In vitro bioassay for human serum follicle-stimulating hormone (FSH) based on L cells transfected with recombinant rat FSH receptor: validation of a model system.

FSH plays a central role in normal reproductive function, i.e. control of follicular maturation in the female and initiation and maintenance of spermatogenesis in the male. The effects of FSH are mediated by its interaction with a specific receptor that belongs to the superfamily of guanine nucleotide-binding protein-coupled receptors. Due to the microheterogeneity of gonadotropins, measurement of immunoreactivity does not necessarily reflect their bioactivity. Mutations in gonadotropin beta-subunits, which affect bioreactivity and/or immunoreactivity of gonadotropins, have been described as causes of infertility, thus highlighting the need for rapid and convenient methods to measure bioactivity. To establish a model system for recombinant in vitro bioassays for FSH that would obviate the use of live animals, we developed a strategy for efficient expression of the rat FSH receptor (FSHR) in L cells. A cell line, FSHR 7/12, was developed that bound [125I]FSH with high affinity (Kd 1.42 nM) and responded to human FSH with an increase in cAMP accumulation. Untreated human serum was found to have an unspecific inhibitory effect on cAMP formation. This effect could be thoroughly avoided by mild heating (10 min at 56 C) of serum samples before addition to cells without detectable loss of FSH immunoactivity or bioactivity. Studies on the hormone-sensitive adenylyl cyclase system of transformed FSHR 7/12 cells and of the parental Ltk- cells showed that the cellular response to FSH was highly specific. Using a parallel line assay design, FSHR 7/12 cells were used to validate a novel recombinant in vitro bioassay relying on intracellular cAMP accumulation as a readout system. Up to 10% of serum could be added to the incubation buffer without leading to nonparallelism to the standard curve. When 70 serum samples of male patients attending an infertility clinic were analyzed, the novel assay system displayed high sensitivity and a close correlation (r > 0.8; P < 0.01) to the established rat Sertoli cell aromatase bioassay and to a highly specific fluoroimmunoassay. When sera of 25 normal menstruating women were analyzed for FSH bioactivity at different stages of the menstrual cycle, a midcycle FSH peak followed by a decline in the late luteal phase could be discerned. The analysis of 26 serum samples of postmenopausal women revealed a close correlation between FSH values obtained by the novel in vitro bioassay and by a fluoroimmunoassay (r = 0.90; P < 0.01). Thus, the present in vitro bioassay represents a sensitive, rapid, and convenient model system to measure bioactive FSH in human serum.