The use of peptides to reconstruct conformational determinants; a brief review.

In many biological processes, defined regions of proteins are involved in selective recognition. These regions can often be mimicked with peptides and are the main targets for vaccine and drug development. The authors review the use of peptides, to define and ultimately mimic defined protein regions of interest. Especially the role of the Pepscan method is emphasized. This method has been proven to be a useful and fast tool in defining protein regions of interest. It is based on the simultaneously synthesis of multiple peptides coupled to solid supports. Hundreds of peptides can be produced and tested in a relatively short period of time. With the construction of random peptide libraries in recombinant DNA systems, it is now even possible to screen for peptidic determinants without the requirement of preliminary knowledge of primary structure. Having this information, the affinity of peptides can be further enhanced with the Pepscan approach. The power of this approach will be illustrated with results from studies on the development of synthetic vaccines and hormone analogues.

Synthetic peptides based upon a three-dimensional model for the receptor recognition site of follicle-stimulating hormone exhibit antagonistic or agonistic activity at low concentrations.

Follicle-stimulating hormone (follitropin, FSH) belongs to a group of closely related glycoprotein hormones that contain two noncovalently linked dissimilar subunits designated alpha and beta. By using synthetic peptides, several receptor interaction sites in these hormones have been identified; however, the peptides have a reduced potency (lowest effective concentration of 10(-4) to 10(-5) M) relative to the hormone itself (10(-8) to 10(-11) M). This suggests that the peptides represent only a portion of a larger recognition site in the intact hormone that comprises parts of both the beta and the alpha chains. To develop peptides that exhibit FSH-antagonistic activity at low concentrations, we have constructed a three-dimensional model for FSH, which is based on an alignment of both the beta and the alpha chains of glycoprotein hormones with thioredoxin, for which x-ray diffraction data are available. This model resulted in the prediction of a conformational receptor-binding site in FSH, in which (parts of) three earlier proposed binding regions on the FSH molecule [namely, the regions FSH alpha-(34-37), with the amino acid sequence SRAY; FSH beta-(40-43), with the amino acid sequence TRDL; and FSH beta-(87-94), the "determinant loop" with the amino acid sequence CDSDSTDC] are located within 10 A of one another. On the basis of this model, peptides have been synthesized in which two of these binding regions are linked by a synthetic amino acid whose length was derived from the model, Ac-TDSDS-NH-(CH2)5-CO-SRAY-NH2 and Ac-SRAY-NH-(CH2)4-CO-TRDL-NH2. Both peptides inhibited FSH-induced cAMP production in Sertoli cells at 1000-fold lower concentrations (10(-7) M) than the peptides Ac-TRDL-NH2, Ac-SRAY-NH2, or Ac-TDSDS-NH2. In another peptide, Ac-TDSDS-NH-(CH2)5-CO-SRAY-NH-(CH2)4-CO-TRDL-NH2, all three binding regions have been linked. This peptide appeared to be a strong agonist of FSH action, as measured by the ability to stimulate cAMP production, at concentrations as low as 10(-7) M. The observation that a synthetic peptide, in which (parts of) three earlier described receptor interaction sites are combined according to the three-dimensional model, can mimic the action of FSH, at 10(-7) M, shows that this model is useful to predict a conformational receptor-binding site in FSH and that combination of only a few amino acid residues from the alpha and beta chains of FSH in a small synthetic peptide is sufficient to transduce a signal upon binding to the receptor.