Ligand-receptor interactions of Gonadotropin-releasing hormone antagonists

Assefa, Daniel

Ligand-receptor interactions of Gonadotropin-releasing hormone antagonists

Thesis / Dissertation

1999

Abstract

The interactions of selected GnRH agonists and antagonists with the GnRH receptor were investigated using site directed mutagenesis and photoaffinity cross/inking of the receptor with a labelled GnRH analog. With the aim of identifying residues involved in ligand binding, the relative affinities of five peptide and one non-peptide GnRH antagonists towards three closely related mammalian GnRH receptors (mouse, sheep and human) were determined using cells transiently expressing GnRH receptors. All of the peptide antagonists [Ac-3-Pro 1 , D-4- F-Phe2 , 2-D-Naf, D-lpr-Lys6 , /pr-Lys8 ]GnRH, [2-Ac-D-NaJ1, D-Me-4-CI-Phe, 3-D-Paf, D-Arg6 , D-AlaNH2 10JGnRH, [2-Ac-D-NaJ1, D-4-CI-Phe2 , 3-D-Paf, Nic-Lys5 , D-Nic-Lys6 , /pr-Lys8 , D-AlaNH/0 JGnRH, [Ac-D-NaJ1, D-4-CI-Phe2 , D-Paf, D-Lys(3-pAc)6, lpr-Lys8 , D-AlaNH2 10JGnRH and [2-Ac-D-NaJ1, D-4-CI-Phe2 , 3-D-Paf, 1-Me-Paf, D-Trp6 , DAlaNH2 10JGnRH were found to have a higher potency for the inhibition of GnRH stimulated inositol phosphate by the mouse and sheep receptors than by human GnRH receptor. The greatest differences in affinity were exhibited by the antagonist, [2-Ac-D-NaJ1, D-Me-4-CI-Phe, 3-D-Paf, D-Arg6 , D-AlaNH2 10]GnRH. This antagonist showed an affinity difference of 54 fold between mouse and human receptors and 68 fold between sheep and human receptors. These results suggest that mutation of residues which differ between human and sheep receptors could be used to identify specific residue(s) involved in the binding of this peptide antagonist. This line of investigation was not pursued further, however, in view of more promising possibilities with non-peptide antagonists. FE 101170, a non-peptide GnRH antagonist was found to have 300-fold higher affinity for the human receptor than the sheep receptor. The regions of the receptor responsible for the differential affinity were explored using point-mutated and human/sheep chimeric GnRH receptors. These receptors had one or more residue of the human GnRH receptor replaced by the corresponding sheep residues. Because of the small size and hydrophobic nature of the antagonist, emphasis was made on TM residues. hF227L and L 78V mutant receptors showed no change in agonist or antagonist (FE 101177) affinity. The substitution of the lone charged residue in TM4 Arg179 by G/y caused a small (4 fold) decrease in affinity to FE 101177. Multiple substitution of the TM4 residues (V160I, 165L, V169/) and of the TM1 residues (ASOT, T51 I, A54T) did not produce any significant changes in agonist or antagonist affinity. The D302N mutant receptor, which lacks the charged Asp residue at position 302, showed a minimal change in its affinity to an agonist and about 12 fold decrease in 3 affinity for FE 101177. More extensive alterations of the receptor were made by constructing receptor chimeras. The substitution of the extracellular N-terminus of the human receptor by the corresponding sheep receptor residues did not cause any change in agonist or antagonist binding. However, when all the sheep residues up to the extracelluar end of TM2 were incorporated, a selective decrease in the affinity for FE 101177 by of 20 fold was observed. Extending the sheep amino acid contribution to the extracellular end of TM4 resulted in a further decrease of the affinity by 40 fold. The results suggest that the interaction of FE 101170 with the GnRH receptor involves a number of residues and that the binding are not fully accounted for by TM domains alone. In addition the results show that the interactions are not purely additive, and that cooperative effects are involved in binding of this ligand. The interaction of the GnRH receptor with peptide antagonists was also investigated by photoaffinity cross/inking of the GnRH receptor. 1 O photoreactive GnRH analogues containing either 4-azidobenzoic acid or 4-azidosalicy/ic acid attached to the c-amino group of lysine or the {J-amino group of diaminopropionic acid were synthesised and characterized. Of these, [N-Azidobenzoyl-Ac-D-Lys1 , D-4-CI-Phe2 , D-Trp3 , D-Arg6 , DAlaNH/0 JGnRH, named PAnt-1, was found to be a high affinity (Ki= 3.8 ± 0.04 nM) antagonist possessing a high cross/inking efficiency of > 70%. The peptide was shown to cross/ink to the GnRH receptor in a specific and functionally relevant manner. [Ac-D-Na/1, N-Azidobenzoy/-D-Dpt, D-Trp3 , D-Arg6 ]GnRH and [Ac-D-Na/1, NAzidosa/isy-D-Dpt, D-Trp3 , D-Arg6 ]GnRH were found to have a moderate affinity. However photolabelling of the GnRH receptor using their iodinated analogs could not be demonstrated. [Ac-D-4-C/-Phe1, Ac-D-4-CI-Phe2 , D-Trp3 , N-Azidobenzoy/-D-Lys6 , D-AlaNH2 10]GnRH had high affinity for the GnRH receptor in the absence of UV induced photolysis. Cross/inking of this peptide with the GnRH was demonstrated. The agonist [D-Ala6 , N-Azidobenzoyl-Om8 ]GnRH was found to have an affinity and potency comparable to that of mammalian GnRH. However, it had a very low efficiency of cross/inking as sustained release of inositol phosphate after cross/inking with the GnRH receptor could not be demonstrated. In addition, photolabelling of the receptor could not be demonstrated with either [125I-D-Tyr, D-Ala6 , N-AzidobenzoylOm8 ]GnRH or the related [125I-D-Ty/, D-Ala6 , N-Azidosalicyl-Om8 ]GnRH. 4 The photoreactive analogs [N-azidobenzoyl-L-Dpl, D-Ala6 ]GnRH, [N-azidobenzoyl-LDrp3 , D-Ala6 ]GnRH and [N-azidosalisyl-Dpr, D-Ala6 ]GnRH, with modifications at residues 2 and 3 residue were found to have low affinity. Wild type human, mouse and sheep GnRH receptors, as well as a panel of mutant GnRH receptors, were used in peptide mapping studies to define the attachment site of the radioiodinated photoreactive analog PAnt-1. Photoaffinitiy cross/inking of the mouse receptor with PAnt-1 followed by deg/ycosylation and digestion with endo proteinase Glu-C yielded a 13 kDa labelled fragment. The mouse EBQ and E90Q mutants yielded larger fragments, indicating the labelled fragment in the mouse receptor consists of residues G/n9 -Glu90. Glu-C digestion of human and sheep receptors cross/inked with PAnt-1 peptide gave a labelled band of approximately 8 kDa, consistent with cleavage at residues Glu68 in both species and at residues Glu8 (human) and G/u11 (sheep). Digestion of the human N19E mutant GnRH receptor yielded a very small labelled fragment indicating that the attachment site of the photoreactive antagonist to the human GnRH receptor lies between residues 12 and 18 in the extracellular N-terrninus. Comparison of Glu-C cleavage products from PAnt-1 cross/inked receptors under reducing and non-reducing conditions directly demonstrated the presence of a disulphide bridge Cys114-Cys196. In addition the results suggested that the cross/inking site was Cys14 in the extracellular N-terrninal domain. These findings suggest that related peptide antagonists bind to the GnRH receptor with their N-terrninus in the vicinity of the disufide bridge near the second extracelluar loop. In Summary, molecular determinants of the binding of peptide and non-peptide antagonists to the GnRH receptor have been investigated in the present thesis. The major findings are: (1) The identification of a peptide GnRH antagonist with differential affinity between human and sheep GnRH receptors, which could be used in future studies to map its binding site. (2) The identification of a set of residues which contribute to the differential affinity of the non-peptide GnRH antagonist FE 101177 towards sheep and human receptors. (3) The synthesis and chracterization of PAnt-1, a novel photoreactive peptide GnRH antagonist with high affinity and cross/inking efficiency (4) The localization of the attachment site of PAnt-1 to a 7-residue segment of the Nterrninal domain of the receptor, with Cys14 the putative cross/inking site.

Keywords

Pathology

Reference:

Collections

PhD / Doctoral

Full item page