Multiple molecular weight forms of immunoreactive alpha-inhibin in human seminal plasma.

The purpose of these studies was to determine whether the total immunoreactive alpha-inhibin protein concentration in seminal plasma correlated with serum gonadotropin levels or semen characteristics and to identify the forms of alpha-inhibin present in seminal plasma. Thirty-eight serum samples from men being evaluated for infertility were selected for study based on their serum hormone profiles and semen parameters. Serum LH and testosterone levels were normal, but FSH levels ranged from normal to hypergonadotropic (> 20 IU/L). Most semen parameters were within normal ranges, but germ cell numbers ranged from normal to azoospermic. Thus, seminal plasma from these men provided a unique opportunity to examine the antigenic forms of alpha-inhibin in individuals in whom strong correlations between inhibin and FSH levels might be predicted because of the observed ranges of FSH levels and germ cell numbers. Seminal plasma alpha-inhibin was characterized by RIA or Western blotting, using an antiserum directed against the N-terminal of the alpha-subunit of mature [32,000 mol wt (M(r))] inhibin. The antiserum recognized the alpha-subunit of dimeric inhibin as well as free alpha-inhibin and alpha-inhibin precursor proteins. Total immunoreactive alpha-inhibin ranged from 8.21-43.99 nmol/L in seminal plasma. However, alpha-inhibin levels were not statistically correlated with serum FSH levels or any of the measured semen parameters (including germ cell number). In contrast, the immunoreactive alpha-inhibin concentration in seminal plasma was negatively correlated (P < 0.01) with the serum LH level. Western blot analyses revealed that multiple forms of immunoreactive alpha-inhibin are present in seminal plasma. The majority of immunoreactivity was associated with monomeric proteins (ranging from 58,000-95,000 M(r)) that were larger than the alpha-subunit (21,000 M(r)) predicted for mature dimeric human inhibin (32,000 M(r)). The relative amounts of individual forms of immunoreactive alpha-inhibin varied among the patients studied, but could not be correlated with other serum or seminal parameters measured. Our observations demonstrate that various monomeric alpha-inhibin proteins are present in human seminal plasma. It is unlikely that these proteins alone or combined with inhibin beta-subunit proteins have identical biological activities. Thus, until assays specific for each of the various forms of immunoreactive alpha-inhibin are developed, their role as well as that of inhibin in the endocrine or local modulation of testicular function cannot be deduced from RIA data alone.