Activin receptor type IIB in rohu (Labeo rohita): molecular characterization, tissue distribution and immunohistochemical localization during different stages of gonadal maturation

Activin receptor type IIB (ActRIIB) is a transmembrane serine/threonine kinase receptor which plays a pivotal role in regulating the reproduction in vertebrates including teleost. Earlier studies have documented its importance in governing gonadal maturation in higher vertebrates. However, reports on the regulation of fish reproductive system by ActRIIB gene are still limited. Here, we report the identification and characterization of ActRIIB cDNA of Labeo rohita, a commercially important fish species of the Indian subcontinent. The full-length gene encoding rohu ActRIIB was cloned and found to be of 1674 bp in length. Functional similarities were evident from evolutionary analysis across vertebrates. Real-time PCR to measure the expression of ActRIIB transcript in rohu revealed significant mRNA levels in gonads followed by non-reproductive tissues, including the brain, pituitary and muscle. With respect to different gonadal maturation stages, predominant expression of ActRIIB mRNA was observed during the pre-spawning phase of both sexes. To further delineate its role in rohu reproduction, a recombinant protein of the extracellular domain of ActRIIB (rECD-ActRIIB) was produced, and polyclonal antibody is raised against the protein for its immuno-localization studies during different gonadal maturation stages. Strong immunoreactivity was noticed in the pre-vitellogenic oocytes which decreased dramatically in the fully mature oocytes. Similarly, the strong and intense immunoreactivity was found in the spermatids and spermatocytes of the immature testis, and eventually the intensity reduced with the progression of the maturation stage. These results provide the first evidence of the presence of ActRIIB in rohu gonadal tissues. Taken together, our observations lay the groundwork for further understanding and investigating on the potential role of ActRIIB in fish reproduction system in the event of gonadal maturation.


Introduction
As aquaculture industry aims to boost production of viable eggs with high survival rate, it is vital to bridge the gaps that still remain in understanding the dynamic processes controlling teleost reproduction. The regulation of oocyte maturation and spermatogenesis in fish is a complex process that involves the interaction of several regulatory molecules. Activin receptors (ActRs), members of transforming growth factor-beta (TGF-β) superfamily, with their unique characteristics of wide range ligand specificities are capable of governing several cellular responses in reproduction and development. They are single-pass transmembrane serine/threonine kinase receptors and are basically of two types, type I and type II, having a small extracellular ligand binding domain and an intracellular kinase domain, which act cooperatively to transduce signals for important physiological and molecular processes such as cell proliferation, growth and differentiation of skeletal tissues and gonadal function (Mathews 1994;Mathews and Vale 1991;Patnaik et al. 2017;Goebel et al 2019). Multiple TGF-β ligands like activin, myostatin and follistatin, in their bio-active dimeric forms bind to the extracellular domain of the type II receptors for signal activation. Both the receptors undergo dimerization followed by phosphorylation of Gly-Ser residues in the GS domain preceding the kinase domain for signal transmission (Huang and Chen 2012;Kubiczkova et al. 2012). Among the many ligands of TGF-β family, activin upon interaction with type II receptors plays a key role in promoting follicle growth, primarily regulating ovarian maturation. Two variants of type II receptors such as IIA (ActRIIA) and IIB (ActRIIB) are known (Tsuchida et al. 1993;Yamashita et al. 1994;Dijke et al. 1996) having differences in binding affinity and signalling strength and produced by differential splicing. Experimental findings suggest that ActRIIB has higher affinity for activin than ActRIIA (Attisano et al. 1992). Despite their structural resemblances, no functional overlap or sequence similarity exist between the two receptors.
Till date, ActRs have been characterized in higher vertebrates, and its importance in regulating vertebrate reproduction and development has also been studied. Biochemical and localization studies have demonstrated the expression of ActRII mRNA in bovine and mouse testis (Ethier et al. 1994;Wu et al. 1994) suggesting their putative role in testis maturation. Experimental evidences of the presence of activin receptor and Smad proteins in oocytes of rat might implicate their role in follicular development (Drummond et al. 2002). Moreover, the expression of ActRIIA/IIB has been found in oocytes of other vertebrates regulating both folliculogenesis and differentiation (Van den Hurk and Van de Pavert 2001).
Recently, it has been demonstrated that activin, TGFβ, Nodal and their respective type II and type I receptors are expressed in sex-and cell-type-specific patterns suggesting their specific role in testis and germ cell development (Miles et al. 2013).
Among teleosts, genes encoding ActRIIA cDNA have been cloned and characterized in zebrafish and grass carp (Nagaso et al. 1999;DiMuccio et al. 2005). The significance of ActRIIA in zebra fish reproductive axis is evident from its expression in the oocyte during embryonic developmental stages (Nagaso et al. 1999). Rohu ActRIIA cDNA has been characterized, and its expression during different reproductive stages has been studied (Patnaik et al. 2017). There are ample evidences that have supported specific role of ActRIIB in regulating skeletal muscle development in fishes (Carpio et al. 2009;Funkenstein et al. 2012). Although most of the studies on ActRIIB are associated with its role in regulating body growth, there are also reports suggesting its importance in regulating teleost reproduction and development. Experimental studies provide the proof of ActRIIB expression in zebrafish oocyte at different developmental stages (Garg et al. 1999;Wu et al. 2000). However, much less is known about the expression of ActRIIB in teleost gonad and its importance in ovary and testis maturation. Therefore, molecular characterization of activin receptors in a large number of non-model fish species may be essential to establish molecular mechanisms underlying its role in fish reproduction.
Indian major carp, Labeo rohita (rohu), is a popular table fish widely cultured in monoculture and polyculture systems of India and adjacent countries.
In the present study, we have identified and characterized ActRIIB to determine its role in rohu reproduction. Furthermore, the expression of ActRIIB during different reproductive stages in gonadal tissue has also been detected and quantified. We have also synthesized a recombinant protein for extracellular domain of rohu ActRIIB (rECD-ActRIIB) and have produced polyclonal antibodies in order to investigate the immunohistochemical localization of the receptor gene in the gonadal tissue during the pre-and post-reproductive stages. These findings in the future would help us to comprehend the importance of the receptor gene in regulating fish reproduction, particularly during gonadal maturation and development in a more defined way.

Sample preparation
Rohu carps collected from the farm of Indian Council of Agricultural Research (ICAR)-Central Institute of Freshwater Aquaculture, Bhubaneswar, India, were used for this study. Fishes were anesthetized with 300 mg/L of tricaine methane sulfonate (MS222; Sigma-Aldrich) before the collection of tissue samples. For studying tissue distribution, the testis, ovary, kidney, heart, liver, intestine, brain, pituitary and muscle were dissected. For expression studies during different reproductive stages, tissue samples were collected from pre-spawning (March-May), spawning (June-August) and post-spawning (September-October) phases of rohu. All the samples were snap-frozen in liquid nitrogen before storage and RNA extraction.
RNA extraction and cloning of ActRIIB cDNA Total RNA extraction and cDNA synthesis were performed as described earlier (Patnaik et al. 2017). In brief, RNA was extracted using TRIzoL reagent from approximately 100 mg to each tissue. Subsequently, DNA was removed by DNAse I treatment followed by column purification. Reverse transcription was performed taking ~ 5 µg of the total RNA from testis tissue, using SuperScript® III cDNA synthesis kit (Invitrogen, USA) as per the manufacturer's instructions. Partial cDNA sequence of rohu ActRIIB was amplified using first-strand cDNA as template. The PCR primers used for this purpose (Supplementary Table 1) were designed from conserved regions of the genes of related species available in the NCBI database (http:// www. ncbi. nlm. nih. gov). The PCR amplification was carried out in 25 µl of reaction volume with the following conditions: initial denaturation at 95 °C for 5 min followed by 30 cycles with 95 °C for 30 s, 58 °C for 30 s, 72 °C for 1 min and finally 72 °C for 7 min. The amplified products were gel purified and cloned into pGEM-T Easy Vector (Promega, USA) and sequenced in ABI 3730 XL automated DNA analyser (Perkin-Elmer Applied Bio system) using SP6 and T7 primers. The verification of partial ActRIIB sequences was done using NCBI-BLAST service.
Gene-specific primers (Supplementary Table 1) were designed from the rohu partial ActRIIB sequence obtained above, and full-length cDNA of rohu ActRIIB was generated by rapid amplification of cDNA ends (RACE) with different primer sets using SMARTerTM RACE cDNA Amplification Kit (Clontech, USA) as described earlier (Patnaik et al. 2017). In brief, touchdown PCR was carried out with different sets of 3' and 5' primers as follows: one cycle of initial denaturation at 94 °C for 2 min followed by five cycles at 94 °C for 30 s, 64 °C for 3 min, five cycles of 94 °C for 30 s, 62 °C for 30 s, 72 °C for 3 min and another 27 cycles at 94 °C for 30 s, 60 °C for 30 s and 72 °C for 3 min. Amplified products were cloned and sequenced as mentioned above.
Quantitative real-time PCR Gonadal tissue expression pattern during different reproductive phases and tissue distribution study were carried out by quantitative real-time PCR (qRT-PCR) as described earlier (Patnaik et al. 2017) using Light Cycler @ 480 II real-time PCR detection system (Roche, Germany). Reactions for expression studies were performed in triplicates using SYBR Green Real-time Master Mix II (Roche Diagnostics, Germany) at an annealing temperature of 58 °C. Possibilities of DNA contamination were excluded by taking negative control in each experiment, and determination of PCR efficiency was carried out by serial dilution. The house keeping gene β-actin was taken as a positive calibrator for normalization of target gene expression. The PCR primers used for gene expression study are presented in Supplementary Table 1. Melting curve analysis was performed to confirm primer specificity, and agarose gel electrophoresis was carried out to check size of transcript.

Expression of Recombinant Extracellular Domain of rohu ActRIIB (rECD-ActRIIB)
The recombinant protein for the extracellular domain was produced using pET28a expression vector. Rosetta DE3 host cells harbouring the recombinant plasmid, pET28rECD-ActRIIB, was transformed and positive colony (kanamycin and chloramphenicol resistant) grown overnight in 3-ml culture tube containing Luria Bertani (LB) broth following the manufacturer's standardized protocol. Thirty ml seed culture was grown by inoculating LB broth with 1% of overnight culture and kept in shaker at 37 °C. The overnight grown seed culture was inoculated in 1 l LB. When the culture attained an absorbance of 0.6 at 600 nm, the cells were induced with 1 mM IPTG for a minimum period of 4 h at 37 °C. Induced culture cells were harvested by centrifugation. The pellet obtained was washed in 20 ml of wash buffer (Tris, 50 mM, NaCl, 300 mM, pH. 7.4, stock concentration, 100 mg/ml). After washing, 0.5 mg/ml lysozyme (stock 100 mg/ml) was added and incubated at room temperature in a shaker at 37 °C for 30 min followed by sonication for 10 min. The pellet obtained after centrifugation was again suspended in wash buffer and again incubated in shaker for another 30 min. The suspension was then centrifuged, and the clear supernatant was collected following protein purification.
Purification of recombinant protein (rECD-ActRIIB) One ml Ni-NTA resin was packed in a Bio-Rad column and washed by passing 10 cv de-ionized water. The column was equilibrated by passing 20 cv equilibration buffer (50 mM Tris, pH 7.4, 8 M urea, 300 mM NaCl). The inclusion body (prepared as per the standard protocol) was solubilized in 50 mM Tris, pH 7.4, 8 M urea, 300 mM NaCl containing 10 mM imidazole. Sample was centrifuged, and the supernatant obtained was loaded onto Ni-NTA resin and incubated for 1 h. After incubation, the supernatant was allowed to pass, and the column was washed by passing 10 cv equilibration buffer with 10 mM imidazole. On column, refolding was done by passing gradient from 8 M urea to 2 M urea in 50 mM Tris, pH 7.4, 300 mM NaCl (200 mL). Final wash was given with 10 cv 2 M urea, 50 mM Tris, pH 7.4, 300 mM NaCl and 25 and 50 mM imidazole. The protein was eluted with 10 cv elution buffer: 2 M urea, 50 mM Tris, pH 7.4, 150 mM NaCl, 5% glycerol and 500 mM imidazole. The eluted fractions were analysed in SDS-PAGE, and fraction containing pure proteins were pooled and dialysed against buffer: 2 M urea, 50 mM Tris, pH 8, 150 mM NaCl and 10% glycerol. Protein was concentrated by using 3 kDa MWCO Centricon (Millipore) at 4000 rpm. The protein was kept in small aliquots at − 80 °C for further use.

SDS-PAGE analysis
SDS-PAGE was carried out according to Laemmli (1970). Samples were mixed with loading buffer in the presence of 5% β-mercaptoethanol and separated in 12.5% mini gels. Protein bands were visualized by Coomassie Brilliant blue staining.

Polyclonal antibody production and validation
Polyclonal Ab against the rECD-ActRIIB peptide of rohu was raised in two rabbits, A and B, for immunohistochemical localization study in gonadal tissue of rohu during pre-and post-reproductive stages. The Ab raised was further validated by western blot and Indirect ELISA.

Western blot analysis
For antibody validation, cell lysate electrophoresed on 12% SDS-PAGE was transferred onto the nitrocellulose membrane in Western transfer buffer (25 mM Tris-Cl pH 8.3, 192 mM glycine and 10% methanol) using Bio-Rad Trans blot apparatus following Towbin et al. (1979). Pre-immune sera and third immune sera of both rabbits A and B were tested on recombinant protein (20 ng) at 1:50,000 dilutions. Five per cent blocking solution skim milk powder in 1X TBST (pH, 8.0) and 1% SKM in TBST was prepared freshly. Antibodies were diluted in 1% blocking solution (~ 2.3 µl in 5 ml blocking solution) and shaken well for at least 15 min. The strips were soaked in methanol for 2 min. The strips were washed twice with 1X TBST for 5 min. The strips were blocked in 5% blocking solution for 15 min while shaking. The strips were washed 2 to 3 times in 1X TBST for 5-10 min. The strips were transferred on the antibody solutions for primary Ab incubation while shaking for 2 h at RT and then at 4 °C. The next day, strips were washed once in 1X TBST for 30 min, and strip blocks were made. The blocks were washed in 1X TBST for 30-45 min with twice change of TBST. TBST was discarded, and secondary Ab (HRP conjugate anti-rabbit IgG) (1:5000 dilution in 1% blocking solution) was added and incubated for exactly 1 h at RT. The blocks were washed in 1X TBST for 2-3 h with 3 to 4 changes to reduce background staining. Chemiluminescence's solution (SuperSignal West Pico™ Chemiluminescent, Pierce), 1:1 solution A and solution B, was prepared and kept for 5 min and was mixed well and kept for 5 min. The strip blocks were soaked well in the solution for 5 min under constant shaking. The strip blocks were arranged on a Hyper film with fluorescent marks on the corner, and transparent film was placed over it. Three exposures at 15 s, 1.5 min and 3 min were taken for observation of the band development pattern. The film was taken out and kept in developer solution and allowed to develop followed by washing in tap water and kept in fixer solution. After 5 to 10 min, the film was taken out and then washed in running water, and the film was dried. Finally, the developed film along with the strips was aligned on the fluorescent paper, and individual lanes and marker were marked.

Indirect ELISA analysis
Indirect ELISA was performed from the immune sera obtained from the immunized rabbits A and B to validate the polyclonal Ab produced. First and third immune sera of rabbit A were tested against the synthesized antigen at 200 ng, at 1:5000 dilution of primary Ab. Similarly, first and third immune sera of rabbit B were tested against synthesized antigen at 200 ng, at 1:5000 dilution of primary Ab. Preimmune sera were used as control in place of primary Ab in both rabbits A and B. Plates read were taken after 15 min of enzyme substrate reaction.

Immunohistochemistry
Rohu testis and ovary collected from immature pre-reproductive and mature reproductive stages were cut into small sections and preserved in 10% NBF (neutral buffer formalin) for 18-24 h. Paraffin wax-embedded tissue blocks made were cut into serial sections in LEICA RM2125 microtome to the desired thickness of 3 microns and mounted to microscopic slides treated with Chicken/Mayer's albumin and kept in room temperature for overnight. The slides were then deparaffinized and rehydrated by routine protocol. The slides were suspended in 3% H 2 O 2 , diluted in methanol and kept in humid chamber for 15 min. Heat-induced antigen retrieval was performed using the Tris-EDTA Buffer (pH 9.0) in a microwave oven for 3 min. The slides were washed three times 3 min each with 1X TBST buffer and kept in slide jar with warm Tris-EDTA buffer for antigen retrieval for 15 min at room temperature followed by again washing. The slides were blocked in boxes by Pap pen marker and then suspended in Sniper (Biocare Medical, USA) for 15 min in humid chamber in order to block and minimize background staining.
The slides were washed in 1X TBST buffer, and the sections were suspended with primary antibody (polyclonal Ab raised against rECD-ActRIIB) in three dilution ratios, i.e. 1:500, 1:750 and 1:1500 (dilution in 1X PBS) and incubated for 90 min in a humidified chamber. The slides were rinsed with 1X TBST and then suspended in HRP-conjugated secondary antibody (UltraTek Anti-polyvalent Biotinylated Ab) for 10 min in humidified chamber followed by washing in 1X TBST buffer. Three hundred microliter substrate buffer (Betazoid DAB chromogen) was applied for 10 min to stain the nuclei of the cells in the sections. Then haematoxylin was added, and the slides were washed under tap water followed by rehydration in xylene for 20-30 min. Slides were mounted with DPX, and the sections were observed for immunostaining using the LEICA light microscope with digital camera. As control, primary antibodies were omitted in a few test sections in each experiment.

Data analysis
Gene expression data obtained from qRT-PCR study in triplicates, along with β-actin, was used to normalize any differences in reverse transcriptase efficiency. Relative gene expression was calculated using Pfaffl method (Pfaffl 2001) based on the threshold cycle (C t ) value. All data are expressed as mean ± SEM and analysed by ANOVA followed by Duncan multiplerange test using SPSS18.0 software.

Molecular cloning and sequence analysis of rohu ActRIIB
Rohu ActRIIB partial cDNA (LrActRIIB) was synthesized from testis RNA using normal PCR that generated a sequence of 939 bp. The complete CDS of LrActRIIB was obtained by RACE PCR strategy that produced a sequence of 1674 bp in length (KX710215) with 27 bp of 5' UTR and 114 bp of 3'UTR (Fig. 1) Fig. 2) revealed that the gene is organized into three major domains, typical of TGF-β superfamily, including a signal peptide (SP, 1-22aa), a short extracellular domain (ECD, 27-117aa), a hydrophobic transmembrane domain (TMD, 136-158aa, Supplementary  Fig. 3) and a large intracellular serine/threonine kinase domain (ICD, 188-476aa). Disulphide bonds formed by the 10 conserved cysteine residues present in rohu ActRIIB govern the folding of the ligand Fig. 1 Sequence information of Labeo rohita ActRIIB. The full-length nucleotide and deduced amino acid sequences of the rohu ActRIIB with numbers are represented on the right side. The start and stop codons are in bold letters. Ten conserved cysteine in extracellular domain are inside rectangular boxes. The two characteristic Ser/Thr kinase motifs are shaded in black. "D319" in bold letters is predicted to be the Ser/Thr kinase active site. "K215" as the ATP binding site is shown in circle. The amino acid sequence shaded in grey and underlined is predicted to be the nucleotide binding site. The PDZ protein binding motif is in the rectangular box binding domain (Fig. 1) which is in agreement with earlier reports (Greenwald et al. 1999;Lin et al. 2006;Walton et al. 2012). The characteristic HRDFKSKN (317-324aa) and GTRRYM (358-363aa) Ser/Thr kinase motifs present in the ICD region and D 319 , respectively, are predicted to be the Ser/Thr kinase active sites. The characteristic ESSI motif essential for mediating functions between ActRIIs and PDZ-domain of activin receptor-interacting proteins (ARIPS) (Tsuchida et al. 2004) is present at the C-terminal end of LrActRIIB. Further, it is predicted that  and K 215 are nucleotide binding site and ATP binding site, respectively. Two potential N-glycosylation sites and potential posttranslational modification (PTM) sites are also found to be present in the putative transmembrane structure of the rohu ActRIIB. It is predicted to have 17 serine phosphorylation sites, 7 threonine phosphorylation sites and 4 tyrosine phosphorylation sites. Secondary structure analysis using SOPMA showed that LrActRIIB holds 43.33% alpha helix, 36.67% random coil, 13.98% extended strands and 6.02% beta turns ( Supplementary Fig. 4). The results of the Ramachandran plot analysis of modelled 3D structure of kinase domain of LrActRIIB showed 92.12% in favourable region and 100% of the residues in the allowed region ( Supplementary Figs. 5 and 6).

Phylogenetic analysis
The evolutionary relationship of rohu gene with other vertebrates was evaluated by constructing a neighbour-joining (NJ) tree using the protein sequences available in the public database. The tree showed two distinct clades, with teleost ActRIIBs forming one major cluster whereas higher vertebrates, i.e. mammalian ActRIIBs forming another major cluster as illustrated in Fig. 2. The teleostean clade was further sub-divided according to their habitat and family. Percentage of similarity and identity of rohu ActRIIB amino acid sequences with other species are presented in Table 1. The highest percentage of similarity and identity (97.45% and 95.88% respectively) were observed between rohu and Ctenopharyngodon idella demonstrating their high degree of relatedness.
Quantitative expression of ActRIIA mRNA during different reproductive and growth stages

Tissue distribution of ActRIIB mRNA
qRT-PCR analysis revealed the expression of ActRIIB transcript in different tissues of juvenile rohu. The expression of ActRIIB mRNA was observed in all tissues studied. The highest expression was found in the gonad and then the brain, pituitary, liver and muscle.
In contrast, the expression in the intestine, heart and kidney is exceptionally low (Fig. 3A, P < 0.05).

Expression profiling of LrActRIIB mRNA during different reproductive stages of rohu
Variation in expression pattern during pre-spawning, spawning and post-spawning stages was carried out. During pre-spawning phase, the expression of LrAc-tRIIB transcript in ovary is observed to be highest. Similarly, in testis during pre-spawning phase, LrAc-tRIIB expression was significantly high followed by a remarkable decline during spawning and post-spawning stages. The expression profile of LrActRIIB in the Fig. 2 Phylogenetic relationship of Labeo rohita ActRIIB. The evolutionary relationship with other ActRIIB protein counterparts was analysed with Mega 6.06 program by bootstrap analysis using neighbour-joining with 1000 replicates gonadal tissue during different reproductive stages of rohu (P < 0.05) is presented in Fig. 3B.

Expression and purification of recombinant extracellular domain of LrActRIIB protein (rECD-ActRIIB)
The recombinant protein, i.e. extracellular domain of LrActRIIB of 113aa, was synthesized and expressed in pET28a vector. The theoretical molecular weight and pI were estimated to be 14.04 kDa and 4.73 kDa, respectively. For the screening of the expressed protein, the supernatant and the pellet isolated from induced cells were run on SDS-PAGE, and the protein of approximately similar molecular weight to the estimated value was found in the supernatant as shown in Fig. 4A. The isolated protein was purified on Ni-NTA column, and column refolding was done by passing gradient from 8 M urea to 2 M urea in 50 mM Tris, pH 7.4 and 300 mM NaCl. After the final wash, the protein was eluted with 10 cv elution buffer, 2 M urea, 50 mM Tris, pH 7.4, 150 mM NaCl, 5% glycerol, 500 mM imidazole and fractions containing pure protein were pooled and dialysed against buffer, 2 M urea, 50 mM Tris, pH 8, 150 mM NaCl and 10% glycerol as shown in Fig. 4B.

Production and validation of polyclonal antibody against rECD-ActRIIB peptide
Polyclonal Ab against the rECD-ActRIIB peptide was produced for immunohistochemical study in gonadal tissue of rohu during pre-spawning and spawning stages. Indirect ELISA was performed from the immune sera obtained from the immunized rabbits A and B to validate the polyclonal Ab produced. The first immune sera of rabbit A tested against the synthesized antigen at 200 ng, at 1:5000 dilution of primary Ab, obtained a value of 1.176 OD at 450 nm. The third immune sera of rabbit A, also tested against synthesized antigen at 200 ng, at 1:5000 dilution of primary Ab, obtained a value of 1.215 OD at 450 nm. Similarly, the first immune sera of rabbit B, tested against synthesized antigen at 200 ng, at 1:5000 dilution of primary Ab, obtained a value of 1.056 OD at 450 nm. The third immune sera of rabbit B, tested against synthesized antigen at 200 ng, at 1:5000 dilution of primary Ab, obtained a value of 1.174 OD at 450 nm as shown in Fig. 5A. The third immune sera of rabbit A were purified and validated by indirect ELISA. Protein A purified Ab (200 ng), tested against 200 ng of antigen, obtained a value of 2.572 OD at 450 nm as shown in Fig. 5B. Pre-immune sera were used as control in place of primary Ab at 1:5000 dilutions. The Ab raised was further validated by western blot analyses of the immune sera derived from the two immunized rabbit, resulting in a single band in the blot with an apparent molecular weight slightly higher than that of the calculated weight of ~ 14 kDa as shown in Fig. 5C. Besides western blot, the purified Ab was further used for immunohistochemical localization study in rohu gonadal tissue during reproductive stages.

Immunohistochemical localization study in gonadal tissue of rohu during pre-spawning and spawning stages
In the pre-vitellogenic ovary, the cytoplasm of oogonia and the primary oocytes displayed strong immunoreactivity of ECD-ActRIIB (Fig. 6A, i-ii). As ovarian maturity progressed, follicle cells (granulosa) surrounding the oocyte showed weak immunoreactivity. Also, specific immunostaining in the follicular cell layer of vitellogenic oocytes and the nuclei (germinal vesicles) were completely negative with no staining (Fig. 6A, iv-vi). In addition, the cytoplasm of oocytes that started to accumulate yolk inclusions gradually lost its immunoreactivity, with the reaction intensity being inversely proportional to the amount of yolk globules in the oocyte. The oocytoplasm of mature oocyte was weak for ActRIIB, whereas immunostain was found in zona pellucida region and follicular cell layer. The ECD-ActRIIB Ab immunoreactivity in the pre-vitellogenic stage was stronger, and the ActRIIB (1, pre-immune sera of rabbit A at 1:50,000 dilution, 2-3rd immune sera of rabbit at A 1:50,000 dilution; 3, preimmune sera of rabbit B at 1:50,000 dilution; 4-3rd immune sera of rabbit B at 1:50 000 dilution) immunostain reduced with maturation of oocyte. In immature male rohu during spermatogenesis in pre-spawning stage, the testis was filled with primary and secondary spermatocyte and spermatids, showing strong immunoreactivity to ECD-ActRIIB Ab. However, during spawning stage, the testis was filled with spermatozoa and was in the stage of spermiating. The immunoreactivity reduced with maturation, and the germ cells in the stage of spermiation showed weak immunostain than prespawning stage as noticed in a similar pattern in the oocytes (Fig. 6B).

Discussion
In the present study, ActRIIB of rohu has been cloned and characterized. ActRIIB cDNA of rohu consists of 510 aa with 89% to 95% similarity to its vertebrate counterparts. The characterized protein in rohu possesses the typical domain architecture of TGF-β type II receptors with conserved cysteine residues in the activin receptor-specific extracellular domain essential for ligand binding (Kingsley 1994;Josso and di Clemente 1997;Shav-Tal et al. 2001). Moreover, Ser/Thr kinase motifs in ICD appeared to be well conserved. The protein sequence, "KARGRFGCV", for nucleotide binding is conserved for type II receptors of TGF-β superfamily. This is striking to observe such high level of structural conservation through evolution, suggesting similar functional role of the gene in fishes including rohu. The PDZ motif essential for receptor binding to PDZ protein for interaction with ARIPS for receptor internalization is conserved for type II receptors. The COOH-terminal sequence of rohu ActRIIA is E-S-S-L (Patnaik et al. 2017), and the COOH-terminal sequence of LrActRIIB is E-S-S-I, both of which are concordant with a class I consensus PDZ binding motif XSX (V/I/L) (where X is any amino acid). Among the various receptor, ActRIIs from different species contain a PDZ binding motif at their COOH-terminus (Shoji et al. 2000;Matsuzaki et al. 2002;Tsuchida et al. 2004). The structural dissimilarity might suggest difference in their receptor functioning upon different ligand interaction. The predicted kinase domain of rohu ActRIIB is of good quality as revealed from the Ramachandran plot analysis and is comparable to the predicted structure of kinase domain of rohu ActRIIA as reported earlier (Patnaik et al. 2017). Phylogenetic analysis showed that the receptor gene is evolutionarily conserved across the vertebrate lineage, signifying that it might share common biological functions with its other counterparts. LrActRIIB formed a major cluster with teleosts depending on their habitat. High degree of similarity/identity value observed between rohu and its counterparts may suggest structural and functional similarity among them. Functional involvement of ActRIIB in diverse biological processes was evident form ubiquitously expression in reproductive as well as non-reproductive tissues. However, high expression level of the gene in the gonad, brain, pituitary, muscle and liver might implicate its potential role in governing physiological traits like reproduction and body growth. Recently, we have also examined the expression of ActRIIA gene in rohu, demonstrating its importance in regulating gonadal maturation and skeletal muscle development (Patnaik et al. 2017). The results of the present study are also in agreement with previous reports in other vertebrates including teleosts (Carpio et al. 2009;Otto and Patel 2010;Østbye et al. 2007). A comparative expression profiling of the transcript has been done in the gonadal tissue of rohu during the reproductive cycle, i.e. pre-spawning, spawning and post-spawning stages. In agreement with previous studies, high level of expression of rohu ActRIIB mRNA in gonad during pre-spawning stage demonstrated its significant role in stimulating oocyte maturation and spermatogenesis (Wu et al. 2000; Van den Hurk and Van de Pavert 2001;Pangas et al. 2002;Silva et al. 2004). However, decreasing LrActRIIB transcript expression towards the post-spawning stage might suggest the effect of feedback inhibition of follicular activin signalling by antagonists like inhibin to decrease FSH release from pituitary as described in earlier reports (Yuen and Ge 2004;Cheng et al. 2007).
Besides qRT-PCR studies showing high expression levels of the receptor gene in gonadal tissue, we also determined the spatial distribution of the immunoreactive ECD-ActRIIB peptide in the ovary and testis of rohu by immunohistochemical study. In the pre-vitellogenic and vitellogenic ovary of rohu, the cytoplasm of oogonia and primary oocytes exhibited strong immunoreactivity of ActRIIB peptide. However, the immunoreactivity significantly decreased and was eventually found to be localized in the follicular cell layer and zona pellucid region when the oocytes became fully mature. This result signifies its role in regulating oocyte maturation as reported in earlier studies (Wu et al. 2000;Cheng et al. 2007). The decrease in intensity of immunoreactivity suggests that ActRIIB mRNA expression may decrease with the development of oocytes. Alternatively, this could be due to the dilution of a constant amount of the mRNA as oocytes become larger. This result is consistent with the reports of previous studies in zebrafish (Garg et al. 1999) as well as in mammals (Cameron et al. 1994). In the localization studies of ECD-ActRIIB peptide in male rohu, strong and intense immunoreactivity was found in spermatids and spermatocytes of immature testis. However, in mature rohu, immunosignals were particularly evident towards the periphery of the testis but with reduced immunostaining centrally. This result is in accordance with previous studies as reported in mammals (Anderson et al. 2002), suggesting its role in regulating spermatogenesis in rohu. In correlation with the present localization study of ECD-ActRIIB peptide in rohu, ActRII and ActRIIB mRNA have also been detected in bovine and rat (Cameron et al. 1994;Ethier et al. 1994;Drummond et al. 2002). Thus, we can conclude that the receptor gene might possess potential role in regulating reproductive physiology of rohu.

Conclusion
The present study reports for the first time the characterization of ActRIIB gene in rohu and its spatial and stage-specific expression during different reproductive phases. Furthermore, recombinant protein, LrActRIIB-ECD, found to be localized in the gonadal tissues of rohu provides the evidence of its potential in promoting oocyte maturation and spermatogenesis. Thus, the information obtained from this study will be of interest for further research towards understanding great details of ActRIIB involvement in teleost reproductive system.