An on-demand, drop-on-drop method for studying enzyme catalysis by serial crystallography

Abstract

Serial femtosecond crystallography has opened up many new opportunities in structural biology. In recent years, several approaches employing light-inducible systems have emerged to enable time-resolved experiments that reveal protein dynamics at high atomic and temporal resolutions. However, very few enzymes are light-dependent, whereas macromolecules requiring ligand diffusion into an active site are ubiquitous. In this work we present a drop-on-drop sample delivery system that enables the study of enzyme-catalyzed reactions in microcrystal slurries. The system delivers ligand solutions in bursts of multiple picoliter-sized drops on top of a larger crystal-containing drop inducing turbulent mixing and transports the mixture to the X-ray interaction region with temporal resolution. We demonstrate mixing using fluorescent dyes, numerical simulations and time-resolved serial femtosecond crystallography, which show rapid ligand diffusion through microdroplets. The drop-on-drop method has the potential to be widely applicable to serial crystallography studies, particularly of enzyme reactions with small molecule substrates.

Fig. 1. Drop-on-drop experimental setup enables accelerated mixing in droplets.

a Overview of the setup. The protein crystals are ejected by an acoustic droplet ejector (ADE, white and dark gray column above tape and on the right-hand side) and synchronized to the XFEL master clock. The burst of substrate drops is ejected by a piezoelectric injector (PEI, blue box above tape, on the left-hand side) that can accommodate disposable cartridges with a range of orifices that dictate the size of the droplets. Reaction time is varied by the speed of the tape drive (10–600 mm s⁻¹) and depends upon the location of the PEI with respect to the ADE droplet intersection and the X-ray interaction points; in the case reported here, we selected time points between 0.1 and 6 s. b Relative fluorescence intensity (crosses) of the calcium-bound forms of Fura Red (salmon) and Fluo-5N (purple) dyes obtained from the scaling of the normalized spectral components and their errors of the fits (Supplementary Fig. 5) compared to reaction simulations of collision-driven hydrodynamic flow (inset) and equilibration by diffusion. Lines correspond to the fitted exponential functions of the experimental data and rise times are indicated. Fluorescence signal of Fura Red rises faster than Fluo-5N due to the lower dissociation constant.

Fig. 2. Drop-on-drop delivery enables the analysis of N-acetyl-D-glucosamine (GlcNAc) binding to the HEWL active site and catalytic activity of CTX-M-15 by serial X-ray crystallography.

a 2mF_o–DF_c electron density maps for HEWL-GlcNAc XFEL structures, displayed at ±1σ contour level and carved at 2 Å around the GlcNAc ligand. The GlcNAc molecule from the 2 s time point structure was used to carve the 200 ms time point map. All maps are at 1.45 Å resolution. b Cartoon representation of the HEWL structure (2 s mixing time point, PDB ID 7BHN). GlcNAc is shown as sticks and colored in yellow. Waters, sodium, and chloride ions are shown as red, purple, and green spheres, respectively. Ligand binding subsites A-D (as in PDB ID 5NJR) are indicated by yellow circles. c, d Close-up active site view of the HEWL structure shown in (b). c mF_o–DF_c polder OMIT difference density map contoured at ±3σ and carved 1.5 Å around the ligand site. Hydrogen bonding network is highlighted. d Superimposition of the 2 s mixing time point structure (yellow and gray) and the single-crystal room-temperature structure of HEWL soaked with GlcNAc obtained by Tanley et al. (light blue, PDB 3TXJ). e 2mF_o–DF_c electron density maps of CTX-M-15 structure displayed at ±1σ and carved at 2 Å around the ertapenem ligand (pink; 1.55, 1.55, and 1.65 Å for 0.6 s, 2, and 10 min, respectively). The 0.6 and 2 s time points are calculated with XFEL data, whereas the 10 min time point is calculated with room temperature, SSX data collected at Diamond Light Source. f, g Close-up view at the active site of the structure obtained after 2 s (f) and 10 min (g) of mixing with ertapenem. Interactions of ertapenem (blue sticks) with residues on the protein main chain (gray sticks) as well as the interactions of the proposed deacylating water (DW, red sphere) with the protein are shown as black dashes. mF_o–DF_c polder OMIT difference density maps are contoured at ±3σ and carved 1.5 Å around the ligand site. Ertapenem has been tentatively modeled and refined into the observed electron density as the (R)-Δ¹-pyrroline tautomer for both 0.6 s and 10 min structures (see Supplementary Note 3). Graphic was created using the UCSF Chimera package.