Immunoradiometric assay for follistatin: serum immunoreactive follistatin levels in normal adults and pregnant women.

Abstract

A sensitive and specific immunoradiometric assay for follistatin was developed using antifollistatin mouse monoclonal and rabbit polyclonal antibodies. The sensitivity of the assay was 0.5 micrograms/L, and cross-reactivities with recombinant human activin A and bovine inhibin were less than 0.1%. The intra- and interassay coefficients of variation were less than 10%, and the recovery rate was about 90% in human serum. The addition of activin A to the same sample resulted in a minimal influence on follistatin recovery, indicating that this assay system can measure the total level of activin-bound and unbound follistatin. Gel filtration analysis of human serum showed that the majority of immunoreactivity was eluted in a larger molecular size position than that of free follistatin, suggesting that the large part of follistatin is bound to other proteins, presumably activins, in serum. Using this assay, immunoreactive follistatin levels in various biological fluids and human sera were examined. The dose-response curves of porcine follicular and amniotic fluids were parallel to the standard curve, and porcine follicular fluid contained extremely high follistatin immunoreactivity (5.6 mg/L). The serum follistatin level in normal human volunteers was 13.3 +/- 4.7 micrograms/L (mean +/- SD; n = 60), with a tendency to increase gradually with age. On the other hand, the serum follistatin level was remarkably elevated in pregnant women (62.7 +/- 35.3 micrograms/L; n = 57), with a positive correlation with weeks of pregnancy. These data indicated that circulating immunoreactive follistatin is detectable in human serum, and the levels vary with physiological conditions such as aging and pregnancy.

Development, validation and application of an ultra-sensitive two-site enzyme immunoassay for human follistatin

“…Several groups agree that, in the serum of pregnant women, total follistatin concentrations rise steadily with increasing length of gestation (Khoury et al 1995, Wakatsuki et al 1996. Our data support these findings (maternal serum follistatin concentrations: first trimester 0·847 0·26 ng/ml, third trimester 2·80 0·93 ng/ml).…”

“…This included normal male serum, hypogonadal male serum, normal female serum, serum samples from pregnant women and serum samples from postmenopausal women. Wakatsuki et al (1996) reported higher levels of follistatin in postmenopausal women. Whilst Gilfillan and Robertson (1994) also reported an increase with age, the magnitude was much smaller than in the Wakatsuki study.…”

“…This seems to be the case since Khoury et al (1995) report serum concentrations of 8·09 ng/ml in the normal human menstrual cycle, compared with 0·5-1·0 ng/ml in the present study. Wakatsuki et al (1996) report human female serum levels as 12·5 ng/ml. It is not known which forms of follistatin are present in most tissues and fluids and thus it is possible that mixtures may be found.…”

“…Gilfillan and Robertson (1994) reported that in their assay the potencies of other forms relative to the bFS 35 kDa standard were as follows: 31 kDa, 4·1, 39 kDa, 2·6 and 45 kDa, 5·9. Wakatsuki et al (1996) developed an assay for FS315 but provided no evidence for lack of reactivity with the other forms. Wang et al (1996) developed an assay for FS288 but provided no evidence that FS315 did not cross-react.…”

Follistatin serum concentrations during full-term labour in women – significant differences between spontaneous and induced labour

“…At least two groups indicate that their assay determines the levels of unbound follistatin (Wang et al 1996, McConnell et al 1998, whilst the other assays measure total (bound and unbound) follistatin (Sugawara et al 1990, Klein et al 1991, Gilfillan & Robertson 1994, Khoury et al 1995, Wakatsuki et al 1996, Evans et al 1998, McConnell et al 1998, de Kretser et al 1999, O'Connor et al 1999.…”