Human follicle-stimulating hormone (hFSH) and luteinizing hormone (hLH) are gonadotropins which are secreted as multiple forms by the pituitary. Evidence supporting the structural and functional heterogeneity of 15 purified hFSH isoforms and 20 purified hLH isoforms from pituitary extracts will be presented. Gonadotropin isoforms were purified by a combination of preparative isoelectric focusing and ion-exchange chromatography. The protein mass of each isoform was determined by amino acid analysis, which also correlated (data for hLH) (r = 0.999, P < 0.001, n = 15) with the UV area under the curve at 280 nm of the isoforms following gel-filtration HPLC. The alpha and beta subunits of FSH and LH were shown to be intact by SDS-PAGE under reducing condition, with no evidence of proteolytic nicking or presence of contaminating proteins. hFSH radioreceptor activity varied over a seven-fold range, and a positive correlation (r = 0.85, P < 0.001, n = 9) was observed between FSH receptor activity and the sialic acid (SA) content (1.5-13.7 mol SA/mol hFSH) of the isoforms, as determined by an HPLC-based microfluorometric assay. FSH in vitro activities varied over a similar range with a high correlation (r = 0.82, n = 15) with receptor activities, suggesting that the initial association of the hormone with the receptor is the key interaction with less differences attributed to subsequent effects in the signaling pathway. A similar result was seen with the hLH isoforms. To explore FSH/LH in vivo, the circulating half-life (LH/FSH) and the in vivo bioactivity (LH) using an acute in vivo assay was investigated. The clearance of hLH and hFSH showed a bi-exponential pattern for all isoform preparations with the proportion of the slower dissociating component (t 1/2 50-60 min) increasing three-fold with increasing sialic acid content of the isoform. The more rapidly cleared component (t 1/2 approx 10 min) is attributed to hepatically cleared gonadotropin, rather than gonadotropin equilibration between body compartments. The in vivo assay procedure for LH was based on the 24 h integrated plasma testosterone levels in rats following administration of graded doses of hLH isoform or standard. A 16-fold range in vivo activities between LH isoforms (n = 14) was observed. A comparison between hLH in vitro and in vivo activities showed a good correlation (r = 0.75) with the slope of the regression line (1.39) not significantly different from unity. These results suggest that in this acute in vivo assay method, the differences in circulating half-lives between hLH isoforms although large is not a key factor in their in vivo activity. However, in chronic in vivo assay systems the differences in clearance rates between isoforms may be important in their subsequent biological response. It is concluded that structural heterogeneity of FSH and LH contributes to functional differences, with a key interaction occurring at the receptor level. The contribution of sialic acid to these activities was also investigated.