Successful analog design approaches to be used on peptides the size of the GHRHs can be aided enormously by an examination of solution conformation, usually by circular dichroism measurements and 2D nuclear magnetic resonance (NMR) studies. It can be
Fig. 2. Molecular modeling of [D-Ala2, Ala8,9,15,16,18,22,24-28]GHRH(1-28)NH (JF-01-40; Table 2) in its projected amphiphilic a-helical state from residues 6-29. The hydrophilic side-chains, consisting of the basic Arg and Lys residues and Tyr residues, are shown in a lighter shade and the hydrophobic residues in the dark shade. Note how they align respectively on two surfaces of the molecule (left panel, amino terminus at the left of the picture). This is particularly evident from the end-on view of the molecule (right panel).
quite hard or impossible to infer a receptor-bound conformation from solution data, however, in the case of GHRH the tremendous tendency of virtually the whole sequence to adopt an a-helical conformation points strongly towards this also holding true when the peptide is complexed to the receptor. Thus, although in water alone there is limited helicity, this increases to 80-90% in 50% aqueous alcohol (23,24) and 65-70% in the presence of phospholipid liposome (23). Helical tendencies of the whole chain were also indicated from two-dimensional proton NMR data (25,26). Furthermore, the a-helix formed appears to be of the amphiphilic type (27,28), in that it has two distinct surfaces on which either hydrophilic and hydrophobic amino-acid side-chains cluster. A molecular model of this effect can be seen in Fig. 2, in which a helically-stabilized, Ala-substituted analog (to be discussed in detail in a later section) was assembled and energy-minimized using the SYBYL program. More sophisticated computer molecular modeling, taking into account solution NMR NOE distance data and restrained molecular dynamics simulation, pointed to a structure having helices from position 6-13 and 16-29, but a more flexible short P-strand from the N-terminus to position 5 (29). In general, this is in agreement with the other data already discussed. This N-terminal flexibility has been investigated by analog studies in which increased conformational restriction is introduced, for instance, by the use of D-amino acids.
A more simplistic approach to computer conformation prediction—Chou-Fasman sequence analysis (30)—has also been useful for delineating helical, p-sheet, and folding motifs in peptides of this size and was used by us initially to provide design information. The results of this technique applied to GHRH(1-29)NH2 are shown in the top panel of Fig. 3 and, again, the high probability of the 16-29 sequence being helical is indicated. This method, however, did imply some structural ambiguity in the 8-11 region, which was an impetus to early analog design strategies aimed at probing this region also.
Was this article helpful?