Follistatin is a glycosylated monomeric protein originally isolated from ovarian follicular fluid based on its ability to specifically inhibit pituitary FSH release. To further explore the physiological role of follistatin, we have expressed recombinant human follistatins with 315 (rhFS-315) and 288 (rhFS-288) amino acids in Chinese hamster ovary cells under the control of the simian virus-40 promoter. The two types of FS originated from alternatively spliced mRNAs and rhFS-315 differed from rhFS-288 by having an extra 27-amino acid sequence at the carboxyl-terminal. The yield of the purified rhFS-315 and rhFS-288 after a single step of affinity chromatography on an activin-coupled Affi-Gel column was 3-5 mg/liter conditioned medium. Using the rhFS-315 and rhFS-288 as molecular mass markers, Western blotting with FS carboxyl-terminal-specific antibodies demonstrated that the majority of native FS isolated from porcine ovarian follicular fluid was neither FS-315 nor FS-288, but was composed of 300 amino acids in various forms of glycosylation. This finding is consistent with our earlier results obtained from tryptic peptide fragment analysis of native FS. Only a very small percentage (less than 1%) of native porcine FS was FS-288. In cultures of rat anterior pituitary cells, rhFS-315 (ED50, 115.2 +/- 16.2 pM) is equipotent to native porcine FS (ED50, 86.7 +/- 14.1 pM) on the suppression of FSH release, but, surprisingly, rhFS-288 (ED50, 9.6 +/- 2.2 pM) is 8-10 times more potent than the native protein, similar to the potency of inhibin-A (ED50, 8.6 +/- 0.9 pM). Interestingly, when the in vivo FSH-suppressing activity of rhFS-288 was compared to that of inhibin-A in 1-week ovariectomized adult rats, it was found that rhFS-288 was more potent and longer acting than inhibin-A. Hence, these results indicate that FS-288 is probably one of the most potent natural FSH suppressors.