Scientists tested many versions of the growth‑hormone‑releasing peptide (GRF) to see how changes in its building blocks affect its ability to turn on an enzyme in rat pituitary cells. They found that swapping a specific amino‑acid at position 2 to its mirror‑image form (D‑Arg) made the peptide work even better, while the same swap in a slightly modified version turned the peptide into a blocker that stops the normal GRF from working.

The efficacy and potency of 14 GH-releasing factor (GRF) analogs, substituted in position 1 to 7, on adenylate cyclase activation in crude homogenates from rat anterior pituitary were related to those of human pancreatic GRF(1-29)-amide and vasoactive intestinal peptide. Among several D-amino acid substitutions, that in position 2 was the only one to yield a super-agonist [with a Kact (concentration required for half-maximal adenylate cyclase activation) 2 times lower than that of GRF(1-29)-NH2]. By contrast, D-isomer substitution in position 1 and 3 was without effect and D-isomer substitution in position 4, 6, or 7 decreased the affinity of the analog. The N-acetylated analog of GRF was as potent and active as the parent peptide, and the identity of the amino acid in position 2 of (N-Ac-Tyr1)-GRF(1-29)-NH2 proved to be determining for enzyme activation, with D-Phe2 and D-Trp2 derivatives acting as partial agonists and the (N-Ac-Tyr1,D-Arg2) analog being an efficient competitive antagonist of GRF(1-29)-NH2. With use of this antagonist, it was possible to demonstrate that GRF and vasoactive intestinal peptide receptors represent distinct entities in the rat anterior pituitary.