This study investigated the ability of insulin and of insect insulin-like peptides (ILPs) to stimulate ovarian steroidogenesis in the blowfly Phormia regina. Bovine insulin was active on ovaries isolated in vitro, which showed an age-dependent sensitivity; this peptide progressively stimulated steroidogenesis in ovaries isolated from the third day after adult molt, but not in younger ones, and had maximal activity after the fifth day. This stimulatory effect was observed equally from females reared in the presence or in the absence of males, excluding a regulatory effect of mating. The mode of action of insulin in blowflies did not involve cAMP, but triggered a specific and well-conserved transduction cascade. In particular, a peroxovanadium compound, known to activate specifically the insulin receptor in mammals, also stimulated blowfly ovarian steroidogenesis in vitro. Conversely, chemicals known to inhibit the mammalian insulin receptor or downstream elements of its signaling pathway, such as LY294002, an inhibitor of phosphatidylinositol 3-kinase (PI3K), were able to prevent the steroidogenic action of bovine insulin on fly ovaries. Extracts from the median neurosecretory cells (MNCs) of blowfly brains, which are known to contain endogenous ILPs, stimulated ovarian steroidogenesis very efficiently and were also sensitive to inhibition by LY294002. These experiments indicated the involvement of PI3K in the mode of action of MNC extracts and substantiated that their endogenous ILPs are involved in the regulation of ovarian steroidogenesis. This conclusion was corroborated by the effects of synthetic bombyxin II, an ILP originating from silkworm MNCs, which also stimulated steroidogenesis in isolated blowfly ovaries. Altogether, these data suggest that insulinlike neurohormones from MNCs play a crucial role as steroidogenic gonadotropins in female flies.

Calcium is frequently involved in the stimulation of steroidogenesis in gonads and endocrine glands, generally in association with cAMP. However, our present observations show that it has the opposite effect in the ovary of the blowfly Phormia regina. Our in vitro experiments first showed that extracellular calcium does not play a role during the stimulation of steroidogenesis in fly ovaries; indeed steroidogenesis was activated in vitro as efficiently in a medium with or without calcium, either by pharmacological compounds mimicking cAMP signaling or by active brain extracts. When calcium was experimentally introduced into biosynthetic cells by ionophores or liberated from internal stores by thapsigargin, it did not activate, but clearly inhibited both basal and acute steroidogenesis respectively in previtellogenic and in vitellogenic ovaries. Our experiments also demonstrated that calcium decreases cAMP concentrations in the ovaries of Phormia, by stimulating its degradation, without modifying its biosynthesis. Moreover, inhibitors of calcium-calmodulin phosphodiesterases (PDEs) increased steroid biosynthesis in vitro, whereas inhibitors of calciuminsensitive PDEs did not. These data thus demonstrate that, in blowfly ovaries, calcium ions inhibit cAMPstimulated steroidogenesis by activating a calmodulinsensitive (type I) PDE.

Previous investigations in the female blowfly Phormia regina have shown that 3-isobutyl-1-methylxanthine (IBMX), a broad spectrum inhibitor of phosphodiesterases (PDEs), fails to mimic the steroidogenic effects of cAMP on ovaries, although it efficiently increases the concentrations of this second messenger. In this study, experiments carried out to clear up this contradiction demonstrated that IBMX, besides its effect on cAMP, also increased cGMP concentrations in blowfly ovary and that these two cyclic nucleotides controlled ovarian steroidogenesis antagonistically. In particular, a selective inhibitor of cGMP-specific PDEs, unlike IBMX, had a very strong negative effect on ovarian steroidogenesis. Moreover, a cGMP analog was able to inhibit steroid biosynthesis in previtellogenic and vitellogenic ovaries, thus affecting basal and acute steroidogenesis respectively. Our observations also demonstrated that cGMP was always present in blowfly ovary, reaching its maximal levels at the end of vitellogenesis, in close correlation with the physiological decrease in ovarian steroidogenesis. Experiments using an inhibitor of protein kinase G clearly indicated that the effects of cGMP were mediated by this enzyme. On the contrary, these effects did not seem to involve cGMPregulated PDEs or ion channels. Our results also indicated that ovarian cGMP concentrations were not controlled by brain factors, suggesting a probable involvement of paracrine/autocrine factors. Nitric oxide (NO) appeared to be a good candidate for such a control, because an NO donor was able to stimulate ovarian cGMP concentrations and to drastically decrease ovarian ecdysteroid biosynthesis in blowflies. These data thus demonstrate, for the first time in invertebrates, a potent role of cGMP in the negative control of ovarian steroidogenesis and suggest a possible co-regulation with NO.