J couplings are essential for measuring RDCs (residual dipolar couplings), now routinely used to deduce molecular structure and dynamics of glycans and proteins. Accurate measurement of (1)J(CH) is critical for RDCs to reflect the true structure and dynamics in the molecule of interest. We report noticeable discrepancies between (1)J(CH) values measured with HSQC type pulse sequences in the (1)H dimension from those measured in the (13)C dimension for 17 sugars and show that these discrepancies arise from strong scalar coupling. In order to determine how to minimize errors in measuring (1)J(CH), we analyze the strong coupling effects in detail using the product operator-formalism and spectral simulations based on the solution of the Liouville equation (not considering relaxation effects) in the presence of strong coupling. We report that the apparent (1)J(CH) measured with 2D HSQC-based sequences in either dimension can be in error by up to 4 Hz and that the values measured in the (1)H dimension can disagree with those in the (13)C dimension by up to 7 Hz. We demonstrate that spectral simulations can reproduce the errors induced by strong coupling and that these can be used to extract true (1)J(CH) values. We find that the (1)J(CH) values measured using a modified Z-filtered coupled HSQC are still affected by strong coupling. We conclude that spectral simulation yields accurate (1)J(CH) with errors as low as 1% in the presence of strong coupling.
Published by Elsevier Inc.