Thyroid hormone-induced protein (TIP) gene expression by3,5,30-triiodothyronine in the ovarian follicle of perch

(Anabas testudineus, Bloch): modulation of 3b-hydroxysteroiddehydrogenase/D5–D4-isomerase enzyme by TIP

Malabika Datta,a R.J. Nagendra Prasad,b A.K. Navneet,a

Sib Sankar Roy,a and Samir Bhattacharyaa,*

a Molecular Endocrinology Laboratory, Indian Institute of Chemical Biology, 4, Raja S. C. Mullick Road, Jadavpur, Kolkata 700 032, Indiab Department of Zoology, Visva Bharati University, Santiniketan 731 235, India

Accepted 7 March 2002

Abstract

Our previous reports had shown that 3,5,30-triiodothyronine (T3) induced the generation of a 52-kDa monomer protein, i.e., TIP

(thyroid hormone-induced protein) in the perch ovarian follicle. TIP, in turn, increased progesterone formation by stimulating

D5-3b-HSD activity (3b-hydroxysteroid dehydrogenase/D5–D4 isomerase) [Eur. J. Endocrinol. 134 (1996) 128–135; Gen. Comp.

Endocrinol. 113 (1999) 212–220]. In the present investigation, perch ovarian follicles were incubated in the absence (control) or the

presence of T3 or gonadotropin (GTH) or human chorionic gonadotropin (hCG). RNAs were isolated and allowed to hybridize

with a radiolabeled TIP oligonucleotide probe prepared on the basis of the N-terminal 17-amino-acid sequence of TIP. Only RNA

from T3-incubated follicles hybridized with the probe, while RNA from control or GTH- or hCG-incubated follicles did not hy-

bridize with the probe. The transcript size of TIP mRNA was �1.8 kb. mRNA isolated from T3-incubated ovarian follicles subjected

to in vitro translation and Western blot analysis clearly identified a 52-kDa protein which was not found with the mRNA from the

control follicles. However, both TIP and GTH stimulated progesterone secretion from perch ovarian follicles in vitro. GTH

stimulation of D5-3b-HSD was due to the stimulation of enzyme protein synthesis as a more than twofold increase in D5-3b-HSD

occurred in response to GTH. But TIP did not stimulate synthesis of D5-3b-HSD protein. However, in vitro incubation of D5-3b-

HSD enzyme with TIP in the presence of NAD and substrate (pregnenolone) greatly stimulated enzyme activity, while incubation

with GTH had no effect, indicating a modulation of D5-3b-HSD protein from a less active to a more active state by TIP. This has

been supported by another observation, in which TIP (52 kDa) and D5-3b-HSD (45 kDa) incubation resulted in a complex of

99 kDa. This suggests a protein–protein interaction in the process of D5-3b-HSD activation by TIP. The present work, therefore,

shows some new and interesting aspects of thyroid hormone regulation of the reproductive control mechanism. � 2002 Elsevier

Science (USA). All rights reserved.

1. Introduction

Involvement of thyroid hormone in the reproductionof vertebrates has been implicated for a long time (Ball,1960; Chakraborti and Bhattacharya, 1984; Gordon andSouthren, 1977; Ichikawa et al., 1974; Leatham, 1972;Southren et al., 1974). A relationship between thyroidhormone and ovarian steroidogenesis has been reported

(Goldman et al., 1993; Gregoraszczuk and Galas, 1998;Gregoraszczuk and Skalka, 1996) but the mechanism bywhich it influences the gonadal steroidogenesis is stillunclear. Reports from our laboratory have shown theexistence of thyroid hormone receptors in perch ovarianfollicular cells (Bandyopadhyay and Bhattacharya,1994; Chakraborti et al., 1986; Maitra and Bhattach-arya, 1989), in goat testicular Leydig cells (Jana andBhattacharya, 1994), and in human corpus luteal cellnuclei (Bhattacharya et al., 1988). The presence of thy-roid hormone receptors has also been reported in por-cine and human granulosa cells (Wakim et al., 1987,

General and Comparative Endocrinology 126 (2002) 334–341

www.academicpress.com

GENERAL AND COMPARATIVE

ENDOCRINOLOGY

* Corresponding author. Fax: +91-33-473-5197/0284.

E-mail address: samir@iicb.res.in (S. Bhattacharya).

0016-6480/02/$ - see front matter � 2002 Elsevier Science (USA). All rights reserved.

PII: S0016 -6480 (02 )00009-6

1993, 1994). Search for the biological relevance of thisreceptor leads us to find that binding of T3 (3,5,30-tri-iodothyronine) to the nuclear receptor induces thegeneration of a putative protein, TIP (thyroid hormone-induced protein), that stimulates the secretion of bio-logically active steroids (Bandyopadhyay et al., 1996;Jana and Bhattacharya, 1994). TIP has been purifiedfrom perch ovarian follicles (Bhattacharya et al., 1996),rat granulosa cells (Bandyopadhyay et al., 1996), andgoat testicular Leydig cells (Jana et al., 1996) by trig-gering its synthesis with T3. TIP is a 52-kDa monomerprotein and its addition to perch ovarian follicles, ratgranulosa cells or goat Leydig cells in vitro greatlystimulated the synthesis and release of progesterone andtestosterone, respectively ( Bandyopadhyay et al., 1996;Bhattacharya et al., 1996; Jana et al., 1996). In ad-dressing the question of how TIP can stimulate prog-esterone or testosterone formation, which is known tobe regulated by gonadotropin, we have observed thatTIP is specifically involved in augmenting the conver-sion of D5 biologically inert steroid precursors to D4

biologically active steroids by enhancing D5-3b-hy-droxysteroid dehydrogenase ðD5-3b-HSD) enzyme ac-tivity (Datta et al., 1999; Prasad et al., 1999).

In this communication, we report T3-induced TIPgene expression in perch ovarian follicles by preparingan oligonucleotide probe based on the N-terminal 17-amino-acid sequence of TIP and in vitro translation ofTIP from T3-incubated perch ovarian follicular mRNAand evidence for protein–protein interaction in TIP-augmented D5-3b-HSD activity.

2. Materials and methods

2.1. Incubation of ovarian follicles

Ovaries collected from perch, Anabas testudineus(Bloch), belonging to the prespawning stage were washedthoroughly in ice-cold oxygenated culture medium (Ea-gle’s minimum essential medium (MEM), GIBCO BRL,Gaithersburg, MD). The mesoovarian covering was cutwith scissors and peeled off with the help of fine forceps.The tunica albuginea and germinal epithelium were re-moved. Separation of loosely attached follicles did notrequire any enzyme treatment. Isolated ovarian follicleswere washed thrice with the culture medium (MEM) andabout 25 mg of ovarian follicles (300 follicles) was in-cubated in 1.0 ml of oxygenated (95% O2=5% CO2)MEM (supplemented with 100 U/ml of penicillin and100 lg=ml of streptomycin). All incubations were carriedout at 30 �C. An initial 2-h preincubation was necessaryfor the follicles to recover from the shock of surgery(Bhattacharya et al., 1996; Datta et al., 1999) and at theend of 2 h, T3 (Sigma Chemical, St. Louis, MO; 100 ng/ml) or GTH (gonadotropin) (1 lg=ml) or TIP (5 lg=ml)

or human chorionic gonadotropin (hCG (1 lg=ml), Sig-ma Chemical) was added and incubated for another 3 h.Incubations in the absence of any of these were taken asthe control. Viability of the ovarian follicles was esti-mated by using the trypan blue (0.1%) dye exclusionmethod and found to be about 95% at the end of incu-bation. TIP was purified from perch ovarian follicles bytreating them with T3 according to the procedure pre-viously described by us (Bhattacharya et al., 1996). GTHwas purified from the pituitary glands of murrel, Channapunctatus, by following a previously described procedure(Banerjee et al., 1989).

2.2. Northern blot

2.2.1. Isolation of RNA for Northern blotTotal RNA from control and T3-, GTH-, or hCG-

treated perch ovarian follicles was isolated by theTriPure isolation reagent (Boehringer Mannheim, Ger-many). Briefly, after 3 h of incubation with the respec-tive hormones, the ovarian follicles were homogenizedin 1.0 ml of the TriPure reagent and incubated for 5 minat room temperature to dissociate the nucleoproteincomplexes and 0.2 ml of chloroform was then added toeach tube, vigorously shaken for 15 s and incubated for15 min at room temperature. These were centrifuged at12,000g for 15 min at 4 �C. The colorless aqueous phasewas collected in separate tubes, 0.5 ml of isopropanolwas added to each tube and centrifuged at 12,000g for10 min at 4 �C, and the pellet obtained was washed with1.0 ml of 75% ethanol. The air-dried pellet was resus-pended in DEPC (Sigma)-treated water and stored at)80 �C until further use.

2.2.2. Polymerase chain reaction (PCR)TIP probe was generated by PCR using sense and

antisense primers synthesized on the basis of the N-terminal 17-amino-acid sequence of TIP (TNRLQGKVALVTGGISH). (We are grateful to Professor Anil K.Lala, Department of Chemistry and BiotechnologyCentre, Indian Institute of Technology, Powai, Mum-bai, India, for the N-terminal sequencing of TIP.) PCRconsisted of 30 cycles of denaturation at 94 �C for 45 s,annealing at 45 �C for 45 s, and extension at 72 �C for45 s. The reaction mixture contained 1� amplificationbuffer (10 mM Tris–HCl (pH 8.4), 50 mM KCl), 2.5 mMMgCl2, 200 lM dATP, dCTP, dTTP, and dGTP, 2.5units of Taq polymerase, 100 ng of liver genomic DNAas template and 4 lM primers. The oligonucleotideDNA (51 bp) so obtained was end labeled and used as aprobe in Northern hybridization.

2.2.3. End labeling of the PCR productThe 51-bp PCR product was labeled with c-32P. To

obtain this, the oligonucleotide DNA was incubatedwith 10� polynucleotide kinase (PNK) buffer (1 ll),

M. Datta et al. / General and Comparative Endocrinology 126 (2002) 334–341 335

polynucleotide kinase (1 ll), and [c-32P][ATP] (5 ll of3000 Ci/mmol) in a total volume of 20.4 ll. These weremixed thoroughly and incubated at 37 �C for 45 min.The mixture was then heated at 68 �C for 10 min to in-activate the enzyme and 40 ll of double distilled water(autoclaved), 240 ll of 5 M ammonium acetate, and750 ll of chilled ethanol were added. The tube was kepton ice for 1 h and then centrifuged at 12,000g for 20 minat 4 �C. The labeled oligonucleotide was obtained as aprecipitate. It was dissolved in sterile double-distilledwater and stored at )20 �C until further use as a probe.

2.2.4. Northern transfer and hybridizationNorthern hybridization was carried out by using the

labeled TIP (51-bp oligonucleotide DNA) probes. RNA(15 lg) isolated from control or T3-, GTH-, or hCG-treated ovarian follicles were subjected to denaturing 1%agarose gel electrophoresis and transferred to a nylonmembrane (Immobilon S, Sigma Chemical) by capillarytransfer using 10� SSC. The membrane was washedwith 10� SSC, UV-crosslinked, and stored until furtheruse. The RNA markers (0.28–6.5 kb) used were pur-chased from GIBCO-BRL. For hybridization, themembrane was first prehybridized for 1 h at 65 �C inhybridization buffer (6� SSC containing 1� Denhardt’ssolution and 1% SDS). After 1 h, the hybridizationbuffer was replaced with fresh buffer and the labeledprobe was added. Hybridization was carried out for 18 hat 65 �C. The membrane was washed thrice for 30 mineach in 2� SSC containing 0.1% SDS at 65 �C, dried,wrapped in a Saran Wrap, and exposed to X-ray film.After 3 days the film was developed.

2.3. In vitro translation

In vitro translation in wheat germ lysate usingmRNA purified from control or T3-treated perch ovar-ian follicles was done according to the method ofComstock et al. (1987) with a little modification.

2.3.1. mRNA isolation for in vitro translationOn termination of 3 h of incubation, the follicles

(control and T3 treated) were harvested and washed toremove the medium completely. mRNA was then iso-lated with the help of a Quick Prep Micro mRNA Pu-rification Kit (Amersham Pharmacia Biotech, USA).The mRNA isolated was stored at )80 �C until furtheruse for in vitro translation.

2.3.2. Preparation of wheat germ lysateWheat germ (1.0 g) was added to 70 ml of chloroform

and stirred for 60 min at 4 �C. This was then filtered onWhatman filter paper and the wheat germ obtained as aresidue was dried at 4 �C. The completely dried materialwas grounded with 0.7 ml of homogenizing buffer(5 mM Hepes, 120 mM magnesium acetate, and 1 mM

dithiothreitol, pH 7.6) with pestle and mortar to make athick paste. Another 5.0 ml of buffer was gradually ad-ded and the suspension was centrifuged at 15,000g for10 min at 4 �C. To the supernatant, 1/50 vol of 0.5 MHepes was added and the solution was dialyzed over-night at 4 �C against dialysis buffer (20 mM Tris,120 mM potassium acetate, 5 mM magnesium acetate,and 1 mM dithiothreitol, pH 7.6). Following dialysis,the solution was centrifuged at 5000g to remove anyunwanted residue and the supernatant was stored inaliquots at )20 �C until further use.

2.3.3. In vitro translation in wheat germ lysateWheat germ lysate (10 ll) was incubated in the pres-

ence of 30 ll of reaction mixture (20 mM Hepes, 2 mMdithiothreitol, 1 mM ATP, 20 lM GTP, 8 mM creatinephosphate, 1.5 mM creatine phosphokinase, and 20 lMconcentrations of 19 different amino acids each). ThemRNA (5.0 lg) from control or T3-treated follicles wasadded to separate sets. The contents were mixed tho-roughly and then incubated at 25 �C for 2 h with constantshaking. After 2 h, the tubes were placed in an ice bathfollowed by centrifugation at 15,000g for 30 min. Thesupernatants were subjected to Western blot where anti-TIP antibody was used to detect TIP on the membrane.

2.3.4. Western blot of translation productThe translated products (from in vitro translation

with mRNAs from control or T3-incubated follicles)were subjected to SDS–PAGE according to the methodof Laemmli (1970) and as described by Bandyopadhyayand Bhattacharya (1993). The translated products wereelectrophoresed at a constant voltage of 60 V. After therun, the gel was washed thoroughly and the proteinswere transferred to PVDF membrane (Immobilon P,Sigma), which was incubated in the presence of ablocking solution (5% nonfat dried milk in PBST) for1 h to block the potential nonspecific binding sites.PVDF membrane containing the transferred protein wasthen washed thrice with PBST (15 min each) and incu-bated overnight with anti-TIP antibody (diluted 1:1000)at 4 �C. Anti-TIP antibody was raised in rabbit againstpurified perch TIP according to a previously describedprocedure (Prasad et al., 1999). This was followed bythree subsequent washes of the membrane with PBST(15 min each) followed by an incubation for 2 h at roomtemperature in the presence of second antibody linked toalkaline phosphatase (diluted 1:500). After 2 h, themembrane was washed thoroughly and then incubatedin the presence of the substrate (BCIP/NBT).

2.4. Progesterone release

To determine the progesterone (P4) released into themedium in response to TIP (5 lg=ml) or GTH (1 lg=ml),perch ovarian follicles were incubated in a manner

336 M. Datta et al. / General and Comparative Endocrinology 126 (2002) 334–341

similar to that described above. On termination of in-cubation, the media were collected and the amount of P4

released into the media was measured by P4 RIA(Mukherjee et al., 1994).

2.5. D5-3b-HSD protein synthesis

Perch ovarian follicles (25 mg) were incubated in 5.0-ml sterile beakers containing 1.0 ml of oxygenated MEM(95% O2=5% CO2) in the presence of GTH (1 lg/incu-bation) or TIP (5 lg/incubation) or in the absence ofeither of them (control). The incubation medium wassupplemented with 1 mM concentrations of 18 differentamino acids and [14C]leucine (300 mCi/mmol). After 3 hof incubation, the follicles were harvested, washedthoroughly (until the washings were free of radioactivecounts), and sonicated (132 kHz, Labsonic 2000, B.Braun, Germany). The sonicated material was centri-fuged at 1000g for 10 min, the supernatant was collected,and 100 ll of 3b-HSD antibody (diluted 1:1000) wasadded. Details of this procedure, including the raising ofperch 3b-HSD antibody, have been described earlier(Prasad et al., 1999). At the end of 24 h, second antibody(goat antirabbit IgG) was added followed by 10% PEG.The pellet obtained after centrifugation was dissolved in100 ll of NCS tissue solubilizer and counted in a liquidscintillation counter (LS 6000 SC, Beckman, USA).Results were expressed as dpm/mg of protein.

2.6. Effect of TIP and GTH on D5-3b-HSD activity in acell-free system

D5-3b-HSD was purified from perch ovarian folliclesby following a method described earlier by Datta et al.(1999). The purified enzyme (10 lg) was incubated in0.01 M phosphate buffer (pH 7.4) for 1 h with purifiedTIP (5 lg/ml) or GTH (1 lg/ml) in the presence of[3H]pregnenolone (100,000 cpm) and 1 mM NAD (totalvolume 1.5 ml). Another incubation with TIP alone wasalso done to check if TIP has any dehydrogenase andisomerase activities by itself. After 1 h the tubes wereimmediately kept in icewater slurry and 2.0 ml of ice-cold dichloromethane was added. Steroids were ex-tracted and the amount of [3H]progesterone formed wasquantified (Datta et al., 1999).

2.7. SDS–PAGE of TIP and D5-3b-HSD incubation invitro

TIP (7.0 lg) and 7.0 lg of D5-3b-HSD were incubatedfor 1 h in the presence of 0.01 M phosphate buffer, pH7.4, containing 1 nmol of NAD and 0.125 lmol ofpregnenolone dissolved in propylene:glycol (1:1 v/v; thevolume of this was reduced to 10 ll to avoid turbidity).The incubation was terminated by adding sample buffer(0.36 M Na-phosphate buffer, pH 7.0, 20% glycerol, and

0.1% bromophenol blue). To determine the bindingpattern of TIP and D5-3b-HSD, SDS–PAGE was per-formed according to the method of Laemmli (1970) withslight modifications. The samples were heated for 2 minat 90 �C and loaded on a 4% stacking gel over a 10%separating gel (2% SDS was used instead of the usual10%). The running buffer contained 0.18 M Na-phos-phate buffer (pH 7.0) and 0.1% SDS. Electrophoresiswas carried out at a constant voltage of 60 V and the gelwas stained with Coommassie blue.

2.8. Estimation of protein

Protein content wherever mentioned was estimatedby the method of Lowry et al. (1951) using bovine serumalbumin as the standard.

2.9. Statistical analysis

Data were analyzed by one-way analysis of variance(ANOVA). Where F values indicated significance,means were compared by a post hoc multiple range test.All values are expressed as means� SEM.

3. Results

3.1. TIP gene expression by T3

Perch ovarian follicles were incubated in the presenceor the absence of T3, follicles were lysed by sonication,and RNAs were isolated and subjected to hybridizationwith the radiolabeled TIP oligonucleotide probe. Fig. 1shows that hybridization could not be detected with theRNA from control follicles, whereas RNA from T3-in-cubated follicles clearly demonstrates hybridization sig-nal. The TIP transcript size was �1.8 kb. Incubationswith GTH or hCG did not produce hybridization signal,indicating the specificity of T3 in expressing the TIP gene.

Fig. 1. Northern hybridization of RNA isolated from perch ovarian

follicles incubated in the absence (control) or the presence of T3

(100 ng/ml) or GTH (1 lg/ml) or hCG (1 lg/ml). Total RNA (15 lg)

was subjected to denaturing agarose (1%) gel electrophoresis and hy-

bridized with the labeled TIP oligonucleotide probe. Lane 1, control;

lane 2, GTH; lane 3, hCG; and lane 4, T3.

M. Datta et al. / General and Comparative Endocrinology 126 (2002) 334–341 337

Since Northern hybridization showed TIP gene expres-sion by T3, in vitro translation was performed withmRNA from ovarian follicles incubated without (con-trol) or with T3. mRNAs collected from these incubateswere added to wheat germ lysates and the translatedproducts were run through SDS–PAGE. Western blotanalysis distinctly identified a 52-kDa protein formed bymRNA from T3-incubated follicles, while this proteincould not be detected with the mRNA from controlfollicles (Fig. 2). Addition of TIP to the follicle incuba-tion significantly increased P4 release; GTH also stimu-lated P4 release to a comparable extent (Fig. 3).

3.2. Differential modulation of 3b-HSD activities by TIPand GTH

To observe the difference between TIP and GTHstimulation on P4 secretion, experiments on D5-3b-HSDprotein synthesis were conducted. Fig. 4 shows that in-cubation of follicles with GTH caused a more thantwofold increase in D5-3b-HSD protein synthesis as de-termined from the radioactivity of the immunoprecipi-tate, while incubation with TIP did not stimulate theD5-3b-HSD protein synthesis even with five times moreTIP than GTH. Incubation of TIP with purified D5-3b-HSD enzyme in vitro greatly augmented the conversionof [3H]pregnenolone to [3H]progesterone, but incuba-tion of GTH with the enzyme did not alter its activity(Table 1). This indicates a modulation of the enzymeactivity in an in vitro cell-free system by TIP, whileGTH has no such effect.

3.3. Interaction between TIP and 3b-HSD

Modulation of D5-3b-HSD protein by TIP in a cell-free system was further clarified by results obtained froman experiment where D5-3b-HSD was incubated withTIP in the presence of NAD and pregnenolone and thenwas subjected to gel electrophoresis in a modified SDS–PAGE. A separate run of purified TIP or D5-3b-HSD

Fig. 2. mRNA isolated from control or T3-treated perch ovarian fol-

licles was subjected to in vitro translation in wheat germ lysate. The

translated product of each treatment was electrophoresed on SDS–

PAGE. Proteins from each lane were transferred to PVDF membrane,

which was then subjected to Western blot using anti-TIP antibody.

Fig. 3. Incubation of perch ovarian follicles in the presence of TIP

(5 lg/ml) or GTH (1 lg/ml) or in the absence of either of them (C). On

termination of incubation at 3 h, the media were collected for the es-

timation of P4 by RIA. Each observation is the mean�SEM of five

independent observations. �P < 0:01 compared with the control.

Fig. 4. Perch ovarian follicles were incubated either with TIP (5 lg/ml)

or with GTH (1 lg/ml) or in the absence of either of them (C) in

the presence of [14C]leucine and 18 other amino acids to monitor

D5-3b-HSD protein synthesis. At the end of the incubation, cells were

lysed by sonication and the D5-3b-HSD protein was precipitated by

anti-D5-3b-HSD-antibody and second antibody. Details of the experi-

ment were described in the text. Each value is the mean� SEM of five

independent observations. �P < 0:01 in comparison to control.

338 M. Datta et al. / General and Comparative Endocrinology 126 (2002) 334–341

was performed concurrently. Fig. 5 demonstrates astrong band of approximately 99 kDa and two bandscorresponding to TIP (52 kDa) and D5-3b-HSD(45 kDa), but these two bands corresponding to TIP andD5-3b-HSD were found to be weaker than their markers,although the same amount of protein was loaded in eachlane. The existence of a 99-kDa band appeared to beoccurring due to protein–protein interaction betweenTIP and D5-3b-HSD.

4. Discussion

This report shows that TIP gene expression in perchovary is specifically induced by T3. We have shownearlier that T3 significantly stimulates progesteroneformation in perch ovarian follicle which could be

blocked by a transcription inhibitor, actinomycin D,indicating the requirement of a protein mediator in thisaugmentory process. This protein, i.e., TIP, has beenpurified to homogeneity, it is a 52-kDa monomer, andits internalization into the follicle causes the stimulationof D5-3b-HSD activity, resulting in greater progesteroneformation (Bhattacharya et al., 1996). Similar molecularsize proteins have also been isolated from rat granulosacells and goat Leydig cells by inducing them with T3;both showed similar function, i.e., stimulation of steroidsecretion from these cells (Bandyopadhyay et al., 1996;Jana et al., 1996). We determined the N-terminal 17-amino-acid sequence of perch TIP and searched ho-mology in the vertebrate protein bank but failed to findone. This indicates TIP to be a novel protein, althoughto arrive at such a conclusion, further verifications, i.e.,complete sequencing of the TIP gene and protein wouldbe required.

TIP is a moderately large sized protein; decipherationof its gene and protein structure would need time.Probably, TIP is a functionally conserved protein; it hasthe same function in fish, rat, goat, and human gonadalcells and has similar molecular sizes in them (Bandyo-padhyay et al., 1996; Bhattacharya et al., 1996; Dattaet al., 1998; Jana et al., 1996). Again, addition of fishTIP to human corpus luteal cells in vitro greatly aug-mented progesterone release (Datta et al., 1998). Allthese indicate TIP to have an evolutionary conservednature. Our earlier observations on TIP suggest that T3

induces the generation of this protein in the gonadalcells, possibly by expressing its gene, as T3 has a ge-nomic receptor and is a known transcriptor (Bandyo-padhyay and Bhattacharya, 1994; Chakraborti et al.,1986; Oppenheimer et al., 1974; Tata, 1974). In thepresent study, we have shown that an oligonucleotideprobe developed on the basis of N-terminal 17-amino-acid sequences of TIP hybridizes only with RNA fromT3-incubated perch ovarian follicles and not with theRNA from control follicles. This appears to be a specifictranscriptional event as piscine GTH and hCG do notproduce hybridization signals. T3, therefore, specificallyinduces the expression of the TIP gene; how it does soremains to be studied in future.

We were confused at the beginning of our studyabout the function of TIP as we found that TIP is actingsimilarly to GTH; it is increasing the secretion of sexsteroids from the gonadal somatic cells by catalyzing theconversion of D5 hydroxysteroid precursors to D4 ketosteroids (Jana and Bhattacharya, 1994; Prasad et al.,1999). This suggests stimulation of D5-3b -HSD, whichis known to be a GTH function. However, augmenta-tion of D5-3b-HSD by thyroid hormone has been shownby others in mammalian gonadal cells (Gregoraszczuk etal., 1999; Hayashi et al., 1987; Simonian, 1986). Ourobservations showed that thyroid hormone does notstimulate this enzyme directly, it induces the generation

Fig. 5. Purified TIP (7.0 lg) and purified D5-3b-HSD (7.0 lg) were

incubated together or alone for 1 h in the presence of NAD (1 nmol)

and 0.125 lmol of pregnenolene. On termination of incubation, they

were subjected to SDS–PAGE. Experimental details are given in the

text. M, molecular weight marker; lane 1, D5-3b-HSD alone; lane 2,

TIP alone; lane 3, TIP plus D5-3b-HSD.

Table 1

Stimulation of D5-3b-HSD activity by TIP in an in vitro cell-free

system

Incubation mixture [3H]Progesterone formed

(nmol/mg protein)

[3H]Pregnenolone ND

3b-HSD+[3H]pregnenolone 14:1 � 0:12

TIP+[3H]pregnenolone ND

3b-HSD+TIP+[3H]pregnenolone 55:7 � 0:32�

3b-HSD+GTH+[3H]pregnenolone 14:3 � 0:32

Note. The incubation mixture contained 0.01 M phosphate buffer,

pH 7.4, and 1 mM NAD in a total volume of 1.5 ml; 10 lg of purified

D5-3b-HSD, 5.0 lg of TIP, and 1.0 lg of GTH were added wherever

mentioned. The D5-3b-HSD activity was determined on the basis of the

conversion of [3H]pregnenolone to [3H]progesterone/mg of enzyme

protein.* P < 0:01 compared to incubation with D5-3b-HSD alone.

M. Datta et al. / General and Comparative Endocrinology 126 (2002) 334–341 339

of TIP, which in turn augments enzyme activity (Ban-dyopadhyay et al., 1996; Bhattacharya et al., 1996; Janaand Bhattacharya, 1994; Prasad et al., 1999). Thequestion, therefore, arises how TIP can stimulate D5-3b-HSD and how its action is different from GTH aug-mentation.

GTH possibly regulates D5-3b-HSD gene expressionsvia cAMP, which exerts its effect through a cAMP re-sponse element, resulting in the increase of D5-3b-HSDmRNA (Chedrese et al., 1990; Payne and Sha, 1991;Simpson et al., 1990). We have also found support in thisdirection; in vitro incubation of perch Leydig cells withGTH causes the stimulation of D5-3b-HSD protein syn-thesis as determined by specific immunoprecipitation,while TIP has no such effect (Prasad et al., 1999). A similarobservation has been made with perch ovarian follicles.On the other hand, use of TIP in a cell-free system, i.e., itsaddition to the purified D5-3b-HSD enzyme, greatlyaugmented enzyme activity (Datta et al., 1999; Prasad etal., 1999). How TIP could stimulate D5-3b-HSD activityin in vitro incubation remains a puzzle.

During this investigation, a clue to resolving thispuzzle appeared to be available; i.e., there is an associ-ation of TIP and D5-3b-HSD proteins, suggesting aprotein–protein interaction between them which mightlead to the activation of D5-3b -HSD from its apparentlyless active state to a more active form. No doubt addi-tional evidence is necessary to elucidate this protein–protein interaction, but it is very intriguing to observethat the incubation of these two pure proteins in thepresence of enzyme cofactor and substrate leads to theactivation of D5-3b-HSD. Formation of a 99-kDa pro-tein out of TIP (a 52-kDa protein) and D5-3b-HSD (a45-kDa protein) is a highly meaningful insight in un-derstanding the mechanism. On the whole, we can nowsuggest that both GTH and thyroid hormone regulateD5-3b-HSD activity; the former does it by stimulatingprotein synthesis, possibly by expressing its gene, whilethe latter induces the expression of the TIP gene andformation of TIP protein, which then modulates enzymeactivity through protein–protein interactions. AlthoughTIP–D5-3b-HSD interaction requires additional studies,the present investigation shows a novel path of repro-ductive regulation.

Acknowledgments

This work was supported by a grant from the De-partment of Science and Technology (SP/SO/CO5/94)under the Ministry of Science and Technology, Gov-ernment of India. We are also indebted to the IndianInstitute of Chemical Biology, Kolkata, West Bengal,India, for extending the facilities for conducting thiswork.

References

Ball, J.N., 1960. Reproduction in female bony fishes. Symp. Zool. Soc.

1, 105–135.

Banerjee, P.P., Bhattacharya, S., Nath, P., 1989. Purification and

properties of pituitary gonadotropic hormone from an Indian

freshwater murrel, Channa punctatus, and carp, Catla catla. Gen.

Comp. Endocrinol. 73, 118–128.

Bandyopadhyay, S., Bhattacharya, S., 1993. Purification and proper-

ties of an Indian major carp (Cirrhinus mrigala, Ham.) pituitary

thyrotropin. Gen. Comp. Endocrinol. 90, 192–204.

Bandyopadhyay, A., Bhattacharya, S., 1994. Purification of putative

thyroid hormone receptor from the ovarian nuclei of freshwater

perch, Anabas testudineus. Fish Physiol. Biochem. 13 (5), 387–398.

Bandyopadhyay, A., Roy, P., Bhattacharya, S., 1996. Thyroid

hormone induces the synthesis of a putative protein in the rat

granulosa cell which stimulates progesterone release. J. Endocrinol.

150, 309–318.

Bhattacharya, S., Banerjee, J., Jamaluddin, M.D., Banerjee, P.P.,

Maitra, G., 1988. Thyroid hormone binds to human corpus

luteum. Experientia 44, 1005–1007.

Bhattacharya, S., Guin, S., Bandyopadhyay, A., Jana, N.R., Halder,

S., 1996. Thyroid hormone induces the generation of a novel

putative protein in piscine ovarian follicle that stimulates the

conversion of pregnenolone to progesterone. Eur. J. Endocrinol.

134, 128–135.

Chakraborti, P., Bhattacharya, S., 1984. Plasma thyroxine levels in

freshwater perch: influence of season, gonadotropins and gonadal

hormones. Gen. Comp. Endocrinol. 53, 176–186.

Chakraborti, P., Maitra, G., Bhattacharya, S., 1986. Binding of

thyroid hormone to isolated ovarian nuclei from freshwater perch,

Anabas testudineus. Gen. Comp. Endocrinol. 62, 239–246.

Chedrese, P.J., Zhang, D., Luu-The, V., Labrie, F., Juorio, A.V.,

Murphy, B.D., 1990. Regulation of mRNA expression of 3b-

hydroxysteroid dehydrogenase in porcine granulosa cells in culture:

a role of protein kinase C pathway. Mol. Endocrinol. 4, 1532–1538.

Comstock, J.P., Chan, L., Means, A.R., 1987. In: Schrader, W.T.,

O’Malley, B.T. (Eds.), Laboratory Methods Manual For Hormone

Action and Molecular Endocrinology, TX, 11th ed, p. 8-1.

Datta, M., Roy, P., Banerjee, J., Bhattacharya, S., 1998. Thyroid

hormone stimulates progesterone release from human luteal cells

by generating a proteinaceous factor. J. Endocrinol. 158, 319–325.

Datta, M., Nagendra Prasad, R.J, Bhattacharya, S., 1999. Thyroid

hormone regulation of perch ovarian 3b-hydroxysteroid dehydro-

genase/D5–D4-isomerase activity: involvement of a 52-kDa protein.

Gen. Comp. Endocrinol. 113, 212–220.

Goldman, S., Dirnfeld, M., Abramovici, H., Kraiem, Z., 1993.

Triiodothyronine (T3) modulates hCG-regulated progesterone

secretion, cAMP accumulation and DNA content in cultured

human luteinized granulosa cells. Mol. Cell. Endocrinol. 96, 125–

131.

Gordon, G., Southren, A.L., 1977. Thyroid hormone effects on steroid

hormone metabolism. N. Y. Acad. Med. 53, 241–259.

Gregoraszczuk, E.L., Skalka, M., 1996. Thyroid hormone as a

regulator of basal and human chorionic gonadotropin stimulated

steroidogenesis by cultured porcine theca and granulosa cells

isolated at different stages of the follicular phase. Reprod. Fertil.

Dev. 8, 961–967.

Gregoraszczuk, E.L., Galas, J., 1998. Triiodothyronine-dependent

activation of cAMP accumulation and steroids secretion by

follicular and thecal cells. Endocr. Regul. 32, 93–98.

Gregoraszczuk, E.L., Kolodziejczyk, J., Rzysa, J., 1999. Triiodothyr-

onine stimulates 3b-hydroxysteroid dehydrogenase activity in the

porcine corpus luteum. Endocr. Regul. 33, 155–160.

Hayashi, M., Maruo, T., Matsuo, H., Mochizuki, M., 1987. Effect of

thyroid hormone on steroidogenic enzyme induction in porcine

340 M. Datta et al. / General and Comparative Endocrinology 126 (2002) 334–341

granulosa cells cultured in vitro. Nippon Naibunpi Gakkai Zasshi

63, 1231–1240.

Ichikawa, M., Mori, T., Kawashima, S., Ueda, K., Shirahata, S., 1974.

Histological changes in the thyroid and interregnal tissue of the

kokanee (Oncorhynchus nerka) during sexual maturation and

spawning. J. Fac. Sci. Univ. 4, 175–182.

Jana, N.R, Bhattacharya, S., 1994. Binding of thyroid hormone to

goat testicular Leydig cell induces the generation of a proteina-

ceous factor which stimulates androgen release. J. Endocrinol. 143,

549–556.

Jana, N.R., Halder, S., Bhattacharya, S., 1996. Thyroid hormone

induces a 52 kDa soluble protein in goat testis Leydig cell which

stimulates androgen release. Biochim. Biophys. Acta 1292, 209–

214.

Laemmli, U.K., 1970. Cleavage of structural proteins during the

assembly of the head of bacteriophage T4. Nature 227, 680–685.

Leatham, J., 1972. Role of thyroid. In: Ballin, H., Glasser, S. (Eds.),

Reproductive Biology, Amsterdam, pp. 857–876.

Lowry, O.H., Rosenbrough, N.J, Randal, R.J., 1951. Protein mea-

surement with Folin phenol reagent. J. Biol. Chem. 193, 265–275.

Maitra, G., Bhattacharya, S., 1989. Seasonal changes of triiodothyr-

onine binding to piscine ovarian nuclei. Zool. Sci. 6, 771–775.

Mukherjee, D., Manna, P.R., Bhattacharya, S., 1994. Functional

relevance of luteinizing hormone receptor in mouse uterus. Eur. J.

Endocrinol. 131, 103–108.

Nagendra Prasad, R.J., Datta, M., Bhattacharya, S., 1999. Differential

regulation of Leydig cell 3b-hydroxysteroid dehydrogenase/D5–D4-

isomerase activity by gonadotropin and thyroid hormone in a

freshwater perch, Anabas testudineus (Bloch). Comp. Biochem.

Physiol. 124, 165–173.

Oppenheimer, J.H., Schwartz, H.L., Surks, M.I., 1974. Tissue differ-

ences in the concentration of triiodothyronine nuclear binding sites

in the rat liver, kidney, pituitary, heart, brain, spleen and testis.

Endocrinology 95, 897–903.

Payne, A.H, Sha, L., 1991. Multiple mechanisms for regulation of 3b-

hydroxysteroid dehydrogenase/D5–D4 isomerase, 17a-hydroxylase/

C17–20 lyase, and cholesterol side chain cleavage cytochrome P450

messenger ribonucleic acid levels in primary cultures of mouse

Leydig cells. Endocrinology 129, 1429–1435.

Simonian, M.H., 1986. ACTH and thyroid hormone regulation of

3beta-hydroxysteroid dehydrogenase activity in human fetal

adrenocortical cells. J. Steroid Biochem. 25, 1001–1006.

Simpson, E.R., Lund, J., Ahlgren, R., Waterman, M.R., 1990.

Regulation of cyclic AMP of the genes encoding steroidogenic

enzymes: when the light shines. Mol. Cell. Endocrinol. 70, C25–

C28.

Southren, A.C., Olivo, J., Gordon, G.G., Vittek, J., Briner, J., Rafii,

F., 1974. The conversion of androgens to estrogens in hyperthy-

roidism. J. Clin. Endocrinol. Metab. 38, 207–214.

Tata, J.R., 1974. Growth and development action of thyroid hormone

at the cellular level. In: Greer, A., Soloman, D.S. (Eds.), Handbook

of Physiology, vol. 3, section VII. American Physiological Society,

Washington, pp. 469–478.

Wakim, N.J., Ramani, N., Rao, Ch.V., 1987. Triiodothyronine

receptor in porcine granulosa cells. Am. J. Obstet. Gynecol. 156,

237–240.

Wakim, A.N., Polizotto, S.L., Buffo, M.J., Marrero, M.A., Burholt,

D.R., 1993. Thyroid hormones in human follicular fluid and

thyroid hormone receptors in human granulosa cells. Fertil. Steril.

59, 1187–1190.

Wakim, A.N., Paljug, W.R., Jasnosz, K.M., Alhakim, N., Brown,

A.B., Burholt, D.R., 1994. Thyroid hormone receptor messenger

ribonucleic acid in human granulosa and ovarian stromal cells.

Fertil. Steril. 62, 531–534.

M. Datta et al. / General and Comparative Endocrinology 126 (2002) 334–341 341