Scientists tried swapping a normal peptide bond for a special chemical link in a growth‑hormone‑releasing peptide (GRF‑1‑29) to see how it changes activity. Most of the modified versions barely activated the hormone release, and one version actually blocked the natural peptide at relatively high concentrations.

Previous peptide structure-activity investigations employing the psi[CH2NH] peptide bond isostere have produced antagonists when inserted into various sequences. These include bombesin, in which the incorporation of Leu13 psi[CH2NH]Leu14 produced a potent antagonist, and tetragastrin, with which Boc-Trp-Leu psi[CH2NH]Asp-Phe-NH2 is an antagonist. In this study, we chose to investigate the effect of this isostere on growth hormone-releasing factor (1-29) amide. Analogues were prepared by solid-phase synthesis and the isosteres incorporated by racemization-free reductive alkylation with a preformed protected amino acid aldehyde in the presence of NaBH3CN. The aldehydes were prepared by the reduction of the protected N,O-dimethyl hydroxamates with LiAlH4 at 0 degrees C. The purified analogues were assayed in a 4-day primary culture of male rat anterior pituitary cells for growth hormone (GH) release. Potential antagonists were retested in the presence of GRF(1-29)NH2. The following results were obtained: At position 5-6, a very weak agonist was produced with much less than 0.01% activity. Incorporation of the isostere in positions 1-2, 2-3, and 6-7 gave weak agonists with approximately 0.1% activity. Agonists with 0.39% and 1.6% activity were produced by incorporation at 10-11 and 3-4, respectively. The analogue [Ser9 psi[CH2NH]Tyr10]GRF(1-29)NH2 was found to be an antagonist in the 10 microM range vs 1 nM GRF and had no agonist activity at doses as high as 0.1 mM.