salmonids exhibited increased sst gene duplications due to salmonid specific genome duplication. Similar to GH, the SST and SSTR exhibited pleiotropic effects in modulating diverse physiological processes including growth and smoltification. Compared to diploid, triploid trout exhibited a better seawater acclimation. The igfbp5 and igfbp6 subtypes were involved in energy redistribution between growth and osmoregulatory regulation.

Conflicts of Interest:
The authors declare no conflict of interest.