Introduction

The oriental river prawn, Macrobrachium nipponense (Crustacea; Decapoda; Palaemonidae), is widely distributed in China and other Asian countries1,2,3. This species is commercially important with an annual aquaculture production of ~ 205,010 tons in 20164. Similar to other Macrobrachium species, male prawns grow faster and reach a larger size at harvest time2. Thus, male prawns are preferred in the M. nipponense aquaculture industry. In addition, the rapid development of testes in the reproductive season is another issue restricting the sustainable development of M. nipponense. Previous studies revealed that the testes of M. nipponense reach sexual maturity within 40 days after hatching5. Thus, inbreeding will occur between the newborn prawns. Inbreeding will lead to a decrease in resistance to adversity in their offspring, small scale of market prawn, and a degradation in germplasm resources. Therefore, it is important to fully understand the mechanism of male sexual differentiation and development with the aim of producing all-male progeny on a commercial scale. This can be accomplished by regulating the process of testes development in M. nipponense.

The androgenic gland is a unique tissue in crustacean species. It has been shown to play an essential role in male sexual differentiation and development in crustacean species. Many studies have reported that the androgenic gland and its secreted hormones promote the driving of male sexual differentiation, the establishment of male sexual characteristics and the development of the testes in crustacean species6,7. For example, the ablation of the androgenic gland from male M. rosenbergii resulted in the sex reversal; a phenomenon that has been termed “neo-female”6,7. The insulin-like androgenic gland hormone (IAG) is an important hormone secreted by the androgenic gland. IAG was shown to promote male sexual differentiation and development in many crustacean species8,9,10. It is the most important male sex-related gene acknowledged in many crustacean species. Knockdown of IAG expression by RNAi in male M. rosenbergii can also result in sex reversal11. IAG, specially expressed in the androgenic gland of M. nipponense and other crustaceans, was also implicated in the maintenance of male secondary sex characteristics, spermatogenesis, reproductive strategies and the regulation of growth8,9,10,12,13,14,15. Recent studies on the androgenic gland have become more prominent in the literature. A series of transcriptomes from the androgenic gland have been constructed in M. nipponense16,17,18, and a series of important genes from the androgenic gland have been implicated as having an essential role in male sexual development19,20,21,22. In addition, the histological observations during different post-larval developmental stages indicated that the development of the androgenic gland has regulatory roles in the development of testes5.

The eyestalk of crustacean species has many neurosecretory hormones. The X-organ–SG complex (XO–SG) was identified as a principal neuroendocrine gland located in the eyestalk23. It stores and releases the crustacean hyperglycemic hormone (CHH) superfamily of neurohormones, including CHH, ion transport peptides (ITP), gonad-inhibiting hormone (GIH), molt inhibiting hormone (MIH) and mandibular organ-inhibiting hormone (MOIH), playing essential roles in reproduction24,25,26, molting27,28,29, metabolism of glucose30,31 and other function32,33,34. Knockdown of GIH by RNAi promoted ovarian development in M. nipponense35. Knockdown of MIH by RNAi promoted molting in M. nipponense36. CHH has been shown to promote testes development in M. nipponense37.

In this study, the vital metabolic pathways and genes involved in male sexual differentiation and development in M. nipponense were selected by performing long-read and next generation transcriptome profiling analysis of the androgenic gland after the ablation of the single-side and double-side eyestalk. The functions of Hydroxysteroid dehydrogenase like 1 (HSDL1), which was predicted to be involved in the mechanism of male sexual development in this study and in a previous study38, were further analyzed in depth by using qPCR analysis and RNAi. This study provides valuable evidence of male sexual differentiation and development in M. nipponense, as well as in other crustacean species.

Results

The expression analysis of Mn-IAG after eyestalk ablation

The mRNA expression of Mn-IAG was measured in three groups: control group (CG), single-side ablation of eyestalk (SS), and double-side ablation of eyestalk (DS) (Fig. 1). The data showed that the mRNA expression of Mn-IAG increased with the time of eyestalk ablation in the SS group and the DS group. The mRNA expression of Mn-IAG was ~ 5-folder higher at day 4 and day 7 than for day 1 in both the SS and DS groups (P < 0.01). However, Mn-IAG expression was only slightly higher at day 7 than day 4 in both the SS and DS groups, showing no significant difference (P > 0.05). Mn-IAG expression in the DS group was ~ 2-folder higher than that in the SS group on the same day and showed significant difference (P < 0.05).

Figure 1
figure 1

Measurement of the expression of Mn-IAG after the ablation of eyestalk. The amount of Mn-IAG mRNA was normalized to the EIF (eukaryotic translation initiation factor 5A) transcript level. Data are shown as mean ± SD (standard deviation) of tissues from three separate individuals. Capital letters indicated expression difference between different days in the same group. * (P < 0.05) and ** (P < 0.01) indicates significant expression difference between different groups at the sample day.

Long-read transcriptome

A total of 22.83 GBs of clean data were generated in the long-read transcriptome. A total of 160,496 high-quality transcripts were obtained with a mean length of 2230 bp. Finally, 49,840 non-redundant transcripts were identified in the long-read transcriptome. These unigenes were annotated, based on the M. nipponense genome, and a total of 43,115 unigenes matched known sequences in the M. nipponense genome. All of the non-redundant transcripts were then compared with the Nr database (non-redundant protein database) and nucleotide sequences in NCBI in order to identify their putative functions (Table S1). A total of 37,355 (74.94%) unigenes were annotated in the Nr database. The other unannotated transcripts represent novel genes whose functions need further investigation. According to the sequence alignment analysis, approximately 4083 (10.95%) unigenes showed the highest similarities with Zootermophsis nevadensis in the Nr database, followed by Daphnia pulex (3,166, 8.49%), M. nipponense (1969, 5.28%) and Stegodyphus mimosarun (1403, 3.76%). The unigenes were further matched with the known genomes of crustacean species, including Litopenaeus vannamei, Eriocheir sinensis, Portunus trituberculatus, and Drosophila melanogaster. A total of 3067, 3263, 3198 and 167 unigenes matched the known proteins with the genome of L. vannamei, E. sinensis, P. trituberculatus, and drosophila fly, respectively. Generally speaking, the unigenes of M. nipponense transcriptome showed the highest sequence identities with that of E. sinensis.

Gene Ontology (GO) and Cluster of Orthologous Groups (COG)analysis aimed to provide a structured vocabulary to describe gene products. A total of 19,673 (39.76%) unigenes were assigned to the GO database comprised of 52 functional groups (Fig. 2). The number of unigenes in each functional group ranged from 1 to 10,057. A total of 13,395 (27.07%) unigenes were highly matched with known proteins in the COG database that were classified into 25 functional groups (Fig. 3). The number of unigenes in each functional group ranged from 1 to 6793. Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis aimed to reveal the regulatory relationship between unigenes in the long-read transcriptome (www.kegg.jp/kegg/kegg1.html). A total of 18,618 (36.72%) unigenes were highly matched known genes in the KEGG database, mapped onto 264 metabolic pathways.

Figure 2
figure 2

Gene ontology classification of non-redundant transcripts. Three categories of GO analysis include biological process (19 functional groups), cellular component (16 functional groups), and molecular function (17 functional groups). The left y-axis and right y-axis indicate the percentage and the number of a specific category of genes existed in the main category, respectively.

Figure 3
figure 3

Cluster of orthologous groups (COG) classification of putative proteins.

Identification of differentially expressed genes

Differentially expressed genes (DEGs) were identified, using the criterion of > 2.0 as up-regulatory genes and < 0.5 as down-regulatory genes, and with a P value < 0.05. A total of 1319 DEGs were identified between CG and SS, including 713 up-regulated genes and 606 down-regulated genes. A total of 2092 DEGs were identified between SS and DS, including 1036 up-regulated genes and 1056 down-regulated genes. A total of 4351 DEGs were found between CG and DS, including 2163 up-regulatory genes and 2188 down-regulatory genes. KEGG analysis revealed that Cell cycle, Cellular Senescence, Oxidative Phosphorylation, Glycolysis/Gluconeogenesis and Steroid Hormone Biosynthesis were the main enriched metabolic pathways in all of these three comparisons.

A total of 15 DEGs were selected from these enriched metabolic pathways, which are listed in Table 1. These genes were differentially expressed in at least two of the three comparisons. Cyclin B3, MAD2A, Polo-like kinase 1, Cyclin A, cyclin-dependent kinase 2 (Cdk2) and Cyclin B were found in the metabolic pathways of Cell cycle and Cellular senescence, which were differentially expressed in all three comparisons. Succinate dehydrogenase complex iron sulfur subunit B Gene (SDHB), Cytochrome c oxidase assembly protein COX11 and Cytochrome c oxidase subunit 7A1 were selected from the metabolic pathways of Oxidative Phosphorylation. Acetyl-coenzyme A synthetase 2-like, Fructose-bisphosphate aldolase and Alcohol dehydrogenase class-3 were differentially expressed in the metabolic pathways of Glycolysis/Gluconeogenesis. Estrogen Sulfotransferase, 3 beta-hydroxysteroid dehydrogenase and HSDL1 were identified from the metabolic pathways of Steroid Hormone Biosynthesis.

Table 1 Identification of important DEGs from transcriptome profiling analysis.

qPCR verification

qPCR analysis was used to verify the expressions of important DEGs in the androgenic gland from the CG, SS, and DS prawns. We selected 10 out of 15 DEGs to verify the accuracy of RNA-seq. The qPCR analysis showed the same expression pattern as the RNA-seq (Fig. 4). Six DEGs from the metabolic pathways of Cell Cycle and Cellular Senescence showed the lowest expressions in CG prawns, and highest expressions in DS prawns, including Cyclin B3 (Fig. 4A), MAD2A (Fig. 4B), Polo-like kinase 1 (Fig. 4C), Cyclin A (Fig. 4D), Cdc2 kinase (Fig. 4E) and Cyclin B (Fig. 4F). The mRNA expression of these DEGs in DS prawns showed significant difference with CG prawns and SS prawns (P < 0.01). The mRNA expression of estrogen sulfotransferase (Fig. 4G) and alcohol dehydrogenase class-3 (ADC-3) (Fig. 4H) showed no significant difference between CG prawns and SS prawns (P > 0.05), whereas a significant difference was observed in DS prawns (P < 0.05). The mRNA expression of HSDL1 (Fig. 4I) and SDHB (Fig. 4J) between SS and DS prawns showed no significant difference (P > 0.05), whereas a significant difference was found in CG prawns (P < 0.05).

Figure 4
figure 4

Verification of the expression of 10 differentially expressed genes (DEGs) between the androgenic gland of CG, SS and DS by qPCR. The amounts of DEGs expression were normalized to the EIF transcript level. Data are shown as mean ± SD (standard deviation) of tissues in three separate individuals. Capital letter indicates expression (P < 0.05).

Expression analysis of Mn-HSDL1

Previous studies have shown that Mn-HSDL1 mRNA showed the highest expression level in the hepatopancreas, followed by the testis, which showed significant difference with other tested tissues (P < 0.05)38. The mRNA expression of Mn-HSDL1 in different prawn developmental stages was measured by using qPCR (Fig. 5). Mn-HSDL1 expression in the larval developmental stages was generally higher than in post-larval-developmental. The highest expression level was observed in larval day 5 (L5), while it showed no significant difference with other tested stages (P > 0.05). During the post-larval developmental stage, the lowest expression level was observed in post-larval day 5 (PL5), and then gradually increased. The highest expression level was observed in PL25♂, which was 3.72 and 1.94-folder higher than PL5 and PL25♀, respectively.

Figure 5
figure 5

Expression characterization of Mn-HSDL1 in different developmental stages. The amount of Mn-HSDL1 mRNA was normalized to the EIF transcript level. Data are shown as mean ± SD (standard deviation) of tissues from three separate individuals. Capital letters indicate expression difference between different samples (P < 0.05).

RNAi analysis of Mn-HSDL1

RNAi was used to analyze the functions of Mn-HSDL1 on male sexual development in M. nipponense. qPCR analysis revealed that the expression of Mn-HSDL1 remained stable in the control group after the injection of Mn-HSDL1 dsRNA (P > 0.05). However, the expression of Mn-HSDL1 significantly decreased by 96% and 90% at day 7 and 14, respectively, following the injection of Mn-HSDL1 dsRNA as compared with the control group (Fig. 6A).

Figure 6
figure 6

Measurement of the expression levels of Mn-HSDL1 and Mn-IAG at different days after Mn-HSDL1 dsRNA injection. EIF transcript level was used to normalize the amount of Mn-HSDL1 and Mn-IAG mRNA expression. Three separate individuals were performed for each tissue, and data are shown as mean ± SD (standard deviation). Capital letters and lowercase indicated expression difference between different days after GFP injection in control group and Mn-HSDL1 dsRNA injection in RNAi group, respectively. * (P < 0.05) and ** (P < 0.01) indicates significant expression difference between the RNAi group and control group at the sample day. (A) Measurement of the expression levels of Mn-HSDL1 at different days after Mn-HSDL1 dsRNA injection. (B) Measurement of the expression levels of Mn-IAG at different days after Mn-HSDL1 dsRNA injection.

The expression of Mn-IAG was also measured in a cDNA template of androgenic gland from the same prawns (Fig. 6B). According to the qPCR analysis, the expression of Mn-IAG at day 1 in the control group was slightly higher than on day 7 or day 14, when it generally remained stable (P > 0.05). In the RNAi group, the expression of Mn-IAG was significantly decreased at day 7 and day 14 after the injection of Mn-HSDL1 dsRNA. Specifically, the expression decreased by ~ 61% and 54% at day 7 and 14, respectively, compared with the control group (P < 0.05).

Histological observations of testes after RNAi

According to histological observations, sperm was the dominant cell type in the testes from the control group, and only a limited number of spermatogonia and spermatocytes were observed (Fig. 7A). The percentages of sperm in Day 1, 7 and14 of control group were 67.90%, 63.64% and 61.24%, respectively (Fig. 7B). In the RNAi group, the number of sperm gradually deceased with the time of Mn-HSDL1 dsRNA treatment. Sperm were rarely found at day 14 after Mn-HSDL1 dsRNA treatment. The percentages of sperm decreased from 57.69% at Day 1 to 1.27% at Day 14 in RNAi group (Fig. 7C). However, the number of spermatogonia increased from 20.85% at Day 1 to 67.89% at Day 14 in RNAi group (Fig. 7C).

Figure 7
figure 7

The histological observations of testis between RNAi and control group. SG: Spermatogonia; SC: spermatocyte; S: sperm; CT: collecting tissue. Scale bars = 20 μm. Data are shown as mean ± SD (standard deviation) from three separate slides. Capital letters indicated statistically significant percentage difference for the same cell type between different days in control group and RNAi group. (A) The histological observations of testis between RNAi and control group. (B) The percentages of different cell types in control group. (C) The percentages of different cell types in RNAi group.

Regulatory effects of Mn-HSDL1 with IGF1, IGF2, CYP11 and PRKAA2

HSDL1 was reported to have regulatory relationship with that of Insulin-like growth factor 1 (IGF1), Insulin-like growth factor 2 (IGF2), Cytochrome P450 (CYP11) and 5′-AMP-activated protein kinase catalytic subunit alpha-2 (PRKAA2) in the previous studies39,40. The regulatory effects of Mn-HSDL1 with Mn-IGF1, Mn-IGF2, Mn-CYP11 and Mn-PRKAA2 were measured in the same cDNA template of RNAi by using qPCR. According to the qPCR analysis, the expressions of Mn-CYP11 and Mn-PRKAA2 were decreased with the decrease of Mn-HSDL1, which showed positive regulatory effects (Fig. 8A,B). However, the expressions of Mn-IGF1 and Mn-IGF2 were increased with the decrease of Mn-HSDL1, which showed negative regulatory effects (Fig. 8C,D).

Figure 8
figure 8

Expression characterization of Mn-CYP11, Mn-PRKAA2, Mn-IGF1 and Mn-IGF2 at different days after Mn-HSDL1 dsRNA injection. The amount of mRNA expression was normalized to the EIF transcript level. Data are shown as mean ± SD (standard deviation) of tissues from three separate individuals. Capital letters indicate statistically significant expression differences between different days after GFP dsRNA injection in control group. Lowercase indicated expression difference between different days after Mn-HSDL1 dsRNA injection in the RNAi group. ** (P < 0.01) indicates a significant expression difference between the RNAi group and control group at the sample day. (A) Expression characterization of Mn-CYP11. (B) Expression characterization of Mn-PRKAA2. (C) Expression characterization of Mn-IGF1. (D) Expression characterization of Mn-IGF2.

Discussion

The eyestalk of crustaceans secretes many neurosecretory hormones that mediate reproduction, molting and metabolism of glucose in crustaceans23,24,25,26,27,28,29,30,31,32,33,34. The important neurosecretory hormones include CHH, ITP, GIH, MIH and MOIH. This study aimed to analyze the effects of eyestalk hormones on male sexual development. qPCR analysis revealed that the mRNA expression of Mn-IAG significantly increased at day 4 and day 7 after eyestalk ablation in both SS group and DS group, compared with day 1 (Fig. 1). The expression in DS group was significantly higher than SS group, which is consistent with previous studies41,42,43. However, the expression between day 4 and day 7 showed no significant difference in both SS group and DS group. IAG has been reported to promote male sexual differentiation and development in crustaceans8,9,10. Thus, the increase in Mn-IAG expression after the eyestalk ablation indicated that the eyestalk has negative effects on male sexual differentiation and development in M. nipponense. This also has a similar mediated function in ovarian development in M. nipponense35.

To date, this is the first long-read transcriptome in M. nipponense. The combination of long-read and next generation transcriptome sequencing can be useful to obtain transcripts with better integrity and quality for further gene structure and functional analysis. The accuracy and length of the transcripts of long-read transcriptome are further improved and optimized through correction by the next generation transcripts. Thus, this method is a suitable strategy for a species without a reference genome. The genes related to male sexual development were predicted to be found in the functional groups of Cell, Cellular Process and Binding in the GO assignment (Fig. 2), and the functional groups of General Function prediction only, Signal Transduction Mechanisms, Posttranslational Modification, Protein Turnover and Chaperones in the COG classification (Fig. 3), which were consistent with previous studies38,44. The gene sequences from this long-read transcriptome provide valuable information for the analysis of gene structure and function.

The number of DEGs between the CG versus DS groups were 4351, which were significantly greater than the number of DEGs between CG versus SS and SS versus DS. This indicates that the ablation of the double-side eyestalk has more regulatory roles on male sexual development in M. nipponense. This was consistent with the qPCR analysis. KEGG analysis revealed that Cell cycle, Cellular Senescence, Oxidative Phosphorylation, Glycolysis/Gluconeogenesis and Steroid Hormone Biosynthesis were the primary enriched metabolic pathways in all three comparisons. Previous studies have predicted the important roles of Oxidative Phosphorylation, Glycolysis/Gluconeogenesis and Steroid Hormone Biosynthesis in the mechanism of male sexual development in M. nipponense38,44. These previous studies included the transcriptome profiling analysis of testis between reproductive season and non-reproductive season38, and transcriptome analysis during the sex-differentiation and development sensitive period44. In these reports, it was predicted that the DEGs from these particular metabolic pathways play function in male sexual development in M. nipponense by providing ATP and promoting the biosynthesis of steroid hormones. The present study revealed that several metabolic pathways and DEGs involved in the immune system are predicted to participate in the mechanism of male sexual development in M. nipponense. These included Cell Cycle and Cellular Senescence. A reasonable explanation is that vigorous male sexual development after ablation of the eyestalks is due to a significant increase in IAG expression. Thus, immune related metabolic pathways are needed, in order to maintain normal testes development. It is suggested that these metabolic pathways digest aged cells and prevent proliferation of damaged copies of DNA.

The transcriptome profiling analysis revealed that Cell Cycle and Cell Senescence are the most enriched metabolic pathways in all three comparisons. The Cell Cycle is a ubiquitous and complex process that ensures correct cell proliferation. This pathway is crucial for the prevention and/or correction of damaged DNA, genetic abnormalities and mutations, with cyclins and cyclin-dependent kinases functioning in this process45,46. Cellular Senescence is defined as irreversible cell cycle arrest caused by different forms of stress. These stresses include telomere shortening, genotoxic stress, mitogens or inflammatory cytokines, the activation of the p53 tumor suppressor gene and/or the cyclin-dependent kinase inhibitor p1647,48. The dramatic enrichment of DEGs in these two metabolic pathways indicates that Cell Cycle and Cell Senescence function in the proofreading process when cells undergo replication. Four DEGs were enriched in both of the Cell Cycle and Cell Senescence categories, including cyclin A, cyclin B, cyclinB3 and Cdk2. Cyclin A is a vital component of the cell-cycle machinery, which can activate two different cyclin-dependent kinases (Cdk1 and Cdk2), functioning in both S-phase and mitosis49,50,51. Cdk1/cyclin B, also known as maturation promoting factor (MPF), is one of the main protein kinases. It activates, and serves as master regulator, for the M-phase transition, phosphorylating and activating other downstream protein kinases, and directly phosphorylating several structural proteins involved in cellular reorganization52,53,54. The Cdk family includes eight Cdk genes that can combine with different types of cyclins to form complexes, regulating the process of cell transition from the G1 phase to the S phase or G2 phase to the M phase and finally exiting from M phase. Cdk2 in particular is a member of a highly conserved family of protein kinases, regulating the eukaryotic cell cycle55,56,57.

Adenosine-triphosphate (ATP), a high-energy compound used as an energy source in nearly all metabolic activities, is essential for male differentiation and development. Therefore, it is of interest that in the present study, Oxidative Phosphorylation and Glycolysis/Gluconeogenesis were the main enriched metabolic pathways in all three comparisons. Oxidative Phosphorylation occurs in the inner membrane of mitochondria of eukaryotic cells or in the cytoplasm of prokaryotes. The energy released from the oxidation of substances in vivo promotes the coupling reaction between adenosine diphosphate (ADP) and inorganic phosphate to synthesize ATP through the respiratory chain58. Glycolysis/Gluconeogenesis promotes the conversion of glucose (C6H12O6) into pyruvate (CH3COCOO−  + H+), releasing free energy to form ATP and reduced nicotinamide adenine dinucleotide59. Three DEGs were selected from Oxidative Phosphorylation and Glycolysis/Gluconeogenesis. SDHB, a DEG that was down-regulated between CG versus SS and CG versus DS. SDHB, was also predicted to be involved in the mechanism of male sexual development in M. nipponense38. SDHB is one of four protein subunits that form succinate dehydrogenase, which catalyzes the oxidation of succinate60,61. Two subunits of cytochrome c oxidase, which function during oxidative phosphorylation, were also differentially expressed. These two subunits included cytochrome c oxidase assembly protein COX11 and cytochrome c oxidase subunit 7A1. Cytochrome c oxidase is located at the end of the cytochrome c system in cellular respiration. This enzyme directly transfers the electrons of respiratory substrates to molecular oxygen through the cytochrome system62,63.

It is widely acknowledged that steroid hormones primarily function in sexual development64,65. Hormones are generally divided into five main classes: glucocorticoids, mineralocorticoids, androgens, estrogens, and progestogens. Natural steroid hormones, which are lipids, are generally synthesized from cholesterol in the gonads and adrenal glands66,67. HSDL1 was differentially expressed between CG versus SS and CG versus DS, indicating that the expressions of HSDL1 is significantly regulated by the ablation of both single-side eyestalk and double-side eyestalk. HSDL1 was also shown to be involved in the mechanism of male sexual development in a previous study38. The short-chain dehydrogenase/reductases family (SDR) is a large enzyme family, which can affect mammalian reproduction, hypertension, neoplasia, and digestion68,69. Hydroxysteroid dehydrogenase is a subfamily of SDR, that functions in sex-determination, establishment and maintenance of secondary sexual characteristics, and the regulation of the endocrine system through catalyzing the metabolism of steroid hormones. HSDL1 is an important gene in the metabolic pathway of steroid hormones70. qPCR verification revealed that the expression pattern of important DEGs from these metabolic pathways were the same as the RNA-seq data (Fig. 4). This is an indicator of the accuracy of the transcriptome profiling analysis.

Both this study and previous studies predict the potentially vital roles of HSDL1 in the mechanism of male sexual development in M. nipponense38. Thus, the function of HSDL1 in male sexual development was also analyzed by using qPCR and RNAi, combined with histological observations. Previous studies have shown that HSDL1 was highly expressed in reproductive tissues (i.e., testes and ovaries) in humans, as revealed by Northern Blot analysis70. In situ hybridization indicated that the expression of HSDL1 was higher in prostate cancer than that in normal prostate tissue. In addition, this gene is involved in the fetal sheep development in the late gestational stages71. The qPCR analysis in different mature tissues revealed that the highest expression level of Mn-HSDL1 was observed in hepatopancreas, followed by testes, while Mn-HSDL1 RNA was rarely detected in other tissues38. Thus, it is predicted that HSDL1 may be involved in testes development in M. nipponense. RNAi was further used to analyze the potential functions of Mn-HSDL1 in testes development. The mRNA expression of Mn-HSDL1 was significantly decreased at day 7 and day 14 after Mn-HSDL1 dsRNA injection (Fig. 6A), indicating the RNAi used was effective in this study. The mRNA expression of Mn-IAG was also measured in the androgenic gland from the same prawn (Fig. 6B). The qPCR analysis revealed that Mn-IAG expression decreased with the decrease of Mn-HSDL1, indicating that HSDL1 has a positive regulatory effect on IAG in M. nipponense. IAG is a hormone, secreted by androgenic gland, promoting male sexual differentiation and development in many crustacean species8,9,10. According to the histological observations, the number of sperm decreased with the time of Mn-HSDL1 dsRNA injection. Compared with the control group, sperm were rarely found at day 14 after Mn-HSDL1 dsRNA injection (Fig. 7). This indicated that HSDL1 has a positive regulatory effect on testes development in M. nipponense. In different developmental stages, expression in the larval developmental stages was generally higher than in the post-larval developmental stages, indicating that HSDL1 may be involved in organ development and metamorphosis of M. nipponense. This is particularly observed in the development of the hepatopancreas, heart, and gills72,73, however, this requires further investigation. Mn-HSDL1 mRNA expression gradually increased from PL5 to PL25. The period from PL5 to PL25 is the sex-differentiation sensitive period5. Thus, the increase from PL5 to PL25 indicates that HSDL1 might function in gonad differentiation and development. In addition, gender is first distinguished at PL25. Expression in PL25♂ was two-fold higher than PL25♀, which also indicates that HSDL1 has a functional role in male sexual development.

Previous studies reported that HSDL1 has a regulatory relationship with IGF1, IGF2, CYP11 and PRKAA2, which participate in the glycolytic and lipogenic pathway39,40. This study identified the positive regulatory effects of Mn-HSDL1 on the expression of Mn-CYP11 and Mn-PRKAA2. CYP11a, which has possible relation to hyperandrogenemia, was reported to be associated with both polycystic ovary syndrome and total testosterone levels in women with polycystic ovary syndrome74. PRKAA2 is a gene that belongs to the AMP-activated protein kinase (AMPK). It received considerable attention because of the activation of both glucose and lipid metabolism and improvement of the insulin sensitivity75,76. However, Mn-HSDL1 negatively affects expression of Mn-IGF1 and Mn-IGF2. A reasonable explanation for this is that silencing Mn-HSDL1 inhibited testis development. Thus, the energy for testis development was used to promote the growth performance77,78.

In conclusion, candidate genes that are possibly involved in the male sexual development were selected through performing long-read and next generation transcriptome sequencing of the androgenic gland after eyestalk ablation in M. nipponense. qPCR analysis revealed that Mn-IAG expression significantly increased after the ablation of both single-side and double-side eyestalks, indicating that the ablation of eyestalk has potential regulatory roles on the process of male sexual development in M. nipponense. The long-read transcriptome generated 49,480 non-redundant transcripts. A total of 1319, 2092 and 4351 DEGs were identified between CG versus SS, SS versus DS and CG versus DS, respectively, indicating that the ablation of double-side eyestalks plays a regulatory role on male sexual development. Cell Cycle, Cellular Senescence, Oxidative Phosphorylation, Glycolysis/Gluconeogenesis and Steroid Hormone Biosynthesis were the main enriched metabolic pathways in all three comparisons, and the important DEGs from these metabolic pathways were identified. qPCR analysis and RNAi analysis of Mn-HSDL1 indicated that HSDL1 has positive regulatory effects on testes development. Overall, this study provided valuable information concerning the mechanisms underlying male sexual development in M. nipponense and potentially other crustacean species as well.

Materials and methods

Ethics statement

Permission was obtained from the Tai Lake Fishery Management Council and the committee of Freshwater Fisheries Research Center during the experimental programs. MS222 anesthesia was used to sedate the prawns and shear the tissues. All experiments were performed in accordance with relevant guidelines and regulations. Authors complied with the ARRIVE guidelines.

Sample collection

A total of 600 healthy male prawns and 20 healthy female prawns of M. nipponense were collected from a wild population in Tai Lake in July, Wuxi, China (120° 13′ 44″ E, 31° 28′ 22″ N). The body weight of male prawns was 3.63–4.94 g and the body weight for females was 3.21–3.45 g. All samples were randomly divided and transferred to three, 500 L tanks and maintained in aerated freshwater for three days. The three groups in this study were: CG, SS, and DS. The androgenic glands were collected from the three groups after 7 days of eyestalk ablation, and immediately preserved in liquid nitrogen until used for long-read and next-generation transcriptomic analysis. Mature tissues that were studied included testes ovaries, hepatopancreas, muscle, eyestalk, gill, heart and brain. One male parent prawn with a body weight of 4.87 g and one female parent prawn with a body weight of 3.45 g were collected from the wild population and mated in the laboratory in order to produce the full-sibs population. Specimens for the different stages of larval and post-larval developmental stages were obtained from the full-sibs population after hatching and collected throughout the maturation process.

Long-read transcriptome analysis

In order to provide sufficient RNA with an aim to establish a reference transcriptome for further analysis, equal amount of androgenic gland tissue from the CG, SS, and DS groups (N ≥ 60) were pooled together to perform the long-read sequencing. According to the manufacturer’s instructions, the UNlQ-10 Column Trizol Total RNA Isolation Kit (Sangon, Shanghai, China) was used to extract total RNA, and an Agilent RNA 6000 Nano kit and chips on a Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA, USA) was used to measure the RNA integrity. A PacBio RSII platform (Pacific Bioscience Inc., Menlo Park, CA, USA) was employed to construct the long-read transcriptome. The detailed procedures for the construction of long-read transcriptome and the analysis of raw sequence data have been well described in our previous study79.

In the next step, the contaminant sequences were removed by stepwise CLC80, and the LRS isoforms were annotated81. Using Blastp, the transcriptome factors were aligned to the PlnTFDB database (http://plntfdb.bio.uni-potsdam.de/v3.0/), the AnimalTFDB database (http://bioinfo.life.hust.edu.cn/AnimalTFDB/), and the CARD database (https://card.mcmaster.ca/) for the selection of genes involved in the mechanism of male sexual development in M. nipponense, using the threshold of E-value ≥ 1e–10. Finally, all Blastp results were processed with BLAST2GO82 for functional annotation. The long-read were annotated in the M. nipponense genome by using Lorean83.

Transcriptomic profiling analysis

The comparative transcriptome analysis of the androgenic gland between the CG, SS and DS groups were performed. In order to ensure the sufficient amount of RNA samples, androgenic glands from at least 30 prawns were pooled to form one biological replicate, and three biological replicates were sequenced for all three groups. Previously published studies have described the experimental process16,42.

Clean reads were assembled into non-redundant transcripts by using the Trinity program (version: trinityrnaseq_r20131110)84. The NR protein, the GO, the COG and the KEGG database were then used to perform the gene annotation, using an E-value cut-off of 10−516. Blast2go software was used for functional annotation by GO terms82. Blast software was employed to perform the functional annotation against the COG85 and KEGG86 database. EB-seq algorithm was used to filter the differentially expressed genes, under the criteria of FDR (False discovery rate) < 0.0587.

qPCR analysis

qPCR was used to measure the relative mRNA expression of Mn-HSDL1 in different developmental stages, as well as for confirmation of DEGs. The Bio-Rad iCycler iQ5 Real-Time PCR System (Bio-Rad) was used to carry out the SYBR Green RT-qPCR assay. The procedure has been well described in previous studies21,22. The primers used for qPCR verification of important DEGs are listed in Table 2. The primers used for qPCR analysis of Mn-HSDL1 are listed in Table 3. EIF was used as a reference gene in this study88. Three replicates were performed for each tissue.

Table 2 Primers used for qPCR verification.
Table 3 Primers used for HSDL1 analysis.

RNA interference (RNAi) analysis

RNAi was performed to analyze the potential regulatory roles of Mn- HSDL1 in male sexual development in M. nipponense. The Snap Dragon tool was used to design the specific RNAi primer with the T7 promoter site (http://www.flyrnai.org/cgibin/RNAifind_primers.pl) shown in Table 1. The Transcript Aid™ T7 High Yield Transcription kit (Fermentas, Inc, USA) was used to synthesize the Mn-HSDL1 dsRNA, according to manufacturer’s instructions. A total of 300 healthy mature male M. nipponense with a body weight of 3.21–4.78 g were collected and divided into two groups. As described in the previous study89,90, prawns from the experimental group were injected with 4 μg/g Mn- HSDL1 dsRNA, while prawns from the control group were injected with an equal volume of GFP dsRNA (control). HSDL1 mRNA expression was investigated in the androgenic gland by qPCR 1, 7 and 14 days after the injection, permitting confirmation of silencing efficiency (N ≥ 5). mRNA expression of Mn-IAG was measured in the same cDNA templates in order to analyze the regulatory relationship between Mn-HSDL1 and Mn-IAG.

Histological observation

The morphological changes in the testes between different days after RNAi treatment were observed by Hematoxylin and eosin (HE) staining. Five testicular samples were collected after 1, 7, and 14 days of RNAi treatment for HE staining. The procedures have been well described in previous studies91,92. Olympus SZX16 microscope was used to observe the slides (Olympus Corporation, Tokyo, Japan). The various cell types were labeled based on morphological analysis5.

Statistical analysis

Quantitative data were expressed as mean ± SD. Statistical differences were estimated by one-way ANOVA followed by LSD and Duncan’s multiple range test. All statistics were measured using SPSS Statistics 23.0. A probability level of 0.05 was used to indicate significance (P < 0.05).