Currently, little is known about the role of TGF-β in regulating ovarian functions in lower vertebrates, such as fish. Recent studies have suggested that TGF-β inhibits steroid production in the goldfish ovary  and oocyte maturation in zebrafish . In the present study, we further examined the cellular mechanisms underlying the inhibitory effect of TGF-β in oocyte maturation. We have shown that TGF-β1 inhibits mRNA expression of 20β-HSD, the key enzyme involved in MIH production, as well as LHR and mPR-β. These novel findings suggest that TGF-β1 inhibits multiple targets in the oocyte maturation pathway, both upstream and downstream of MIH.
One of the potential targets of TGF-β1 identified in this study is 20β-HSD. We found that TGF-β1 inhibited basal and hCG-induced 20β-HSD mRNA levels. TGF-β1 alone induced a dose- and time-dependent inhibitory effect on 20β-HSD mRNA expression. Treatment with hCG increased 20β-HSD mRNA levels; however, in the presence of TGF-β1, the effect of hCG was blocked. 20β-HSD activity in the ovary and its stimulation by gonadotropins and cAMP-enhancing drugs has been demonstrated in many species [10, 11, 24–28]. Our finding that hCG stimulates 20β-HSD is consistent with studies in mammals [24, 28] and a recent study in Nile Tilapia, which reported an increase in the mRNA expression of 20β-HSD in response to hCG treatment . The observation of decreased 20β-HSD mRNA levels after TGF-β1 treatment suggests that TGF-β1 may inhibit 20β-HSD activity, leading to a decrease in MIH production. This notion is supported by a recent report that TGF-β1 inhibits the conversion of 17α-HP to DHP in the goldfish . It is possible that one of the actions of TGF-β1 is to decreases MIH production, and thus inhibits oocyte maturation.
In this study, we observed that hCG increased LHR but had no effect on FSHR mRNA levels. Several studies conducted in fish on the effect of gonadotropin on LHR and FSHR expression have yielded inconsistent results. In African and Channel catfish, it has been reported that hCG treatment caused an activation of the LHR and cAMP mediated pathways, as well as a slight increase in FSHR mRNA levels [29–31]. However, a two-fold increase in FSHR expression, but no change in LHR expression in pre-maturational follicles of rainbow trout in response to treatment with partially purified gonadotropins has also been reported . The reason for such discrepancy is unclear, however, it could be due to species specificity or variation in treatment conditions.
An interesting finding from this study is that TGF-β1 differentially regulates FSHR and LHR mRNA levels. We observed that basal FSH receptor mRNA levels were increased, while basal- and hCG-induced LH receptor mRNA levels were decreased, upon incubation of follicles with TGF-β1. TGF-β1 has been reported to regulate LH and FSH binding sites in mammals, however, whether TGF-β1 has a stimulatory or inhibitory effect appears to be species-dependant. TGF-β1 decreased basal and FSH-induced LH binding sites in porcine granulosa cells but enhanced FSH-induced LH binding in rat granulosa cells . FSH induced FSH binding in porcine granulosa cells and this effect was attenuated by TGF-β1. On the other hand, FSH decreased its own binding levels in rat granulosa cells and this effect was also blocked by TGF-β1 . Our finding that TGF-β1 increased FSHR mRNA levels is consistent with previous studies in rat granulosa cells [34, 35]. However, our observation that TGF-β1 decreased LHR mRNA levels is opposite to studies in rat  and chicken  granulosa cells. In these studies, it was reported that TGF-β1 induced LHR mRNA expression. Whether this is due to species variation as in the case of rat and pig or to the difference in follicle development between fish and higher vertebrates awaits more studies in fish species. The finding that TGF-β1 inhibited LHR mRNA expression suggests that the inhibitory effect of TGF-β1 on hCG-induced oocyte maturation may be due, in part, to the downregulation of LH receptor by TGF-β1. Since FSH is known to be a major regulator of oocyte growth [10, 11], by stimulating FSH receptor expression TGF-β1 may play a role in promoting follicle growth. These findings, together with our previous observation that TGF-β1 mRNA levels are higher in growing follicles than in maturing follicles, suggest that TGF-β1 may stimulate follicle development and inhibit precocious oocyte maturation in the zebrafish ovary. This possibility will be investigated further in the future.
A recent study has shown that microinjection of zebrafish oocytes with antisense oligonucleotides to either mPR-α or mPR-β or both causes similar marked decreases in the rates of oocyte maturation, suggesting that both subtypes are obligatory for oocyte maturation in zebrafish . In this study, we observed a strong induction of mPR-β mRNA levels by hCG and an inhibitory effect of TGF-β1 on both basal and hCG-induced mPR-β mRNA expression. These findings support the role of mPR-β in oocyte maturation and suggest that one of the mechanisms by which TGF-β1 inhibits hCG- and MIH-induced oocyte maturation is by the downregulation of MIH receptors, specifically mPR-β. However, we found that neither hCG nor TGF-β1 had an effect of mPR-α mRNA levels, suggesting that the two membrane progestin receptors are under differential regulation. It has been reported that hCG caused an increase in the mPR protein expression in sea-trout oocytes . The sea-trout mPR has a higher homology to mPR-α (80%) than to mPR-β (46%). It remains to be determined if hCG regulates mPR-α and mPR-β protein levels in the zebrafish. Recently, Kazeta et al. (2005) reported that hCG did not change mRNA levels of mPR-α and mPR-β at 5 and 10 h after hCG treatment . The difference between this and our studies may be due to the duration of hCG treatment as 18 h was used in our study.