Conformational dynamics are essential for protein functions like enzyme catalysis and allosteric regulation. Studying these motions at atomic resolution is crucial for understanding biological mechanisms and for protein engineering. However, techniques like X-ray crystallography and cryo-EM provide ensemble-averaged data, obscuring details of spatiotemporal coupling and transient intermediate states. Time-resolved X-ray crystallography (TRX) offers a solution by capturing high-resolution structural snapshots as the protein relaxes from a perturbed state. Existing TRX methods often rely on protein-specific perturbations like photocaged ligands or rapid mixing. This research presents a more universally applicable approach: coupling a solvent-based temperature jump (T-jump) with TRX. This leverages the inherent link between temperature and macromolecular dynamics, enabling the study of proteins whose intrinsic dynamics cannot be easily triggered by light or mixing. The study uses lysozyme as a model system to demonstrate the method's effectiveness in visualizing intrinsic conformational dynamics and the influence of inhibitor binding on these dynamics.

The study of protein dynamics has been a significant focus in structural biology. While techniques like X-ray crystallography have provided static structural information, understanding the dynamic nature of proteins is essential. Time-resolved crystallography has emerged as a powerful tool in this regard, allowing researchers to observe structural changes over time. However, challenges remain in developing universal methods to initiate these changes. The use of photoactive proteins and photocaged ligands has proven useful in some contexts. But the universal nature of thermal fluctuations in the solvent suggests that temperature manipulation could be a widely applicable perturbation. Previous studies using temperature jump with small-angle X-ray scattering have shown some success, and multitemperature crystallography experiments have demonstrated that varying temperature alters conformational equilibrium, laying the groundwork for the present approach. This study builds upon this prior work by combining temperature jump with serial femtosecond crystallography (SFX) for high-resolution, time-resolved observation of protein dynamics.

This research used a pump-probe configuration combining an infrared (IR) laser with high-resolution SFX. A nanosecond pulsed IR laser, tuned to excite the O-H stretching mode of water, was used to induce a rapid temperature jump in lysozyme microcrystals. A custom-built microfluidic device delivered crystals to the pump-probe interaction region. Ultrafast XFEL pulses probed the sample at various time delays (20 ns, 20 μs, and 200 μs) after the T-jump. Diffraction images were collected using a multi-panel CCD detector. Data processing involved a pipeline for real-time hit identification and integration, followed by more rigorous offline processing and merging. To confirm the T-jump, the isotropic diffuse scattering signal from the solvent was analyzed using singular value decomposition (SVD). Changes in unit cell dimensions and atomic B-factors were also assessed. Difference electron density maps, calculated by subtracting dark-state measurements from T-jump measurements, were used to visualize time-resolved structural changes. Weighted difference maps were generated using a scheme developed to improve estimations of difference structure factors from noisy data. The spatial distribution of changes was quantified using the integrated absolute difference density above a noise threshold (IADDAT). Modeling of time-dependent conformational changes was performed using distinct approaches for nanosecond and microsecond timescales. For nanosecond timescale changes, a simulation of B-factor increases was generated and compared to experimental maps. For microsecond timescales, alternative conformations were manually built and refined against extrapolated structure factor magnitudes (ESFMs). Analogous experiments were performed with lysozyme bound to chitobiose, a natural inhibitor, to assess the effect of inhibitor binding on the observed dynamics.

The T-jump successfully introduced a temperature change of approximately 21 K, confirmed by the analysis of diffuse scattering and unit cell expansion. At short time delays (20 ns), widespread atomic vibrations were observed, consistent with increased harmonic motion due to heat transfer from the solvent. At longer time delays (20–200 µs), these vibrations evolved into localized, coordinated motions in functionally relevant regions, particularly near the active site. The analysis of IADDAT values revealed non-uniform spatial distribution of difference density, highlighting specific regions of enhanced dynamics. Modeling of the observed changes revealed short-amplitude motions (e.g., rotamer flips) on fast timescales and larger motions (e.g., backbone shifts) on slower timescales. Notably, the most significant motion observed was a shift of loop 97-100, located near the active site and known to be involved in lysozyme's hinge-bending motion. Experiments with chitobiose-bound lysozyme showed that the inhibitor blocked the transition from fast vibrations to larger-scale conformational changes observed in the apo enzyme. The results suggested that inhibitor binding stabilizes a ‘closed’ conformation, impeding the microsecond functional motions linked to the catalytic cycle. The overall findings provide a model where T-jump initially excites rapid, short-amplitude motions which subsequently dissipate into larger, functionally relevant conformational changes on a microsecond timescale.

This study demonstrates that IR laser-induced T-jump is a viable perturbation method for TRX at atomic resolution. The results show that the method can successfully capture both fast vibrational movements and slower, functionally relevant conformational changes in lysozyme. The observation of altered dynamics upon chitobiose binding strongly indicates that the larger-scale movements observed are functionally important. The consistency of the observed dynamics with previous studies on lysozyme provides further validation for the method. The use of T-jump as a universal perturbation method offers significant advantages for TRX, extending the applicability of the technique beyond proteins readily manipulated by other methods.

This research successfully demonstrates the use of infrared laser-induced temperature jump as a universal perturbation method for time-resolved X-ray crystallography. The study provides a detailed model for the dynamics of lysozyme, highlighting the transition from rapid atomic vibrations to larger-scale, functionally significant motions. The impact of inhibitor binding on these dynamics further validates the approach's ability to probe functional conformational changes. Future improvements in data collection speed and refinement techniques will enhance the method's temporal resolution, enabling even more detailed exploration of macromolecular dynamics across various protein systems.

The current study has some limitations. The temporal resolution, while significantly improved compared to previous approaches, could be further enhanced. The refinements against extrapolated structure factor magnitudes, while successful in identifying high-energy states, have limitations due to the inherent inaccuracies of the scalar approximation. Further development of refinement methods that simultaneously handle both extrapolated structure factor magnitudes and phases is necessary. The focus on lysozyme as a model system warrants further application of the technique to other protein systems and broader investigation of the functional implications of the observed dynamics.