Background

Many important economic traits in animals, including growth rate, are complex traits controlled by quantitative trait loci (QTL) [1]. QTL analysis is an effective method to identify genomic regions controlling specific traits, which can further be used to find candidate genes or molecular markers linked to the traits for marker-assisted selection [2]. However, in the past, QTL studies were based on a limited number of markers with limited map** resolution, resulting in the detection of QTL regions that typically contained thousands of candidate genes [3]. To narrow down the boundaries of target QTL and precisely locate candidate genes, expensive, laborious, and time-consuming fine map** and positional cloning strategies are often required [4]. The introduction of next-generation sequencing (NGS) technologies has made it easier to develop hundreds of thousands of single nucleotide polymorphism (SNP) markers. The combination of large-scale data generated by NGS and powerful computational tools has resulted in significant technological advancements in QTL map** from low to high resolution [5]. After fine map**, the number of candidate genes within the QTL can be reduced to hundreds [6]. However, identifying the causal genes for the traits remains a major challenge due to the expensive and time-consuming procedures involved in validating the candidate genes [7]. This limitation greatly restricts the study of quantitative traits. Furthermore, this independent analysis alone cannot fully explain the genetic regulatory mechanisms underlying complex traits. Recently, a method for identifying candidate genes, known as expression quantitative trait locus (eQTL) analysis or genetical genomics, has been developed [8]. eQTL analysis applies genetic linkage analysis of whole-genome expression patterns obtained through microarray or NGS technologies to study genetic variation in gene expression. In eQTL analysis, mRNA transcript abundance is treated as a quantitative trait and mapped as eQTL [9]. eQTL analysis can construct gene regulatory networks to further investigate the coordinated functions of genes controlled by common eQTLs, thereby elucidating the genetic basis of complex traits at the gene expression regulation level [10]. eQTL has been applied in studies across multiple species and has been proven to be an effective method for identifying candidate genes [11,12,13].

Growth rate is one of the most important economic traits in aquaculture animals and is closely related to the economic benefits of aquaculture. Although the growth rate trait in shrimp is polygenically controlled quantitative trait, the genetic regulatory mechanism underlying this trait is still largely unknown [16,17]. However, these QTL intervals contain a large number of candidate genes, and the major regulatory genes for the growth rate trait remain unknown [18].

Shrimp are economically important aquaculture animals that provide a significant protein source for humans, while also playing essential ecological roles in marine and freshwater environments. The Pacific white shrimp (L. vannamei) is the most globally important species, accounting for over 50% of total shrimp production [19]. To gain deeper insights into the genetic regulatory mechanisms of growth in L. vannamei and search for candidate genes associated with growth rate trait, we constructed a genetic linkage map of L. vannamei using RNA sequencing. For the first time, we explored candidate genes related to growth rate in L. vannamei through integrated analysis of phenotypic quantitative trait loci (QTL) and eQTL. Candidate genes associated with growth were subsequently identified and validated through gene silencing and shrimp growth experiments. This study provides critical new genomic information for shrimp breeding and investigations of L. vannamei. The method outlined in this study for identifying candidate genes through comprehensive analysis of QTL and eQTL may also provide valuable preferences for other species.

Methods

Preparation of map** family for eQTL analysis

The shrimp used in this experiment were obtained from Guangxi Shrimp Breeding Engineering Technology Research Center (Nanning, Guangxi, China). The map** family was created through artificial insemination. To elaborate, a female shrimp and a male shrimp from a family achieved through eight generations of full-sib matings, originally derived from an SPF strain obtained from the Hawaii Institute of Oceanography) were mated. The resulting offspring comprised the map** family. All of these offspring were cultured together in an indoor pool measuring 12 m×6 m and were fed a standard diet. After five months of cultivation, the family, consisting of two parents and their 268 offspring, was collected for genetic map**. The hepatopancreas tissue was promptly removed from each shrimp and stored in liquid nitrogen for subsequent RNA extraction. The body weights of the shrimp were measured using an electronic scale.

RNA extraction, library construction, and RNA-Seq

Total RNA was extracted from the hepatopancreas by grinding them in liquid nitrogen and using the TRIzol® LS Reagent (Ambion, US) following the manufacturer’s protocols. Genomic DNA was removed using DNase I (Ambion, US). The concentration of RNA in each sample was measured using a NanoDrop ND-2000 Spectrophotometer (Thermo Scientific, USA) and adjusted to equivalent concentrations across samples using ultrapure water. RNA quality was assessed using a Bioanalyser 2100 (Agilent, USA), and samples with an RNA Integrity Number (RIN) value > 7 were used to construct cDNA libraries for sequencing. The cDNA libraries for RNA-seq sequencing were prepared using the NEBNext Ultra RNA Library Prep Kit (Illumina, US) manufacturer’s the manufacturer’s instructions and sequenced on the Illumina HiSeq2500 platform.

SNP identification and linkage map construction

The software STAR [20] was used to align clean reads from each sample to a full-length transcriptome dataset of L. vannamei that we previously constructed [25]. The significance thresholds of LOD scores were calculated based on 10,000 permutations with an experiment-wise significance level of < 0.05 QTLs with a maximum LOD score greater than or equal to the significance threshold were considered statistically significant. Confidence intervals corresponding to a LOD score decrease of one or two on either side of the likelihood peak were calculated for each significant QTL. The proportion of variance explained by a QTL peak (Expl) was calculated using the formula: \(\text{E}\text{x}\text{p}\text{l}=1-{10}^{-2\text{L}\text{O}\text{D}/\text{N}}\) ,Where N is the number of samples [26].

Map** of eQTLs

The expression levels of genes were calculated by aligning them to the L. vannamei genome, and the mapped reads were used to calculate the expression profile for each shrimp sample. The expression of genes was represented by Reads Per Kilobase per Million mapped reads (RPKM) values. Only genes with complete expression (genes with RPKM = 0 were considered missing) of 80% or above were used for eQTL analysis. The eQTL map** was conducted using the method similar to the QTL map** analysis, but with gene expression data as the trait. The map** was also analyzed using the MapQTL 5.0 software [25].

Co-location of QTLs and eQTLs

When an eQTL for a gene was located within 5 cM upstream or downstream of the genetic distance of the gene, it was considered a cis-eQTL [1). The total distance of the map was 6205.75 cM, the average marker interval was 1.12 cM, and the estimated completeness of the markers on the map was 99.89% (Supplementary Material Table S2). To validate the map, a sequence similarity Blastn search was conducted between 4053 map** sequences from the map and our previously published genome of L. vannamei [24]. In total, 3890 hits were returned. According to the Blastn results, 44 linkage groups in the map were aligned to 44 pseudochromosomes of the genome, and the number of linkage groups was consistent with that of the previously reported maps of L. vannamei [29,31], confirming the successful construction of the map.

Fig. 1
figure 1

The high-density linkage map of L. vannamei based on RNA sequencing, illustrating the genetic distances between SNP markers. SNP markers are represented by black lines

Full size image

QTL analysis

To identify QTLs associated with L. vannamei growth rate, the body weights of 268 individual shrimp were used as an indicator of growth rate. The average weight of the shrimp was 22.36 ± 3.72 g (Supplementary Material Table S3). To obtain significant QTLs, strict logarithm of the odds (LOD) thresholds was used (24.0), with a total of 11 QTLs being significantly correlated with weight traits and distributed on nine linkage groups (Fig. 2). Within these QTL intervals, 50 genes were included, containing 75 SNP markers (Supplementary Material Table S4).

Fig. 2
figure 2

QTL analysis for body weight in L. vannamei. The blue and red traces represent the logarithm of odds (LOD) values and the contribution rate of the linkage, respectively. The gray line indicates the LOD threshold (24.0; p = 0.05). The numbers on the x-axis indicate the linkage group identifier

Full size image

eQTL analysis

To perform eQTL map**, each gene was aligned to the L. vannamei genome to quantify expression, comprising a total of 19,641 genes. After removing genes with missing expression values, the remaining 11,503 genes were subjected to eQTL analysis. A total of 117,525 eQTLs were identified, including 12,088 trans-eQTLs and 105,437 cis-eQTLs that were widely distributed across different linkage groups (Fig. 3).

Fig. 3
figure 3

The distribution of eQTLs. The blue trace represents the number of identified eQTLs and the numbers on the x-axis indicate linkage group identifiers

Full size image

Co-localization of QTLs and eQTLs

Genes whose cis-eQTLs overlapped with the 11 weight-related QTL regions were first screened, leading to the identification of 7 such genes (Table 1; Fig. 4a). Correlations between the expression levels of these genes and shrimp weights were then conducted (Supplementary Material Table S5). The expression levels of four genes were significantly and positively correlated with the body weights of 268 shrimp that were used for genetic linkage map construction. In addition, two genes were negatively correlated with body weights, and one did not exhibit a significant correlation (Table 1; Fig. 4b). The four positively correlated genes (2329_0|path2, 5952_686|path0, C10972/f3p0/1885, and 545_0|path20) were consequently considered candidate genes related to the growth rates of shrimp. Annotation by comparison to the NR database indicated that the genes encoded “40S ribosomal protein S5”, “protein SMG7”, “metalloreductase STEAP4-like (Steap4)”, and “nascent polypeptide-associated complex alpha” products.

Fig. 4
figure 4

Co-location and regression analyses for the candidate trait gene c10972/f3p0/1885. (a) The cis-eQTL of c10972/f3p0/1885 gene overlap** with QTL intervals on linkage group 30. The blue trace represents the LOD values of the QTL for the linkage, while the red trace represents the contribution rate of the QTL for the linkage. The gray trace represents the LOD values of eQTLs of the linkage. (b) Regression analysis for the correlation between expression levels of the c10972/f3p0/1885 gene and the body weights of 268 L. vannamei used for genetic linkage map construction

Full size image
Table 1 Genes with cis-eQTLs overlap** with QTL intervals
Full size table

Silencing of candidate genes

To explore whether the aforementioned candidate genes were related to shrimp growth, the corresponding dsRNA for each gene were synthesized. The synthesized dsRNAs were evaluated using agarose gel electrophoresis after dilution, confirming their expected sizes (Fig. 5a). Gene expression was then interrupted by injecting the dsRNAs into shrimp, and samples were collected at 0, 24, 48, and 72 h after injection for real-time PCR analysis of mRNA. The expression of two candidate genes in the experimental group (2329_0|path2, Fig. 5b; and c10972/f3p0/1885, Fig. 5d) was significantly affected, while the expression of the other two genes was not significantly inhibited (Fig. 5c, e). Shrimp in the experimental, dsRNA-egfp, and control groups were raised for 30 days, and their body weights were evaluated accordingly (Supplementary Material Table S6). The average weight gain of shrimp injected with dsRNA-c10972/f3p0/1885 was significantly higher than that of the control groups injected with dsRNA-egfp or 0.9% NaCl (Fig. 5f). However, no significant differences in weight gain were observed among shrimp in groups injected with dsRNA corresponding to the other three candidate genes (Fig. 5f). Thus, shrimp growth was inhibited after silencing the c10972/f3p0/1885 gene (Steap4), suggesting that this gene is closely related to shrimp growth.

To further investigate the correlation between Steap4 gene expression levels and shrimp growth rate, two shrimp strains with different growth rates (Fig. 5g) were selected to evaluate gene expression profiles. Thirty shrimps were randomly collected from each strain, RNAs were extracted, and the expression levels of the Steap4 gene were measured using fluorescent quantitative PCR (Supplementary Material Table S7). Significant differences in the average expression levels of Steap4 gene were observed between the two different L. vannamei strains. The average expression levels of the Steap4 gene in the strain (LV-B) with high growth rates were statistically significantly higher than in the strain (LV-S) with low growth rates (Fig. 5h). Moreover, a significant correlation was observed between individual body weight and the expression level of Steap4 gene in both LV-B and LV-S strains (Pearson’s R = 0.530, p = 0.003, and Pearson’s R = 0.577, p = 0.001, respectively) (Fig. 5i, j). Therefore, the expression of the Steap4 gene is an important factor influencing the growth rate of L. vannamei.

Fig. 5
figure 5

Silencing of candidate trait genes in L. vannamei by RNA interference. (a) Agarose gel electrophoretogram showing in vitro dsRNA synthesis. M: DNA Marker 2000; lane 1: dsRNA-2329_0|path2; lane 2: dsRNA-5952_686|path0; lane 3: dsRNA-C10972/f3p0/1885; lane 4: dsRNA-545_0|path20; lane 5: dsRNA-egfp. mRNA levels of different dsRNAs injected into shrimp, including (b) dsRNA-2329_0|path2, (c) dsRNA-5952_686|path0, (d) dsRNA-C10972/f3p0/1885, and (e) dsRNA-545_0|path20. mRNA levels are shown in comparison to the control group injected with dsRNA-egfp at 0, 24, 48, and 72 h after injection. (f) Comparison of weight gain in L. vannamei 30 days after injection with different dsRNAs. The data were statistically analyzed using paired Student’s t-tests. Asterisks denote statistical significance (*: p < 0.5; **: p < 0.01). (g) Comparison of average body weight between LV-B and LV-S shrimp strains. The data were statistically analyzed using paired Student’s t-tests. Asterisks denote statistical significance (**: p < 0.01). (h) Comparison of the average expression level of Steap4 genes in LV-B and LV-S shrimp strains. The data were statistically analyzed using paired Student’s t-tests. Asterisks denote statistical significance (**: p < 0.01). (i) Correlation between individual body weight and the expression level of Steap4 genes in LV-B strain. (j) Correlation between individual body weight and the expression level of Steap4 genes in LV-S strain

Full size image

Discussion

The development of high-throughput sequencing technologies has led to an increase in the number of genetic linkage maps constructed using genoty**-by-sequencing methods [32]. This method involves sequencing genomic DNA, identifying SNPs, and constructing linkage maps [32]. In this study, we utilized RNA sequencing to construct a SNP linkage map for L. vannamei and simultaneously measure gene expression levels for eQTL analysis. Previous eQTL analyses primarily relied on gene chips [33], which are limited for non-model organisms. Therefore, we used RNA sequencing to measure gene expression levels instead. Transcriptomic sequencing was conducted on 270 individuals of the map** population of L. vannamei to evaluate gene expression levels. This enabled the identification of abundant SNPs, which were used to construct a high-density genetic linkage map of L. vannamei with an average marker distance of 1.12 cM. Our map had a higher marker density compared to previous genetic maps of L. vannamei constructed using the amplified fragment length polymorphism (AFLP) or simple sequence repeats (SSR) markers, where the average marker distance was 7.6–15.1 cM [34, 35]. The density of our map was slightly lower than previous maps of L. vannamei constructed using specific length amplified fragment sequencing (SLAF-seq) method, where the average marker distance was 0.37–0.4 cM [36, 37]. However, a specific advantage of our map is its ability to obtain gene expression levels for eQTL analysis.

Using the constructed genetic linkage map, we identified 11 QTLs associated with L. vannamei growth rates. However, these QTL regions contained hundreds of genes. To further identify genes potentially responsible for growth rates, we conducted genome-wide eQTL analysis. This analysis revealed seven cis-eQTLs of genes that overlapped with the 11 QTLs. Four of these genes showed significant and positive correlations with shrimp body weight, suggesting their potential involvement in growth rates. Subsequent RNA interference experiments confirmed that only the expression of the STEAP4 gene significantly affected growth rates. Previous studies have suggested a relationship between STEAP4 and the growth of the red swamp crayfish (Procambarus clarkii), where STEAP4 expression was higher in larger populations of P. clarkii compared to smaller populations [38]. Additionally, larger P. clarkii populations had higher lipid contents, indicating that STEAP4 promotes growth rates by regulating lipid metabolism [38]. STEAP4 belongs to the six transmembrane epithelial antigen of prostate (STEAP) family and is involved in regulating iron and copper homeostasis within cells [39]. It has also been implicated in the regulation of glucose and lipid metabolism in various tissues, such as adipose tissue, liver, muscle, and pancreatic beta-cells, by facilitating insulin-mediated glucose uptake and storage [40]. Studies in mice and humans have suggested that alteration of STEAP4 expression may contribute to metabolic disorders like obesity, insulin resistance, and type 2 diabetes [40]. In our study, a combination of QTL and eQTL analyses indicated that STEAP4 is related to L. vannamei growth, which was confirmed by gene silencing and shrimp growth experiments. Given the role of STEAP4 in fat metabolism, we propose that it promotes the growth rate of L. vannamei by regulating fat metabolism. Therefore, STEAP4 could be considered a gene associated with shrimp growth rates.

Conclusions

In conclusion, we generated a genetic linkage map of L. vannamei and conducted comprehensive QTL and eQTL analyses across the entire genome. These analyses revealed the presence of a gene closely associated with shrimp growth rates, providing valuable genetic resources for shrimp breeding.