Development and application of a bioassay for follicle-stimulating hormone : a thesis presented in partial fulfilment of the requirements for the degree of Master of Science in Physiology at Massey University

Abstract

Follicle-stimulating hormone (FSH) is involved in the regulation and maintenance of vital reproductive processes, such as gametogenesis, follicular development and ovulation. Produced in the anterior pituitary, FSH is a glycoprotein hormone that exists as a family of isohormones. Follicle-stimulating hormone concentrations have traditionally been measured by radioimmunoassay (RIA). However, results generated using RIA are a determination of the immunological activity of FSH. The potential of FSH to generate a biological response cannot be measured by RIA. Therefore, the identification of physiologically significant differences in the activity of these isoforms requires the use of assay systems that can differentiate between the biological activity of the FSH isoforms. Commonly used assays for measuring the biological activity of FSH are based on the measurement of aromatase activity in cultured rat Sertoli cells following stimulation with FSH. However, these assays have an inherently high ethical cost involved due to the use of primary tissue culture. In addition, the variation in these assays associated with differences between animals is difficult to eliminate. Recently a bioassay for human FSH has been described based on FSH stimulation of cyclic AMP production by a Chinese hamster ovary (CHO) cell line stably expressing the human FSH receptor (FSH-R). The purpose of this study was to evaluate the potential usefulness of this CHO FSH-R cell line expressing the human receptor for FSH to be used as a bioassay to measure the biological activity of ovine FSH. The receptor cell line bioassay described in this study is based on the ability of FSH to stimulate cAMP production by cultured CHO FSH-R cells. Optimisation of the culture system to enable the bioactivity of ovine FSH to be measured by bioassay was undertaken. This involved optimising the density of cultured cells, the time in culture and time exposed to FSH and the most suitable dose range for FSH. The influence of matrix effects, such as those exerted by serum was also investigated. The specificity of the assay towards FSH was also determined as was the sensitivity, accuracy and precision of the assay. No stimulation of cAMP production was seen in CHO FSH-R cells following treatment with α-FSH, β-FSH, LH, TSH, GH, prolactin or vasopressin at concentrations up to 10 μg/ml. Although the methodology used differed slightly depending on the presence or absence of serum, all assayed were performed using the following methods and materials. Freshly thawed FSH-R cells were bulked up in culture, and aliquots of 1 x 105 to 5 x 105 cells/well dispensed into 48 well culture dishes and incubated overnight at 37°C. The assay culture media was then replaced with 0.25 ml fresh media (α-MEM + 0.1% BSA + 0.25 mM 3-isobutyl-1-methyl-xanthine) containing varying doses of NIH-FSH-RP2 (RP2) FSH preparations or FSH containing samples, and the cells incubated for 4 hours at 37°C. The assay culture media was then removed and stored frozen at -20°C until assayed for cyclic adenosine monophosphate (cAMP) by RIA. Once optimal assay conditions were determined, the CHO FSH-R cell bioassay was used to measure FSH concentrations in ovine serum, pituitary extracts and medium from cultures of ovine pituitary cells. It was found that the concentrations of FSH in serum from intact sheep was close to the detection limit of the assay. Thus, while FSH concentrations could be measured in serum from some sheep, other animals had concentrations that were too low to be accurately measured by the bioassay in its present form. The assay was, however, well suited to measuring FSH concentrations in serum from sheep that had elevated concentrations of FSH. In one study, FSH concentrations measured by the bioassay were compared to those measured by RIA in sheep that had been ovariectomised and then hypophysectomised. It was found that the profile of FSH concentrations following hypophysectomy was similar whether measured by RIA or by bioassay (R2=0.7513), though absolute concentrations sometimes differed. This suggested that the immunoassay and bioassay were not always measuring the same characteristics of FSH. The assay was also used to measure FSH concentrations in samples of ovine hypophyseal venous blood. However, the results obtained for these samples indicated a poor correlation between FSH concentrations obtained by bioassay and RIA. Levels of bioactive FSH in hypophyseal venous blood fluctuated markedly and were up to 10-fold higher than the associated RIA concentrations. The CHO-cell bioassay was also found to be very suitable for measuring pituitary concentrations of FSH. In one study, pituitary extracts underwent chromatography and the separated isoforms of FSH were analysed by bioassay and RIA. Again, there was excellent correlation (R2=0.9328) between the concentrations of FSH measured both assay types. However, some differences were apparent suggesting a discrepancy in the biological and immunological characteristics of different FSH isoforms. The bioassay was also used to measure FSH concentrations in media from pituitary cells in tissue culture where serially diluted samples displayed good parallelism with the RP2 FSH standard curve. Results of this study demonstrate that the CHO FSH-R cell bioassay is suitable for measuring the biological activity of ovine FSH in a variety of biological fluids. The use of a permanent cell line eliminates the high ethical cost associated with primary tissue culture that other bioassay systems have. The inherent variation associated with culture systems utilising tissue from different sources is also avoided. The sensitivity of the bioassay is suitable for measuring FSH in surgically altered sheep or hypophyseal blood concentrations where FSH levels are generally higher than those in the peripheral circulation. In addition to blood samples, the bioassay is also excellent for monitoring FSH activity in pituitary extracts and in media from tissue culture. However, the sensitivity of the bioassay currently does not always allow measurement of bioactive FSH concentrations in serum samples with low FSH levels. In summary, the CHO FSH-R cell bioassay described in this study offers a useful alternative to RIA and other bioassays for monitoring the biological activity of ovine FSH and its isoforms in various biological fluids. It is concluded that this convenient and robust bioassay may have considerable application in future investigations of ovine FSH bioactivity.