The redclaw crayfish used in this experiment were taken from the Balilidian breeding base of the Zhejiang Institute of Freshwater Fisheries (Huzhou, Zhejiang Province). The average body weight and body length were 60 ± 0.5 g and 14.5 ± 0.2 cm, respectively. Tissues, including the heart (Hea), hepatopancreas (Hep), ovary (Ov), testis (Te), muscle (Mu), intestine (In), and gill (Gi), were dissected from three males and three females. Embryos at different developmental stages, including fertilized eggs (I), cleavage and blastula (II), gastrula (III), nauplius (IV), eye pigments forming (V), and prehatching (VI), were collected from gravid crayfish. Description of the embryonic development of C. quadricarinatus was based on a previously published method (Meng et al., 2001). Juveniles that cultured in the pool of a greenhouse were collected at different body length stages. All the samples were immediately frozen in liquid nitrogen and transferred to −80°C for nucleic acid extraction.

Total RNA were extracted from various tissues and embryonic samples using Trizol Reagent (KangWei, Beijing) according to the manufacturer’s protocol. The RNA quality was detected by electrophoresis on 1% agarose gel, and the concentration was measured by NanoDrop 2000c (Thermo Scientific, United States). The isolated RNA was treated with RNase-free DNase I (Promega, United States). Approximately 1 μg RNA was used to synthesize cDNA using the HiFiScript cDNA First-strand Synthesis Kit (KangWei, Beijing).

Genes encoding putative Tra2 proteins were screened from the transcriptomes of C. quadricarinatus gonadal tissues that were constructed in our lab (not published) by using F. chinensis Tra2 sequences (accessions: JQ239126, JQ239127, and JQ239128) as a query. Three sequences were found to be similar to members of the tra2 homolog in F. chinensis. Additionally, three pairs of gene-specific primers were designed to obtain the full-length transcripts of three Cqtra2 homologs. PCRs using the cDNA template from the ovary were performed under the following conditions: one cycle of 95°C for 5 min; 32 cycles including denaturation at 95°C for 30 s, annealing at 56°C for 30 s, and extension at 72°C for 90 s; one cycle of 72°C for 7 min. The PCR products were separated using 1.0% agarose gels, ligated to the pMD18-T vector (TAKARA, Japan), and transformed into Escherichia coli competent cells (KangWei, Beijing) for sequencing.

Furthermore, the partial genomic sequence of CqTra2 was cloned using a DNA template that was extracted from ovarian tissue via the genome walking method according to the protocol of the GenomeWalker Universal Kit (Clontech, Japan). Specific reverse primers (SRP1 and SRP2) were designed from previously obtained cDNA sequences, and the method for obtaining, purifying, and sequencing PCR products was as previously described. All primers used in this section are shown in Table 1.

The open reading frame and amino acid sequences of CqTra-2 were predicted using online software¹. The secondary structure was analyzed by GOR4², and multiple alignments among the different Tra-2 proteins were carried out by BLAST³. The phylogenetic tree was constructed using the Neighbor-Joining method and MEGA 5.0 software. The amino acid sequences of Tra-2 from other species (Table 2) were obtained from the NCBI website.

qRT-PCR assays were employed to measure gene expression levels using cDNA that was prepared from different tissues, embryos, and individuals with different body lengths. The specific primers were designed for three mature tra-2 isoforms (Table 1). qRT-PCR was performed on a LightCycler^®480 system (Roche, Switzerland) in a 10-μL reaction mix containing 1 μL of cDNA template, 0.5 μL of each primer (10 μM), 3 μL of water, and 5 μL of SYBR Premix Ex Taq^TM (TaKaRa, China). 18S-rRNA was used as the reference gene, and each test was performed in triplicate. The melting curves were analyzed after amplification to identify the specific product in all PCRs. The threshold cycle (Ct) values via the 2^–ΔΔCt method were calculated using the qRT software provided for the LightCycler^® 480 System (Tan et al., 2017). A histogram for fold comparison of different samples was generated by inputting the 2^–ΔΔCt values of different samples into the GraphPad Prism 5 program software (Roche, Switzerland). Statistical analyses were performed by SPSS 19.0, and significant differences were determined using a Pearson test.

Fragments of CqTra-2 (350 bp) and EGFP (359 bp) containing T7 promoter were synthesized by the GenScript company and subcloned into the vector pUC57. The recombinant plasmids were digested at HindIII or EcoRI, and the purified product was used as the template for dsRNA synthesis with a MEGA script RNAi kit (Thermo Fisher Scientific, United States). For in vivo gene knockdown, 5 μg/g dsRNA was injected by microsyringe (0–50 μL, Ningbo Zhenhai Sanai instrument factory) into a single undifferentiated crayfish (juvenile crayfish in which the secondary sex characteristics could not be distinguished) (n = 10) with an average weight of 0.2 g and average body length of 2 cm. After injection, all crayfish were returned to the culture tanks for 2 weeks until the cephalothoraxes were dissected. Total RNA was extracted, and the first-strand cDNA synthesis of each sample was carried out as described above. To determine the RNAi effect, the expression level of Cqdsx (Accession No. MK342618) was investigated by qRT-PCR quantification, using the qPCR method described above. Three replicates were used for analysis. The results were expressed as the mean ± SEM, and a Student’s t-test was used to analyze the difference between groups.

Statistical analyses were performed using SPSS software version 13.0. Data were expressed as mean ± SD (n = 3), and statistical significance was determined by one-way ANOVA. Significance was set at P < 0.05.

By searching homologous sequences submitted to the NCBI GenBank, the full-length cDNA sequence of Tra-2 was obtained from the transcriptome database constructed in our lab and validated by Sanger sequencing. As a result, three different Tra-2 isoforms were identified from a cDNA library of gonad tissues, including CqTra-2A, CqTra-2B, and CqTra-2C. The full-length cDNA of CqTra-2A was 1207 bp, containing an 822 bp open reading frame (ORF) that encoded 273 amino acids. CqTra-2B was 1225 bp in length, with an 804 bp ORF encoding 267 amino acids. The full-length CqTra-2C cDNA sequence consisted of 880 bp nucleotides with 615 ORF and 204 amino acids (Figure 1A). We have submitted the full-length cDNA sequences of CqTra-2 to GenBank (Accession no. MN393520, MN393521, and MN393522).

A large number of repeat sequences existed in the intron region during genomic DNA amplification using Genome Walking in C. quadricarinatus, presenting a great obstacle to obtaining the full-length Tra2 sequence. Sequence alignment analysis revealed that three Cqtra2 homologs had a consistent 5′-UTR, and their difference chiefly appeared in the terminal region of the cDNA sequence. Therefore, the gene structures with a partial exon-intron organization pattern were analyzed from the different sites to the gene terminal were analyzed (data from DNA sequence not shown). Sequence analysis revealed that the three CqTra2 cDNA isoforms were in the same genomic locus, with differences in the splice pattern of the last three exons. As shown in Figure 1B, CqTra-2A contained exon A and exon C, and CqTra-2B contained exon A and exon B, while the three exons were absent from CqTra-2C.

The multiple alignment of amino acid sequences by BLAST showed that the identity of all homologs of CqTra2 was very high, and domain prediction using the online software GOR4 revealed that they all possessed a highly conserved RNA recognition motif (RRM) that was shared with other Tra2 proteins (Figure 2A). Based on the clustalW algorithm alignment of 21 Tra-2 members (Table 2), the phylogenetic tree was constructed by the neighbor-joining method with 1000 bootstrap replicates using MEGA 5 software. The phylogenetic tree showed that the Tra2 protein was highly conserved among shrimp and crab species, including C. quadricarinatus, M. rosenbergii, F. chinensis, E. sinensis, and M. nipponense. Curiously, D. pulex, part of the scientific class Crustacea, was classified into a single clade with Tra-2 sequences from Aedes aegypti that belonged to insects (Figure 2B).

A quantitative real-time PCR assay was used to detect the expression level of three different splice variants of CqTra-2 in adult tissues. Tissue distribution analysis revealed that the three splice isoforms were all highly expressed in the ovary and weakly expressed in the testis, while low expression was observed in other tissues, including the hepatopancreas, muscle, gill, intestine, and heart (Figure 3). In the gonads, expression profiling of CqTra-2B and CqTra-2C showed sex differences with a higher expression level in the ovary than in the testis (P < 0.05), but CqTra-2A mRNA abundance was not significantly different between female and male gonads (P > 0.05). Additionally, CqTra-2 expression did not significantly differ between adult male and female tissues of C. quadricarinatus.

To investigate the temporal expression pattern of the CqTra2 gene during embryogenesis and juvenile development, we measured the relative expression level at various stages using RT-PCR. All CqTra2 transcripts were detected at a low level in fertilized eggs, at a high level in the cleavage and blastula stage, and at a peak in the prehatching stage. The mRNA expression level of CqTra-2B was obviously higher than CqTra-2A and CqTra-2C at each stage of embryonic development (Figure 4). Determining the initial stages of sex differentiation is crucial to study sexual regulation mechanisms. The expression levels of CqTra-2 at different developmental stages of juveniles were analyzed by qRT-PCR. The results showed that the CqTra-2 expression level exhibited a peak when the body length of juveniles reached 3 cm (Figure 5). Interestingly, CqTra-2B/C mRNA abundances were significantly different between female and male individuals, suggesting that this period may be the key point for the sex differentiation of C. quadricarinatus.

Considering that CqTra-2 genes exhibited a dimorphic expression pattern in gonad tissues, we utilized RNAi to examine the function of CqTra-2 in female sex differentiation. DNA fragments of CqTra2 and EGFP containing the T7 promoter were directly synthesized and cloned into the vector for dsRNA transcription in vitro. An RNAi assay was performed through injection of CqTra2- and EGFP-dsRNA into the cephalothoraxes of the juveniles. The results showed that the transcription levels of CqTra2 with the specific primers CqTra2-RT-F and CqTra2-RT-R by qRT-PCR was decreased by approximately 85%, indicating that dsRNA-mediated gene silencing was successful. In crustaceans, doublesex and its homologous genes were important regulators of sexual differentiation. Here, we also observed a significant decrease in Cqdsx after knockdown of CqTra-2 (Figure 6). Thus, it was determined that CqTra2 might be the upstream regulator of Cqdsx.