Amide Hydrogen-Deuterium Exchange in Isotopically Mixed Water

Hydrogen-deuterium exchange (HDX) is a well-established method for studying protein conformational dynamics. In pure deuterium oxide (D2O), the classical Linderström-Lang (LL) model has been used to interpret HDX data by measuring protection factors. However, when experiments are conducted in isotopically mixed water—such as a 50% H2O/D2O solution—quantitative interpretation becomes challenging. This is because back exchange and isotope effects are not fully accounted for in the LL model. Prior research has shown that protection factors derived from the LL model in mixed solvents may not accurately reflect true protein dynamics. This gap motivated the development of a new framework to address the limitations of existing models in mixed isotopic environments.

The aim of this research is to introduce a generalized Linderström-Lang (GLL) framework that improves the accuracy of HDX data interpretation in isotopically mixed water. The specific problem addressed is the inability of the classical LL model to account for both forward and reverse exchange in mixed solvents. This limitation leads to systematic underestimation of protection factors, which are critical for understanding protein conformational behavior. The motivation for this study stems from the need for a more precise analytical approach to extract meaningful data from HDX experiments. By developing the GLL model, the researchers aim to provide a tool that can recover true protection factors even in mixed solvent conditions. This approach is particularly valuable for experiments where pure D2O is not feasible or where mixed solvents are necessary for experimental design.

The study introduces a generalized Linderström-Lang (GLL) model that incorporates both forward and reverse exchange processes in isotopically mixed water. The model is derived analytically to describe equilibrium enrichment and protection factors in mixed solvents. The researchers applied the GLL framework to hydrogen-deuterium exchange nuclear magnetic resonance (HDX/NMR) experiments on the molecular chaperone DNAJB1 in 50% D2O. The model accounts for isotope effects and changes in protection upon isotopic substitution. The analytical solutions were validated by comparing results from the GLL model with those from the classical LL model. The study also demonstrates how a single HDX experiment in a mixed solvent can provide multiple types of information, including protection factors and fractionation factors. The approach relies on precise measurement of deuteration levels and careful modeling of exchange rates.

Application of the GLL model to HDX/NMR experiments on DNAJB1 in 50% D2O revealed that protection factors derived from the GLL model match those observed in pure D2O. In contrast, using the classical LL model in mixed solvents leads to systematic underestimation of protection factors. The GLL framework successfully recovers true protection factors by accounting for back exchange and isotope effects. The model also provides fractionation factors that are sensitive to the local hydrogen-bonding environment. These findings suggest that the GLL model offers a more accurate interpretation of HDX data in mixed solvents. The results demonstrate that a single experiment in isotopically mixed water can yield both protection and fractionation factors simultaneously. The study shows that ignoring back exchange in mixed solvents results in significant errors in protection factor estimation. The GLL model thus enables more reliable analysis of conformational dynamics and local stability in proteins.

The researchers conclude that the generalized Linderström-Lang (GLL) model provides a more accurate framework for interpreting hydrogen-deuterium exchange (HDX) data in isotopically mixed water. The model accounts for both forward and reverse exchange processes, which are not captured by the classical LL model. The study demonstrates that protection factors derived from the GLL model in mixed solvents match those observed in pure D2O. This suggests that the GLL model can recover true protection factors even when experiments are conducted in mixed solvents. The researchers also highlight that a single HDX experiment in a mixture can yield both protection and fractionation factors. These findings support the use of the GLL model for more precise analysis of protein conformational dynamics. The study shows that ignoring back exchange in mixed solvents leads to systematic underestimation of protection factors. The GLL model thus offers a valuable tool for improving the accuracy of HDX data interpretation.