The Pacific white shrimp Litopenaeus vannamei is a major global marine economic species. L. vannamei is a euryhaline species; although the optimal salinity range has not been determined, it is generally believed that a salinity of 15–35 is more conducive to growth and survival(Reference Romano and Zeng1–Reference Decamp, Cody and Conquest4). The shortage of aquaculture areas and the increase in shrimp epidemics and pollution in coastal areas(Reference Saoud, Davis and Rouse5) have led to the rapid development of extensive and intensive farming systems using underground saline waters. Alternative farming conditions bring new challenges to shrimp aquaculture. Various adverse effects of low salinity have been reported, including a lower specific growth rate, feed consumption(Reference Yan, Wang and Cao6), reduced weight gain(Reference Decamp, Cody and Conquest4) and impaired survival(Reference Laramore, Laramore and Scarpa3,Reference Li, Chen and Zeng7) . Low salinity reduces the diffusion of ions to body tissues or the water environment and results in water absorption in the tissues, causing swelling and damage to the cells(Reference Davis, Saoud and McGraw8). In addition, a change in environmental osmotic stress can cause tissue damage due to the production of free radicals (ROS)(Reference Li, Chen and Zeng9). These adverse effects have caused slow growth, low digestion and absorption capacity(Reference Yan, Wang and Cao6), weak immunity(Reference Xu, Li and Liu10) and disruption of gut microbiota homeostasis(Reference Zhang, Sun and Liu11), ultimately increasing the risks of death and disease. Therefore, it is crucial to alleviate the adverse effects of low salinity stress on L. vannamei.
Se is a necessary trace mineral element to support growth, oxidation resistance, absorption and utilisation of Ca, glucose, lipids and vitamins in animals. Se is an important component of the enzyme glutathione peroxidase (GPx)(Reference Rotruck, Pope and Ganther12). GPx catalyses the conversion of hydrogen peroxide and fatty acid hydroperoxides into water and fatty acid alcohol by using reduced glutathione, thereby protecting cell membranes against oxidative damage(Reference Dröge13,Reference Lin and Shiau14) . Therefore, Se is an important nutrient that supports the antioxidant system in scavenging excess reactive oxygen species (ROS). Se deficiency can decrease GPx activity, increase free radical content, reduce mobility and increase the permeability of biofilms(Reference Mansour, Goda and Omar15), but excessive Se is toxic to animals(Reference Hamilton16–Reference Yu, Fu and Huang18). At present, the optimal Se level of many aquatic animals has been reported(Reference Han, Xie and Liu19–Reference Atencio, Moreno and Jos21), and these studies show that adding adequate levels of dietary Se can improve growth performance, stress tolerance and immunity. For L. vannamei, appropriate dietary Se levels (0·45 mg/kg) are beneficial to maintain normal antioxidant capacity(Reference Yu, Fu and Huang18). In addition, a previous study found that organic Se (l-selenomethionine) is a more suitable Se source for L. vannamei (Reference Yu, Xia and Han22). However, little is known about the effects of Se on shrimp under low salinity stress. The optimal dietary Se supplementation level and the mechanism of Se tolerance of shrimp under low salinity stress remain unclear.
Therefore, this study investigated the effects of dietary l-selenomethionine on the growth, antioxidant capacity, hepatopancreas health and stress resistance of L. vannamei. The results explore the regulatory mechanism of Se to alleviate the adverse effects of low salinity stress on L. vannamei and provide a reference for dietary manipulation of inland white shrimp at low salinity.
Materials and methods
Diet preparation
The purified experimental basal diet contained 38·67 % crude protein and 4·68 % total lipid. The ingredients and proximate composition of the basal diet are shown in Table 1. Three different levels of l-selenomethionine (CAS: 3211-76-5, Rhawn) were added to the experimental basal diets of 0·4 (control), 0·8 and 1·2 mg/kg Se. The analytical Se content of feeds was determined by hydride atomic fluorescence spectrometry (GB/T 13883-2008, National Standard of the People’s Republic of China) and consisted of average of the two results. The analytical Se levels in each experimental diet were 0·41, 0·84, and 1·14 mg/kg, respectively. All dry ingredients were ground, weighed, blended, sifted through 60 mesh screens, mixed with oils and deionised water (300 ml/kg diet) and manufactured into pellets (2 mm diameter) using a press pelleting machine. The pellets were air-dried in a cool place until 10 % moisture was reached and then stored in a −20°C refrigerator.
Table 1. The compositions of the experimental basal diets
* Casein: crude protein ≥92 %, total lipid ≤2 %, Hua Ling Dairy Corporation, Lanzhou, China.
† Vitamin premix (per kg of diet): vitamin A: 4800 mg; l-ascorbyl-2-polyphosphate 35 % Active C: 35·71 g; folic acid: 0·18 g; biotin: 0·05 g; riboflavin 3 g; DL Ca-pantothenate L: 5 g; pyrodoxine HCl B6:1 g; vitamin B12:0 002 g; thiamine HCl: 0·5 g; Menadione K3:2 g; DL-α-tocopheryl acetate: 20 mg; inositol: 5 g; nicotinamide: 5 g; vitamin D: 8000 mg.
‡ Mineral premix (per kg of diet): ZnSO4·H2O: 20·585 g; Ca(IO3)2:0 117 g; CuSO4·5H2O: 0·625 g; MnSO4·H2O: 1·225 g; MgSO4·H2O: 39·86 g; CoCl2:0 01 g; FeSO4·H2O: 11·179 g; CaHPO4·2H2O: 166·442 g.
Rearing and experimental conditions
L. vannamei were obtained from a commercial shrimp farm in Wenchang. First, shrimp were acclimated for 14 d in seawater (salinity 31) in experimental cylindrical tanks (500 l). Then, three-fourth of the shrimp were acclimated to salinity 3 by decreasing two units/d. After reaching a salinity of 3 and feeding for 3 d, shrimp with similar body weights (1·20 ± 0·14 g) were randomly divided into four groups with 3 replicates/group and 30 shrimp/tank. Shrimp were reared in two culture conditions (standard and low salinity) and fed diets supplemented with increasing levels of Se (0·41, 0·84 and 1·14 mg/kg Se) for 56 d, resulting in four treatments, namely, 0·41 standard salinity (31; positive control), 0·41 low salinity (3, negative control), 0·84 low salinity and 1·14 low salinity. Shrimp were hand-fed with the experimental diets 4 times a day (08:00, 12:00, 16:00 and 20:00) with a total daily ration of 4 % of the body weight. Half of the seawater in the positive control and treatment groups was replaced daily at 13:00. Water quality was maintained at 28 ± 2°C, pH at 8·0 ± 0·3 and nitrite at 0·08 ± 0·02 mg/l. Feces and undigested feed were discharged along with the circulating seawater.
Sample collection
To avoid postpandrial responses, shrimp were fasted for 24 h before sampling at the end of the 8-week trial(Reference Bu, Wang and Lin23). All shrimp were anesthetised on ice. Body and hepatopancreas weights and body lengths were measured. Five shrimp/tank were preserved at −20°C to determine proximate body composition. Three hepatopancreases were dissected and placed in 4 % paraformaldehyde for 24 h. Other hepatopancreases were rapidly immersed in liquid nitrogen and stored at −80°C for antioxidant enzymes and transcriptome analysis. The animal ethics protocol was approved by the East China Normal University Experimental Animal Ethics Committee (No. F20201001).
The following formulas were used to calculate the growth parameters:
$$\begin{gathered}
{\text{Survival}}\left( \% \right) = \left( {{\text{final }}{\mkern 1mu} {\text{shrimp }}{\mkern 1mu} {\text{number/initial }}{\mkern 1mu} {\text{shrimp }}{\mkern 1mu} {\text{number}}} \right) \\
\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\times 100 \\\end{gathered}$$
$$\begin{gathered}
{\text{Weight }}{\mkern 1mu} {\text{gain}}{\mkern 1mu} \left( \% \right) = \left( {{\text{final }}{\mkern 1mu} {\text{weight - initial }}{\mkern 1mu} {\text{weight}}} \right){\text{/initial }}{\mkern 1mu} {\text{weight}} \\
\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\times 100 \\
\end{gathered}$$
$$\begin{gathered}
{\text{Hepatosomatic }}{\mkern 1mu} {\text{index}}{\mkern 1mu} \left( \% \right) = {\text{hepatopancreas }}{\mkern 1mu} {\text{weight/final }}{\mkern 1mu} {\text{weight}} \\
\times 100 \\
\end{gathered}$$
$$\begin{gathered}
{\text{Condition }}{\mkern 1mu} {\text{factor}}{\mkern 1mu} \left( \% \right) = {\text{final }}{\mkern 1mu} {\text{weight/lengt}}{{\text{h}}^3} \times 100,{\mkern 1mu} \\
{\text{where }}{\mkern 1mu} {\text{weight }}{\mkern 1mu} {\text{is }}{\mkern 1mu} {\text{in }}{\mkern 1mu} {\text{g }}{\mkern 1mu} {\text{and }}{\mkern 1mu} {\text{length }}{\mkern 1mu} {\text{is }}{\mkern 1mu} {\text{in }}{\mkern 1mu} {\text{cm}}. \\\end{gathered}$$
Proximate composition
Analytical methods were used for moisture, crude lipid, crude protein and ash(Reference Horwitz24). Four shrimp in each group and four fully ground and mixed diets were dried at 105°C to a constant weight. Crude protein content was determined by a Dumas nitrogen analyzer (Elementar Rapid N Excess). Crude lipid content was measured by the Soxhlet method (Buchi 36680) after extraction with ether. The ash content was determined by burning the sample in a muffle furnace for 10 h at 550°C.
Enzyme assays and biochemical analysis
Six hepatopancreas samples in each group (two samples per tank) were homogenised in normal saline (0·86 %) and centrifuged at 4°C at 3000 rpm for 10 min. The supernatant concentration of the homogenate was determined by the weight to volume ratio. The activities of superoxide dismutase and catalase, GPx activity, total antioxidant capacity, malondialdehyde content (MDA) and total protein concentration and the contents of lactic acid and glucose in serum were detected using a commercial kit (Nanjing Jiancheng Co., Ltd.). The results were recorded on a microplate reader (Epoch, BioTek). The specific experimental method was described in a previous study(Reference Yu, Fu and Huang18).
Hepatopancreas histology
The hepatopancreases from three shrimp per treatment were dehydrated with different gradients of alcohol (80 %, 90 %, 100 %), cleared with xylene and embedded in paraffin to make solid wax blocks. These solid wax blocks were cut into 5-μm thick continuous section blocks in each section by a rotary microtome (Leica RM2125). The segmented tissues were stained with hematoxylin–eosin and examined under a microscope (Nikon, Eclipse 200).
Total RNA isolation and transcriptome analysis
Total RNA was isolated using TRIzol reagent (Invitrogen) from the hepatopancreases (three mixed samples) of nine shrimp randomly selected from each group. Sequencing was performed according to a previous procedure(Reference Yu, Fu and Huang18). The cDNA library was sequenced with an Illumina HiSeq™ 4000 instrument. The sequencing data were stored in the NCBI SRA. The registration number is SRP310771.
The number of segments per million segments in 1000 bases was used to calculate the expression of single genes to determine the differentially expressed genes (DEG) for the three groups: 0·41 (low salinity) group (shrimp in the normal seawater were used as the salinity control for the shrimp fed the control diet under low salinity), 0·84 group (shrimp fed the control diet under low salinity were used as the control for those fed the diet with 0·84 mg/kg Se) and 1·14 group (shrimp fed the control diet under low salinity were used as the control for those fed the diet with 1·14 mg/kg Se). The R statistical software package edgeR (empirical analysis of digital gene expression in R, http://www.bioconductor.org/packages/2.12/bioc/html/edgeR.html) was used for RSEM analysis (http://deweylab.biostat.wisc.edu/RSEM). An error rate <0·05 was used as the threshold P value to judge the significant difference in gene expression level. Gene set enrichment analysis (www.broadinstitute.org/gsea) was used to identify differential pathways between the control diet in the normal seawater and low salinity seawater, 0·41 and the diet with 0·84 mg/kg Se under low salinity, and 0·41 and 1·14 mg/kg Se under low salinity.
A comprehensive value > 0·2 was obtained, and the top 50 DEG related to metabolism were screened. Cytoscape version 3.7.1(Reference Cline, Smoot and Cerami25) and the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) database(Reference Szklarczyk, Franceschini and Wyder26) were used to construct a protein–protein interaction network to obtain known and predicted protein interactions. The confidence score cut-off applied for interactions was 0·9 (highest confidence). Single nodes, doublets and triplets detached from the main network cluster were removed.
Statistical analysis
All data were calculated as the mean ± standard error (se). Data were checked for normality and homogeneity of variances. Two hypotheses were tested: (1) Student’s t test was used to compare the positive control (standard salinity seawater) group with negative control (low salinity) group and other two low salinity groups (i.e., 0·41 standard salinity v. 0·41 low salinity; 0·41 standard salinity v. 0·84 low salinity; and 0·41 standard salinity v. 1·14 low salinity); and (2) one-way ANOVA was used in shrimp fed diets with 0·41, 0·84 and 1·14 mg/kg Se under low salinity. SPSS statistics 22.0 software (IBM Inc.) was used to conduct statistical analyses. A multiple comparison test (Duncan) was performed to compare the significant differences among the low salinity treatments. P values <0·05 were considered statistically significant.
Results
Growth performance and whole-body proximate composition
Individual comparisons between the positive control and the rest of the dietary treatments demonstrated a reduced survival of shrimp fed the negative control diet (P < 0·05), while no significant differences were found in the 0·84 and 1·14 mg/kg Se groups (Fig. 1(a)). Compared with the seawater group, low salinity stress significantly decreased the weight gain of shrimp fed the 0·41 and 1·14 mg/kg Se diets (P < 0·01), while there was no significant change in the 0·84 mg/kg Se group. Shrimp fed the 0·84 mg/kg Se diet exhibited a significant decreased in hepatosomatic index (P < 0·01), while the 0·41 and 1·14 mg/kg Se groups showed a significant decrease in hepatosomatic index compared with the seawater group (P < 0·05, Fig. 1(b) and (c)). No significant change was found in the condition factor among all treatment groups (P > 0·05; Fig. 1(d)). The ash content of shrimp fed 0·84 mg/kg Se under low salinity stress was significantly lower than that in the salinity control, and the contents of crude protein, total lipids and moisture were not significantly different among all groups (P > 0·05, Table 2).
Fig. 1. Effects of different dietary Se levels of l-selenomethionine on (a) survival; (b) weight gain; (c) hepatosomal index; (d) condition factor of L. vannamei. Different letters (a, b) indicate significant differences (P < 0·05) among the three low salinity groups. Asterisks (*) represent a significant difference of P < 0·05 between the low salinity treatments and the seawater control. Double asterisks (**) represent a significant difference of P < 0·01 between the low salinity treatments and the seawater control.
Table 2. Proximate whole body composition of L. vannamei fed different diets
Three whole shrimp bodies from each of the three tanks were tested. The results represent the means ± SE.
a,bIndicates a significant difference between Se levels (P < 0·05).
Antioxidant capacity and related biochemical components
The results of antioxidant parameters and biochemical components are shown in Fig. 2. The total antioxidant capacity in shrimp fed the negative control diet was numerically lower than that in shrimp fed the positive control diet, but the difference was insignificant. No significant difference was found in the total antioxidant capacity (P > 0·05; Fig. 2(a)). The MDA content in shrimp fed the negative control diet was significantly higher than that in shrimp fed the positive control diet (P < 0·05), and the MDA content gradually decreased with increasing Se supplementation to a similar level (Fig. 2(b)). Low salinity did not affect the superoxide dismutase and catalase activities of shrimp fed the positive control diet, while superoxide dismutase activity was significantly improved, and catalase activity was inhibited by the diet with 1·14 mg/kg Se compared with other groups (P < 0·05; Fig. 2(c) and (d)). Low salinity significantly inhibited GPx activity (P < 0·05), and the increase in Se levels numerically enhanced GPx activity, but the difference was insignificant (Fig. 2(e)). Compared to the positive control, low salinity significantly increased the lactic acid content in the serum (P < 0·05), and the diets with 0·84 or 1·14 mg/kg Se corrected this effect (P < 0·05; Fig. 2(f)). Low salinity also significantly decreased the glucose content in the serum (P < 0·05), and the diet with 0·84 mg/kg Se significantly increased the glucose content under low salinity stress (P < 0·05; Fig. 2(g)).
Fig. 2. Effects of different dietary Se levels of l-selenomethionine on the hepatopancreas antioxidative ability of L. vannamei. (a) T-AOC; (b) SOD activity; (c) CAT activity; (d) GPx activity; (e) MDA content; (f) serum lactic acid content; (G) serum glucose content. The different letters (a, b) indicate significant differences (P < 0·05) among groups. Asterisks (*) represent a significant difference of P < 0·05 between the low salinity treatments and the seawater control. T-AOC, total antioxidant capacity; SOD, superoxide dismutase; CAT, catalase; MDA, malondialdehyde content.
Hepatopancreas histology
The hepatopancreas histology of L. vannamei is shown in Fig. 3. Salinity significantly affected the four kinds of cells in hepatopancreas tubules: embryonalzellen cells (E cells), restzellen cells (R cells), fibrillenzellen cells (F cells) and blasenzellen cells (B cells). The hepatopancreas tubules of shrimp in seawater were arranged neatly, and the B cells were full in shape, evenly distributed and abundant (Fig. 3(a)), while the B-cell radius decreased in shrimp fed the negative control diet under low salinity (Fig. 3(b)). In the shrimp-fed diets with 0·84 and 1·14 mg/kg Se in low salinity, the hepatopancreas tubules were arranged more closely. A larger size of B cells was found in shrimp fed the diet with 0·84 mg/kg Se (Fig. 3(c)), while tighter spaces between the hepatopancreas tubules were observed in shrimp fed 1·14 mg/kg Se (Fig. 3(d)).
Fig. 3. Effects of different dietary Se levels of l-selenomethionine on hepatopancreas histology. (a) Seawater group; (b) diet with 0·41 mg/kg Se under low salinity; (c) diet with 0·84 mg/kg Se under low salinity; (d) diet with 1·14 mg/kg Se diet under low salinity. The illustration at the upper right shows four types of cells visible in a cross-section of the hepatopancreas tubule.
Hepatopancreas transcriptome analysis
The RNA sequence analysis of the hepatopancreas produced 536,394,450 raw reads and 80,995,561,950 raw bases. A total of 528,036,952 clean reads and 77,445,110,324 clean bases were produced. Clean reads were compared to the genome of L. vannamei, and the ratio was 88·29 %–93·60 %. The Venn diagram showed common and unique DEG among the three groups (Fig. 4(a)). In total, 555 differentially expressed genes were identified in the three low-salinity feeding groups (Fig. 4(b)). The negative control contained seventy-one upregulated and ninety-eight downregulated genes (Fig. 4(c)). The diet with 0·84 mg/kg Se group contained 153 upregulated and 162 downregulated genes (Fig. 4(c)). The diet with 1·14 mg/kg Se group contained eighty upregulated and ninety-one downregulated genes (Fig. 4(c)). Gene set enrichment analysis was used to identify genes at the overall level and provide clues for significantly related DEG and biological processes (Table 3). Compared with the positive control, the DEG in the shrimp fed the negative control diet at low salinity were significantly enriched in metabolic pathways, such as glutathione metabolism, fructose and mannose metabolism, retinol metabolism and oxidative phosphorylation. Compared with the shrimp fed 0·84 mg/kg Se at low salinity, the DEG in the negative control diet at low salinity were significantly enriched in the metabolic pathways of retinol metabolism, steroid hormone biosynthesis and multiple carbohydrate metabolism-related pathways (e.g., pentose, glucuronic acid conversion, biosynthesis of other O-glycans and biosynthesis of glutenin-type O-glycans). The differentially expressed genes in shrimp fed 1·14 mg/kg Se at low salinity were significantly enriched in fructose and mannose metabolism, oxidative phosphorylation, glutathione metabolism, steroid hormone biosynthesis and multiple carbohydrate metabolism (e.g., biosynthesis of other O-glycans and biosynthesis of mucin-type O-glycans). The heatmap shows the changes in all DEG in each group and clusters the genes with the same expression pattern into three types (Fig. 4(d)).
Fig. 4. Transcriptome analysis of the effect of different dietary Se levels on the hepatopancreas in L. vannamei. (a) Venn diagram; (b) Volcano diagram; (c) Up- and downregulated differentially expressed genes in each group; (d) Heatmap of all differentially expressed genes in the three groups. Genes were clustered into three expression patterns. GSEA, Gene set enrichment analysis.
Table 3. Metabolism-related gene set enrichment analysis
NES, normalised enrichment score. The enrichment score calculated for each gene subset was standardised according to the size of the gene set to obtain the standardised enrichment score.
The |NES| is the analysis result, shows that the higher the value, the better the reliability.
Protein–protein interaction analysis of DEG
We further analysed the interactions between the DEG. Protein products from fifty metabolism-related DEG produced a total of forty-seven links (Fig. 5). The names of nodes are the KEGG Orthologs names of the genes, and the full descriptions and corresponding KEGG pathways are shown in Table 4. The DEG ndufv1 and ost3, followed by atp6s14, had the highest scores for betweenness centrality, indicating that they play an important connecting role in the corresponding protein clusters. Among them, ndufv1 has the highest degree of connection with other proteins. Its connected protein clusters are associated with oxidative phosphorylation. The ost3 and atp6s14 protein clusters were associated with protein processing pathways in the endoplasmic and mTOR signalling pathways, respectively.
Fig. 5. Three main protein–protein interaction network clusters of differentially expressed genes in three comparisons. Nodes represent genes, and edges represent the interaction between the two proteins that express genes. The more connections there are, the larger the node area. The colour of the most connected gene in each cluster was darkened, indicating that this gene may be the core gene in the cluster. The detailed notes of each node are shown in Table 4.
Table 4. The annotation of DEG with the highest connectivity in the protein interaction network
DEG, differentially expressed gene.
Discussion
Growth performance
Excessive deviation of environmental osmotic pressure from the isotonic point can adversely affect the growth and survival of organisms. L. vannamei has an extraordinary tolerance to low salinity, but the negative effects of low salinity on the growth of L. vannamei should be considered. For example, acute salinity changes can directly lead to an increase in the mortality of L. vannamei (Reference Li, Chen and Zeng7). In the 50 d culture experiments at high, medium and low salinity (32, 17 and 3), the weight gain in L. vannamei at salinity 17 was significantly higher than that at salinity 3, and the lowest survival occurred at salinity 3(Reference Zhang, Zhang and Li27). In this study, the survival, weight gain and hepatosomatic index of shrimp fed different levels of l-selenomethionine under low salinity stress were significantly lower than those in the salinity control, except for the 0·84 mg/kg Se group. With the increase in dietary Se, the survival of shrimp under low salinity stress gradually increased. The weight gain of shrimp fed 0·84 mg/kg Se was equivalent to that of the positive control, indicating that increasing the dietary Se level can promote the growth of shrimp in a low salinity environment. This is the same change as the Se requirement of fish under stress(Reference Rider, Davies and Jha28). Based on weight gain, 0·84 mg/kg Se is the optimal Se requirement for L. vannamei at salinity 3.
Antioxidant capacity
Diets containing 0·84 and 1·14 mg/kg Se displayed strong antioxidant capacity in this study. Se plays a vital role in the antioxidant defense system of animals(Reference Dawood, Basuini and El Yilmaz29). It is the active center of the antioxidant enzyme GPx. Lipid peroxidation caused by ROS produces MDA, which normally acts as a marker of oxidative stress. Generally, salinity stress can cause an increase in ROS, resulting in MDA production. The antioxidant enzyme system developed by animals during evolution can reduce the generation of oxidative stress by scavenging ROS(Reference Rotruck, Pope and Ganther12,Reference Jovanovic, Grubor-Lajsic and Djukic30) . However, there was no significant difference in GPx activity under low salinity, which was speculated to be because 0·41 mg/kg Se in the diet had reached the activity threshold of GPx and could not continue to increase its activity. Similar results were found in rainbow trout(Reference Rider, Davies and Jha28). This result also indicates that GPx activity can be used as a marker of Se when animals are deficient in Se, and when the maximum requirement of Se is exceeded, GPx activity cannot reflect tissue Se levels. The increased ROS can also cause tissue damage. However, the same histological changes can occur in hepatopancreas tubules(Reference Li, Chen and Zeng9). Se has a limited effect on hepatopancreas histology under low salinity stress. The increased numbers of B and R cells may be a passive change in tissue histology to provide more active energy for metabolism in response to low salinity stress. Metabolic changes may be the cause of low weight gain at low salinities.
The Gene set enrichment analysis results showed that “glutathione metabolism” was significantly downregulated in L. vannamei fed the control diet after 8 weeks of low salinity stress, while it was significantly upregulated in shrimp fed the diet with 1·14 mg/kg Se. Glutathione is a functional substance in animals that can maintain normal immune function, antioxidant capacity and detoxification. Previous studies have found that appropriate Se supplementation contributes to the synthesis of glutathione(Reference Yu, Fu and Huang18), which is consistent with the results of this study, suggesting that Se supplementation can contribute to glutathione synthesis. Se supplementation is beneficial to support the antioxidant system. In addition, “retinol metabolism”-related genes in shrimp fed the positive control diet were significantly upregulated, while those fed the diet with 0·84 mg/kg Se were significantly downregulated. Known as vitamin A, retinol is an essential lipid-soluble antioxidant in animals. The upregulation of “retinol metabolism” under salinity stress indicates that retinol is involved in alleviating low salinity stress as a supplement to antioxidant capacity. In addition, previous studies have shown that Se supplementation can improve immune capacity by regulating “retinol metabolism” and “glutathione metabolism” in chickens, which is consistent with these results(Reference Liu, Huang and Nie31). Therefore, Se can reduce oxidative stress by regulating antioxidant-related pathways.
Energy regulation
Under chronic low salinity stress (salinity 3), L. vannamei juveniles require additional energy to resist osmotic pressure from nutrient metabolism(Reference Chen, Li and Li32). After low salinity stress for 8 weeks, fructose and mannose metabolism and oxidative phosphorylation-related genes were significantly upregulated in shrimp fed the negative control diet compared with the positive control. However, this effect was corrected by higher Se supplementation with shrimp-fed diets with 0·84 or 1·14 mg/kg Se displaying significantly downregulated fructose and mannose metabolism and oxidative phosphorylation-related genes. Fructose and mannose metabolism is one of the crucial processes of carbohydrate metabolism in animals and can provide energy for the body. Therefore, carbohydrate supplementation can directly meet the higher energy requirements of aquatic animals under stress conditions, especially under low salinity(Reference Chen, Li and Li32). On the other hand, Se deficiency can disrupt glucose homeostasis in the body, resulting in a disruption of energy metabolism(Reference Labunskyy, Lee and Handy33,Reference Seale, Hashimoto and Kurokawa34) . ‘Oxidative phosphorylation’ is the coupling reaction of ADP and inorganic phosphoric acid to synthesise ATP through the energy released during oxidation. Oxidative phosphorylation activity can be regulated by tissue metabolism to maintain energy metabolic homeostasis. Changes in the concentration of oxidative phosphorylation maximally enhance energy metabolism(Reference Phillips, Covian and Aponte35). Additionally, as a crustacean, the oxidative phosphorylation pathway-related DEG of the Chinese mitten crab Eriocheir sinensis were significantly enriched, which is the same as this study’s result(Reference Li, Wang and Li36).
In the protein–protein interaction analysis, ndufv1 in all DEG related to oxidative phosphorylation played an important connecting role. As a product of the ndufv1 gene, the NDUFV1 protein is the core subunit of the mitochondrial membrane respiratory chain NADH dehydrogenase. The main function of NADH dehydrogenase is to generate ROS. As a product of the ndufv1 gene, the protein NDUFV1 is the core subunit of NADH dehydrogenase in the mitochondrial membrane respiratory chain. Downregulated expression of ndufv1 may lead to loss of energy supply or oxidative stress(Reference Jacquemin, Margiotta and Kasahara37,Reference Wang, Wang and Lin38) . Cytochrome c oxidase submit 4 (cox4) was also found in DEG. Cytochrome c is an essential component of the electron transport chain (ETC) in mitochondria and is involved in energy metabolism(Reference Brown39). When zebrafish were exposed to microcystin-LR, the expression of ndufv1 and cox2 was significantly inhibited, the body changed to anaerobic respiration, and mitochondria were damaged(Reference Chen, Hu and He40), while Se-methylselenocysteine containing Se can interfere with the protein expression related to energy metabolism disorder, amino acid metabolism abnormality, synaptic dysfunction and oxidative stress caused by disease in mice(Reference Du, Shi and Zhao41). These results indicated that Se could alleviate energy metabolism disorders under stress.
Overall, the above results combined with body weight gain, glucose content, lactate content in serum and downregulated oxidative phosphorylation-related genes suggest that Se can alleviate low salinity stress by regulating energy metabolism.
Hormone regulation
Se supplementation also downregulated the ‘steroid hormone biosynthesis’ pathway in L. vannamei under low salinity stress. Steroid hormones play an essential role in regulating normal metabolism, immune function and osmotic pressure(Reference Frye42–Reference Gupta and Lalchhandama44). Low salinity was found in a previous study to cause changes in steroid metabolism(Reference Chen, Li and Li32). The metabolic mechanism of tea Se polysaccharide in rats was closely related to the metabolic mechanism regulating steroid biosynthesis(Reference Shao, Song and Zhao45), and steroid hormones were involved in regulating osmotic pressure in whales(Reference Birukawa, Ando and Goto46). This suggests that Se may alleviate low salinity stress by regulating the metabolism of steroid hormones. However, the mechanism by which Se participates in hormone regulation in L. vannamei needs further investigation.
Conclusion
In conclusion, Se could effectively alleviate the physiological effects of low salinity stress on L. vannamei by regulating antioxidant capacity (e.g., retinol metabolism, glutathione metabolism and antioxidant enzyme activity), hormone regulation via the control of steroid hormone synthesis and energy metabolism (e.g., carbohydrate metabolism and oxidative phosphorylation). Furthermore, our findings suggest an adequate supplementation of Se at 0·84 mg/kg (l-selenomethionine) to maintain weight gain and enhance stress resistance under low salinity stress.
Acknowledgements
This study was supported by grants from the Research and Development Program Projects in Key Areas of Guangdong Province [grant number 2020B0202010001], the National Key R & D Program of China [grant number 2018YFD0900400], the National Natural Foundation of China [grant number 32060832] and the initial fund from the Hainan University for R & D [grant number KYQD(ZR)21091, KYQD(ZR)1736].
Q. Y., F. H. and E. L. conceived the experiment; Q. Y. formulated the experimental diets; Q. Y., F. H. and E. L. conducted the experimental trials and laboratory analysis; F. H., A. R., J. Q., L. C. and E. L. conducted polish work of this manuscript. All authors contributed to and approved the manuscript.
There are no conflicts of interest.
