Gentle, fast and effective crystal soaking by acoustic dispensing

Abstract

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.

Keywords: Diamond Light Source I04-1; Structural Genomics Consortium; XChem; acoustic droplet ejection; crystal soaking; fragment screening.

Figure 1

Acoustic droplet ejection allows precise and selective targeting of crystallization droplets. (a) Schematic representation of acoustic droplet ejection for crystal soaking using the Labcyte Echo. (b) Crystallization droplets (200 nl initial volume) containing protein crystals with acoustically added solvents. Initially, 135 nl of 100 mM compound in DMSO was dispensed with the offset targeting approach to the indicated location (yellow ‘X’). Later, 50 nl of cryoprotecting solvent (ethylene glycol) was added at the same location. (c) The TeXRank interface showing a crystallization drop containing a single JMJD2D crystal. Clicking a location records the acoustic dispensing target for both compound-containing solvents and cryoprotectants (DMSO and ethylene glycol, respectively, in this study). The yellow ‘X’ and xy coordinates have been added for clarity. The expanded section shows the ranked plot of crystal images.

Figure 2

Positional precision of acoustic transfer allows exploration of the influence of complex dispensing patterns on the solvent tolerance of protein crystals. (a) The requested pattern of dispensing and (b) the resulting 2.5 nl drop pattern within a subwell of a sitting-drop plate. (c–h) Schematic representations of different dispensing patterns showing a crystallization droplet (blue circle), a protein crystal (yellow diamond) and the solvent target locations (red spots) within a sitting-drop well (outer circle). Dispensing patterns investigated were (c) an offset location away from the crystal, (d) direct targeting of the crystal, (e) a ring pattern around the drop edge (15 target locations) or multiple ring patterns across the drop, increasing in target density and final DMSO concentration, with (f) 20% final DMSO concentration (20 target locations), (g) 40% (43 target locations) or (h) 60% (120 target locations).

Figure 3

Offset targeting increases solvent tolerance and enables longer soaking times. Survival rate of JMJD2D crystals (percentage of diffracting crystals) from soaking in DMSO after acoustic transfer targeted at the crystal (orange) or targeted away from the crystal (blue). Crystals were soaked for 1 h (lighter colousr) or overnight (darker colours).

Figure 4

Compound diffusion after offset targeting through a droplet takes just several minutes. (a) Schematic illustration of diffusion (black arrows) from a crystallization droplet (blue circle) soaked with the offset targeting approach (the red circle represent the location of dispensed solvent). (b) Reduced diffusion of methylene blue dye through a crystallization drop that has a skin.

Figure 5

Variation of soaking times and concentration can be used to improve the visibility of the ligand. Plots showing the number of ligands detected (strong, medium and weak binding ligands) as a function of (a) concentration or (b) time after crystal soaking using acoustic transfer (from 100 mM stock). The concentration series in (a) were soaked for a fixed time of 4 h, while the time series in (b) were soaked at a fixed nominal concentration of 20 mM. Soaks were performed in duplicate for each condition, leading to 30 X-ray diffraction data sets for (a) and 36 data sets for (b). (c) Electron-density maps (PanDDA maps: event maps are shown in blue at 2σ, Z-maps are shown in green/red at ±3σ; 1.3–1.4 Å resolution) from the minimum experimental conditions (time or concentration series) required to detect ligand binding. The maps are generated prior to refinement of the ligand with the model. The ligand coordinates have been modelled into the electron-density maps for clarity. The PanDDA reported background density correction (BDC) values are shown (Pearce et al., 2016 ▸).

Figure 6

Individual adjustment of solvent concentration and soaking time is essential for successful soaking experiments. The working soaking conditions selected and used for 16 recent protein targets which were investigated as part of the XChem user program at the Diamond Light Source. The graphs show (a) the nominal final DMSO concentration used for fragment screening and (b) the soaking time selected for each project. A total of over 18 000 crystals were soaked in the course of these experiments.