Potential of Time-Resolved Serial Femtosecond Crystallography Using High Repetition Rate XFEL Sources

Crystal structure of a bacterial photoactivated adenylate cyclase determined by serial femtosecond and serial synchrotron crystallography

OaPAC is a recently discovered blue-light-using flavin adenosine dinucleotide (BLUF) photoactivated adenylate cyclase from the cyanobacterium Oscillatoria acuminata that uses adenosine triphosphate and translates the light signal into the production of cyclic adenosine monophosphate. Here, we report crystal structures of the enzyme in the absence of its natural substrate determined from room-temperature serial crystallography data collected at both an X-ray free-electron laser and a synchrotron, and we compare these structures with cryo-macromolecular crystallography structures obtained at a synchrotron by us and others. These results reveal slight differences in the structure of the enzyme due to data collection at different temperatures and X-ray sources. We further investigate the effect of the Y6W mutation in the BLUF domain, a mutation which results in a rearrangement of the hydrogen-bond network around the flavin and a notable rotation of the side chain of the critical Gln48 residue. These studies pave the way for picosecond-millisecond time-resolved serial crystallography experiments at X-ray free-electron lasers and synchrotrons in order to determine the early structural intermediates and correlate them with the well studied picosecond-millisecond spectroscopic intermediates.

A comparative review of time-resolved x-ray and electron scattering to probe structural dynamics

The structure of molecules, particularly the dynamic changes in structure, plays an essential role in understanding physical and chemical phenomena. Time-resolved (TR) scattering techniques serve as crucial experimental tools for studying structural dynamics, offering direct sensitivity to molecular structures through scattering signals. Over the past decade, the advent of x-ray free-electron lasers (XFELs) and mega-electron-volt ultrafast electron diffraction (MeV-UED) facilities has ushered TR scattering experiments into a new era, garnering significant attention. In this review, we delve into the basic principles of TR scattering experiments, especially focusing on those that employ x-rays and electrons. We highlight the variations in experimental conditions when employing x-rays vs electrons and discuss their complementarity. Additionally, cutting-edge XFELs and MeV-UED facilities for TR x-ray and electron scattering experiments and the experiments performed at those facilities are reviewed. As new facilities are constructed and existing ones undergo upgrades, the landscape for TR x-ray and electron scattering experiments is poised for further expansion. Through this review, we aim to facilitate the effective utilization of these emerging opportunities, assisting researchers in delving deeper into the intricate dynamics of molecular structures.

Towards understanding the crystallization of photosystem II: influence of poly(ethylene glycol) of various molecular sizes on the micelle formation of alkyl maltosides

The influence of poly(ethylene glycol) (PEG) polymers H-(O-CH2-CH2)p-OH with different average molecular sizes p on the micelle formation of n-alkyl-β-D-maltoside detergents with the number of carbon atoms in the alkyl chain ranging from 10 to 12 is investigated with the aim to learn more about the detergent behavior under conditions suitable for the crystallization of the photosynthetic pigment-protein complex photosystem II. PEG is shown to increase the critical micelle concentration (CMC) of all three detergents in the crystallization buffer in a way that the free energy of micelle formation increases linearly with the concentration of oxyethylene units (O-CH2-CH2) irrespective of the actual molecular weight of the polymer. The CMC shift is modeled by assuming for simplicity that it is dominated by the interaction between PEG and detergent monomers and is interpreted in terms of an increase of the transfer free energy of a methylene group of the alkyl chain by 0.2 kJ mol-1 per 1 mol L-1 increase of the concentration of oxyethylene units at 298 K. Implications of this effect for the solubilization and crystallization of protein-detergent complexes as well as detergent extraction from crystals are discussed.

Application of density matrix Wigner transforms for ultrafast macromolecular and chemical x-ray crystallography

Time-Resolved Serial Femtosecond Crystallography (TR-SFX) conducted at X-ray Free Electron Lasers (XFELs) has become a powerful tool for capturing macromolecular structural movies of light-initiated processes. As the capabilities of XFELs advance, we anticipate that a new range of coherent control and structural Raman measurements will become achievable. Shorter optical and x-ray pulse durations and increasingly more exotic pulse regimes are becoming available at free electron lasers. Moreover, with high repetition enabled by the superconducting technology of European XFEL (EuXFEL) and Linac Coherent Light Source (LCLS-II) , it will be possible to improve the signal-to-noise ratio of the light-induced differences, allowing for the observation of vibronic motion on the sub-Angstrom level. To predict and assign this coherent motion, which is measurable with a structural technique, new theoretical approaches must be developed. In this paper, we present a theoretical density matrix approach to model the various population and coherent dynamics of a system, which considers molecular system parameters and excitation conditions. We emphasize the use of the Wigner transform of the time-dependent density matrix, which provides a phase space representation that can be directly compared to the experimental positional displacements measured in a TR-SFX experiment. Here, we extend the results from simple models to include more realistic schemes that include large relaxation terms. We explore a variety of pulse schemes using multiple model systems using realistic parameters. An open-source software package is provided to perform the density matrix simulation and Wigner transformations. The open-source software allows us to define any arbitrary level schemes as well as any arbitrary electric field in the interaction Hamiltonian.

Kilohertz droplet-on-demand serial femtosecond crystallography at the European XFEL station FXE

X-ray Free Electron Lasers (XFELs) allow the collection of high-quality serial femtosecond crystallography data. The next generation of megahertz superconducting FELs promises to drastically reduce data collection times, enabling the capture of more structures with higher signal-to-noise ratios and facilitating more complex experiments. Currently, gas dynamic virtual nozzles (GDVNs) stand as the sole delivery method capable of best utilizing the repetition rate of megahertz sources for crystallography. However, their substantial sample consumption renders their use impractical for many protein targets in serial crystallography experiments. Here, we present a novel application of a droplet-on-demand injection method, which allowed operation at 47 kHz at the European XFEL (EuXFEL) by tailoring a multi-droplet injection scheme for each macro-pulse. We demonstrate a collection rate of 150 000 indexed patterns per hour. We show that the performance and effective data collection rate are comparable to GDVN, with a sample consumption reduction of two orders of magnitude. We present lysozyme crystallographic data using the Large Pixel Detector at the femtosecond x-ray experiment endstation. Significant improvement of the crystallographic statistics was made by correcting for a systematic drift of the photon energy in the EuXFEL macro-pulse train, which was characterized from indexing the individual frames in the pulse train. This is the highest resolution protein structure collected and reported at the EuXFEL at 1.38 Å resolution.

Deep residual networks for crystallography trained on synthetic data

The use of artificial intelligence to process diffraction images is challenged by the need to assemble large and precisely designed training data sets. To address this, a codebase called Resonet was developed for synthesizing diffraction data and training residual neural networks on these data. Here, two per-pattern capabilities of Resonet are demonstrated: (i) interpretation of crystal resolution and (ii) identification of overlapping lattices. Resonet was tested across a compilation of diffraction images from synchrotron experiments and X-ray free-electron laser experiments. Crucially, these models readily execute on graphics processing units and can thus significantly outperform conventional algorithms. While Resonet is currently utilized to provide real-time feedback for macromolecular crystallography users at the Stanford Synchrotron Radiation Lightsource, its simple Python-based interface makes it easy to embed in other processing frameworks. This work highlights the utility of physics-based simulation for training deep neural networks and lays the groundwork for the development of additional models to enhance diffraction collection and analysis.

A phenomenological model of the X-ray pulse statistics of a high-repetition-rate X-ray free-electron laser

Many coherent imaging applications that utilize ultrafast X-ray free-electron laser (XFEL) radiation pulses are highly sensitive to fluctuations in the shot-to-shot statistical properties of the source. Understanding and modelling these fluctuations are key to successful experiment planning and necessary to maximize the potential of XFEL facilities. Current models of XFEL radiation and their shot-to-shot statistics are based on theoretical descriptions of the source and are limited in their ability to capture the shot-to-shot intensity fluctuations observed experimentally. The lack of accurate temporal statistics in simulations that utilize these models is a significant barrier to optimizing and interpreting data from XFEL coherent diffraction experiments. Presented here is a phenomenological model of XFEL radiation that is capable of capturing the shot-to-shot statistics observed experimentally using a simple time-dependent approximation of the pulse wavefront. The model is applied to reproduce non-stationary shot-to-shot intensity fluctuations observed at the European XFEL, whilst accurately representing the single-shot properties predicted by FEL theory. Compared with previous models, this approach provides a simple, robust and computationally inexpensive method of generating statistical representations of XFEL radiation.

Insight into the structural dynamics of light sensitive proteins from time-resolved crystallography and quantum chemical calculations

The structural dynamics underlying molecular mechanisms of light-sensitive proteins can be studied by a variety of experimental and computational biophysical techniques. Here we review recent progress in combining time-resolved crystallography at X-ray free electron lasers and quantum chemical calculations to study structural changes in photoenzymes, photosynthetic proteins, photoreceptors, and photoswitchable fluorescent proteins following photoexcitation.

Novel approach to push the limit of temporal resolution in ultrafast electron diffraction accelerators

Ultrafast electron diffraction techniques that employ relativistic electrons as a probe have been in the spotlight as a key technology for visualizing structural dynamics which take place on a time scale of a few femtoseconds to hundreds femtoseconds. These applications highly demand not only extreme beam quality in 6-D phase space such as a few nanometer transverse emittances and femtosecond duration but also equivalent beam stability. Although these utmost requirements have been demonstrated by a compact setup with a high-gradient electron gun with state-of-the-art laser technologies, this approach is fundamentally restricted by its nature for compressing the electrons in a short distance by a ballistic bunching method. Here, we propose a new methodology that pushes the limit of timing jitter beyond the state-of-the-art by utilizing consecutive RF cavities. This layout already exists in reality for energy recovery linear accelerator demonstrators. Furthermore, the demonstrators are able to provide MHz repetition rates, which are out of reach for most conventional high-gradient electron guns.