Guide to serial synchrotron crystallography

Abstract

Serial crystallography (SX) is an emerging technique that can be used to determine the noncryogenic crystal structure of macromolecules while minimizing radiation damage. Applying SX using pump-probe or mix-and-inject techniques enables the observation of time-resolved molecular reactions and dynamics in macromolecules. After the successful demonstration of the SX experimental technique with structure determination in serial femtosecond crystallography using an X-ray free electron laser, this method was adapted to the synchrotron, leading to the development of serial synchrotron crystallography (SSX). SSX offers new opportunities for researchers to leverage SX techniques, contributing to the advancement of structural biology and offering a deeper understanding of the structure and function of macromolecules. This review covers the background and advantages of SSX and its experimental approach. It also discusses important considerations when conducting SSX experiments.

Keywords: Radiation damage; Room temperature; Serial crystallography; Serial synchrotron crystallography; Time-resolved study; X-ray crystallography.

Graphical abstract

Fig. 1

Limitations of conventional macromolecular crystallography. (A) Radiation damage during room temperature data collection: When X-rays are exposed to a single crystal at room temperature, the diffraction intensity is reduced due to X-ray radiation damage and heating effects. This causes the collection of incomplete data sets or poor diffraction quality. (B) Cryocrystallography experiment: Crystals were immersed in a cryoprotectant solution and exposed to X-rays in a cryogenic environment (e.g., 100 K under a liquid nitrogen stream). (C) Experimental limitations of macromolecular crystallography methods.

Fig. 2

Serial crystallography using an X-ray Free Electron Laser (XFEL) and synchrotron X-ray. (A) XFELs provide unprecedented intense X-ray pulses at the femtosecond level. Currently, there are five hard X-ray XFEL facilities worldwide capable of conducting serial femtosecond crystallography (SFX) experiments. One feature of XFELs is that X-rays can be delivered to only one beamline at a time. Although X-rays can be distributed to multiple beamlines using optical instruments, this reduces the photon flux. Consequently, the availability of XFEL beamtime for conducting SFX research is highly limited. (B) Synchrotrons are available in many countries, providing an advantage in terms of accessibility and beamtime allocation compared with XFEL facilities. (C) The experimental setup for the SFX and SSX is nearly identical. Crystals are continuously delivered to the X-ray interaction point in a noncryogenic environment. Time-resolved SX experiments using methods such as pump-probe or mix-and-inject allow the observation of reaction mechanisms. (D) Using high-speed detectors, numerous diffraction data were collected to determine the three-dimensional structure.

Fig. 3

Overview of the experimental setup and sample delivery methods for serial synchrotron crystallography. (A) Photo of SSX experimental setup for injection-based sample delivery. The original figures were obtained from a previous study (Kim and Nam, 2022) and have been modified. This method uses a high-viscosity extruder (HVE) injector or a syringe to extrude an injection stream containing crystals embedded in a viscous medium. (B) Photo of SSX experimental setup for the fixed-target (FT) scanning experiment: Sample holders containing crystal samples are scanned in vertical and horizontal directions during data collection. The original figures were obtained from a previous study (Park et al., 2020) and have been modified. (C) Schematic drawing of the injection-based sample delivery devices. This method involves the use of a high-viscosity extruder (HVE) injector or a syringe to extrude an injection stream containing crystals embedded in a viscous medium. Typically, the HVE injector and syringe samples are extruded using an HPLC pump or mechanical pressure, respectively. The sample extruded from the HVE injector can be focused using the gas such as helium. Various factors, including crystal concentration, beam size, injection stream stability, and hit rate during data collection, determine the injection flow rate. (Inset: photo of an injection stream generated via a syringe) (D) Drawing of the fixed-target (FT) scanning device. This method uses sample holders with regular holes for crystals to settle into without relying on gravity. In general, these sample holders are enclosed in a film to prevent dehydration of the crystal solution. Sample holders containing crystals can be scanned vertically and horizontally, or translation with oscillation can be applied during data collection. (E) Tape-drive-based hybrid method: The crystal suspension was injected from the injector and placed on polyimide tape. The tape winds up, delivering the crystal to the X-ray interaction region. The sample delivery methods described above have been successfully applied to SSX and time-resolved SSX experiments with optical lasers. (F) Photo of lysozyme crystals in 60% (v/v) monoolein for the injection-based sample delivery method. (G) Photo of lysozyme crystals in nylon mesh-based sample holder for the FT scanning method.

Fig. 4

General data processing procedure of the serial crystallography experiment. (A). Collected images from SSX experiment include crystal diffraction images and images without crystal diffraction or having weak diffraction intensity. (B). Hit-finding. Based on the hit-finding parameters (signal-to-noise ratio, number of peaks, etc), images with weak or no diffraction intensity and unwanted diffraction patterns were filtered through the hit-finding program such as Cheetah, or Psocake. Although this step is not absolutely essential, it can significantly reduce processing time for subsequent steps, such as indexing, and help maintain data storage efficiency by removing unnecessary data. (C) Indexing. Bragg peaks were indexed, and crystal information was obtained from the crystal diffraction pattern. No indexing or diffraction data with different crystal forms were excluded. (D) Integration. Fully integrated intensities were obtained using Monte-Calro integration with partially integrated intensities. The hkl file containing reflection information is utilized to determine the crystal structure. (E) Structure determination, validation, and deposition of structure factors and coordinates.