N. Tjandra et al., ANISOTROPIC ROTATIONAL DIFFUSION OF PERDEUTERATED HIV PROTEASE FROM N-15 NMR RELAXATION MEASUREMENTS AT 2 MAGNETIC, Journal of biomolecular NMR, 8(3), 1996, pp. 273-284
N-15 NMR relaxation times in perdeuterated HIV-1 protease, complexed w
ith the sub-nanomolar inhibitor DMP323, have been measured at 600 and
360 MHz H-1 frequency. The relative magnitudes of the principal compon
ents of the inertia tensor, calculated from the X-ray coordinates of t
he protein-drug complex, are 1.0:0.85:0.44. The relation between the T
-1/T-2 ratios observed for the individual backbone amides and their N-
H orientation within the 3D structure of the protease dimer yields a r
otational diffusion tensor oriented nearly collinear to the inertia te
nser. The relative magnitudes of its principal components (1.00:1.11:1
.42) are also in good agreement with hydrodynamic modeling results. Th
e orientation and magnitude of the diffusion tensors derived from rela
xation data obtained at 360 and 600 MHz are nearly identical. The anis
otropic nature of the rotational diffusion has little influence on the
order parameters derived from the N-15 T-1 and T-2 relaxation times;
however, if anisotropy is ignored, this can result in erroneous identi
fication of either exchange broadening or internal motions on a nanose
cond time scale. The average ratio of the T-1 values measured at 360 a
nd 600 MHz is 0.50+/-0.015, which is slightly larger than the value of
0.466 expected for an isotropic rigid rotor with tau(c)=10.7 ns. The
average ratio of the T-2 values measured at 360 and 600 MHz is 1.14+/-
0.04, which is also slightly larger than the expected ratio of 1.11, T
his magnetic field dependence of the T-1 and T-2 relaxation times sugg
ests that the spectral density contribution from fast internal motions
is not negligible, and that the chemical shift anisotropy of peptide
backbone amides, on average, is larger than the 160 ppm value commonly
used in N-15 relaxation studies of proteins.