Steroids 92 (2014) 81–89

Contents lists available at ScienceDirect

Steroids

journal homepage: www.elsevier .com/locate /s teroids

Molecular cloning and expression analysis of 17b-hydroxysteroiddehydrogenase 1 and 12 during gonadal development, recrudescenceand after in vivo hCG induction in catfish, Clarias batrachus

http://dx.doi.org/10.1016/j.steroids.2014.09.0090039-128X/� 2014 Elsevier Inc. All rights reserved.

⇑ Corresponding author. Tel.: +91 40 23134562/4662; fax: +91 40 23010307/23010120.

E-mail addresses: bsksl@uohyd.ernet.in, senthilkumaranb@yahoo.com(B. Senthilkumaran).

Anbazhagan Rajakumar, Balasubramanian Senthilkumaran ⇑Laboratory of Molecular Endocrinology and Reproductive Biology, Department of Animal Biology, School of Life Sciences, University of Hyderabad, P.O. Central University,Hyderabad 500046, India

a r t i c l e i n f o a b s t r a c t

Article history:Received 20 June 2014Received in revised form 28 August 2014Accepted 23 September 2014Available online 16 October 2014

Keywords:17b-HSDGonadal ontogenyTeleostsReproductive cyclehCG inductionT and E2 levels

In teleosts, the levels of steroids during critical period of sex differentiation are critical for gonadogenesis.Hence, steroidogenesis and expression of steroidogenic enzyme genes are very critical for gonadal devel-opment and function. In this regard, 17b-HSDs are important as they are involved in both 17b-estradiol(E2) and testosterone (T) biosynthesis. Full length cDNAs of 17b-HSD 1 (1791 bp) and 12 (1073 bp) werecloned from catfish gonads which encodes a protein of 295 and 317 amino acids, respectively. To under-stand the importance of these enzymes in teleost reproduction, mRNA expression was analyzed duringgonadal development, seasonal reproductive cycle and after human chorionic gonadotropin (hCG) induc-tion. Phylogenetic analysis revealed that the 17b-HSD 1 and 12 share high homology with their respective17b-HSD forms from other teleosts and both the 17b-HSD forms belong to short chain dehydrogenase/reductase family. Tissue distribution analysis showed that the 17b-HSD 1 expression was higher in ovaryand gills, while 17b-HSD 12 was higher expressed in testis, ovary, brain, intestine and head kidney com-pared to other tissues analyzed. Developing and mature ovary showed higher expression of 17b-HSD 1,while 17b-HSD 12 was higher in testis than the ovary of corresponding stages. Further, 17b-HSD 1 and12 transcripts together with E2 and T levels were found to be modulated during different phases of theseasonal reproductive cycle. Expression of 17b-HSD 1 and 12 was upregulated after hCG induction whichshows possible regulation by gonadotropin. Our findings suggest that 17b-HSD 1 and 12 might playimportant role in regulating gonadal development and gametogenesis through modulation of sex steroidlevels.

� 2014 Elsevier Inc. All rights reserved.

1. Introduction

17b-hydroxysteroid dehydrogenases (17b-HSDs) are multi-functional enzymes which catalyze the interconversion between17b-hydroxy- (active) and 17-keto- (inactive) steroids and therebyregulate the specific levels of substrates for sex steroid biosynthe-sis [1,2]. Most 17b-HSDs belong to short chain dehydrogenase/reductase (SDR) and aldo–ketoreductase (AKR) protein superfami-lies [3,4] which has been isolated and characterized in manymammals and few teleosts. 17b-HSDs share SDR conserved motifs,which consist of the cofactor binding domain (TGxxxGxG), the tet-rad of active site (N-S-Y-K), the structural stabilization domainNNAG, and the PGxxxT domain [4–7]. Although most 17b-HSDs

share high homology in sequence and structure, they differ in sub-strate specificity, catalytic coenzyme preference, reaction directionand tissue distribution [8,9,2].

17b-HSD 1 is a key enzyme catalyzing the final step of the syn-thesis of 17b-estradiol (E2). In most vertebrates, the most promi-nent site of expression is the ovary, but in zebrafish it can alsobe found at low levels in other tissues like skin, muscle and eye[10]. 17b-HSD 1 is predominantly involved in interconversion ofestrone (E1) to E2 in rat, mice and teleosts [11]. Interestingly inthe Nile tilapia, 17b-HSD 1 can also catalyze the conversionbetween androstenedione and testosterone (T) but less efficiently[11]. 17b-HSD 12 is the most recent addition to this enzyme familyand subsequent phylogenetic analysis revealed its close relation-ship to 17b-HSD 3 and was suggested that 17b-HSD 12 would bean ancestor of 17b-HSD 3 [12]. Zebrafish contains two paralogouscopies of 17b-HSD 12 (A and B) and both are expressed widely inseveral tissues of both developing and adult fish [10,12].

82 A. Rajakumar, B. Senthilkumaran / Steroids 92 (2014) 81–89

17b-HSD 12 has been shown to be involved in the biosynthesis ofE2 and the elongation of the essential very long fatty acids [13].17b-HSD 12 was first characterized as a 3-ketoacyl-CoA reductasewhich is specifically involved in long and very long-chain fatty acidbiosynthesis [14] in mammals.

17b-HSDs are required for the production of E2 and T and thusindirectly involved in the process of sex differentiation and game-togenesis in fish [12,15,16]. It is well known that sex steroids playimportant roles in the process of sex differentiation and gameto-genesis of non-mammalian vertebrates like teleosts [17]. Hence,the expression and regulation of steroidogenic enzyme genes arecritical for steroidogenesis as well as gonadal differentiation anddevelopment. Further, the analysis of 17b-HSDs, in an annuallyreproducing teleost was never attempted. Such an attempt mightprovide important clues to understand gonadal developmenttogether with recrudescence.

Teleosts having seasonal reproductive cycle are good models forstudying gonadal regulation, as gonadal maturation and regressionoccurs seasonally. On this perspective, the present study used anannually reproducing catfish, Clarias batrachus as a teleost model.Catfish takes around a year to mature and undergo seasonalreproduction. The reproductive cycle of C. batrachus comprises ofpreparatory (February–April), pre-spawning (May–June), spawn-ing (July–August) and post-spawning/resting (September–January)phases by comparing pituitary gonadal relationship [18]. To under-stand the possible involvement of steroids and the expression ofsteroidogenic enzyme genes on gonadal development togetherwith recrudescence, the present study was conducted. Full lengthcDNA encoding 17b-HSD 1 and 12 were cloned and expressionwas analyzed during ontogeny, seasonal cycle and after humanchorionic gonadotropin (hCG) induction. In addition, the levels ofT in serum and testicular tissue and E2 in serum during differentstages of reproductive cycle were measured.

Table 1List of primers used for cDNA cloning and qPCR analysis.

S.no. Primer name Primer sequence 50 to 30 Usage

1. 17b-HSD1 DF1 TSATYACCGGYTGCTCCTCKGGMAT RT-PCR2. 17b-HSD1 DR1 GTGTKSACYGSSCCRCACT RT-PCR3. 17b-HSD1 5P ACACCAGGATGTCCACTCGGCCCTCG RACE4. 17b-HSD1 5N AGGTCCCTCACACTCTCCAGCAAGCGC RACE5. 17b-HSD1 3P GTCACAGGCAGTATGGGAGGGCTACAAGG RACE6. 17b-HSD1 3N AGGGTGTCTGTGAGAGTCTGGCCAT RACE7. 17b-HSD1 ORFF ATGGAGTGGAAGGTAGTTCTC ORF cloning8. 17b-HSD1 ORFR TTATTTCTGTGCTCTTATTTC ORF cloning9. 17b-HSD1 RTF GACATCCTGGTGTGTAATGCAGG qPCR

10. 17b-HSD1 RTR CTGCCTGTGACCAGGATCCGT qPCR11. 17b-HSD12DF12 GTTGTBCANGGDRSYCANKMYGG RT-PCR12. 17b-

HSD12DR12GAGAAGAADKYYACAAAVGYYTT RT-PCR

13. 17b-HSD12 5P CGATGGCAAAGCCACGCCGTGCAAGC RACE14. 17b-HSD12 5N TCCGTAAGTAGCCCCCGTCACAAC RACE15. 17b-HSD12 3P GATGACTCGGCTTGTGCTGCCTAAGATGG RACE16. 17b-HSD12 3N AGCGGCATGTACCCTGTTCCACTC RACE17. 17b-HSD12

ORFFATGGAGGCGGCAGATATTTTG ORF cloning

18. 17b-HSD12ORFR

TTAGCCCTGCTTTTGCTTCT ORF cloning

19. 17b-HSD12 RTF AGCCATCGAGAGCAAGTACCATGT qPCR20. 17b-HSD12 RTR AAGCCGAGTCATCTGACAAACCGA qPCR21. 18S rRNA F GCTACCACATCCAAGGAAGGCAGC qPCR22. 18S rRNA R CGGCTGCTGGCACCAGACTTG qPCR

IUPAC nucleic acid codes: R – A or G; Y – C or T; S – G or C; W – A or T; K – G or T; M– A or C; B – C or G or T; D – A or G or T; H – A or C or T; V – A or C or G; N – any base.

2. Materials and methods

2.1. Animals

In vitro fertilization was carried out during the breeding season(July–August) of catfish and the obtained catfish hatchlings werereared till adulthood under ambient photothermal conditions asdescribed previously [19]. Sampling was done at different timepoints from 0, 10, 20, 30, 40, 50, 100, 150, 200 and 250 days posthatch (dph) and adult (2 year) catfish. For 0 dph whole trunk wasused for total RNA preparation, while for 10, 20, 30 and 40 dphmesonephric gonadal complex was used. During gonadal develop-ment, the sex of the gonads was determined by histological analy-sis. Testes appear like a slender thread while ovary appears as atransparent pouch devoid of sperm and oocytes, respectively.Histological examination of gonads showed that the gonadal differ-entiation into testis and ovary occurs around 45–50 dph. Hencefrom 50 dph onwards to adult, testis and ovary were dissectedfor total RNA preparation. Adult C. batrachus reared for two yearin the outdoor tanks before it was sacrificed for experiments onthe seasonal cycle and hCG (Pubergen, Uni-Sankyo Pvt., Ltd., India)induction. Prior to sacrifice, blood was collected from differentstages of catfishes by caudal puncture and the collected bloodwas centrifuged at 2,500 g and 20 �C for 20 min for serum separa-tion. For tissue distribution analysis, different tissues from bothmale and female catfishes (gonads, brain, liver, head kidney, gills,intestine, muscle, heart and spleen) were collected during pre-spawning phase of catfish reproductive cycle and all the catfisheswere sacrificed by anesthetizing with 100 mg/L of MS 222 (Sigma,St. Louis, MO, USA) in mild ice-cold water by following the guide-lines of institutional animal ethics committee.

2.2. Molecular cloning of full length 17b-HSD cDNAs from catfishgonads

Total RNA was prepared from catfish gonads using SigmaTRI-reagent. Quality and quantity assessment were made by usinga NanoDrop spectrophotometer (ND-2000, NanoDrop Technologiesand Wilmington, Delaware, USA) and cDNA synthesis was carriedout using 5 lg total RNA, oligo (dT) and SuperScript� III accordingto the manufacturer’s instructions (Invitrogen, Carlsbad, CA, USA).The cDNA quality was checked by amplifying b-actin. A set ofdegenerate primers were designed individually for 17b-HSD 1and 12 based on the available nucleotide sequences of both 17b-HSD 1 and 12 of teleosts from the NCBI GenBank database. PCRamplification was performed at 94 �C (1 min) and 35 cycles of94 �C (30 s), 53 �C (30 s), 72 �C (1 min) and 72 �C (10 min) usingdegenerate primers (Table 1) of either 17b-HSD 1 (DF1 and DR1)or 17b-HSD 12 (DF12 and DR12). The partial cDNA fragmentsobtained were cloned into pGEM�-T easy vector (Promega,Madison, WI, USA) and sequenced bi-directionally using dye termi-nator cycle sequencing method in an ABI 3730 DNA analyzer(Applied Biosystems, Foster City, CA, USA). RACE strategy was per-formed to obtain the full length 50 and 30 cDNA of both 17b-HSD 1and 12 using the gene-specific primers (GSP) designed based on thesequence information of the respective partial cDNA of 17b-HSD 1and 12. The 50 and 30 RACE cDNA templates were prepared usingSMARTer™ RACE cDNA amplification kit (Clontech, MountainView, CA, USA) according to the manufacturer’s protocol. Touch-down PCR reactions were performed to obtain 50 and 30 ends usingthe advantage polymerase kit (Clontech) and gene specific primers(Table 1) as well as using anchor primers, universal primer A mixand nested universal primer by following manufacturer’s universalthermal cycling conditions (Clontech). The obtained cDNA frag-ments were cloned into pGEM�-T easy vector (Promega) andsequenced bi-directionally. In the 30RACE of 17b-HSD 12 two frag-ments were amplified which upon analysis, showed that both are17b-HSD 12 but a variant form. All the overlapping DNA fragmentswere assembled and the full length cDNA of 17b-HSD 1 (1791 bp

A. Rajakumar, B. Senthilkumaran / Steroids 92 (2014) 81–89 83

with ORF of 888 bp) and 17b-HSD 12 normal type (1073 bp withORF of 954 bp) and variant (921 bp with ORF of 690 bp) wereobtained. For ORF cloning and further analysis only 17b-HSD 12normal type was used. ORF primers were designed for 17b-HSD 1and 12 and the cDNAs were cloned into pGEM�-T easy vector(Promega) and sequenced bi-directionally. The analysis ofsequence homology of the cloned cDNA sequences to 17b-HSD 1and 12 of other vertebrates were carried out by using NCBI-BLASTsearch (http://blast.ncbi.nlm.nih.gov/Blast.cgi) and the assembledsequences were translated using EditSeq of Lasergene 7.1.0 (DNA-STAR, Madison, WI, USA). The analysis of homology of the deducedamino acid sequence of catfish 17b-HSD 1 and 12 to 17b-HSD 1 and12 of other teleosts were carried out by using ClustalW multiplealignment tool using default parameters.

2.3. Phylogenetic analysis

The phylogenetic tree containing both 17b-HSD 1 and 12 wasconstructed by ClustalW (http://clustalw.ddbj.nig.ac.jp/) usingthe neighbor-joining method with Gonnet protein weight matrixand a bootstrap analysis with 1000 replicates was used to assessthe strength of the nodes in the tree. The phylogenetic tree wasdisplayed using the TreeView software package version 1.6.6. Thefollowing are the GenBank accession numbers of amino acidsequences used for constructing the phylogenetic tree: 17b-HSD1 [Danio rerio (NM_205584), Oreochromis niloticus(NM_001279795), and Oryzias latipes (XM_004071297), Anguillajaponica (AY498620), Mus musculus (NM_010475), Rattus norvegi-cus (NM_012851), Bos taurus (NM_001102365), Homo sapiens(NM_000413) and C. batrachus (KM034751)] and 17b-HSD 12 [D.rerio 12a (NM_200881), D. rerio 12b (NM_199613), Salmo salar12b (BT045689), Mus musculus (NM_019657), Rattus norvegicus(NM_032066), Bos taurus (NM_001101307), Homo sapiens(NP_057226) and C. batrachus (JN848590)].

2.4. Quantitative real-time PCR (qPCR)

Total RNA isolation was carried out using TRI-reagent� (Sigma)as per the manufacturer’s protocol. Reverse transcription was per-formed using 1 lg of total RNA, random hexamers and verso�

reverse transcriptase (Thermo Scientific Inc., Waltham, MA, USA).All qPCR primers were designed for the amplicon length of �150–250 bp and all reactions were performed in triplicate for three dif-ferent samples using gene specific primers (Table 1) and powerSYBR� Green PCR master mix (Applied Biosystems) in an AppliedBiosystems 7500 fast thermal cycler according to the manufac-turer’s universal thermal cycling conditions. Melting-curve analysiswas performed to check the specificity of PCR amplification. Further,the samples were run in agarose gel. The bands were eluted and theamplicons were sequenced and verified. All assays were performedwith no template controls which yielded no amplification. Cyclethreshold (Ct) values were obtained from the exponential phase ofPCR amplification and 17b-HSD 1 or 12 expressions were normalizedagainst the expression of 18S rRNA to generate a DCt value (Ct of tar-get gene � Ct of reference gene). The change in the gene expressionwas calculated using 2�DCt method.

2.5. Enzyme immune assay (EIA)

Serum and tissue levels of T were measured by using EIA kit[20] and for serum levels of E2 Cayman’s EIA kit was used by fol-lowing respective manufacturer’s protocol. Serum from both maleand female catfishes and testicular tissue collected from catfishesduring different stages of recrudescence were stored at �80 �Cbriefly until assayed. The dilutions of series of serum samplesand/or tissue extracts from catfish with differing hormone concen-

trations were linear with standards of T and E2. Intra and interassay variations were within the limits specified in the manufac-turer’s protocol. The cross reactivity of T antisera to other naturallyoccurring C27-, C21-, C19- and C18- steroids is less than 0.1%. Thecross- reactivity of the E2 antisera for E1 is 12%, estradiol-17-glu-coronide (10%), and estriol (0.3%).

2.6. hCG induction

During preparatory phase (March), adult catfishes (male andfemale) weighing about 170–200 g, were collected for hCG induc-tion experiments. For each time point (0, 6, 12, 18 and 24 h) 3 maleand female fishes (in total 30 fishes) were maintained in the glasstank and acclimatized for a fortnight. Later, all catfishes were intra-peritoneally injected a single dose (1000 IU/kg body weight) ofhCG. Fishes were sacrificed at end of the time points in a 6 h inter-val up to 24 h, by immersing them in ice-cold water with MS-222,to collect testis. The age matched control fishes were injected withfish physiological saline and sacrificed for every time point similarto the hCG group. This procedure was repeated again with differentbatches of catfishes. The samples were snap-frozen in liquidnitrogen and stored at �80 �C briefly. Total RNA isolation, reversetranscription and qPCR analysis were performed as describedearlier, to analyze the changes in the expression pattern of17b-HSD 1 and 12 after hCG induction.

2.7. Statistical analysis

All data are presented as mean of different samples withStandard Error of the Mean (SEM). All data were compared byMann–Whitney Rank Sum Test or Kruskal–Wallis one-way analysisof variance (ANOVA) by ranks followed by Student–Newman–Keuls(SNK) post hoc test. All statistical analyses were performed usingSigmaPlot 11.0 software (Systat Software Inc., Chicago, IL, USA). Aprobability of P < 0.05 was considered statistically significant.

3. Results

3.1. Cloning of 17b-HSD 1 and 12 from catfish gonads

A partial cDNA of 521 bp encoding 17b-HSD 1 was cloned fromcatfish ovary and a different partial cDNA of 479 bp encoding17b-HSD 12 was cloned from catfish testis. Based on the sequenceinformation, gene specific RACE primers were designed and usedfor 50 and 30 RACE. The cDNA fragments obtained through 50

(801 bp) and 30 RACE (728 bp) for 17b-HSD 1 and 50 (346 bp) and30 RACE (372 and 252 bp) for 17b-HSD 12 were cloned and verifiedusing Lasergene software 7.1.0 (DNASTAR) and confirmed as17b-HSD 1 and 12 using NCBI-BLAST and ClustalW. After assemblingof all the cDNA fragments corresponding to 17b-HSD 1 and 12, fulllength cDNA of 1791 bp (17b-HSD 1), 1073 bp (17b-HSD 12 regulartype), and 921 bp (17b-HSD 12 variant) with open reading framesof 888 bp (17b-HSD 1), 954 bp (17b-HSD 12 regular type), and690 bp (17b-HSD 12 variant) were obtained. The cDNA sequenceswere submitted to GenBank, 17b-HSD 1 (KM034751) 17b-HSD 12(JN848590) and 17b-HSD 12 variant (KM034752). The 17b-HSD 1and 12 (normal type) ORF were cloned from catfish testis,sequenced bi-directionally and confirmed with the assembled fulllength 17b-HSD 1 and 12 cDNA of other vertebrates. 17b-HSD 12normal type was used for all the expression and phylogeneticanalyses.

3.2. Sequence and phylogenetic analyses

The multiple alignment of deduced 17b-HSD 1 (Fig. 1) and 12(Fig. 2) amino acid sequence in comparison with other species

Fig. 1. Multiple alignment of catfish 17b-HSD 1 with 17b-HSD 1 from other teleosts. The multiple alignment was created using ClustalW2 (http://www.ebi.ac.uk/Tools/msa/clustalw2/) and high homologous regions are shaded using Jalview 2.8 and important domains are shown as rectangles. The GenBank accession numbers of sequences usedare: Danio rerio (NM_205584), Oreochromis niloticus (NM_001279795), and Oryzias latipes (XM_004071297), Anguilla japonica (AY498620) and Clarias batrachus (KM034751).Protein-BLAST of catfish 17b-HSD 1 showed coenzyme binding motif, active sites, catalytic center and other important motifs specific for SDR superfamily.

84 A. Rajakumar, B. Senthilkumaran / Steroids 92 (2014) 81–89

revealed considerable homology with 17b-HSD 1 and 12 of otherteleosts specifically at coenzyme binding motif, active sites,catalytic center and other important motifs specific for SDR super-family (Figs. 1 and 2). Phylogenetic analysis revealed that catfish17b-HSD 1 and 12 in comparison with other vertebrate counter-parts showed considerable homology with 17b-HSD 1 and 12 ofother teleosts by forming separate clades (Fig. 3). 17b-HSD 1 of cat-fish was more similar to D. rerio (75.6%) while 17b-HSD 12 showedhigh homology with D. rerio 17b-HSD 12b (84.9%) and 12a (65.6%).

3.3. Tissue distribution of 17b-HSD 1 and 12 in different tissue of adultcatfish

Expression of 17b-HSD 1 was highest in ovary followed by gills,head kidney, muscle, heart and intestine (Fig. 4A) while 17b-HSD12 was expressed ubiquitously with higher levels in testis, ovary,brain, intestine and head kidney (Fig. 4B).

3.4. Expression of 17b-HSD 1 and 12 during gonadal ontogeny

qPCR analysis of 17b-HSD 1 and 12 at different stages of catfishgonadal development showed that both male and female gonadsexpress 17b-HSD 1 and 12 differentially throughout gonadal devel-opment till maturity. The expression of 17b-HSD 1 was higher infemales compared to males of corresponding age group (Fig. 5A).In contrast, 17b-HSD 12 was higher in males than the females dur-ing gonadal development and at maturity (Fig. 5B).

3.5. Expression of 17b-HSD 1 and 12 during different phases of theseasonal cycle

The qPCR analysis showed that the expression of 17b-HSD 1 wassignificantly higher (P < 0.05) during pre-spawning phase of catfishovarian cycle (Fig. 6A) compared to other phases. The expression of17b-HSD 12 was higher in preparatory and pre-spawning than theother phases of catfish testicular cycle (Fig. 6B).

3.6. Changes in expression of 17b-HSD 1 and 12 in the gonads afterhCG induction

The hCG induction during mid-preparatory phase (March) ofcatfish reproductive cycle showed, significantly enhanced17b-HSD 1 expression in the ovary and 17b-HSD 12 expression inthe testis during entire duration of the study when compared to0 h in the ovary and the testis respectively. The 17b-HSD 1 expres-sion increased significantly (P < 0.05) with maximum expression at18 h after hCG induction in the ovary (Fig. 7A), while the expres-sion of 17b-HSD 12 was maximum at 24 h (Fig. 7B) in testis.

3.7. Levels of T and E2 during different phases of recrudescence

The levels of T was found to be higher in the serum (Fig. 8A) ofcatfishes of pre-spawning (2.22 ± 0.39 ng/ml) and spawning(2.02 ± 0.20 ng/ml) than the rest of the phases analyzed. In testic-ular tissue, a similar pattern with high levels of T (Fig. 8B) duringpre-spawning phase (30.75 ± 1.67 pg/mg tissue) and spawning

Fig. 2. Multiple alignment of catfish 17b-HSD 12 with 17b-HSD 12 from other teleosts. The multiple alignment was created using ClustalW2 (http://www.ebi.ac.uk/Tools/msa/clustalw2/) and high homologous regions are shaded using Jalview 2.8 and important domains are shown as rectangles. The GenBank accession numbers of sequencesused are as follows: Danio rerio 12a (NM_200881), Danio rerio 12b (NM_199613), Salmo salar 12b (BT045689) and Clarias batrachus (JN848590). Protein-BLAST of catfish17b-HSD 12 showed coenzyme binding motif, active sites, catalytic center and other important motifs specific for SDR superfamily.

A. Rajakumar, B. Senthilkumaran / Steroids 92 (2014) 81–89 85

phase (21.15 ± 1.90 pg/mg tissue) was observed, while the T levelsdropped to 14.42 ± 1.04 pg/mg tissue in the post spawning phase.The levels of serum E2 was found to be higher (Fig. 9) in the pre-spawning phase (2.10 ± 0.15 ng/ml) followed by preparatory(1.12 ± 0.17 ng/ml), spawning (1.07 ± 0.24 ng/ml) and post spawn-ing (0.50 ± 0.15 ng/ml) phases of catfish.

4. Discussion

In this study, we report molecular cloning of catfish 17b-HSD 1and 12 and their expression during early gonadal development, dif-ferent phases of the seasonal reproductive cycle and after hCGinduction. Besides, modulation of T and E2 levels during differentphases of the seasonal reproductive cycle was also recorded.

The cDNA sequences of 17b-HSD 1 and 12 contains all the impor-tant motifs generally observed in the SDR super family and showshigh similarity to their respective protein sequences of teleosts.To further confirm that the isolated sequence belongs to 17b-HSD1 and 12 respectively, phylogenetic analysis was performed whichimplied distinct clades with their respective 17b-HSD forms fromdifferent teleost counterparts. 17b-HSD 1 and 12 of catfish showedhigh homology with zebrafish 17b-HSD 1 and 12b, respectively.Two forms of 17b-HSD 12 were cloned from zebrafish [10] andBLAST search also retrieved two 17b-HSD 12-like genes in both fuguand tetraodon genomes. But in the catfish, only one form of

17b-HSD 12 and a variant could be cloned successfully even aftervarying the reaction conditions in RT-PCR and RACE. Based on thesequence analysis, 17b-HSD 12 (both normal and variant) clonedfrom catfish showed high homology with 17b-HSD 12b from otherteleosts. The amino acid variation in the variant do not affect impor-tant catalytic sites and hence the variant form has minimal or noimportance. The possible reason for existence of variant might bedue to whole genome duplication events in fishes, which resultedin many isoforms and variants of genes [21]. In order to gain furtherinsight on the functional role of 17b-HSD 1 and 12, we analyzedtheir tissue distribution during pre-spawning phase of catfishreproductive cycle. 17b-HSD 1 was expressed predominantly inthe ovary, while a weaker expression was found in the gills, muscle,head kidney, heart and intestine. Similarly, higher 17b-HSD 1expression was found in the ovary and low levels were seen in tis-sues like skin, muscle and eye of zebrafish [10]. In the Nile tilapia,17b-HSD 1 is involved in both E2 and T interconversion and thussuggested to be a multifunctional enzyme that might be involvedin many different reactions in vivo in the steroidogenic pathwaydepending on its spatial and temporal expression pattern [11].However, in our present study, expression of 17b-HSD 1 is higherin the ovary when compared to gills, and thus suggesting no possi-ble role in testis or other steroidogenic tissues of catfish. Gills areinvolved in the production of 17a,20a- and 17a,20b-dihydroxy-4-pregnen-3-one (17a,20b-DP) and 11-ketotestosterone (11-KT) in

Fig. 3. Phylogenetic analysis of catfish 17b-HSD 1 and 12. The phylogenetic analysis was performed using ClustalW (http://clustalw.ddbj.nig.ac.jp/) by neighbor-joiningmethod and a bootstrap analysis with 1000 replicates was used to assess the strength of the nodes in the tree. The phylogenetic tree was generated using the TreeViewsoftware package version 1.6.6. Please refer Section 2.3 for GenBank accession numbers of sequences used for constructing the phylogenetic tree.

Fig. 4. Relative mRNA levels of (A) 17b-HSD 1 and (B) 17b-HSD 12 in different tissues of adult catfish. All data were expressed as mean ± SEM. Means with different lettersdiffer significantly (P < 0.05; Kruskal–Wallis one way ANOVA on ranks followed by SNK test).

86 A. Rajakumar, B. Senthilkumaran / Steroids 92 (2014) 81–89

teleosts [22]. This raises the possibility that the gill of the catfishmight produce several steroidal metabolites or E2. More in depthanalyses are required to prove this contention. Expression of 17b-HSD 12 was higher in the testis, ovary, brain, intestine and head kid-ney compared to other tissues. This shows the possible involvementof 17b-HSD 12 in gonadal as well as extra-gonadal steroidogenesistogether with their involvement in lipid metabolism. The higherexpression of 17b-HSD 12 in intestine shows their probable involve-ment in fatty acid metabolism since at first, human 17b-HSD 12 was

characterized as a ketoacylreductase due to its homology with theyeast enzyme YBR159w, which is active in fatty acid elongation[14].

In the Nile tilapia, 17b-HSD 1 is not involved in the early steroi-dogenesis during sex differentiation while its role in gametogene-sis seems essential [11]. However, in this study, the expression ofboth 17b-HSD 1 and 12 was detected much earlier during gonadaldifferentiation and development and then on sex-specific dimor-phic expression was evident, which shows possible involvement

Fig. 5. Relative mRNA levels of (A) 17b-HSD 1 and (B) 17b-HSD 12 in different stages of gonadal development. M-Male; F-Female. All data were expressed as mean ± SEM. *indicates means with significantly higher 17b-HSD 1 and 12 mRNA levels when compared with the corresponding opposite sex of the same age group (*, P < 0.05; Mann–Whitney rank sum test).

Fig. 6. Relative mRNA levels of (A) 17b-HSD 1 and (B) 17b-HSD 12 in the gonads of catfish during different phases of the reproductive cycle. All data were expressed asmean ± SEM. Means with different letters differ significantly (P < 0.05; Kruskal–Wallis one way ANOVA on ranks followed by SNK test).

A. Rajakumar, B. Senthilkumaran / Steroids 92 (2014) 81–89 87

of 17b-HSD 1 and 12 in the gonadal differentiation and develop-ment. 17b-HSD 1 is primarily involved in E2 production [11], while17b-HSD 12 is involved in fatty acid elongation in addition to E2

production [14,23]. 17b-HSD 1 expression was higher duringpre-spawning phase of ovarian cycle, while 17b-HSD 12 was highduring both preparatory and pre-spawning phases. This showsthe involvement of E2 in oocyte growth and in early stages ofgametogenesis. The involvement of 17b-HSD 12 in steroid and lipidmetabolism makes it an important candidate in regulation of ste-roidogenesis through substrate limitation. Further, based on theinvolvement of 17b-HSD 12 in fatty acid elongation, it can behypothesized that, the enzyme might also be involved in polyun-saturated fatty acid (PUFA) biosynthesis. A x-6 PUFA, arachidonicacid (AA) and its metabolites, eicosanoids are involved in variousreproductive processes including gametogenesis, maturation andrelease [24–27]. A recent study [13] using embryonic stem cellsof knockout mouse showed that 17b-HSD 12 deficiency causedinsufficient synthesis of AA suggesting that 17b-HSD 12 is involvedin AA synthesis. Wade et al. [25] reported that AA stimulates Tproduction in goldfish. Such a possibility might exist incatfish, wherein levels of T were increased during pre-spawning

phase, the period where expression of 17b-HSD 12 was higher. E2

in serum was higher during pre-spawning phase of ovarian cyclewhich corresponds to higher expression of 17b-HSD 1. T producedby thecal cells is converted into E2 by cytochrome P450 aromatasein the granulosa cells [28]. It is well known that E2 and T areinvolved in gametogenesis and decreased during gamete matura-tion during which progestin levels might get elevated [29,30].Gonadal steroidogenesis can be stimulated with hCG which actmore similar to LH by binding to LH receptors present in thegonads of teleosts [31–33]. hCG stimulate the production of E2, Tand 11-KT [31,33] both directly and indirectly through an increasein the expression of terminal steroidogenic enzyme genes like 11b-hydroxylase and 11b-hydroxysteroid dehydrogenase [34,35]. Thepresent study demonstrated that increase in the expression of17b-HSD 1 and 12 after hCG induction showed the possible regula-tion of E2 levels by hCG. Androgens including T can stimulate17a,20b-DP production by down regulating P450c17 activity inthe Japanese eel and catfish [29,36,37]. This may also explain theincreased expression of 20b-hydroxysteroid dehydrogenase incatfish [30]. PUFAs and its metabolites together with E2 areinvolved in various reproductive processes including

Fig. 7. Relative mRNA levels of (A) 17b-HSD 1 and (B) 17b-HSD 12 in the gonads ofcatfish at different time intervals after hCG induction, All data were expressed asmean ± SEM. */** indicates means with significantly higher 17b-HSD 1 and 12 mRNAlevels compared with the 0 h (**, P < 0.001, *, P < 0.05, Kruskal–Wallis one wayANOVA on ranks followed by SNK test).

Fig. 9. Levels of E2 in the serum of catfish during different phases of ovarian cycle.All data were expressed as mean ± SEM. Means with different letters differsignificantly (P < 0.05; Kruskal–Wallis One Way ANOVA on ranks followed by SNKtest).

88 A. Rajakumar, B. Senthilkumaran / Steroids 92 (2014) 81–89

gametogenesis, gamete maturation, ovulation and spermiation.Hence, there is a possibility of involvement of 17b-HSD 1 and 12in steroid and lipid metabolism, which might be thereby regulatingreproductive processes in teleosts.

Fig. 8. Levels of T in (A) serum and (B) testicular tissue of catfish during different phaseletters differ significantly (P < 0.05; Kruskal–Wallis one way ANOVA on ranks followed

5. Conclusions

The present study showed the sexual dimorphic and differentialexpression of 17b-HSD 1 and 12 during gonadal development andrecrudescence in catfish together with E2 and T levels. Higher lev-els of E2 and T together with dynamic expression of 17b-HSD 1 and12 during gonadal development and reproductive cycle show theirpossible involvement in gametogenesis and gamete maturation. Inaddition, the present work also demonstrated a possible involve-ment of gonadotropin in the regulation of 17b-HSD 1 and 12.

Funding and Acknowledgments

A grant-in-aid (F. No. 34-412/2008) from University GrantsCommission to BS supported this work completely. We thank Dr.T.G. Shrivastav, National Institute of Health & Family Welfare, Indiafor generously providing the ‘T’ EIA kit. BS is a recipient ofDepartment of Biotechnology-TATA innovation fellowship-2014which is acknowledged. AR is thankful to Council of Scientific

s of testicular cycle. All data were expressed as mean ± SEM. Means with differentby SNK test).

A. Rajakumar, B. Senthilkumaran / Steroids 92 (2014) 81–89 89

and Industrial Research, India for Junior and Senior ResearchFellowships. We also acknowledge the support of Department ofScience and Technology-FIST (Phase II) to the Department ofAnimal Biology at School of Life Sciences, University of Hyderabad.We profoundly thank the two anonymous reviewers for theirvaluable suggestions while revising the manuscript.

References

[1] Adamski J, Jakob FJ. A guide to 17b-hydroxysteroid dehydrogenases. Mol CellEndocrinol 2001;171:1–4.

[2] Moeller G, Adamski J. Integrated view on 17b-hydroxysteroid dehydrogenases.Mol Cell Endocrinol 2009;301:7–19.

[3] Oppermann UC, Salim S, Tjernberg LO, Terenius L, Jornvall H. Binding ofamyloidb-peptide to mitochondrial hydroxyacyl-CoA dehydrogenase (ERAB):regulation of an SDR enzyme activity with implications for apoptosis inAlzheimer’s disease. FEBS Lett 1999;451:238–42.

[4] Penning TM. Hydroxysteroid dehydrogenases and pre-receptor regulation ofsteroid. Hum Reprod Update 2003;9:193–205.

[5] Lukacik P, Kavanagh KL, Oppermann U. Structure and function of human 17b-hydroxysteroid dehydrogenases. Mol Cell Endocrinol 2006;248:61–71.

[6] Marchais-Oberwinkler S, Henn C, Moller G, Klein T, Negri M, Oster A, et al. 17b-hydroxysteroid dehydrogenases (17b-HSDs) as therapeutic targets: proteinstructures, functions, and recent progress in inhibitor development. J SteroidBiochem Mol Biol 2011;125:66–82.

[7] Oppermann U, Filling C, Hult M, Shafqat N, Wu X, Lindh M, et al. Short-chaindehydrogenases/reductases (SDR): the 2002 update. Chem Biol Interact2003;143–144:247–53.

[8] Poirier D. Inhibitors of 17b-hydroxysteroid dehydrogenases. Curr Med Chem2003;10:453–77.

[9] Prehn C, Möller G, Adamski J. Recent advances in 17b-hydroxysteroiddehydrogenases. J Steroid Biochem Mol Biol 2009;114:72–7.

[10] Mindnich R, Deluca D, Adamski J. Identification and characterization of 17b-hydroxysteroid dehydrogenases in the zebrafish. Danio rerio. Mol CellEndocrinol 2004;215:19–30.

[11] Zhou LY, Wang DS, Senthilkumaran B, Yoshikuni M, Shibata Y, Kobayashi T,et al. Cloning, expression and characterization of three types of 17b-hydroxysteroid dehydrogenases from the Nile tilapia, Oreochromis niloticus. JMol Endocrinol 2005;35:103–16.

[12] Mindnich R, Moller G, Adamski J. The role of 17b-hydroxysteroiddehydrogenase. Mol Cell Endocrinol 2004;218:7–20.

[13] Rantakari P, Lagerbohm H, Kaimainen M, Suomela JP, Strauss L, Sainio K, et al.Hydroxysteroid (17 b) dehydrogenase 12 is essential for mouse organogenesisand embryonic survival. Endocrinology 2010;151:1893–901.

[14] Moon YA, Horton JD. Identification of two mammalian reductases involved inthe two-carbon fatty acyl elongation cascade. J Biol Chem 2003;278:7335–43.

[15] Guiguen Y. Implication of steroids in fish gonadal sex differentiation and sexinversion. Curr Top in Steroid Res 2000;3:127–43.

[16] Nagahama Y. Molecular mechanisms of sex determination and gonadal sexdifferentiation in fish. Fish Physiol Biochem 2005;2–3:105–9.

[17] Nagahama Y. Gonadal steroid hormones: major regulators of gonadal sexdifferentiation and gametogenesis in fish. In: Norberg B, Kjesbu OS, TarangerGL, Andersson E, Stefansson SO, editors. Proceedings of the 6th InternationalSymposium on Reproductive Physiology of fish. Norway: University of Bergen;2000. p. 211–22.

[18] Joy KP, Singh MS, Senthilkumaran B, Goos HJTh. Pituitary-gonadal relationshipin the catfish Clarias batrachus (L): a study correlating gonadotrophin-II andsex steroid dynamics. Zool Sci 2000;17:395–404.

[19] Rajakumar A, Singh R, Chakrabarty S, Murugananthkumar R, Laldinsangi C,Prathibha Y, et al. Endosulfan and flutamide impair testicular development inthe juvenile Asian catfish, Clarias batrachus. Aquat Toxicol 2012;110–111:123–32.

[20] Shrivatsav TG, Chaube SK, Kariya KP, Kumar D. Heterologous enzyme linkedimmunosorbent assay for measurement of testosterone in serum. JImmunoassay Immunochem 2012;33:253–69.

[21] Meyer A, Schartl M. Gene and genome duplications in vertebrates: the one-to-four (-to eight in fish) rule and the evolution of novel gene functions. Curr OpinCell Biol 1999;11:699–704.

[22] Kime DE, Ebrahimi M. Synthesis of 17,20a- and 17,20b-dihydroxy-4-pregnen-3- ones, 11-ketotestosterone and their conjugates by gills of teleost fish. FishPhysiol Biochem 1997;17:117–21.

[23] Luu-The V, Tremblay P, Labrie F. Characterization of type 12 17b-hydroxysteroid dehydrogenase, an isoform of type 3 17b-hydroxysteroiddehydrogenase responsible for estradiol formation in women. Mol Endocrinol2006;20:437–43.

[24] Goetz FW, Berndtson A, Ranjan M. Ovulation: mediators at the ovarian level.In: Pang P, Schreibman M, editors. Vertebrate endocrinology, fundamentalsand biomedical implications. New York: Academic Press; 1991. p. 127–203.

[25] Wade MG, Van der Kraak G, Gerrits MF, Ballantyne JS. Release andsteroidogenic actions of polyunsaturated fatty acids in the goldfish testis.Biol Reprod 1994;51:131–9.

[26] Lenzi A, Gandini L, Maresca V, Rago R, Sgro P, Dondero F, Picardo M. Fatty acidcomposition of spermatozoa and immature germ cells. Mol Hum Reprod2000;6:226–31.

[27] Patiño R, Yoshizaki G, Bolamba D, Thomas P. Role of arachidonic acid andprotein kinase C during maturation inducing hormone-dependent meioticresumption and ovulation in ovarian follicles of Atlantic croaker. Biol Reprod2003;68:516–23.

[28] Nagahama Y. Endocrine regulation of gametogenesis in fish. Int J Dev Biol1994;38:217–29.

[29] Rajakumar A, Senthilkumaran B. Sperm maturation in teleosts: role ofandrogens and progestins in our present understanding to emerging newconcepts. In: Senthilkumaran B, editor. Sexual plasticity and gametogenesis infishes. New York: Nova Biomedical Inc.; 2013. p. 377–400.

[30] Sreenivasulu G, Senthilkumaran B, Sridevi P, Rajakumar A, Rasheeda MK.Expression and immunolocalization of 20b-hydroxysteroid dehydrogenaseduring testicular cycle and after hCG induction, in vivo in the catfis, Clariasgariepinus. Gen Comp Endocrinol 2012;175:48–54.

[31] Bogerd J, Blomenröhr M, Andersson E, van der Putten HHAGM, Tensen CP,Vischer HF, et al. Discrepancy between molecular structure and ligandselectivity of a testicular follicle-stimulating hormone receptor of the Africancatfish (Clarias gariepinus). Biol Reprod 2001;64:1633–43.

[32] Shiraishi T, Ohta K, Yamaguchi A, Yoda M, Chuda H, Matsuyama M.Reproductive parameters of the chub mackerel Scomber japonicas estimatedfrom human chorionic gonadotropin-induced final oocyte maturation andovulation in captivity. Fish Sci 2005;71:531–42.

[33] Paul S, Kundu S, Pramanick K, Bandyopadhyay A, Mukherjee D. Regulation ofovarian steroidogenesis in vitro by gonadotropin in common carp Cyprinuscarpio: interaction between calcium- and adenylate cyclase-dependentpathways and involvement of ERK signaling cascade. J Mol Endocrinol2010;45:207–18.

[34] Jiang JQ, Kobayashi T, Ge W, Kobayashi H, Tanaka M, Okamoto M. Fishtesticular 11b-hydroxylase: cDNA cloning and mRNA expression duringspermatogenesis. FEBS Lett 1996;397:250–2.

[35] Rasheeda MK, Kagawa H, Kirubagaran R, Dutta-Gupta A, Senthilkumaran B.Cloning, expression and enzyme activity analysis of testicular 11b-hydroxysteroid dehydrogenase during seasonal cycle and after hCGinduction in air-breathing catfish Clarias gariepinus. J Steroid Biochem MolBiol 2010;120:1–10.

[36] Cavaco JEB, Blijswijk Van B, Leatherland JF, Goos HJT, Schulz RW. Androgen-induced changes in leydig cell ultrastructure and steroidogenesis in juvenileAfrican catfish Clarias gariepinus. Cell Tissue Res 1999;297:291–9.

[37] Miura T, Higuchi M, Ozaki Y, Ohta T, Miura C. Progestin is an essential factorfor the initiation of the meiosis in spermatogenetic cells of the eel. Proc NatlAcad Sci USA 2006;103:7333–8.