Conversion of human choriogonadotropin into a follitropin by protein engineering.

Abstract

Human reproduction is dependent upon the actions of follicle-stimulating hormone (hFSH), lutenizing hormone (hLH), and chorionic gonadotropin (hCG). While the a subunits of these heterodimeric proteins can be interchanged without effect on receptor-binding specificity, their (3 subunits differ and direct hormone binding to either LH/CG or FSH receptors. Previous studies employing chemical modifications of the hormones, monoclonal antibodies, or synthetic peptides have implicated hCG (-subunit residues between Cys-38 and Cys-57 and corresponding regions of hLHP and hFSH(3 in receptor recognition and activation. Since the P subunits of hCG or hLH and hFSH exhibit very little sequence similarity in this region, we postulated that these residues might contribute to hormone specificity. To test this hypothesis we constructed chimeric hCG/hFSH ( subunits, coexpressed them with the human a subunit, and examined their ability to interact with LH and FSH receptors and hormone-specific monoclonal antibodies. Surprisingly, substitution of hFSH(3 residues 33-52 for hCG(3 residues 39-58 had no effect on receptor binding or stimulation. However, substitution of hFSH( residues 88-108 in place of the carboxyl terminus of hCG(3 (residues 94-145) resulted in a hormone analog identical to hFSH in Its ability to bind and stimulate FSH receptors. The altered binding specificity displayed by this analog is not attributable solely to the replacement of hCGI3 residues 108-145 or substitution of residues in the "determinant loop" located between hCG(3 residues 93 and 100.Follicle-stimulating hormone (FSH; follitropin), luteinizing hormone (LH; lutropin), chorionic gonadotropin (CG; choriogonadotropin), and thyroid-stimulating hormone (TSH; thyrotropin) constitute a family of heterodimeric glycoproteins that regulate the development and function of the ovary, testis, and thyroid (1). Within a species the a subunit is encoded by a single gene and can be interchanged between hormones without effect on receptor binding, whereas the (3 subunits differ and direct binding specificity (2, 3). Both subunits are required for full biological activity. On the basis of sequence differences between the P subunits and the results of chemical modification studies it has been proposed that amino acid residues between Cys-93 and Cys-100 in human CG ( subunit (hCG() and the corresponding regions of the other hormones form a "determinant loop" which directs hormone specificity (4). Recent studies have demonstrated that synthetic peptides containing the determinant loop region from hCG(3 (5) or hFSH,8 (6) inhibit the binding of 125I-labeled hormone to receptors, albeit at high concentrations (10-100 EM). The FSH peptide also stimulates granulosa cell steroidogenesis. In addition, phosphorylation of hCG(3 residue Thr-97 disrupts binding to LH receptors (7).The region of the ( subunit between Cys-38 and Cys-57 (hCG numbering) has also been proposed to play an important role in receptor binding and stimulation. Synthetic peptides containing hCG, hLH (5...

Receptor-selective determinants in catfish gonadotropin seat-belt loops

“…In human gonadotropins, the net charge difference between their determinant loops determines the ability of hCG and hFSH to interact with the LH-R (Campbell et al, 1991;Moyle et al, 1994;Han et al, 1996). Accurate comparison of superimposed crystal structures of hCG and hFSH revealed significant differences in surface charge characteristics between the hFSH␤ and hCG␤ determinant loops (Fox et al, 2001): three Asp residues in FSH form a negatively charged patch on one side of the determinant loop.…”

“…amino acid residue 93 of human FSH␤; see Fig. 3), the determinant loop of human FSH␤ is not essential for FSH binding to the hFSH-R, but restrains the interaction with the hLH-R (Lindau-Shepard et al, 1994;Campbell et al, 1991;Moyle et al, 1994). The determinant loop of hCG␤, on the other hand, and in particular its net positive charge, was found to be crucial for LH-R-stimulating activity, without influencing FSH-R-recognition Han et al, 1996).…”

“…3), this region appeared to be of utmost importance for FSH activity in all species tested. Placing the human FSH␤ Cys 11-12 loop in the context of hCG, allowed this hCG analog to interact with the FSH-R (Campbell et al, 1991;Moyle et al, 1994), whereas substituting the FSH␤ Cys 11-12 loop with the corresponding hCG␤ loop significantly impaired the binding to the FSH-R . Subsequent Ala-scanning analysis delineated the amino acid triplet Arg-Gly-Leu as being crucial for FSH-R binding (Lindau-Shepard et al, 1994).…”

“…Based on the property to bind both FSH-R and LH-R, Moyle and co-workers (Campbell et al, 1991;Moyle et al, 1994;Han et al, 1996;Moyle et al, 1998) proposed that an hCG␤ chimera that harbors its native, positively charged determinant loop in conjunction with the hFSH␤ Cys 11-12 region represents an omnipotent ancestral ␤-subunit. According to this model, LH␤/CG␤-like hormones and FSH␤/TSH␤-like hormones have evolved from this common ancestor, predominantly through sequence divergence of their seat-belt loops.…”

Only a Portion of the Small Seatbelt Loop in Human Choriogonadotropin Appears Capable of Contacting the Lutropin Receptor

“…This led to the speculation, which was made several years before it could be tested experimentally, that the charge of this loop "determines" receptor binding specificity (11). Replacing the positively charged residues in the hCG seatbelt loop with their hFSH counterparts reduced LHR binding 10 -15-fold but did not convert hCG to a follitropin (4). Changing the specificity of hCG required replacing the C-terminal half of its seatbelt, which we term the strap, with its hFSH counterpart (4,5).…”