Studies of the conformational behaviour and preferential interactions with opiate receptors of the cis and trans forms of [Dmet2,pro5]enkephalin and [Dmet2,pro5]enkephalinamide by 1H and 13C NMR, theoretical calculations and 13C relaxation measurements.

Abstract

[DMet2, Pro5]Enkephalin (I) and [DMet2, Pro5]enkephalinamide (II), two of the more potent opioid peptides, have been studied by 1H and 13C NMR and by theoretical calculations. Their biological activity is tentatively related to their conformational behaviour. A cis⇌ trans equilibrium around the Phe-Pro bond is brought about by the terminal prolyl residue. The cis/trans ratios lie around 50/50 for peptide I (55%cis in dimethylsulfoxide, 35 %cis in water) and around 20/80 for peptide II, the more potent compound, in both the solvents. Almost all the signals, both in 1H and 13C spectra, give separate resonances for the cis and trans rotamers, in dimethylsulfoxide as well as in water. A temperature variation allows one to determine the rotational barrier (ΔG+TC, ∼ 81.5 kJ/mol) for the cis⇌trans interconversion. The analysis of the vicinal coupling constants and theoretical calculations suggests that Tyr-dMet-Gly-Phe-Pro exists as a class of highly puckered conformations with short end-to-end distances at r= 0.4 nm and r= 0.6 nm for the trans and cisconfigurations respectively. The populations of the side-chain rotamers are predominantly trans/gauche (tg) for the tyrosine residue. This feature, which is quite general in enkephalins, is related to the crucial role of the tyramine moiety for opioid activity. The orientation of the Phe-4 side-chain is independent of the spatial disposition of the following prolyl residue. pH titrations provide evidence for the existence of head-to-tail interactions. Hence, in (2H6)dimethylsulfoxide, and to a lesser extent in 2H2O, the titration of the Pro-5 residue is clearly felt on the Tyr-1 moiety, whereas the deprotonation of this latter residue leads to changes in the chemical shifts of the pyrrolidine ring protons. The β protons of the Tyr-1 and Phe-4 residues are strongly magnetically nonequivalent; the opposite is true for Met-enkephalin where they show the same chemical shifts below pH 6. The titration effects suggest that no strong conformational change occurs during the amino or carboxyl group titration and indicate an important steric hindrance on the NMR time scale for the two aromatic side chains. The 13C NMR spectra of peptide I in (2H6)dimethylsulfoxide and in a mixture of (2H6)dimethylsulfoxide/2H2O (507sol;50) exhibit for almost all the resonances well-separated signals, corresponding to the cis and trans conformers. The large changes in 13C chemical shifts occurring Qn the backbone and side-chain carbons during the titration steps confirm the through-space effects already deduced from the proton titration curves. Hence, several C-α (Met-2, Phe-4) and C-β (Met-2, Pro-5) atoms are affected by both the titrations of the two terminal parts of the peptide. Such effects were not observed in Met-enkephalin or Leu-enkephalin. The 13C relaxation times for the cis and trans rotamers were compared, in (2H6)dimethylsulfoxide, under strictly the same conditions. All the NT1 values are small and lie in the same range both for the carbons of the peptide backbone and for the carbons of the lateral chains. The restriction in the segmental motion of the sidechains is shown by the NT1 values of the 13C of the pyrrolidine ring and by the C-β and C-γ of the methionine residue which lie near those of the backbone carbons. No significant internal rotation of the aromatic ring of the tyrosyl residue seems to occur. A restricted degree of freedom for the C-α-C-β carbons is shown for the phenylalanyl residue which contrasts with Metenkephalin, Leu-enkephalin and a larger peptide such as α-endorphin. All these results, both in dimethylsulfoxide and in water seem to exclude a random conformation for the compounds I and II. The titration effects are related to strong compression effects due to the presence of a d met-2 residue wedged between two relatively inflexible aromatic residues. Moreover. the results indicate that, despite the absence of a classical β-turn, these compounds exhibit folded structures induced by their high content of hydrophobic side chains. The conformational behaviour of peptides I and II can be related to their high specificity for the μ-receptor site. Hence, it can be noticed that, in the trans form of the peptides the first part, Tyr-DMet-Gly, may adopt the angular requirements (ψ1=−169°, φ2= plus;83°, ψ2=+53°, φ3=+150°) of the active conformation that we recently proposed for the interaction at the μ-receptor site. Moreover it is possible that the trans form, more populated in the highly potent compound II, might better represent the active form in the interaction at the receptor site.