Tryptophan sidechain dynamics in hydrophobic oligopeptides determined by use of 13C nuclear magnetic resonance spectroscopy.

Two oligopeptides, t-boc-LAWAL-OMe and t-boc-LALALW-OMe, were synthesized for the purpose of examining the sidechain dynamics of the tryptophan residue in hydrophobic environments by 13C nuclear magnetic resonance and fluorescence spectroscopy. In both peptides, the tryptophan sidechain was greater than 95% enriched with 13C at the C delta 1 position. Spin-lattice relaxation time (T1) and steady-state nuclear Overhauser effect (NOE) data were obtained at 50.3 and 75.4 MHz for both peptides in CD3OD, and at 75.4 MHz for t-boc-LALALW-OMe in lysolecithin-D2O micelles. We have adapted the model-free approach of G. Lipari and A. Szabo (1982, J. Am. Chem. Soc. 104:4546) to interpret the 13C-NMR data. Computer-generated curves based on experimental data obtained at a single frequency demonstrate relationships between an effective correlation time for tryptophan sidechain motion (tau e), a generalized order parameter (sigma) describing the extent of motional restriction, and an overall correlation time for the peptide (tau m). Assuming predominantly dipolar relaxation, least-squares fits of the dual frequency relaxation data provide values for these parameters for both peptides. The contribution of chemical shift anisotropy (CSA), however, is also explicitly assessed in the data analysis, and is shown to perturb the predicted sigma, tau e, and tau m values and to decrease chi(2) values observed in nonlinear least-squares analysis of the data. Because of uncertainty in the contribution of CSA to the relaxation of the indole ring 13C delta 1 atom, nonlinear least-squares analysis of the relaxation data were performed with and without inclusion of a CSA term in the appropriate relaxation equations. Neglecting CSA, an overall peptide correlation time of 0.69 ns is predicted for t-boc-LAWAL-OMe in CD3OD at 20 degrees C compared with 1.28 ns for t-boc-LALALW-OMe. Given these tau m values and taking into account the effect of measurement error in the T1 and NOE data, the internal dynamics of the tryptophan residue of t-boc-LAWAL-OMe in this isotropic environment are described by a range of tau e values from 70 to 112 ps and sigma values between 0.22 and 0.36. Similarly, for t-boc-LALALW-OMe, 68 less than or equal to tau e less than or equal to 93 ps and 0.09 less than or equal to sigma less than or equal to 0.17. The Ch-terminal position of the tryptophan residue in the hexapeptide may account for its lower order parameter. In lysolecithin micelles, the model-free approach applied tot-boc-LALALW-OMe predicts a Te between 0.87 and 1.08 ns, and an order parameter range of 0.72-0.80, assuming an average Tm of 14 ns (Saunders, L. 1966. Biochim. Biophys. Acta. 125:70) for a typical peptide-micelle complex. In this case, measurement of only two 13C relaxation parameters at a single frequency yields sufficient information, plotted in the form of a composite T1-NOE solution curve, to constrain the allowed values of the model-free motional parameters within a relatively narrow range. The predicted range of eV and Te values for the peptide-micelle complex demonstrate that both the rate and spatial mobility of the indole moiety are markedly restrained in the anisotropic micelle environment relative to free methanol solution. Steady-state fluorescence anisotropy measurements made on the peptides dissolved in methanol or with synthetic lysolecithins in water were used to calculate apparent order parameters for tryptophan motion; these values agree well with order parameters calculated from 13C NMR data. The reported results are relevant to the issue of protein dynamic events occurring on the picosecond time scale predicted by molecular dynamics simulations.