Editorial: Inhibin in the Male--Progress at Last

Inhibin is a glycoprotein hormone, which inhibits the production and/or secretion of the pituitary gonadotropins, preferentially FSH. (1). The ability of aqueous rodent testicular extracts to inhibit the formation of castration cells in the anterior pituitary, whereas organic solvent extracts failed to do so, provided the basis of the “inhibin hypothesis” more than 60 years ago (2). Subsequently, both human seminal plasma (3) and bovine testicular extracts (4) were shown to suppress specifically FSH levels in castrate rats and sheep, respectively, consistent with the possibility that the testis was capable of inhibin production and confirmed by the selective FSH suppression induced by the infusion of recombinant human inhibin into several species (5, 6). Cultured rat Sertoli cells were shown to be capable of secreting bioassayable inhibin, and it was thus postulated that the testicular source was the Sertoli cell (7). Inhibin was first purified from ovarian sources (for review, see Ref.8), and bovine ovarian inhibin was used as the antigen in one of the first inhibin RIAs to be fully characterized and applied to the study of inhibin physiology (9), an RIA known widely as the Monash assay. The cloning of human inhibin showed that there were two major species, inhibins A and B, dimers which shared a common α-subunit, disulphide-linked to one of two different β subunits,β A or βB. Both of these inhibins were shown to be detectable by the RIA, as were peptides related to the inhibinα -subunit. While the application of the RIA to studies of inhibin physiology in the female has generated data consistent with a physiological role for inhibin in women (10), problems arose with data generated in men. Initial studies as reviewed (11) were promising. Immunoreactive inhibin levels were elevated at the time of hypothalamo-pituitary-testicular axis activation in young boys and rose during normal human puberty, being correlated with rising levels of the gonadotropins and testosterone. Levels fell with increasing age in men, were suppressed by exogenous testosterone administration, and responded to the administration of both FSH and human CG, suggesting the possibility of dual gonadotropic control, a somewhat unexpected finding. In contrast, in states of disordered spermatogenesis, leading to human male infertility, no correlation could be found between FSH and immunoreactive inhibin, with levels in severe spermatogenic failure, e.g. Klinefelter’s syndrome, often being normal or even raised (12). Elaboration of the inhibin hypotheses had postulated that it was a factor produced by the seminiferous epithelium (specifically the Sertoli cell), involved in the regulation of FSH, whereas Leydig cell testosterone was the primary feedback factor for LH. Testosterone was known to be capable of FSH suppression. The failure to find a correlation between FSH and inhibin in disorders of the seminiferous epithelium was puzzling. When it became clear that inhibin α-subunit related peptides were present in the circulation (13) and that cultured rat testicular Leydig cells could secrete immunoreactive inhibin and could be stimulated by LH (14), an explanation for the paradoxical findings in Klinefelter’s syndrome and other forms of male infertility could be postulated. It was possible that bioactive inhibin normally secreted by the Sertoli cell was absent but that the Leydig cells, and where evident, Sertoli cells remaining in the testis of severely infertile men, were capable of producing immunoreactive inhibin species without biological activity, hence leading to elevated FSH levels. The recent and clearly very important development of assays specific for the dimeric inhibins, A (15) and B (16), has shed some light on the situation, particularly the recent demonstration that inhibin B, which is inducible by exogenous FSH (17), is the only inhibin detectable in adult men and that its levels are very low in men with Klinefelter’s syndrome and other forms of severe male infertility. A recent and rapidly growing body of evidence is consistent with inhibin B being the relevant physiological inhibin involved in FSH negative feedback in the male. In this issue (18), evidence is reported that inhibin B is present in the circulation of the juvenile Rhesus monkey whose pituitary-testicular axis has been driven by intermittent iv GnRH infusion, though it is not clear from the data presented that FSH specifically stimulated levels of inhibin B in this situation. While persuasive evidence is accumulating that inhibin B may represent a physiologically relevant species, no clear cut explanation has been found for the quantitatively much larger amounts of α-subunit related immunoreactivity present in male serum and apparently responsive to FSH stimulation as shown both in the human male and in the present report in juvenile Rhesus monkeys. What is the function of α subunit-related peptides? Do they necessarily arise from the testis? Some evidence suggests that they may not (19). Further complexities have been identified by the recognition that inhibin in serum is present in a variety of molecular weight forms due, in the main, to cleavage at specific sites on both subunits. Thus, inhibin is found in human serum as the mature 30K (αβ) dimer with extensions of either or both subunits resulting in additional mol wt species of 50–70K and 90–110K (20). It would also appear that inhibin A and B are processed differentially in serum, with the α-subunit precursor of inhibin B more readily cleaved than that of inhibin A (21). Recent studies by Mason (22) have showed that the in vitro biological activity of noncleavable high molecular weight forms of inhibin produced by recombinant expression was markedly reduced, suggesting that processing of inhibins may be critical for biological activity. It would thus appear that inhibin biological activity is dependent on the degree of subunit processing either in the circulation or at the pituitary. While results from the present paper highlight some of these structural complexities, it does appear that there are differences between the human and monkey. In the human, α-subunit immunoassays including the Monash RIA are able to readily detect high molecular weight forms of inhibin (20, 21), whereas in the monkey, even though high molecular weight forms are measurable as seen in Figs. 1 and 2 of Majumdar et al. (18), they are seemingly poorly detected, at least in stored serum (Fig. 4). It will clearly be important to resolve the basis for these apparent differences particularly if there are questions of assay specificity. Over a number of years, Plant and colleagues have undertaken a series of studies using elegant physiological models to explore the regulation by testicular factors of FSH secretion in monkeys. Earlier studies examined the effects of inhibin immunoneutralization (23) and recombinant inhibin administration on FSH secretion (6), as well as the contribution of testosterone to suppressing FSH. In this issue (18), these authors have continued these studies by exploring the stimulatory effects of of FSH and LH (human CG) on inhibin production. FSH showing a clear stimulatory effect on inhibin levels in contrast to the absence of effects of CG despite supraphysiological circulating testosterone levels. These studies have established the overall importance of inhibin, and the limited importance of testosterone in regulating FSH in the monkey and extend studies undertaken in other species. The present discussion has focused only on the characteristics of the long loop feedback system between the pituitary and the testis. The local intrapituitary production of the inhibin-related peptides (11) no doubt complicates any of the interpretations proposed. It is also interesting to comment on the complexity of the pituitary-testicular feedback system. Not only is it clear that a variety of inhibin species is present in the circulation, at least some of which appear very likely to be of testicular origin, it is also clear that both FSH and LH exist in multiple isoforms of various biological and immunological activity. It is difficult to postulate reasons for the apparent complexity of the reproductive axis control system, in contrast with what appear to be relatively much simpler systems in the control of thyroid and adrenocortical function.