Effects of J couplings and unobservable minor states on kinetics parameters extracted from CEST data

Chemical exchange saturation transfer (CEST) experiments have emerged as a powerful tool for characterizing dynamics and sparse populated conformers of protein in slow exchanging systems. We show that J couplings and 'invisible' minor states can cause systematic errors in kinetics parameters and chemical shifts extracted from CEST data. For weakly coupled spin systems, the J coupling effect can be removed using an approximation method. This method is warranted through detailed theoretical derivation, supported by results from simulations and experiments on an acyl carrier protein domain. Simulations demonstrate that the effect of 'invisible' minor states on the extracted kinetics parameters depends on the chemical shifts, populations, exchange rates of the 'invisible' states to the observed major or minor state and exchange models. Moreover, the extracted chemical shifts of the observed minor state can also be influenced by the "invisible" minor states. The presence of an off-pathway folding intermediate in the acyl carrier protein domain explains why the exchange rates obtained with a two-state model from individual residues that displayed only two obvious CEST dips varied significantly and the extracted exchange rates for ¹⁵N and ¹³CO spins located in the same peptide bond could be very different. The approximation method described here simplifies CEST data analysis in many situations where the coupling effect cannot be ignored and decoupling techniques are not desirable. In addition, this study also raises alerts for 'invisible' minor states which can cause errors in not only kinetics parameters but also chemical shifts of the observed minor state.