Insight into the Regulatory Relationships between the Insulin-Like Androgenic Gland Hormone Gene and the Insulin-Like Androgenic Gland Hormone-binding Protein Gene in Giant Freshwater Prawns (Macrobrachium rosenbergii)

Giant freshwater prawns (Macrobrachium rosenbergii) are commonly found throughout the world. The size of the male giant freshwater prawn is much larger than that of the female. Therefore, understanding the molecular mechanism that underlies the sexual differentiation of M. rosenbergii is of both commercial and scientific importance. Insulin-like androgenic gland hormone (IAG) plays a key role in the differentiation of sex in M. rosenbergii. Although IAG has been investigated, the regulatory relationship between IAG and its binding protein partner, the insulin-like androgenic gland hormone-binding protein (IAGBP), has not been studied in M. rosenbergii. Here, we cloned and characterized the IAGBP from M. rosenbergii (Mr-IAGBP) for the very first time. Transcriptomic analysis showed that Mr-IAGBP mRNA was detected in a wide array of tissues with the highest expression found in the androgenic gland. The importance of IAG in male development was further demonstrated by an increase in IAG transcripts during the development of the androgenic gland and Mr-IAG was only highly transcribed in the androgenic gland of M. rosenbergii. Interestingly, we found that the Mr-IAG gene expression started during the 20th-day larva after hatching stage (LH20), followed (20th-day post-larval stage, PL20) by a gradual elevation of Mr-IAGBP levels. The levels of both genes peaked at the adult stage. The relationship between Mr-IAGBP and Mr-IAG was further analyzed using RNA interference. The injection of Mr-IAGBP double-stranded RNA (dsRNA) significantly reduced the transcription of Mr-IAG, while the amount of Mr-IAGBP mRNA and the translation of IAGBP protein was significantly reduced by the injection of Mr-IAG dsRNA. These results revealed that IAGBP is involved in IAG signaling. Furthermore, our data supports the hypothesis that (IAG and IAGBP)-IAG receptor signaling schemes exist in M. rosenbergii. Our results will provide important information for the further study of determining the sex of M. rosenbergii.


Introduction
Giant freshwater prawn (Macrobrachium rosenbergii) is a commercially important species of freshwater prawn, which is widely cultured worldwide [1]. Because there has been sexual dimorphism in most economic crustaceans [2], growth in a monosex environment enables them to allocate more energy to growth instead of sexual activity. Thus, there could be a marked rise in the yield and reduction in the length of the reproductive cycle of the M. rosenbergii if there is a development of the monosex culture [3].
Males have an exclusive endocrine organ known as the androgenic gland (AG), which was initially observed in crustaceans of swimming blue crab (Callinectes sapidus) [4]. Previous studies have proven that AG plays a crucial role in sexual and/or morphotypic differentiation in crustaceans [5][6][7][8][9][10]. Moreover, AG is responsible for masculinity if it is implanted in females and vice versa if removed from males, which exhibits the main role in the differentiation of male sexual characteristics in crustaceans [5,[11][12][13][14]. Several androgenic effects of the AG are attributed to another hormone that is encoded by an insulin-like androgenic gland hormone (IAG) gene secreted as a proteinaceous hormone [15][16][17][18]. The IAG gene expressed specifically in the AG of males in Cherax quardricarinatus [15] and Penaeus monodon [16]. Recently, an increase in the number of IAGs existing in various Decapod species has been recognized [19][20][21]. In addition, a previous study suggested that insulin-like androgenic gland hormone-binding protein (IAGBP) modulates and aids the signaling of IAG in the overall route in Macrobrachium nipponense [22]. Rosen found that the Insulin-like growth factor-binding protein (IGFBP) in Cherax quadricarinatus (Cq-IGFBP) is the first IGFBP family member to specifically interact with IAG [23]. Chandler described in great detail the identification and characterization of the IGFBP in Sagmariasus verreauxi (Sv-IGFBP) [24].
The researchers found that silencing of Mr-IAG led to the cessation of testicular spermatogenesis and spermatogenic cell development in the ampulla of the end of the sperm duct, accompanied by hypertrophy and hyperplasia of the androgenic gland (AG), so it was concluded that Mr-IAG plays an important role in spermatogenesis and the development of primary and secondary male sexual characteristics in M. rosenbergii [18,25,26]. Silencing the Mr-IAG gene in M. rosenbergii juveniles by the repeated injection of Mr-IAG double-stranded RNA leads to functional sex reversal [27,28]. In addition, the role of IAG in orange claw male (oc) and blue claw male (bc) morphotype transformation has also been confirmed [29]. Despite the examination of the sequences and expressive profiles of IAG, the regulatory mechanism of IAG is not fully understood. A previous study suggested that IAG and the insulin-like androgenic gland hormone-binding protein (IAGBP) transcription may be regulated by feedback inhibition in crustaceans [22]. Information on the regulatory relationship between IAG and IAGBP is still limited. Detailed studies on the regulatory relationship between IAG and IAGBP would help to clarify the signaling schemes of Mr-IAG.
The aim of the present study is to investigate the regulatory relationship between IAG and IAGBP in M. rosenbergii, which reflects the roles of these two genes in gonadal development of M. rosenbergii. The IAGBP gene in M. rosenbergii was cloned and characterized using a de novo transcriptomic library. Subsequently, the expression patterns of Mr-IAGBP and Mr-IAG in different types of tissues and at different stages of development were investigated. Finally, the relationship between Mr-IAGBP and Mr-IAG was demonstrated by gene silencing using double-stranded RNA (dsRNA). Our results clarified the transcriptional relationship between the Mr-IAGBP and Mr-IAG genes, which will lay the foundations for further studying of the sexual determination of M. rosenbergii.

Multiple Alignment and Phylogenetic Analysis
The Mr-IAGBP peptide sequence that was discovered was re-examined in more depth using ClustalW to check for similarities with other crustaceans (Figure 2A). Vertebral IGFBP sequences were also aligned ( Figure 2B). The most identical Mr-IAGBP was found with M. nipponense (81.16%),

Multiple Alignment and Phylogenetic Analysis
The Mr-IAGBP peptide sequence that was discovered was re-examined in more depth using ClustalW to check for similarities with other crustaceans (Figure 2A). Vertebral IGFBP sequences were also aligned ( Figure 2B). The most identical Mr-IAGBP was found with M. nipponense (81.16%), and the least identical with C. quadricarinatus (48.57%). IGFBP sequences were found to be similar to vertebrates, (at least 65.03%) between the Anolis carolinensis and Geospiza fortis, and the highest similarity was between Alligator mississippiensis and Pelodiscus sinensis (79.91%). The results of sequence similarities were constructed with the neighbor-joining method and a phylogenetic tree. Figure 3 demonstrates how two clades of division occur from peptides of the vertebrates and invertebrates: one consists of IAGBP from crustaceans, while the other consists of known IGFBPs from vertebrates and arachnids. and the least identical with C. quadricarinatus (48.57%). IGFBP sequences were found to be similar to vertebrates, (at least 65.03%) between the Anolis carolinensis and Geospiza fortis, and the highest similarity was between Alligator mississippiensis and Pelodiscus sinensis (79.91%). The results of sequence similarities were constructed with the neighbor-joining method and a phylogenetic tree. Figure 3 demonstrates how two clades of division occur from peptides of the vertebrates and invertebrates: one consists of IAGBP from crustaceans, while the other consists of known IGFBPs from vertebrates and arachnids.

Recombinant Mr-IAGBP Protein Expression, Purification, and Polyclonal Antibodies Analysis
The open reading frame (ORF) of Mr-IAGBP was cloned into a pET-32a vector, transformed to BL21 (DE3), and the recombinant protein fused with His-tag columns was purified and analyzed with SDS-PAGE gel. After double digestion of the recombinant plasmid, two distinct bands appeared, and the size of the target band was consistent with that predicted ( Figure 4A). As shown in the SDS-PAGE gel, a distinct band was identified with a size of 47.48 kDa of inclusion bodies ( Figure 4B), corresponding to the His-IAGBP recombinant fusion protein (the Mr-IAGBP protein about 29.48 kDa, and the control pET-32a about 18 kDa). Using the His-tag primary antibody, the recombinant Mr-IAGBP protein has a single and distinct band around 47.48 kDa, and the size is as predicted, indicating that the recombinant Mr-IAGBP protein contains a His-tag ( Figure 4C). The purified His-IAGBP recombinant protein was injected into the rabbit to produce polyclonal antibodies, which was used as the primary antibody to anti the total protein of the M. rosenbergii AG for Western blot. Mr-IAGBP caused the rabbit polyclonal antibodies to react. This, as well as a specific positive band on Western blot nitrocellulose membrane of about 29.48 kDa, ensured the expected molecular weight of Mr-IAGBP protein. This, in turn, also confirmed the successful preparation of the rabbit polyclonal antibodies ( Figure 4D). The titer of the polyclonal antibody serum was determined by enzyme-linked immunosorbent assay, and the polyclonal antibody titer was not less than 1:51,200 ( Figure 4E). The open reading frame (ORF) of Mr-IAGBP was cloned into a pET-32a vector, transformed to BL21 (DE3), and the recombinant protein fused with His-tag columns was purified and analyzed with SDS-PAGE gel. After double digestion of the recombinant plasmid, two distinct bands appeared, and the size of the target band was consistent with that predicted ( Figure 4A). As shown in the SDS-PAGE gel, a distinct band was identified with a size of 47.48 kDa of inclusion bodies ( Figure 4B), corresponding to the His-IAGBP recombinant fusion protein (the Mr-IAGBP protein about 29.48 kDa, and the control pET-32a about 18 kDa). Using the His-tag primary antibody, the recombinant Mr-IAGBP protein has a single and distinct band around 47.48 kDa, and the size is as predicted, indicating that the recombinant Mr-IAGBP protein contains a His-tag ( Figure 4C). The purified His-IAGBP recombinant protein was injected into the rabbit to produce polyclonal antibodies, which was used as the primary antibody to anti the total protein of the M. rosenbergii AG for Western blot. Mr-IAGBP caused the rabbit polyclonal antibodies to react. This, as well as a specific positive band on Western blot nitrocellulose membrane of about 29.48 kDa, ensured the expected molecular weight of Mr-IAGBP protein. This, in turn, also confirmed the successful preparation of the rabbit polyclonal antibodies ( Figure 4D). The titer of the polyclonal antibody serum was determined by enzyme-linked immunosorbent assay, and the polyclonal antibody titer was not less than 1:51,200 ( Figure 4E).  * * represents very significant difference, p < 0.01; * * * represents extremely significant difference, p < 0.001).

Tissue and Spatial Distributions of Mr-IAGBP and Mr-IAG Transcripts
The distributions of tissue of Mr-IAGBP and Mr-IAG ( Figure 5) were examined by the qRT-PCR method. The Mr-IAGBP mRNA was found in eight different tissues. The level of transcription was highest in the AG, but similar levels were identified in muscle tissue and the hepatopancreas. Moreover, this value was the lower in the eyestalk ( Figure 5B). Mr-IAG mRNA was located in the AG and hepatopancreas only, where the AG had comparatively higher levels than the hepatopancreas ( Figure 5A). In other examined tissues, such as the heart, testis, nerve cord, eyestalk, brain tissue, and muscles, Mr-IAG expression was not found at all.

Tissue and Spatial Distributions of Mr-IAGBP and Mr-IAG Transcripts
The distributions of tissue of Mr-IAGBP and Mr-IAG ( Figure 5) were examined by the qRT-PCR method. The Mr-IAGBP mRNA was found in eight different tissues. The level of transcription was highest in the AG, but similar levels were identified in muscle tissue and the hepatopancreas. Moreover, this value was the lower in the eyestalk ( Figure 5B). Mr-IAG mRNA was located in the AG and hepatopancreas only, where the AG had comparatively higher levels than the hepatopancreas ( Figure 5A). In other examined tissues, such as the heart, testis, nerve cord, eyestalk, brain tissue, and muscles, Mr-IAG expression was not found at all. Levels of mRNA in Mr-IAG and Mr-IAGBP were calculated while the developmental stage was ongoing from embryo to adult, and this was accomplished using qRT-PCR ( Figure 6). We did not detect Mr-IAG mRNA in embryonic tissues or within 20 days of larva hatching (LH20) stage ( Figure  6A), and the expression of Mr-IAG was detected in post-larval stage and adulthood. Specifically, there was transcription of Mr-IAG detected at the 20th-day larva after hatching (LH20) stage, and the expression level of Mr-IAG mRNA from LH20 to the adult stage was always increased. However, the transcription of Mr-IAGBP were detected from the cleavage stage to adulthood rather than from LH20 to adulthood, and the expression level of Mr-IAGBP mRNA from PL20 to the adult stage was always increased ( Figure 6B).  Levels of mRNA in Mr-IAG and Mr-IAGBP were calculated while the developmental stage was ongoing from embryo to adult, and this was accomplished using qRT-PCR ( Figure 6). We did not detect Mr-IAG mRNA in embryonic tissues or within 20 days of larva hatching (LH20) stage ( Figure 6A), and the expression of Mr-IAG was detected in post-larval stage and adulthood. Specifically, there was transcription of Mr-IAG detected at the 20th-day larva after hatching (LH20) stage, and the expression level of Mr-IAG mRNA from LH20 to the adult stage was always increased. However, the transcription of Mr-IAGBP were detected from the cleavage stage to adulthood rather than from LH20 to adulthood, and the expression level of Mr-IAGBP mRNA from PL20 to the adult stage was always increased ( Figure 6B).

Tissue and Spatial Distributions of Mr-IAGBP and Mr-IAG Transcripts
The distributions of tissue of Mr-IAGBP and Mr-IAG ( Figure 5) were examined by the qRT-PCR method. The Mr-IAGBP mRNA was found in eight different tissues. The level of transcription was highest in the AG, but similar levels were identified in muscle tissue and the hepatopancreas. Moreover, this value was the lower in the eyestalk ( Figure 5B). Mr-IAG mRNA was located in the AG and hepatopancreas only, where the AG had comparatively higher levels than the hepatopancreas ( Figure 5A). In other examined tissues, such as the heart, testis, nerve cord, eyestalk, brain tissue, and muscles, Mr-IAG expression was not found at all. Levels of mRNA in Mr-IAG and Mr-IAGBP were calculated while the developmental stage was ongoing from embryo to adult, and this was accomplished using qRT-PCR ( Figure 6). We did not detect Mr-IAG mRNA in embryonic tissues or within 20 days of larva hatching (LH20) stage ( Figure  6A), and the expression of Mr-IAG was detected in post-larval stage and adulthood. Specifically, there was transcription of Mr-IAG detected at the 20th-day larva after hatching (LH20) stage, and the expression level of Mr-IAG mRNA from LH20 to the adult stage was always increased. However, the transcription of Mr-IAGBP were detected from the cleavage stage to adulthood rather than from LH20 to adulthood, and the expression level of Mr-IAGBP mRNA from PL20 to the adult stage was always increased ( Figure 6B).

Transcription in Mr-IAG-dsRNA-and Mr-IAGBP-dsRNA-injection Groups
RNAi was used to describe the correlation between the Mr-IAG and Mr-IAGBP in vivo transcriptions. dsRNA was injected into M. rosenbergii, along with the target gene or DEPC water. This injection significantly decreased the transcription levels in Mr-IAG and Mr-IAGBP, as proved by the qRT-PCR (Figures 7 and 8). The mRNA levels in Mr-IAG and Mr-IAGBP also reduced greatly in the Mr-IAGBP-dsRNA-injected group, to 30.5% ( Figure 7A) and 15.6% ( Figure 7B) of control group prawns, with (p < 0.01). Mr-IAG mRNA levels were also seen to drop significantly (p < 0.01) in the Mr-IAG-dsRNA-injected prawns, with a new value of 13.4% ( Figure 8A), whereas the Mr-IAGBP mRNA levels reduced to 65.6%, 69.7%, and 68.5%, in muscle, the AG, and the epatopancreas, respectively (p < 0.01) ( Figure 8B). In other tissues (heart, testis, eyestalk, nerve cord, and brain), the mRNA levels were not much different for the two test groups and the control group ( Figure 8B).

Transcription in Mr-IAG-dsRNA-and Mr-IAGBP-dsRNA-injection Groups
RNAi was used to describe the correlation between the Mr-IAG and Mr-IAGBP in vivo transcriptions. dsRNA was injected into M. rosenbergii, along with the target gene or DEPC water. This injection significantly decreased the transcription levels in Mr-IAG and Mr-IAGBP, as proved by the qRT-PCR (Figures 7 and 8). The mRNA levels in Mr-IAG and Mr-IAGBP also reduced greatly in the Mr-IAGBP-dsRNA-injected group, to 30.5% ( Figure 7A) and 15.6% ( Figure 7B) of control group prawns, with (p < 0.01). Mr-IAG mRNA levels were also seen to drop significantly (p < 0.01) in the Mr-IAG-dsRNA-injected prawns, with a new value of 13.4% ( Figure 8A), whereas the Mr-IAGBP mRNA levels reduced to 65.6%, 69.7%, and 68.5%, in muscle, the AG, and the epatopancreas, respectively (p < 0.01) ( Figure 8B). In other tissues (heart, testis, eyestalk, nerve cord, and brain), the mRNA levels were not much different for the two test groups and the control group ( Figure 8B).  Mr-IAG and Mr-IAGBP mRNA levels were normalized to β-actin. The tissues included: heart (Ht), testis (Te), eyestalk (Es), nerve cord (Nc), muscle (Mu), androgenic gland (Ag), hepatopancreas (Hp), and brain (Br). qRT-PCR data are shown as means ± SE (standard error). * p < 0.05; * * p < 0.01.

Western Blotting and Immunohistochemistry Analyses
To further examine the relationship between Mr-IAGBP protein and Mr-IAG protein in M. rosenbergii, we examined the expression levels of Mr-IAGBP protein in the heart, testis, nerve cord, eyestalk, androgenic gland, muscle, hepatopancreas, and brain tissues of the control groups (DEPC water) and experimental groups (injection of dsRNA-Mr-IAG) by Western blotting and immunohistochemistry. Heart, eyestalk, testis, nerve cord, androgenic gland, muscle, hepatopancreas, and brain tissue proteins were extracted from the control groups and experimental groups. Western blotting showed that there were more obvious positive signals in the androgenic

Transcription in Mr-IAG-dsRNA-and Mr-IAGBP-dsRNA-injection Groups
RNAi was used to describe the correlation between the Mr-IAG and Mr-IAGBP in vivo transcriptions. dsRNA was injected into M. rosenbergii, along with the target gene or DEPC water. This injection significantly decreased the transcription levels in Mr-IAG and Mr-IAGBP, as proved by the qRT-PCR (Figures 7 and 8). The mRNA levels in Mr-IAG and Mr-IAGBP also reduced greatly in the Mr-IAGBP-dsRNA-injected group, to 30.5% ( Figure 7A) and 15.6% ( Figure 7B) of control group prawns, with (p < 0.01). Mr-IAG mRNA levels were also seen to drop significantly (p < 0.01) in the Mr-IAG-dsRNA-injected prawns, with a new value of 13.4% ( Figure 8A), whereas the Mr-IAGBP mRNA levels reduced to 65.6%, 69.7%, and 68.5%, in muscle, the AG, and the epatopancreas, respectively (p < 0.01) ( Figure 8B). In other tissues (heart, testis, eyestalk, nerve cord, and brain), the mRNA levels were not much different for the two test groups and the control group ( Figure 8B).  Mr-IAG and Mr-IAGBP mRNA levels were normalized to β-actin. The tissues included: heart (Ht), testis (Te), eyestalk (Es), nerve cord (Nc), muscle (Mu), androgenic gland (Ag), hepatopancreas (Hp), and brain (Br). qRT-PCR data are shown as means ± SE (standard error). * p < 0.05; * * p < 0.01.

Western Blotting and Immunohistochemistry Analyses
To further examine the relationship between Mr-IAGBP protein and Mr-IAG protein in M. rosenbergii, we examined the expression levels of Mr-IAGBP protein in the heart, testis, nerve cord, eyestalk, androgenic gland, muscle, hepatopancreas, and brain tissues of the control groups (DEPC water) and experimental groups (injection of dsRNA-Mr-IAG) by Western blotting and immunohistochemistry. Heart, eyestalk, testis, nerve cord, androgenic gland, muscle, hepatopancreas, and brain tissue proteins were extracted from the control groups and experimental groups. Western blotting showed that there were more obvious positive signals in the androgenic gland, hepatopancreas, and muscle tissues of the control groups than in the experimental groups Mr-IAG and Mr-IAGBP mRNA levels were normalized to β-actin. The tissues included: heart (Ht), testis (Te), eyestalk (Es), nerve cord (Nc), muscle (Mu), androgenic gland (Ag), hepatopancreas (Hp), and brain (Br). qRT-PCR data are shown as means ± SE (standard error). * p < 0.05; * * p < 0.01.

Western Blotting and Immunohistochemistry Analyses
To further examine the relationship between Mr-IAGBP protein and Mr-IAG protein in M. rosenbergii, we examined the expression levels of Mr-IAGBP protein in the heart, testis, nerve cord, eyestalk, androgenic gland, muscle, hepatopancreas, and brain tissues of the control groups (DEPC water) and experimental groups (injection of dsRNA-Mr-IAG) by Western blotting and immunohistochemistry. Heart, eyestalk, testis, nerve cord, androgenic gland, muscle, hepatopancreas, and brain tissue proteins were extracted from the control groups and experimental groups. Western blotting showed that there were more obvious positive signals in the androgenic gland, hepatopancreas, and muscle tissues of the control groups than in the experimental groups ( Figure 9). In addition, the same tissue samples were detected by immunohistochemistry. The results showed that different degrees of positive signals were detected in the heart, testis, eye stalk, nerve cord, muscle, androgen gland, hepatopancreas, and brain tissues in the control groups and experimental groups. The results of immunohistochemistry were consistent with those of Western blotting (Figure 10).  Figure 9). In addition, the same tissue samples were detected by immunohistochemistry. The results showed that different degrees of positive signals were detected in the heart, testis, eye stalk, nerve cord, muscle, androgen gland, hepatopancreas, and brain tissues in the control groups and experimental groups. The results of immunohistochemistry were consistent with those of Western blotting ( Figure 10).    Figure 9). In addition, the same tissue samples were detected by immunohistochemistry. The results showed that different degrees of positive signals were detected in the heart, testis, eye stalk, nerve cord, muscle, androgen gland, hepatopancreas, and brain tissues in the control groups and experimental groups. The results of immunohistochemistry were consistent with those of Western blotting ( Figure 10).

Discussion
Previously, a great deal of research has focused on the identification and functional expression of IAG in M. rosenbergii [18,26,30,31]. However, there are still very few studies on the regulatory mechanism of IAG. As its binding protein companion, IAGBP may assist the signaling of IAG in the overall route in crustaceans [22]. Therefore, the regulation strategies of Mr-IAG and Mr-IAGBP at transcriptional and protein levels require further study.
In this study, an IAGBP encoding gene was isolated from M. rosenbergii. The predicted ORF of Mr-IAGBP (276 aa) was found to be similar to that of M. nipponense IAGBP (GenBank: KJ831645, 252 aa, 81.16% identity). The family of IGFBP precursors are proteins that are rich in cysteine-prepeptides containing 16-20 csyteines-and have the same organization of structure as two conserved domains (Nand C-terminal domains), while the central region, being a variable, separates them [32,33]. The current research identifies the 18 cysteines in the Mr-IAGBP precursor, where the N-terminal domain 12 cysteines are also included, along with four from the variable central region and the C-terminal domain 2. The existence of 12 out of 18 cysteines from the N-terminal domain suggests an increased structure of that domain, which includes a maximum of six disulfide bonds [33]. Mr-IAGBP is seen be surrounded by other crustaceans in the phylogenetic analysis, but the vertebrates created a detached clade. C. quadricarinatus [23] and M. nipponense [22] gave results that were similar. However, IGFBP sequences were seen to have greater similarity among the vertebrates than among the crustaceans. These differences could be attributed to the various functions of crustaceans as compared to vertebrates.
The expression patterns of Mr-IAGBP in various tissues suggest that Mr-IAGBP exhibits the highest expression in the androgenic gland, indicating that this gene is associated with the development of the androgenic gland. In addition, we also discovered a wide distribution of Mr-IAGBP, which suggests that IAG and IAGBP may be synthetized in different cell types within a gland and Mr-IAGBP may have extensive biological functions. Mr-IAG was specifically highly expressed in the androgenic gland which is consistent with previous findings [26]. The expression pattern of Mr-IAGBP and Mr-IAG at various stages of the development of M. rosenbergii was also investigated. Our results showed that Mr-IAG mRNA was undetected until the LH20 stage during embryonic development. This is different from previous results where Mr-IAG was expressed from the PL20 stage [26]. The difference between the two results could be due to the expression levels of Mr-IAG being detected at more developmental stages. In addition, there was a difference in the results for M.nipponense and Fenneropenaeus chinensis. IAG transcripts were detectable during embryonic development in M. nipponense and F. chinensis [13,22]. Interestingly, we discovered that levels of Mr-IAGBP transcripts increased significantly later (PL20) than those of Mr-IAG (LH20). In previous studies, it has been observed that fully functional sex reversal of M. rosenbergii males could be achieved by ablation of the androgenic gland into neofemales, and the success rate is greater if the ablation is performed earlier, during PL20 [5,26].
The injection of dsRNA, which targets the Mr-IAGBP or Mr-IAG, reduced the transcriptional level of Mr-IAGBP or Mr-IAG in correspondence with the degradation of the RNA target. Remarkably, the levels of Mr-IAGBP mRNA and IAGBP protein were considerably reduced by the injection of Mr-IAG dsRNA, and Mr-IAG mRNA levels in the androgenic gland were significantly reduced by the injection of Mr-IAGBP dsRNA compared to the controls. These results revealed that IAGBP is associated with IAG signaling. Previous RNAi studies of M. rosenbergii and C. quadricarinatus have indicated that IAG might regulate its own secretion by feedback inhibition [18,34]. It is therefore possible that Mr-IAG and Mr-IAGBP transcription may be regulated through feedback inhibition. Sharabi found that the most significant effect of IAG receptor gene silencing were hypertrophy of AG and the increased production of Mr-IAG, and abnormally rich immature sperm cells can be seen at the end of sperm [35]. Moreover, Qing confirmed the interactions between IAG receptor gene and IAG in male Fenneropenaeus chinensis [36]. Based on the above results, we support the idea that there may be (IAG and IAGBP)-IAG receptor signaling schemes in M. rosenbergii.

Tissue Sampling, RNA Isolation, and Reverse Transcription
Both post-larvae (2-3 cm) and adult males (25-30 cm, male external characters observed by the naked eye) of M. rosenbergii were taken from Jin Yang Aquaculture Co. Ltd., Guangzhou, China. Each sample was transported to the laboratory for breeding and contained in a 500 L tank of aerated freshwater. The quality of the water was guaranteed by running the 500 L volume through a biofilter. The samples collected from Jin Yang Aquaculture Co. Ltd., Guangzhou, China were divided into six groups. Each group of samples collected included 16 various stages of development, including embryos (zoea stage, blastula stage, nauplius stage, gastrula stage, cleavage stage) [37][38][39], larvae (first, 10th, 20th, and 30th days after larva hatching, 1st, 10th, 20th, 30th, 60th, 90th, and the 120th post-larval days) [40,41], as well as male adults. The sex of the 1st and 20th post-larval day prawns are not distinguishable, therefore, a mixed group was taken. The extracted overall RNA was collected at various developmental stages, as well as separate types of tissues (heart, testis, eyestalk, nerve cord, muscle, androgenic gland, hepatopancreas, and brain) from adult prawns. RNAiso Plus Reagent (TaKaRa, Dalian, China) was used in accordance with the original manufacturer's protocol. The RNA was isolated and RNase-free DNase I (Sangon, Shanghai, China) treatment was administered to remove any possible contamination by genomic DNA. Using a biophotometer (Eppendorf, Hamburg, Germany), the concentration of the total RNA in each sample was noted, after which 1% agarose gel was used per 2 µL for electrophoresis to evaluate all the sample's integrity [42]. Each RNA had their first-strand cDNA developed to make real-time reverse transcription-quantitative polymerase chain reaction (qRT-PCR) possible; this was done with 1 µL of the total RNA, 4 µL of 5× iScript reaction mix (Bio-Rad, Hercules, CA, USA), and 1 µL iScript reverse transcriptase in a final volume of 20 µL. The incubation period was 5 min at 85 • C. Reverse transcribed cDNA was stored until needed at 20 • C.

Gene Cloning of Mr-IAGBP and Bioinformatics Analyses
The transcript fragment of Mr-IAGBP (Table S1) was first obtained through a de novo transcriptomic library, and with this fragment a complete cDNA of Mr-IAGBP was cloned by PCR. Table S2 lists every primer used in this study. PCR amplification was executed under the following conditions: 5 min for pre-denaturation at 95 • C, then 35 cycles of 10 s at 95 • C, 30 s at 55 • C, and 1 min at 72 • C, followed by post-extension for 10 min at 72 • C, and then kept at 4°C. Electrophoresis was conducted on the PCR fragments with 1% agarose gels. The amplified cDNA fragments were then cloned by insertion into a pMD18-T vector, followed by sequencing with M13 primers in forward or reverse. After verification of the final sequences, they were exposed to the cluster analysis in NCBI [43].
Emboss (http://emboss.Bioinformatics/) and SMART (http://smart.embl-heidelberg.de) tools predicted the protein domains and the complete amino acid (aa) sequence of the Mr-IAGBP ORF. The BLAST program (http://www.ncbi.nlm.nih.gov/blast) analyzed the similarity in the sequences of the pre-determined nucleotide sequences. CBS prediction servers (http://www.cbs.dtu.dk/services) predicted the signal peptides, O-linked and N-linked glycosylation sites, and phosphorylation sites. The Clustal X 2.0 program created various alignments of Mr-IAGBP, while the DNAMAN software package (Lynnon Biosoft, Quebec, QC, Canada) generated the breakpoint analyses in detail. The phylogenetic tree was constructed on the basis of ORF aa sequences of Mr-IAGBP proteins by the neighbor-joining (NJ) method [44] using the molecular evolution genetics analysis (MEGA 6.0) software [45]. The bootstrap test was conducted on the basis of 10,000 pseudo-replications to judge the dependability of the phylogenetic tree [46].

Construction of Recombinant Mr-IAGBP Plasmid, Expressing and Purification
In order for the recombinant Mr-IAGBP protein to show phenotypical features, the primers expressed in Table S2 were designed to magnify the full-length ORF of Mr-IAGBP. The PCR products were disinfected and injected into the pMD-19T vector, and subsequently confirmed by sequencing. After, the 1% agarose gel electrophoresis and target products were decontaminated using a TaKaRa Agarose Gel DNA Purification KitVer.2.0 (TaKaRa, Kyoto, Japan). T4 DNA ligase was used to connect the Mr-IAGBP fragment to the empty pET-32a vector and kept overnight at 4°C. The expression plasmid pET-32a-Mr-IAGBP was transformed to E.coli BL21(DE3) (TIANGEN, Beijing, China) and then cultured in Amp + LB at 37 • C at a speed of 200 rpm. Isopropyl-β-D-thiogalactopyranpside (IPTG) was added to create a final concentration of 0.5 mmol/L and then cultured at 37 • C at a speed of 200 rpm for 4 h, and then centrifuged at 4 • C at a speed of 12,000 rpm for 10 min. Sediment cells were resuspended in PBS and purified with the His Band Resin columns (Sangon Biotech, Shanghai, China), as per protocol. The concentrations of recombinant pET-32a-Mr-IAGBP protein were determined according to the method specified for the Bradford Protein Assay Kit (Beyotime, Shanghai, China). The purified recombinant protein was separated in a 12% SDS-PAGE gel electrophoresis.

Rabbit Polyclonal Antibodies Against Recombinant Mr-IAGBP and Western-blot
The Frdbio Bioscience & Technology company (Wuhan, China) constructed the rabbit polyclonal antibodies against recombinant protein of Mr-IAGBP. The total protein concentrations from the lysates of tissues were prepared. The total protein quantity was 30 µg, which was divided in 12% SDS-PAGE, and then transferred onto a nitrocellulose membrane (Bio-Rad, Hercules, CA, America). TBST was used containing 137 mM NaCl, 20 mM Tris, and 1% Tween-20, at pH 7.6 to block the membranes, as it contained 3% skimmed milk, and was left for 2.5 h at room temperature. Following this, the membranes were put into incubation overnight at 4°C with the rabbit anti-Mr-IAGBP (1/1000 dilution) primary antibody. The membranes were then rinsed for 5 min in three repetitions with TBST and also incubated with HRP-conjugated secondary antibody goat anti-rabbit IgG (1/10,000 dilution) for 55 min. Another three repeats for 5 min washing with TBST were done [42]. This led to the revealing and measurement of the immunoreactive bands by chemiluminescence (ECL Western Blotting Substrate, Solarbio, Beijing, China) and ChemiScope 6000 (CliNX, Shanghai, China), respectively.

qRT-PCR Analysis of Mr-IAGBP and Mr-IAG
A quantitative real-time PCR (qRT-PCR) assay in Roche LightCycler 480 (Roche, Branchburg, NJ, USA) was used to calculate the levels of transcription of Mr-IAGBP and Mr-IAG, in the different standards set in this study. Table S2 shows all the primers used. AceQ ® qPCR SYBR ® Green Master Mix (Vazyme, Nanjing, China) was used to conduct qRT-PCR. The final volume was 20 µL, and this was reacted in three batches, with 1 µL cDN, 1 µL primer of each specificity, 10 µL Green Master Mix, and finally 7 µL ddH 2 O. The conditions provided were 95 • C for 3 min, 40 cycles at 95 • C for 15 s, 60 • C for 30 s, and 72 • C for 20 s, and finally 4 • C for 5 min. The ratio of expression in correlation of the target genes was standardized with an internal reference β-actin gene [42,43], and the 2 −∆∆CT method was used to calculate expression levels of Mr-IAGBP and Mr-IAG [47].

Double-Stranded RNA Preparation
A MEGAscript T7 Kit (Ambion, Foster, CA, USA) was used to synthesize dsRNA, and the SnapDragon-dsRNA Design (http://www.flyrnai.org/cgi-bin/RNAi_find_primers.pl) was used to design primers that aid RNA interference (RNAi), with cDNA sequences of Mr-IAGBP and Mr-IAG as templates. For the in vitro transcription, the templates were made using gene-specific primers and then PCR, along with the T7 polymerase promoter sequence on their 5´ends (Table S2).
To produce 20 µL in vitro transcription, 150 ng of template was used, and the mixture of reaction was incubated for 2 h at 37 • C, after which extraction of synthesized RNA was done with phenol/chloroform. After ethanol precipitation, the RNA was suspended again in DEPC water.

In Vivo Mr-IAG and Mr-IAGBP Silencing
RNAi technology was used to examine the correlation between Mr-IAGBP and Mr-IAG at the level of transcription. Three groups of healthy adult prawns (four months post hatching) were made with n = 20 for each group, and each wet prawn weighed between 60 g and 80 g. One group was injected with Mr-IAG dsRNA, one with Mr-IAGBP dsRNA, and the third with DEPC water, the latter of which was the control group. For two weeks, each prawn was injected twice per week with 5 µg per gram of body weight dsRNA in the sinus of their fifth walking leg [18,26,29,34]. Group two, Mr-IAGBP-dsRNA, and the control group had their AGs collected for extraction of RNA. From group one, Mr-IAG-dsRNA, as well as the control group, eight types of tissue were sampled. qRT-PCR was used to define the relative level of mRNA in both the test groups.

Immunohistochemistry Assay
An immunohistochemistry (IHC) assay method was modified for this study and then applied [42,43,48]. In short, various tissues (heart, testis, eyestalk, nerve cord, muscle, androgenic gland, hepatopancreas, and brain) were set at 4 • C for 24 h in 4% paraformaldehyde in paraffin. At the end, the samples were divided into sections of 4 µm and then baked for 2 h at 60 • C. These sections were deparaffinized after being shifted to slides by applying xylene, after which they were rehydrated. To retrieve antigens, these slides were all submerged in a buffer of EDTA used for antigenic retrieval and then put into the microwave for 15 min. After this, 3% hydrogen peroxide (H 2 O 2 ) in methanol was applied, and the slides were incubated for 1 h with bovine serum albumin (BSA) at room temperature to prevent non-specific bonding. These tissue sections were again incubated in 1% blocking solution (1/200 dilution) with rabbit anti-Mr-IAGBP primary antibody for a whole night at 4 • C. In the morning, three washes with TBST were done, and the glass slides were incubated for 50 min with HRP-conjugated goat anti-rabbit IgG (1/1000 dilution) at standard room temperature. A DAB substrate solution was used to develop them (Guge Biotech, Wuhan, China). The final step was to counterstain the sections with hematoxylin before mounting and photographing them (ECLIPSE E100, Nikon, Tokyo, Japan).

Statistical Analysis
All data were reported in the form of means ± SE (standard error; n = 9). SPSS 22.0 software was used to analyze the statistics, and the statistical significance was given by one-way ANOVA, after which Duncan's multiple range test was done. Significance was set at p < 0.05.

Conclusions
To summarize, we first cloned and characterized the Mr-IAGBP. Multiple sequence alignment and phylogenetic analyses revealed the association and evolutionary relationship of the IAGBP genes with both the vertebrate and invertebrate species. We further demonstrated that the important roles of Mr-IAGBP throughout androgenic gland development and the levels of Mr-IAG transcripts increased earlier (LH20) than those of Mr-IAGBP (PH20). Furthermore, RNA interference revealed that there may be a positive regulatory relationship between Mr-IAG and Mr-IAGBP in M. rosenbergii. Therefore, we suggest that Mr-IAG in combination with Mr-IAGBP may help regulating the development of androgenic gland tissue in M. rosenbergii. Our data support the hypothesis that (IAG and IAGBP)-IAG receptor signaling schemes exist in M. rosenbergii.