INTRODUCTION ABSTRACT

BIOLOGY OF REPRODUCTION 67, 1080–1086 (2002) DOI 10.1095/biolreprod.102.003905 Ovarian Carbonyl Reductase-Like 20b-Hydroxysteroid Dehydrogenase Shows...
3 downloads 0 Views 770KB Size
BIOLOGY OF REPRODUCTION 67, 1080–1086 (2002) DOI 10.1095/biolreprod.102.003905

Ovarian Carbonyl Reductase-Like 20b-Hydroxysteroid Dehydrogenase Shows Distinct Surge in Messenger RNA Expression During Natural and Gonadotropin-Induced Meiotic Maturation in Nile Tilapia1 Balasubramanian Senthilkumaran,3,6 Cheni-Chery Sudhakumari,3 Xiao-Tian Chang,3 Tohru Kobayashi,3,6 Yuichi Oba,4 Guijun Guan,3,6 Yasutoshi Yoshiura,5 Michiyasu Yoshikuni,3,6 and Yoshitaka Nagahama2,3,6 Laboratory of Reproductive Biology,3 National Institute for Basic Biology, Okazaki 444-8585, Japan Graduate School of Bioagricultural Sciences,4 Nagoya University, Nagoya 464-8601, Japan Inland Station, National Research Institute of Aquaculture,5 Tamaki, Mie 519-0423, Japan CREST,6 Japan Science Technology Corporation, Kawaguchi, Saitama 332-0012, Japan ABSTRACT

INTRODUCTION

Meiotic maturation in fish is accomplished by maturationinducing hormones. 17a,20b-Dihydroxy-4-pregnen-3-one (17a,20b-DP) was identified as the maturation-inducing hormone of several teleosts, including Nile tilapia. A cDNA encoding 20b-hydroxysteroid dehydrogenase (20b-HSD), the enzyme that converts 17a-hydroxyprogesterone to 17a,20b-DP, was cloned from the ovarian follicle of Nile tilapia. Genomic Southern analysis indicated that 20b-HSD probably exists as a single copy in the genome. The Escherichia coli-expressed cDNA product oxidized both carbonyl and steroid compounds, including progestogens, in the presence of NADPH. Carbonyl reductase-like 20b-HSD is broadly expressed in various tissues of tilapia, including ovary, testis, and gill. Northern blot and reverse transcription polymerase chain reaction analyses during the 14-day spawning cycle revealed that the expression of 20b-HSD in ovarian follicles is low from Day 0 to Day 8 after spawning and is not detectable on Day 11. Distinct expression was evident at Day 14, the day of spawning. In males, 20bHSD expression was observed continually in mature testes but not in immature testes of 30-day-old fish. In vitro incubation of postvitellogenic immature follicles (corresponding to Day 11 after spawning) with hCG induced the expression of 20bHSD mRNA transcripts within 1–2 h, followed by the final meiotic maturation of oocytes. In tissues such as gill, muscle, brain, and pituitary, however, hCG treatment did not induce any changes in the levels of mRNA transcripts. Actinomycin D blockade of hCG-induced 20b-HSD expression and final oocyte maturation demonstrated the involvement of transcriptional factors. The carbonyl reductase-like 20b-HSD plays an important role in the meiotic maturation of tilapia gametes.

Meiotic maturation is a vital event for sexual maturation in lower vertebrates in which gonadotropin (GTH) and gonadal steroid mediators play pivotal roles [1]. Steroid mediators known as maturation-inducing hormones (MIHs) are synthesized in the follicular layer of the ovary to initiate maturational events under the influence of LH (also known as GTH-II in teleosts) [1]. In several teleosts, 17a,20b-dihydroxy-4-pregnen-3-one (17a,20b-DP) is considered an MIH [1]. A two-cell model of the process has been proposed. In this model, an LH surge stimulates the theca cells to produce 17a-hydroxyprogesterone concurrently with the rapid elevation of 20b-hydroxysteroid dehydrogenase (20bHSD) activity in the granulosa cells to convert 17a-hydroxyprogesterone to 17a,20b-DP [1, 2]. In sciaenid fishes, 17a,20b,21-trihydroxy-4-pregnen-3-one is considered the MIH [3]. Nevertheless, 20b-HSD is the key enzyme in the production of MIH in many teleosts. Furthermore, 20bHSD activity in the ovary and its elevation in relation to treatment with GTH- and cAMP-enhancing drugs has been demonstrated in many species [1, 4, 5]. Identification of the 17a,20b-DP receptor and production of 17a,20b-DP in teleost testes indicate the importance of 17a,20b-DP vis a´ vis 20b-HSD in the male reproductive system [6–8]. Recently, carbonyl reductase-like 20b-HSD cDNA was cloned from rainbow trout (Oncorhynchus mykiss) and ayu (Plecoglossus altivelis) ovaries [9, 10]. Fish 20b-HSD is very similar to the carbonyl reductase obtained from humans, pigs, rats, and rabbits [11–15]. These cDNA clones are thought to belong to the short-chain steroid dehydrogenase/reductase (SDR) superfamily and to catalyze the NADPH reduction of carbonyl compounds and steroids [9, 10, 12, 16–20]. Despite wide substrate affinities, carbonyl reductase/ 20b-HSD has been implicated in reproduction [9, 10, 12, 14]. In rat ovary, Espey et al. [21] demonstrated a temporal pattern of expression of carbonyl reductase during GTHinduced ovulation. Preliminary studies from our laboratory indicated a role for carbonyl reductase-like 20b-HSD in meiotic maturation of trout and ayu [9, 10]. The significance of the similarity between 20b-HSD and carbonyl reducatse is unclear at present. GTH-induced elevation in carbonyl reductase activity in rat [14] or 20b-HSD activity in fish [5] and presumably the steroids synthesized by them [1, 9, 10] mediate the action of LH to initiate final oocyte maturation (FOM) or ovulation. Nevertheless, the expression pattern of carbonyl reductase/20b-HSD transcripts and

gametogenesis, granulosa cells, ovary, ovulation, steroid hormones This work was supported in part by Grants-in-Aids for Research for the Future (JSPS-RFTF 96L00401) to Y.N. from the Japan Society for the Promotion of Science, for Scientific Research from the Japanese Ministry of Education, Science, Sports, and Culture, Japan, and CREST of JST (Japan Science and Technology). This work was also supported by fellowships from the Japan Society for the Promotion of Science for Young Scientists to B.S., C-C.S., and X-T.C. 2 Correspondence: Yoshitaka Nagahama, Laboratory of Reproductive Biology, National Institute for Basic Biology, Okazaki 444-8585, Japan. FAX: 81 564 55 7556; e-mail: [email protected] 1

Received: 25 January 2002. First decision: 12 February 2002. Accepted: 25 April 2002. Q 2002 by the Society for the Study of Reproduction, Inc. ISSN: 0006-3363. http://www.biolreprod.org

1080

20b-HSD IS REQUIRED FOR MEIOTIC MATURATION

the regulatory mechanisms during the ovarian/testicular cycle have yet to be studied in any of the lower vertebrates, including teleosts. In this regard, fishes that reproduce frequently in a short period can be considered more useful than those reproducing once each year or less frequently. Conversely, fishes that spawn daily would not be of use because their reproductive cycles proceed too rapidly. Nile tilapia (Oreochromis niloticus) is a gonochoristic fish with an average ovulatory cycle of 14–18 days. Males remain in the spermiating stage throughout the reproductive cycle after sexual maturation. Tilapia is an excellent model to study reproduction in teleosts. Recently, Tacon et al. [22] identified 17a,20b-DP as the MIH for Nile tilapia. Hence, attempts to isolate a 20b-HSD cDNA clone from this species were thought to be feasible. In the first part of the present investigation, we isolated and characterized a 20bHSD cDNA clone derived from tilapia ovary. Subsequently, using Northern blot and reverse transcription polymerase chain reaction (RT-PCR) analyses, changes in 20b-HSD mRNA expression were examined during the natural ovarian cycle and hCG-induced FOM in vitro. Follicles were treated with actinomycin D prior to hCG treatment to confirm whether the de novo synthesis of 20b-HSD is transcriptionally regulated. RT-PCR analysis of 20b-HSD was also performed using the testicular RNA obtained at two different stages to determine the importance of 20b-HSD in the sexual maturation of male gametes. We also tested the efficacy of 17a,20b-DP for inducing FOM in Nile tilapia. MATERIALS AND METHODS

Animals Nile tilapia were reared in the laboratory in large tanks with a recirculating aerated freshwater system. Fish were fed ad libitum with commercial food pellets and maintained in natural light under constant temperature (248C 6 18C). Under these conditions, tilapia spawn every 2 wk (average of 14–18 days), and ovarian follicles were obtained only from cycling females after careful consideration of follicle (oocyte) stages.

RT-PCR Amplification of a cDNA Fragment Homologous to 20b-HSD and Construction and Screening of an Ovarian cDNA Library Degenerate primers based on the nucleotide sequence of other vertebrate 20b-HSD/carbonyl or oxidoreductases were designed to clone partial ovarian 20b-HSD cDNA from postvitellogenic follicles at the migratory germinal vesicle (GV) stage of Nile tilapia: sense, 59-CAG AGT GGT GAA TGT (C/A/G)TC (C/T/A)AG C-39; antisense, 59-CCT GCC ATG TC(A/G) GT(T/G) C(T/G)G ACC-39. An ovarian follicular cDNA library was constructed using 5 mg of poly(A)1 RNA obtained from vitellogenic ovaries with a lZAP vector system (Stratagene, La Jolla, CA) and was packaged into UNI-ZAP XR vector using the Gigapack II Gold packaging extract (Stratagene). Screening for carbonyl reductase-like 20b-HSD clones was carried out by hybridization under high-stringency conditions [23] using the RT-PCR-amplified partial cDNA fragment encoding the tilapia ovary carbonyl reductase-like 20b-HSD as probe. After three rounds of screening, positive clones were obtained. In vitro excision and rescue of pBluescript phagemids were performed according to the manufacturer’s protocol.

DNA Sequencing and Analysis Double-stranded DNA was sequenced on both the 59 and the 39 ends using vector-based and/or gene-specific primers in an automated sequencer with a fluorescent dye terminator (Perkin-Elmer/Applied Biosystems, Chiba, Japan). The nucleotide and amino acid sequences were analyzed using Lasergene software (release 3.05; DNASTAR, Madison, WI) and the BLAST network service of the National Center for Biotechnology Information. The sequence data discussed in this study have been submitted to GenBank (accession no. AF439713).

1081

Genomic Southern Analysis Genomic DNA was prepared from mature testes, digested individually with BamHI, HindIII, KpnI, PstI, SacI, and XhoI, segregated in an 0.8% agarose gel, and capillary-transferred to a Hybond N1 nylon membrane (Pharmacia Amresham Biotech, Buckinghamshire, U.K.). The membrane was air dried and cross-linked with ultraviolet light prior to hybridization with the randomly radiolabeled open reading frame (ORF) of the tilapia carbonyl reductase-like 20b-HSD cDNA used as probe. A standard protocol [23] was utilized to perform Southern blot hybridization and washing of the probes. The membrane filter was analyzed in a PhosphorImager (Fuji BAS2000, Fuji, Tokyo, Japan) after briefly exposing it to PhosphorImager plates (Fuji-MacIII BAS MP2040; Fuji) and was exposed for 2–3 days to x-ray film with an intensifying screen at 2808C.

Production and Purification of Tilapia Ovarian Carbonyl Reductase-Like 20b-HSD Recombinant Protein To produce 20b-HSD recombinant protein, two specific primers were designed to introduce BspMI and KpnI sites at the 59 and 39 ends, respectively, in the ORF region of the tilapia carbonyl reductase-like 20b-HSD, according to the manufacturer’s protocol. After appropriate restriction digestion, purified PCR products were inserted into a bacterial expression vector, pETBlue 2 (Novagen, Madison, WI) at the NcoI and KpnI sites and were expressed in the Escherichia coli Tuner (DE3) pLacI strain. To take advantage of the HIS-tag associated with the pETBlue 2 vector system, the inserted ORF region is devoid of a stop codon. The expression plasmid was verified by nucleotide sequence analysis for proper orientation before it was used for recombinant protein production. The soluble fraction of the 20b-HSD recombinant protein was prepared according to the manufacturer’s instructions. The induction of the recombinant protein was verified by 7% SDS-PAGE. To purify the 20b-HSD recombinant protein, a precharged Ni-21 His-Bind resin (capacity 8 mg/ml, ProBound resin; Invitrogen, Life Technologies, Carlsbad, CA) column was utilized. The HIS-binding and elution of the recombinant protein was performed using imidazole-Tris-NaCl buffer at appropriate concentrations as described in the supplier’s protocol. The purified recombinant protein was examined by SDS-PAGE for purity before dialysis and after lyophilization. A similar protocol was followed to produce recombinant proteins from the truncated clone of 20b-HSD. A soluble fraction of protein from the pETBlue 2 vector containing the HIS-tag was also prepared. All the proteins were then analyzed for catalytic activity.

Detection of Enzymatic Activity in the Recombinant Protein by Spectrophotometry The enzymatic properties of 20b-HSD and carbonyl reductase were analyzed by methods described previously [11, 12, 17]. Purified recombinant proteins of tilapia carbonyl reductase-like 20b-HSD (about 50–100 mg) and the truncated clone were utilized to determine the specific activity and substrate specificity, using the substrates shown in Table 1. The soluble protein fraction of pETBlue 2 was tested as an internal control. Enzymatic activity (20b-HSD/carbonyl reductase) was measured as the decrease in absorbance at 340 nm in the presence of 0.08 mM of NADPH at 258C in a DU7400 spectrophotometer (Beckman, Fullerton, CA).

Northern Blot and RT-PCR Analyses Northern and RT-PCR analyses were performed to identify tissue distribution and testicular and ovarian expression patterns of carbonyl reductase-like 20b-HSD during natural and/or GTH-induced meiotic maturation. To execute the RT-PCR, 1 mg of total RNA was reverse transcribed for 1 h at 428C with oligo-dT18 and Superscript-II (GIBCO-BRL, Gaithersburg, MD) to obtain the first strand cDNA template, and then ORF-specific primers (sense, 59-ATG GTC TCT TTT TAC ATG TCA ACC-39; antisense, 59-CCA CTT CTG AAC GGT CTT TTC-39) were used in a PCR to obtain the tilapia carbonyl reductase-like 20b-HSD amplicon. Northern blot analysis was performed using 5 mg of poly(A)1 RNA obtained from ovarian follicles [23]. The membrane filter was analyzed using a Fuji BAS2000 PhosphorImager, and the signals were quantified, especially for follicular mRNA samples collected at different stages of the ovarian cycle, using the Fuji-MacBAS software (version 1.0). The carbonyl reductaselike 20b-HSD signal was expressed as a ratio to b-actin. The blot was subsequently stripped and rehybridized with a partial tilapia b-actin cDNA probe amplified from an ovarian follicular cDNA template to serve as control for loading variations.

1082

SENTHILKUMARAN ET AL. TABLE 1. Specific activity and relative velocity of carbonyl reductaselike 20b-HSD of Nile tilapia.

Substrate p-Nitrobenzaldehyde Menadione 5a-Dihydrotestosterone 5b-Dihydrotestosterone Progesterone 5b-Progesterone 17a-Hydroxyprogesterone

Specific activity Concentra(nmol mg tion (mM) protein21 min21) 0.5 0.25 0.05 0.05 0.02 0.02 0.02

34.72 26.04 5.0 3.21 0.86 0.9 1.51

Relative velocity (%) 100 75 14.4 9.2 2.5 2.6 4.3

SEM. The ratio of carbonyl reductase-like 20b-HSD mRNA to b-actin mRNA was multiplied by 103 and compared for the midvitellogenic, fullgrown immature, and mature ovarian stages using a one-way ANOVA. Differences among ovarian stages were considered significant at P , 0.05. Similar analysis was also employed for GVBD data.

RESULTS

Molecular Cloning and Characterization of Tilapia Ovarian Carbonyl Reductase-Like 20b-HSD cDNA

FIG. 1. Southern blot analysis of genomic DNA extracted from mature testis of tilapia, digested with restriction endonucleases, and probed with the ORF region of the carbonyl reductase-like 20b-HSD cDNA clone. Each lane was designated with the names of the restriction enzymes. The location of lDNA digested with HindIII (HindIII dig. lDNA) was used as a marker to indicate the size (23.13, 9.42, 6.56, 4.36, 2.32, and 2.02 kilobases, ethidium bromide).

17a,20b-DP, hCG, and Actinomycin D Treatments In Vitro 17a,20b-DP was tested for its potency to induce germinal vesicle breakdown (GVBD) or FOM in Nile tilapia. Oocytes with centrally located GV (full-grown immature oocytes) were incubated in triplicate in Ringer solution containing various concentrations of 17a,20b-DP. For the in vitro hCG (100 IU/ml) and actinomycin D (5 mg/ml) treatments, postvitellogenic follicles (oocytes with a centrally located GV), gill, and muscle were obtained from fish at about 11–12 days after spawning. Ringer solution and hCG were added to the ovarian follicles obtained from the same fish to avoid experimental and individual variations. Similarly, the follicles of the same fish used for the actinomycin D experiment were also checked simultaneously for the ability of hCG to induce 20b-HSD expression. Before and during experimentation with hCG, a sample of follicles was visually checked for the presence of a centrally located or migratory GV. Accomplishment of GVBD or FOM and centrally located GV were observed after fixation of follicles (oocytes) with 20% trichloroacetic acid for 30 min, followed by dehydration and brief clearing with ethanol and xylene, respectively. Total and poly(A)1 RNAs were prepared from ovarian follicles, gill, and muscle tissues at 0, 1, 2, 6, 18, and 36 h after hCG/ Ringer solution treatment in vitro. In similar experiments, brain and pituitary were also treated with hCG for 2 h. The experiment was repeated three times with three different female fish. The actinomycin D treatment was given 2 h before hCG treatment to possibly block transcription. Although actinomycin D is known for its broad-based action to inhibit geneinduced metabolic responses, we used this drug in this study to block the transcription process. The time points for the collection of ovarian follicles after the combined treatment with actinomycin D and hCG were similar to those described for hCG treatment except for an additional collection before the actinomycin D treatment (22 h).

Statistical Analysis for Northern Hybridization and GVBD Data The Northern blot data for ovarian follicular carbonyl reductase-like 20b-HSD expression (mRNA) at different stages of the ovarian cycle and GVBD data after 17a,20b-DP induction were expressed as the mean 6

We performed RT-PCR using degenerate oligonucleotide primers to isolate a cDNA fragment of carbonyl reductaselike 20b-HSD from tilapia ovary. The fragment (;313 base pairs [bp]) was about 78–83% identical at the nucleotide and amino acid levels to the corresponding region of the carbonyl reductase-like 20b-HSD obtained from trout and other vertebrates. Approximately 106 independent plaques from a tilapia ovary cDNA library were screened using the RT-PCR-generated cDNA fragment as probe. Fourteen positive clones were obtained and sequenced. All the clones were similar in nucleotide sequence except for a single truncated clone with a partial 59 end. Nucleotide sequences at both the 59 and 39 ends of these clones matched completely within the overlapping region. The 1486-bp cDNA comprised a putative initiation codon (ATG: 182 bp), an ORF, two stop codons (TGA), and the 39 untranslated region, including putative polyadenylation signals 27 and 60 bp upstream of the poly(A)1 tail (GenBank accession no. AF439713). The ORF encodes a protein of 280 amino acids with a calculated molecular mass of 30.9 kDa. A second potential initiation site 16 bp downstream from the first ATG was evident. The ORF region of the truncated clone starts from this site. The truncated clone had the partial 26bp 59 leading sequence of the full-length clone, but the other regions (remaining ORF and 39 end) matched exactly. Tilapia ovarian carbonyl reductase-like 20b-HSD shares 62–70% overall sequence identity with other carbonyl reductase-like 20b-HSDs (data not shown). The homology was highest (100%) in the coenzyme binding or Rossmann fold (GlyXXXGlyIleuGly; amino acid residues 15–21) and substrate pocket (TyrXXXLys; residues 197–201) domains. A phylogenetic tree was constructed to depict the relationship of the various carbonyl reductase-like 20b-HSD amino acids. Tilapia 20b-HSD occupied a unique position in teleosts and was grouped with mammalian counterparts (data not shown). Genomic Southern Analysis

Genomic Southern analysis possibly identified a single band (Fig. 1) in BamHI-, HindIII-, PstI-, and SacI-digested products of genomic DNA after probing with the ORF region of the carbonyl reductase-like 20b-HSD.

20b-HSD IS REQUIRED FOR MEIOTIC MATURATION

1083

FIG. 2. RT-PCR analysis of tissue distribution pattern of carbonyl reductase-like 20b-HSD (upper panel) and b-actin (lower panel) of tilapia. The positive control consists of product amplified from the plasmid DNA of the carbonyl reductase-like 20b-HSD cDNA clone; the negative (2ve) control contains no cDNA template. ØX174-HincII-digested DNA ranging from 1057 bp to 79 bp was used as a marker. Ant. Kidney, anterior kidney; Post. Kidney, posterior kidney; Mature Ova. Fol., mature ovarian follicle.

Carbonyl Reductase-Like 20b-HSD Enzymatic Activity of the cDNA Product Based on spectrophotometric analysis, the purified recombinant protein isolated from E. coli expressed with the cDNA product oxidized both carbonyl and steroid compounds in the presence of NADPH (Table 1). The recombinant protein derived from the truncated 20b-HSD cDNA clone (275 amino acid residues) also exhibited the same ability (data not shown), comparable to the full-length clone-derived protein product. However, the soluble protein or peptide derived from the pETBlue 2 vector (with the HIS-tag) did not show any catalytic activity. Nitrobenzaldehyde and menadione seem to be the best substrates; however, the cDNA product efficiently catalyzed different steroid hormones, including progestogens. The recombinant protein also degraded 17a-hydroxyprogesterone efficiently at a higher concentration (0.5 mM). Tissue Distribution Pattern of Carbonyl Reductase-Like 20b-HSD in Tilapia Carbonyl reductase-like 20b-HSD cDNA was distributed in a wide variety of tissues collected from fish undergoing FOM, with the highest level of expression in gonadal tissues, followed by gill and other tissues (Fig. 2, upper panel). No signal was perceived in the liver, indicating that any possible expression was below the limit of detection. The expression of carbonyl reductase-like 20b-HSD was detected in both ovary and isolated follicles. The expression of b-actin indicated the functional integrity of the cDNA templates (Fig. 2, lower panel). Changes in mRNA Levels of Carbonyl Reductase-Like 20b-HSD in Tilapia During the Reproductive Cycle In the tilapia ovary, two categories of ovarian follicles can be easily distinguished throughout the reproductive cy-

FIG. 3. Northern blot analysis of poly(A)1 RNA (5 mg) from ovarian follicles of individual tilapia at various stages of the ovarian cycle: VOF, vitellogenic ovarian follicles; IOF, full-grown immature ovarian follicles (postvitellogenic follicles); MOF, mature ovarian follicles. A) The location of one of the RNA ladder standards was used as a marker to indicate the size of the transcript. B) The mean (6SEM) expression (arbitrary units) of carbonyl reductase-like 20b-HSD mRNA/b-actin mRNA in the ovarian follicles. Different superscripts on top of the histogram indicate significant differences (P , 0.05) between the mean values.

cle. The largest follicles, which are in the midvitellogenic to full-grown stage (0–11 days after spawning), undergo FOM at Day 14. The second largest follicles, those in the early vitellogenic stage, enter into the midvitellogenic stage in the subsequent cycle. In this study, at Day 0 the follicles were further divided into two groups based on their size and diameter. Unlike females, male fish remain in the maturational stage without undergoing any serial cyclic pattern of gonadal development after initial sexual maturation. Poly(A)1 RNA (5 mg) isolated from ovarian follicles in individual ovaries at the midvitellogenic, full-grown immature (centrally located GV), and mature (migratory GV) stages were visualized using the radiolabeled ORF region of the carbonyl reductase-like 20b-HSD cDNA (Fig. 3A). A single band of ;1.5 kilobases was detected in the Northern blot analysis. The amount of carbonyl reductase-like 20b-HSD mRNA decreased drastically from the midvitellogenic stage to the postvitellogenic stage and then increased rapidly at the migratory GV stage or FOM. The expression of carbonyl reductase-like 20b-HSD/b-actin was significantly higher during FOM (n 5 3 for each group; ANOVA, P , 0.05) than during other stages (Fig. 3B). Consistent with our Northern blot results, RT-PCR analysis (Fig. 4) at different stages of the ovarian cycle also revealed that the expression of 20b2HSD is low from Day 0 to Day

1084

SENTHILKUMARAN ET AL.

FIG. 4. RT-PCR analysis of the expression of carbonyl reductase-like 20b-HSD and b-actin at different stages of the ovarian cycle in tilapia. The negative control is shown only for carbonyl reductase-like 20b-HSD but not for b-actin. Other details are as in Figure 2.

8 after spawning and is not detectable at Day 11. Distinct 20b-HSD expression was evident in the ovarian follicle on the day of spawning (Day 14; n 5 6). Carbonyl reductaselike 20b-HSD expression was observed in mature testes but not in immature testes (30 days old; Fig. 5). Effects of In Vitro hCG and Actinomycin D Treatments on mRNA Levels of Carbonyl Reductase-Like 20b-HSD in Full-Grown Immature Follicles of Tilapia and Induction of GVBD by 17a,20b-DP

In vitro incubation of oocytes with 17a,20b-DP induced GVBD in a dose-dependent manner (ANOVA, P , 0.05); GVBD was 64% 6 0.9%, 82% 6 1.7%, 91% 6 2.4%, 94% 6 1.6%, and 98% 6 1.3% for 0.01, 0.1, 0.5, 1, and 10 mM concentrations of 17a,20b-DP, respectively. Treatment with Ringer solution alone induced only 1.2% 6 0.6% GVBD. Full-grown (postvitellogenic) immature follicles underwent FOM when incubated with hCG for 15–20 h (data not shown). Consistent with our previous results, carbonyl reductase-like 20b-HSD expression was undetectable in the full-grown immature follicles. Both Northern blot (data not shown) and RT-PCR (Fig. 6, A and B) analyses revealed that in vitro incubation of postvitellogenic immature follicles with hCG induced the appearance of carbonyl reduc-

FIG. 6. RT-PCR analysis of the expression of carbonyl reductase-like 20b-HSD and b-actin in ovarian follicles at different times after treatment with hCG (A) or saline (Ringer solution) (B) alone and in combination with actinomycin D (C). Other details are as in Figures 2 and 4 (for C).

tase-like 20b-HSD mRNA within 1–2 h. The expression after 2 h of hCG treatment showed decline; however, the expression did not reach a nadir even after 36 h or after GVBD. Actinomycin D treatment completely blocked both the hCG-induced appearance of carbonyl reductase-like 20b-HSD mRNA (Fig. 6C) and the occurrence of FOM in postvitellogenic immature follicles, indicating that the synthesis of 20b-HSD occurs through a mechanism dependent on RNA synthesis. Tissues such as gill, muscle, brain, and pituitary (data not shown) did not show any changes in the expression of 20b-HSD after incubation with hCG. DISCUSSION

FIG. 5. RT-PCR analysis of the expression of carbonyl reductase-like 20b-HSD and b-actin in immature and mature testis of tilapia. Other details are as in Figures 2 and 4.

We isolated a carbonyl reductase-like 20b-HSD cDNA clone from a tilapia ovarian cDNA library. Tilapia ovarian carbonyl reductase-like 20b-HSD is grouped with mammalian complements; however, it occupies a unique position among teleosts in the phylogenetic tree. Site-directed mutagenesis and x-ray crystallographic studies demonstrat-

20b-HSD IS REQUIRED FOR MEIOTIC MATURATION

ed that the putative coenzyme binding domain or Rossmann fold, GlyXXXGlyIleuGly, and the region responsible for nucleophilic attack of the substrate or substrate pocket, TyrXXXLys, are mandatory for the functional activity of the SDR superfamily of enzymes [19, 24–26]. These regions are strictly conserved in tilapia carbonyl reductaselike 20b-HSD. Two types of carbonyl reductase-like 20bHSD have been isolated from rainbow trout [9] and rats [14]. However, a single type exists in ayu [10], pigs [12], and humans [11]. Genomic Southern analysis confirmed and extended support for the presence of a single copy of the gene for carbonyl reductase-like 20b-HSD in tilapia. Spectrophotometric analysis revealed that the purified recombinant protein from E. coli harboring the carbonyl reductase-like 20b-HSD cDNA isolated from tilapia ovary oxidized both carbonyl and steroid compounds, including progestogens, in the presence of NADPH. Although we were unable to identify the product, similar studies in ayu and trout and amino acid sequence homology of tilapia 20b-HSD protein with these products support the assumption that tilapia 20b-HSD catalyzes 17a-hydroxyprogesterone into 17a,20b-DP. The RT-PCR analysis revealed an abundant expression of carbonyl reductase-like 20b-HSD mRNA in ovary and ovarian follicles and in testis. However, the expression of 20b-HSD mRNA was also found in other tissues, such as gills, brain, and kidney, which is consistent with the previously reported distribution pattern [9, 10] of carbonyl reductase-like 20b-HSD. The significance of such an ubiquitous distribution is presently unclear. Nevertheless, hCG treatment did not induce any changes in the expression of 20b-HSD in extragonadal tissues, indicating that gonadal 20b-HSD alone is the prime candidate for gamete maturation. The presence of carbonyl reductase-like 20b-HSD in extragonadal tissues suggests a divergent role for this enzyme. In mammals, the carbonyl reductase of the sperm membrane was compared with the moonlighting proteins, which are characterized by a change of function according to cellular localization [27]. The expression of carbonyl reductase-like 20b-HSD in gills and interrenal tissues indicates an involvement in the production of 17a,20b-DP for release as a pheromone [28, 29]. The lower level of expression in tissues such as heart and muscle, which do not convert 17a-hydroxyprogesterone to 17a,20b-DP, indicates that an alternative role is also played by the enzyme. In mammals, carbonyl reductase-like 20b-HSDs are involved in the metabolism of xenobiotics, and they reduce biologically active compounds such as steroid hormones and prostaglandins, depending on tissue requirements [18–20]. This wide range of action may explain the wide distribution pattern of carbonyl reductase-like 20b-HSD. Despite the divergent actions of carbonyl reductase-like 20b-HSD, it is considered important for oocyte maturation and ovulation [9, 10, 12, 14, 21]. More specifically, the GTH/hCG/eCG-induced rise in MIH production [1, 30, 31] in different vertebrates indicates a probable mediation of 20b-HSD/carbonyl reductase [1, 2, 14, 21, 32]. However, the timing or onset of the GTH-induced 20b-HSD/carbonyl reductase expression for MIH synthesis has not been clearly established in lower vertebrates. The present investigation is the first to categorically demonstrate the temporal pattern of expression of 20b-HSD (a SDR superfamily enzyme) during the ovarian/spawning cycle. In a previous report, we indicated only a rise in 20b-HSD transcripts in the ayu ovarian follicle undergoing FOM [10]. In teleosts, there is a temporally discrete and massive

1085

production of MIH (17a,20b-DP) just prior to oocyte maturation [1]. Recently, a rise in 17a,20b-DP has been demonstrated in both mouth-brooding and normal female Nile tilapia during FOM [22]. Our study on 17a,20b-DP-induced GVBD response confirmed that the rise in this steroid in tilapia is associated with FOM. Concomitantly, Northern blot and RT-PCR analyses revealed a discrete temporal pattern of carbonyl reductase-like 20b-HSD expression during FOM. The expression of ovarian carbonyl reductase in rats distinctly increases after the induction of ovulation [21], although the transcripts were localized in thecal cells. More recently, Espey et al. [32] reported the expression of another carbonyl reductase-like enzyme, 3ahydroxysteroid dehydrogenase, in immature rat ovary after hCG induction, ultimately suggesting a role in ovulation. These results and those of the present study tend to support the hypothesis that the temporal expression of carbonyl reductase/20b-HSD plays a role in FOM and ovulation processes in vertebrates. In the present study, we demonstrated for the first time the appearance and distinct increase of 20b-HSD mRNA in ovarian follicles 1–2 h after hCG treatment in vitro. This finding corroborates a recent finding in eels of a rise in 20b-HSD enzymatic activity after hCG treatment [5]. The hCG-induced carbonyl reductase-like 20b-HSD expression in postvitellogenic immature follicles did not reach a nadir, even after GVBD, which may explain the presence of 17a,20b-DP in circulation in spawned tilapia [22]. In amago salmon (Oncorhynchus rhodurus), de novo synthesis of 20b-HSD consists of gene transcriptional events within 6 h after exposure to GTH and cAMP and translational events within 6–9 h [1, 4]. The blockade of both hCG-induced 20b-HSD expression and FOM by actinomycin D in the present study assures the transcriptional regulation of carbonyl reductase-like 20b-HSD and MIH production. Because actinomycin D blocks all gene-induced metabolic responses, use of an antisense primer that specifically blocks 20b-HSD transcription may provide more insight. A role for 17a,20b-DP has also been implicated in male sexual maturation in many teleosts [6–8]. Distinct expression of 20b-HSD in mature but not immature testes (at least in ;30-day-old fish) may indicate a role for 20b-HSD in the production of 17a,20b-DP. Concomitantly in hamsters, carbonyl reductase activity in the sperm-zona pellucida complex appears to be important for the fertilization process [27]. Further study is required to identify the onset and importance of 20b-HSD expression during sperm maturation. In the present study, we demonstrated that the 20b-HSD gene is expressed in tilapia ovary and testis in a stagespecific manner during meiotic maturation. The increased expression of 20b-HSD in postvitellogenic follicles during FOM and after hCG treatment indicates that this enzyme plays an essential role in final gamete maturation. Actinomycin D treatment results indicate that the process may be regulated by transcription factors. The absence of any kind of change in the expression of the 20b-HSD gene after hCG treatment in gill, muscle, brain, and pituitary further supports the differential functions of carbonyl reductase-like 20b-HSD in gonads and other tissues. ACKNOWLEDGMENTS We thank our colleagues Dr. Y. Shibata and Dr. R. Horiguchi for their help with SDS-PAGE analysis and recombinant protein purification. B.S. and C.C.S are thankful to Dr. S. Ganesh (Brain Science Institute, RIKEN,

1086

SENTHILKUMARAN ET AL.

Wako, Japan) and Dr. K. Gen (National Research Institute of Aquaculture, Tamaki, Mie, Japan) for technical suggestions. 16.

REFERENCES 1. Nagahama Y. 17a,20b-Dihydroxy-4-pregnen-3-one, a maturation-inducing hormone in fish oocytes: mechanisms of synthesis and action. Steroids 1997; 62:190–196. 2. Young G, Adachi S, Nagahama Y. Role of ovarian thecal and granulosa layers in gonadotropin-induced synthesis of a salmonid maturation-inducing substance (17a,20b-dihydroxy-4-pregnen-3-one). Dev Biol 1986; 118:1–8. 3. Thomas P. Hormonal control of final oocyte maturation in sciaenid fishes. In: Davey KG, Peter RE (eds.), Perspectives in Comparative Endocrinology. Ottawa: National Research Council of Canada; 1994: 619–625. 4. Nagahama Y, Kagawa H, Young G. Stimulation of 17a,20b-dihydroxy-4-pregnen-3-one production in the granulosa cells of amago salmon, Oncorhynchus rhodurus, by cyclic nucleotides. J Exp Zool 1985; 236:371–375. 5. Kazeto Y, Adachi S, Yamauchi K. 20b-Hydroxysteroid dehydrogenase of the Japanese eel ovary: its cellular localization and changes in the enzymatic activity during sexual maturation. Gen Comp Endocrinol 2001; 122:109–115. 6. Nagahama Y. Endocrine regulation of gametogenesis in fish. Int J Dev Biol 1994; 38:217–229. 7. Todo T, Ikeuchi T, Kobayashi T, Kajiura-Kobayashi H, Suzuki K, Yoshikuni M, Yamauchi K, Nagahama Y. Characterization of a testicular 17a,20b-dihydroxy-4-pregnen-3-one (a spermiation-inducing steroid in fish) receptor from a teleost, Japanese eel (Anguilla japonica). FEBS Lett 2000; 465:12–17. 8. Planas JV, Swanson P, Dickhoff WW. Regulation of testicular steroid production in vitro by gonadotropins (GTH I and GTH II) and cyclic AMP in coho salmon (Oncorhynchus kisutch). Gen Comp Endocrinol 1993; 91:8–24. 9. Guan G, Tanaka M, Todo T, Young G, Yoshikuni M, Nagahama Y. Cloning and expression of two carbonyl reductase-like 20b-hydroxysteroid dehydrogenase cDNAs in ovarian follicles of rainbow trout (Oncorhynchus mykiss). Biochem Biophys Res Commun 1999; 255: 123–128. 10. Tanaka M, Nakajin S, Kobayashi D, Fukuda S, Guan G, Todo T, Senthilkumaran B, Nagahama Y. Teleost ovarian carbonyl reductaselike 20b-hydroxysteroid dehydrogenase: potential role in the production of maturation-inducing hormone during final oocyte maturation. Biol Reprod 2002; 66:1498–1504. 11. Wermuth B, Bohren KM, Heinemann G, von Wartburg JP, Gabbay KH. Human carbonyl reductase: nucleotide sequence analysis of a cDNA and amino acid sequence of the encoded protein. J Biol Chem 1998; 263:16185–16188. 12. Tanaka M, Ohno S, Adachi S, Nakajin S, Shinoda M, Nagahama Y. 20b-Hydroxysteroid dehydrogenase of neonatal pig testis: purification and some properties. J Biol Chem 1992; 267:13451–13455. 13. Wermuth B, Mader-Heineman G, Ernst E. Cloning and expression of carbonyl reductase from rat testis. Eur J Biochem 1995; 228:473–479. 14. Aoki H, Okada T, Mizutani T, Numata Y, Minegishi T, Miyamoto K. Identification of two closely related genes, inducible and noninducible carbonyl reductases in the rat ovary. Biochem Biophys Res Commun 1997; 230:518–523. 15. Gonzales B, Sapra A, Rivera H, Kaplan WD, Yam B, Forrest GL.

17. 18. 19. 20.

21.

22.

23. 24. 25.

26.

27. 28. 29. 30.

31. 32.

Cloning and expression of the cDNA encoding rabbit liver carbonyl reductase. Gene 1995; 154:297–298. Nakajin S, Ohno S, Aoki M, Shinoda M. 20b-Hydroxysteroid dehydrogenase of neonatal pig testis: cofactor requirement and stereospecificity of hydrogen transfer from nicotinamide adenine dinucleotide phosphate reduced form. Chem Pharm Bull (Tokyo) 1989; 37:148– 150. Iwata N, Inazu N, Satoh T. Immunological and enzymological localization of carbonyl reductase in ovary and liver of various species. J Biochem 1990; 107:209–212. Wermuth B. Purification and properties of an NADPH-dependent carbonyl reductase from human brain. J Biol Chem 1981; 256:1206– 1213. Penning TM. Molecular endocrinology of hydroxysteroid dehydrogenases. Endocr Rev 1997; 18:281–305. Takada H, Uchida A, Kamizono R, Otagiri M, Imamura Y. Further studies on 20b-hydroxysteroid dehydrogenase with reductase-like activity present in liver microsomes of male rats. Life Sci 2000; 67: 2075–2082. Espey LL, Yoshioka S, Russel D, Ujioka T, Vladu B, Skelsey M, Fujii S, Okamura H, Richards JS. Characterization of ovarian carbonyl reductase gene expression during ovulation in the gonadotropin-primed immature rat. Biol Reprod 2000; 62:390–397. Tacon P, Baroiller JF, Le Bail PY, Prunet P, Jalabert B. Effect of egg deprivation on sex steroids, gonadotropin, prolactin, and growth hormone profiles during the reproductive cycle of the mouthbrooding cichlid fish Oreochromis niloticus. Gen Comp Endocrinol 2000; 117: 54–65. Sambrook J, Russell DW. Molecular Cloning: A Laboratory Manual, 3rd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press; 2001. Chen Z, Lu L, Shirley M, Lee WR, Chang SH. Site-directed mutagenesis of glycine-14 and two ‘‘critical’’ cysteinyl residues in Drosophila alcohol dehydrogenase. Biochemistry 1990; 29:1112–1118. Varughese KI, Xuong NH, Kiefer PM, Matthews DA, Whiteley JM. Structural and mechanistic characteristics of dihydropteridine reductase, a member of the Tyr-(Xaa)3-Lys-containing family of reductases and dehydrogenases. Proc Natl Acad Sci U S A 1994; 91:5582–5586. Guan G, Todo T, Tanaka M, Young G, Nagahama Y. Isoleucine-15 of rainbow trout carbonyl reductase-like 20b-hydroxysteroid dehydrogenase is critical for coenzyme (NADPH) binding. Proc Natl Acad Sci U S A 2000; 97:3079–3083. Montfort L, Frenette G, Sullivan R. Sperm-zona pellucida interaction involves carbonyl reductase activity in the hamster. Mol Reprod Dev 2002; 61:113–119. Kime DE, Ebrahimi M. Synthesis of 17,20a- and 17,20b-dihydroxy4-pregnen-3-ones, 11-ketotestosterone and their conjugates by gills of teleost fish. Fish Physiol Biochem 1997; 17:117–121. Ebrahimi M, Scott AP, Kime DE. Extragonadal production of 17,20dihydroxy-4-pregnen-3-ones in vitro in cyprinid fish. Gen Comp Endocrinol 1996; 104:296–303. Kobayashi T, Sakai N, Adachi S, Iwasawa H, Asahina K, Nagahama Y. 17a,20a-dihydroxy-4-pregnen-3-one is the naturally occuring spermiation-inducing hormone in the testis of a frog, Rana nigromaculata. Endocrinology 1993; 133:321–327. Richards JS, Russell DL, Robker RL, Dajee M, Alliston TN. Molecular mechanisms of ovulation and lutenization. Mol Cell Endocrinol 1998; 145:47–54. Espey LL, Yoshioka S, Ujioka T, Fujii S, Richards JS. 3a-Hydroxysteroid dehydrogenase messenger RNA transcription in the immature rat ovary in response to ovulatory dose of gonadotropin. Biol Reprod 2001; 65:72–78.