High throughput screening using acoustic droplet ejection to combine protein crystals and chemical libraries on crystallization plates at high density

Bragg Spot Finder (BSF): a new machine-learning-aided approach to deal with spot finding for rapidly filtering diffraction pattern images

Macromolecular crystallography contributes significantly to understanding diseases and, more importantly, how to treat them by providing atomic resolution 3D structures of proteins. This is achieved by collecting X-ray diffraction images of protein crystals from important biological pathways. Spotfinders are used to detect the presence of crystals with usable data, and the spots from such crystals are the primary data used to solve the relevant structures. Having fast and accurate spot finding is essential, but recent advances in synchrotron beamlines used to generate X-ray diffraction images have brought us to the limits of what the best existing spotfinders can do. This bottleneck must be removed so spotfinder software can keep pace with the X-ray beamline hardware improvements and be able to see the weak or diffuse spots required to solve the most challenging problems encountered when working with diffraction images. In this paper, we first present Bragg Spot Detection (BSD), a large benchmark Bragg spot image dataset that contains 304 images with more than 66 000 spots. We then discuss the open source extensible U-Net-based spotfinder Bragg Spot Finder (BSF), with image pre-processing, a U-Net segmentation backbone, and post-processing that includes artifact removal and watershed segmentation. Finally, we perform experiments on the BSD benchmark and obtain results that are (in terms of accuracy) comparable to or better than those obtained with two popular spotfinder software packages (Dozor and DIALS), demonstrating that this is an appropriate framework to support future extensions and improvements.

Room-temperature crystallography reveals altered binding of small-molecule fragments to PTP1B

Much of our current understanding of how small-molecule ligands interact with proteins stems from X-ray crystal structures determined at cryogenic (cryo) temperature. For proteins alone, room-temperature (RT) crystallography can reveal previously hidden, biologically relevant alternate conformations. However, less is understood about how RT crystallography may impact the conformational landscapes of protein-ligand complexes. Previously, we showed that small-molecule fragments cluster in putative allosteric sites using a cryo crystallographic screen of the therapeutic target PTP1B (Keedy et al., 2018). Here, we have performed two RT crystallographic screens of PTP1B using many of the same fragments, representing the largest RT crystallographic screens of a diverse library of ligands to date, and enabling a direct interrogation of the effect of data collection temperature on protein-ligand interactions. We show that at RT, fewer ligands bind, and often more weakly - but with a variety of temperature-dependent differences, including unique binding poses, changes in solvation, new binding sites, and distinct protein allosteric conformational responses. Overall, this work suggests that the vast body of existing cryo-temperature protein-ligand structures may provide an incomplete picture, and highlights the potential of RT crystallography to help complete this picture by revealing distinct conformational modes of protein-ligand systems. Our results may inspire future use of RT crystallography to interrogate the roles of protein-ligand conformational ensembles in biological function.

Wall-free droplet microfluidics for probing biological processes by high-brilliance X-ray scattering techniques

Here we review probing biological processes initiated by the deposition of droplets on surfaces by micro- and nanobeam X-ray scattering techniques using synchrotron radiation and X-ray free-electron laser sources. We review probing droplet evaporation on superhydrophobic surfaces and reactions with substrates, basics of droplets deposition and flow simulations, droplet deposition techniques and practical experience at a synchrotron beamline. Selected applications with biological relevance will be reviewed and perspectives for the latest generation of high-brilliance X-ray sources discussed.

Controllable Droplet Ejection of Multiple Reagents through Focused Acoustic Beams

Acoustic droplet ejection (ADE) technology has revolutionized fluid handling with its contactless and fast fluid transfer. For precise droplet ejection and stable droplet coalescence at the target substrates for further detection, the input power of the ADE system needs to be adjusted. Currently, the existing power control method depends on scanning the source fluid wells one by one, which cannot afford precise and highly efficient droplet velocity adjustment, and the complicated operation caused by the repeated power evaluation processes for thousands of fluid transfers will waste much time. We propose a new method, which realizes the controllable ejection of multiple reagents by analyzing the effect of the product of kinematic viscosity and surface tension of the reagents on the droplet initial velocity. The experimental results obtained by ejecting dimethyl sulfoxide coincide well with the predicted results, and the relative error in the droplet initial velocity is mostly less than 8%. On the basis of the input power prediction method proposed in this paper, the ADE system is successfully constructed for continuous dispensing of polystyrene microspheres as cell surrogates, which provided an advanced liquid handling solution for research in biochemistry and other fields.

Towards an Optimal Sample Delivery Method for Serial Crystallography at XFEL

The advent of the X-ray free electron laser (XFEL) in the last decade created the discipline of serial crystallography but also the challenge of how crystal samples are delivered to X-ray. Early sample delivery methods demonstrated the proof-of-concept for serial crystallography and XFEL but were beset with challenges of high sample consumption, jet clogging and low data collection efficiency. The potential of XFEL and serial crystallography as the next frontier of structural solution by X-ray for small and weakly diffracting crystals and provision of ultra-fast time-resolved structural data spawned a huge amount of scientific interest and innovation. To utilize the full potential of XFEL and broaden its applicability to a larger variety of biological samples, researchers are challenged to develop better sample delivery methods. Thus, sample delivery is one of the key areas of research and development in the serial crystallography scientific community. Sample delivery currently falls into three main systems: jet-based methods, fixed-target chips, and drop-on-demand. Huge strides have since been made in reducing sample consumption and improving data collection efficiency, thus enabling the use of XFEL for many biological systems to provide high-resolution, radiation damage-free structural data as well as time-resolved dynamics studies. This review summarizes the current main strategies in sample delivery and their respective pros and cons, as well as some future direction.

Increasing the throughput of crystallization condition screens: Challenges and pitfalls of acoustic dispensing systems

Advances in contactless acoustic liquid transfer technologies have unlocked opportunities to substantially increase the throughput of crystallization screens and decrease the consumption of reagents and consumables. Acoustic energy transfer enables crystallization experiments to be set up precisely and rapidly on a nanoliter scale. Nonetheless, adapting acoustic transfer methods to a diverse range of crystallization conditions and their physicochemical idiosyncrasies remains a major bottleneck for true universality of this technique. Even though the reagent limitations still remain an issue, we present a straightforward protocol for setting up crystallization experiments by acoustic transfer using a Labcyte Echo 550 instrument, with a focus on the technical limitations of this method, including reagent compatibilities, spatial resolution and downscaling limits.

•

Set up crystallization screens in a small scale with reliable drop volumes as low as 50 nl

•

Overview of commonly used crystallographic screen compatibility with acoustic dispensing

•

Comparison of instrument calibrations and settings and its effects on error rate and screen reproducibility

Reducing sample consumption for serial crystallography using acoustic drop ejection

Efficient sample delivery is an essential aspect of serial crystallography at both synchrotrons and X-ray free-electron lasers. Rastering fixed target chips through the X-ray beam is an efficient method for serial delivery from the perspectives of both sample consumption and beam time usage. Here, an approach for loading fixed targets using acoustic drop ejection is presented that does not compromise crystal quality, can reduce sample consumption by more than an order of magnitude and allows serial diffraction to be collected from a larger proportion of the crystals in the slurry.

Miniaturization of the Whole Process of Protein Crystallographic Analysis by a Microfluidic Droplet Robot: From Nanoliter-Scale Purified Proteins to Diffraction-Quality Crystals

To obtain diffraction-quality crystals is one of the largest barriers to analyze the protein structure using X-ray crystallography. Here we describe a microfluidic droplet robot that enables successful miniaturization of the whole process of crystallization experiments including large-scale initial crystallization screening, crystallization optimization, and crystal harvesting. The combination of the state-of-the-art droplet-based microfluidic technique with the microbatch crystallization mode dramatically reduces the volumes of droplet crystallization reactors to tens nanoliter range, allowing large-scale initial screening of 1536 crystallization conditions and multifactor crystallization condition optimization with extremely low protein consumption, and on-chip harvesting of diffraction-quality crystals directly from the droplet reactors. We applied the droplet robot in miniaturized crystallization experiments of seven soluble proteins and two membrane proteins, and on-chip crystal harvesting of six proteins. The X-ray diffraction data sets of these crystals were collected using synchrotron radiation for analyzing the structures with similar diffraction qualities as conventional crystallization methods.

Strategies for sample delivery for femtosecond crystallography

Highly efficient data-collection methods are required for successful macromolecular crystallography (MX) experiments at X-ray free-electron lasers (XFELs). XFEL beamtime is scarce, and the high peak brightness of each XFEL pulse destroys the exposed crystal volume. It is therefore necessary to combine diffraction images from a large number of crystals (hundreds to hundreds of thousands) to obtain a final data set, bringing about sample-refreshment challenges that have previously been unknown to the MX synchrotron community. In view of this experimental complexity, a number of sample delivery methods have emerged, each with specific requirements, drawbacks and advantages. To provide useful selection criteria for future experiments, this review summarizes the currently available sample delivery methods, emphasising the basic principles and the specific sample requirements. Two main approaches to sample delivery are first covered: (i) injector methods with liquid or viscous media and (ii) fixed-target methods using large crystals or using microcrystals inside multi-crystal holders or chips. Additionally, hybrid methods such as acoustic droplet ejection and crystal extraction are covered, which combine the advantages of both fixed-target and injector approaches.

A Simple Technique to Improve Microcrystals Using Gel Exclusion of Nucleation Inducing Elements

A technique is described for generating large well diffracting crystals from conditions that yield microcrystals. Crystallization using this technique is both rapid (crystals appear in <1 h) and robust (48 out of 48 co-crystallized with a fragment library, compared with 26 out of 48 using conventional hanging drop). Agarose gel is used to exclude nucleation inducing elements from the remaining crystallization cocktail. The chemicals in the crystallization cocktail are partitioned into high concentration components (presumed to induce aggregation by reducing water activity) and low concentration nucleation agents (presumed to induce nucleation through direct interaction). The nucleation agents are then combined with 2% agarose gel and deposited on the crystallization shelf of a conventional vapor diffusion plate. The remaining components are mixed with the protein and placed in contact with the agarose drop. This technique yielded well diffracting crystals of lysozyme, cubic insulin, proteinase k, and ferritin (ferritin crystals diffracted to 1.43 Å). The crystals grew rapidly, reaching large size in less than one hour (maximum size was achieved in 1–12 h). This technique is not suitable for poorly expressing proteins because small protein volumes diffuse out of the agarose gel too quickly. However, it is a useful technique for situations where crystals must grow rapidly (such as educational applications and preparation of beamline test specimens) and in situations where crystals must grow robustly (such as co-crystallization with a fragment library).

The structures of penicillin-binding protein 4 (PBP4) and PBP5 from Enterococci provide structural insights into β-lactam resistance

The final steps of cell-wall biosynthesis in bacteria are carried out by penicillin-binding proteins (PBPs), whose transpeptidase domains form the cross-links in peptidoglycan chains that define the bacterial cell wall. These enzymes are the targets of β-lactam antibiotics, as their inhibition reduces the structural integrity of the cell wall. Bacterial resistance to antibiotics is a rapidly growing concern; however, the structural underpinnings of PBP-derived antibiotic resistance are poorly understood. PBP4 and PBP5 are low-affinity, class B transpeptidases that confer antibiotic resistance to Enterococcus faecalis and Enterococcus faecium, respectively. Here, we report the crystal structures of PBP4 (1.8 Å) and PBP5 (2.7 Å) in their apo and acyl-enzyme complexes with the β-lactams benzylpenicillin, imipenem, and ceftaroline. We found that, although these three β-lactams adopt geometries similar to those observed in other class B PBP structures, there are small, but significant, differences that likely decrease antibiotic efficacy. Further, we also discovered that the N-terminal domain extensions in this class of PBPs undergo large rigid-body rotations without impacting the structure of the catalytic transpeptidase domain. Together, our findings are defining the subtle functional and structural differences in the Enterococcus PBPs that allow them to support transpeptidase activity while also conferring bacterial resistance to antibiotics that function as substrate mimics.

Multi-criteria protein structure comparison and structural similarities analysis using pyMCPSC

Protein Structure Comparison (PSC) is a well developed field of computational proteomics with active interest from the research community, since it is widely used in structural biology and drug discovery. With new PSC methods continuously emerging and no clear method of choice, Multi-Criteria Protein Structure Comparison (MCPSC) is commonly employed to combine methods and generate consensus structural similarity scores. We present pyMCPSC, a Python based utility we developed to allow users to perform MCPSC efficiently, by exploiting the parallelism afforded by the multi-core CPUs of today’s desktop computers. We show how pyMCPSC facilitates the analysis of similarities in protein domain datasets and how it can be extended to incorporate new PSC methods as they are becoming available. We exemplify the power of pyMCPSC using a case study based on the Proteus_300 dataset. Results generated using pyMCPSC show that MCPSC scores form a reliable basis for identifying the true classification of a domain, as evidenced both by the ROC analysis as well as the Nearest-Neighbor analysis. Structure similarity based “Phylogenetic Trees” representation generated by pyMCPSC provide insight into functional grouping within the dataset of domains. Furthermore, scatter plots generated by pyMCPSC show the existence of strong correlation between protein domains belonging to SCOP Class C and loose correlation between those of SCOP Class D. Such analyses and corresponding visualizations help users quickly gain insights about their datasets. The source code of pyMCPSC is available under the GPLv3.0 license through a GitHub repository (https://github.com/xulesc/pymcpsc).

Using sound pulses to solve the crystal-harvesting bottleneck

Crystal harvesting has proven to be difficult to automate and remains the rate-limiting step for many structure-determination and high-throughput screening projects. This has resulted in crystals being prepared more rapidly than they can be harvested for X-ray data collection. Fourth-generation synchrotrons will support extraordinarily rapid rates of data acquisition, putting further pressure on the crystal-harvesting bottleneck. Here, a simple solution is reported in which crystals can be acoustically harvested from slightly modified MiTeGen In Situ-1 crystallization plates. This technique uses an acoustic pulse to eject each crystal out of its crystallization well, through a short air column and onto a micro-mesh (improving on previous work, which required separately grown crystals to be transferred before harvesting). Crystals can be individually harvested or can be serially combined with a chemical library such as a fragment library.

Nanoliter Centrifugal Liquid Dispenser Coupled with Superhydrophobic Microwell Array Chips for High-Throughput Cell Assays

Microfluidic systems have been regarded as a potential platform for high-throughput screening technology in drug discovery due to their low sample consumption, high integration, and easy operation. The handling of small-volume liquid is an essential operation in microfluidic systems, especially in investigating large-scale combination conditions. Here, we develop a nanoliter centrifugal liquid dispenser (NanoCLD) coupled with superhydrophobic microwell array chips for high-throughput cell-based assays in the nanoliter scale. The NanoCLD consists of a plastic stock block with an array of drilled through holes, a reagent microwell array chip (reagent chip), and an alignment bottom assembled together in a fixture. A simple centrifugation at 800 rpm can dispense ~160 nL reagents into microwells in 5 min. The dispensed reagents are then delivered to cells by sandwiching the reagent chip upside down with another microwell array chip (cell chip) on which cells are cultured. A gradient of doxorubicin is then dispensed to the cell chip using the NanoCLD for validating the feasibility of performing drug tests on our microchip platform. This novel nanoliter-volume liquid dispensing method is simple, easy to operate, and especially suitable for repeatedly dispensing many different reagents simultaneously to microwells.

Applications of X-Ray Micro-Beam for Data Collection

Micro-diffraction tools for macromolecular crystallography, first developed at the end of 1990s and now an integral part of many synchrotron beamlines, enable some of the experiments which were not feasible just a decade or so ago. These include data collection from very small samples, just a few micrometers in size; from larger, but severely inhomogeneous samples; and from samples which are optically invisible. Improved micro-diffraction tools led to improved signal-to-noise ratio, to mitigation of radiation damage in some cases, and to better-designed diffraction experiments. Small, micron-scale beams can be attained in different ways and knowing the details of the implementation is important in order to design the diffraction experiment properly. Similarly, precision, reproducibility and stability of the goniometry, and caveats of detection systems need to be taken into account. Lastly, to make micro-diffraction widely applicable, the sophistication, robustness, and user-friendliness of these tools are just as important as the technical capabilities.

Gentle, fast and effective crystal soaking by acoustic dispensing

The steady expansion in the capacity of modern beamlines for high-throughput data collection, enabled by increasing X-ray brightness, capacity of robotics and detector speeds, has pushed the bottleneck upstream towards sample preparation. Even in ligand-binding studies using crystal soaking, the experiment best able to exploit beamline capacity, a primary limitation is the need for gentle and nontrivial soaking regimens such as stepwise concentration increases, even for robust and well characterized crystals. Here, the use of acoustic droplet ejection for the soaking of protein crystals with small molecules is described, and it is shown that it is both gentle on crystals and allows very high throughput, with 1000 unique soaks easily performed in under 10 min. In addition to having very low compound consumption (tens of nanolitres per sample), the positional precision of acoustic droplet ejection enables the targeted placement of the compound/solvent away from crystals and towards drop edges, allowing gradual diffusion of solvent across the drop. This ensures both an improvement in the reproducibility of X-ray diffraction and increased solvent tolerance of the crystals, thus enabling higher effective compound-soaking concentrations. The technique is detailed here with examples from the protein target JMJD2D, a histone lysine demethylase with roles in cancer and the focus of active structure-based drug-design efforts.

Acoustic Injectors for Drop-On-Demand Serial Femtosecond Crystallography

X-ray free-electron lasers (XFELs) provide very intense X-ray pulses suitable for macromolecular crystallography. Each X-ray pulse typically lasts for tens of femtoseconds and the interval between pulses is many orders of magnitude longer. Here we describe two novel acoustic injection systems that use focused sound waves to eject picoliter to nanoliter crystal-containing droplets out of microplates and into the X-ray pulse from which diffraction data are collected. The on-demand droplet delivery is synchronized to the XFEL pulse scheme, resulting in X-ray pulses intersecting up to 88% of the droplets. We tested several types of samples in a range of crystallization conditions, wherein the overall crystal hit ratio (e.g., fraction of images with observable diffraction patterns) is a function of the microcrystal slurry concentration. We report crystal structures from lysozyme, thermolysin, and stachydrine demethylase (Stc2). Additional samples were screened to demonstrate that these methods can be applied to rare samples.