Changes in an enzyme ensemble during catalysis observed by high-resolution XFEL crystallography

Enzymes populate ensembles of structures necessary for catalysis that are difficult to experimentally characterize. We use time-resolved mix-and-inject serial crystallography at an x-ray free electron laser to observe catalysis in a designed mutant isocyanide hydratase (ICH) enzyme that enhances sampling of important minor conformations. The active site exists in a mixture of conformations, and formation of the thioimidate intermediate selects for catalytically competent substates. The influence of cysteine ionization on the ICH ensemble is validated by determining structures of the enzyme at multiple pH values. Large molecular dynamics simulations in crystallo and time-resolved electron density maps show that Asp17 ionizes during catalysis and causes conformational changes that propagate across the dimer, permitting water to enter the active site for intermediate hydrolysis. ICH exhibits a tight coupling between ionization of active site residues and catalysis-activated protein motions, exemplifying a mechanism of electrostatic control of enzyme dynamics.


Fig. S1 .
Fig. S1.Schematic of wild-type ICH catalysis-activated motions.(A) Wild-type ICH is shown in ribbon cartoon with the active site highlighted in pink.(B) Upon formation of the thioimidate intermediate, changes in active site hydrogen bonds allow sampling of a shifted conformation of helix H (green highlight).

Fig. S2 .
Fig. S2.Examples of temperature-dependent differences in G150T ICH.Fo(274 K synchrotron)-Fo(298 K XFEL) isomorphous difference electron density maps at 1.3 Å resolution are contoured at 3 s, with positive features in green and negative ones in red.(A) Difference electron density peaks in these residues agree with the small shifts observed when a 100 K cryogenic structure (purple bonds) is superimposed on the 298 K XFEL structure (gold bonds).(B) The populations of C67 alternate conformations are altered by the change in temperature between the synchrotron and XFEL datasets, as shown by difference electron density peaks near the sulfur atom (yellow).(C) The populations of alternate conformations of C101 are altered by change in temperature, as shown by peaks near the sulfur atom (yellow).

Fig. S3 .
Fig. S3.Difference electron density features between synchrotron and XFEL datasets are reproducible.Fo(274 K synchrotron)-Fo(298 K XFEL) isomorphous difference electron density maps at 1.3 Å resolution are contoured at 2.8-3 s (corresponding to 0.17-0.18e -/Å 3 ) with positive features in green and negative ones in red.(A-C) show the same region near C67 with Fo(274 K synchrotron)-Fo(298 K XFEL) isomorphous difference electron density maps calculated from three independent synchrotron datasets.(D-F) show the region near L49 with Fo(274 K synchrotron)-Fo(298 K XFEL) isomorphous difference electron density maps calculated from three independent synchrotron datasets.The difference electron density features agree closely between the three datasets.

Fig. S4 .
Fig. S4.The G150T mutation causes ICH to populate conformations that the wild-type enzyme samples during catalysis.Fo(15s)-Fo(0s) isomorphous difference electron density maps calculated for wild-type ICH from a prior experiment (33) at 1.55 Å resolution are contoured at 2.7s (corresponding to 0.15 e -/Å 3 ).Wild-type ICH is shown in slate gray and G150T ICH is shown in gold bonds.(A-D) Selected examples where the conformational shifts evident between the wildtype and G150T ICH structures agree well with the difference electron density map peaks observed during catalysis in wild-type ICH.In all panels, the G150T ICH structure (gold) overlaps with positive difference electron density peaks.

Fig. S5 .
Fig. S5.pH-dependent shifts of the helix H conformation in wild-type ICH recapitulate changes resulting from the G150T mutation.The structure of G150T ICH is shown in red bonds and structures of wild-type (WT) ICH at the indicated pH values are shown in bonds ranging from yellow (pH=4.2) to purple (pH=8.3).As pH decreases, wild-type ICH adopts a conformation that is highly similar to G150T ICH, consistent with changes expected from protonation of C101.

Fig. S6 .
Fig. S6.pH-dependent conformational heterogeneity in helix H is asymmetric in the wildtype ICH dimer.2mFo-DFc electron density is shown contoured at 0.9s for wild-type ICH at pH=5.4 (1.33 Å resolution, green; A, B) and pH=6.0 (1.2 Å resolution, blue; C, D).The dominant conformer is shown in solid bonds and the minor conformer is semi-transparent.(A,B) At pH=5.4,protomer A is in the fully shifted conformation and protomer B is a mixture of shifted and strained helix conformations.(C,D) At pH=6.0, protomer A is a mixture of shifted and strained helix conformations and protomer B is in the fully strained conformation.In general, the helix in protomer A has a stronger tendency to sample the shifted (G150T-like) conformation.

Fig. S7 .
Fig. S7.Thioimidate intermediate forms in G150T ICH 30 seconds after mixing with pnitrophenyl isocyanide (p-NPIC) substrate.The 1.3 Å resolution Fo(30s)-Fo(0s) isomorphous difference electron density map (2.5s, green), mFo-DFc omit difference electron density map calculated before the model of the p-nitrophenyl thioimidate was included (3.0s, purple), and the 2mFo-DFc omit electron density map calculated before the model of the thioimidate was included (0.8s, blue) all show clear evidence for the covalent modification of C101.The chemical structure of the intermediate is shown to the right.

Fig. S8 .
Fig. S8.Correlated changes in protein and solvent during catalysis in G150T ICH.The 1.3 Å resolution Fo(30s)-Fo(0s) isomorphous difference electron density map is contoured at 3.0s with positive features in green and negative ones in red.Helix H, the IJ linker region from the other protomer (purple, indicated with apostrophe), and surrounding solvent (red spheres, labeled "wat") all display spatially correlated changes in conformation upon intermediate formation during catalysis.

Fig. S9 .
Fig. S9.Ramachandran plot of I152 in wild-type (WT) and G150T ICH.(A) I152 backbone torsions are in a forbidden region of the Ramachandran plot in wild-type ICH (green dot, phi and psi torsion values labeled).This residue is modeled into high-quality electron density and thus the backbone geometry of this residue is not a modelling error.(B) The G150T mutation causes a shift of I152 backbone torsions into the favored region of Ramachandran plot, relieving backbone strain and shifting the position of helix H, which is C-terminal to I152.

Fig. S10 .
Fig. S10.Isomorphous difference map features during catalysis are localized and overlap with areas of elevated ADPs.(A) Difference ADP values between the thioimidate structure and the free enzyme (Bdiff=B30s-B0s) are shown, with scale indicated at the lower left.The IJ linker and helix H ADPs are notably elevated upon thioimidate formation.(B) The Fo(30s)-Fo(0s) difference map features were integrated at the absolute difference density above a noise threshold (IADDAT).The threshold was 3s (corresponding to 0.17 e -/Å 3 ) within 2.5 Å of the model, which was averaged per residue and mapped onto the dimer.The high IADDAT values in helix H and the IJ linker agree with the pattern of elevated ADPs in (A).(C) Residue-averaged main chain values for the ADPs (Bave) are plotted vs. residue number.Helix H and the IJ linker (labeled) have markedly higher average mainchain ADPs upon thioimidate formation during G150T ICH catalysis compared to the resting free enzyme, as mapped onto the G150T ICH dimer in (A).

Fig. S11 .
Fig. S11.Anisotropic ADPs in G150T ICH change in magnitude but not direction upon thioimidate formation.(A,B) Anisotropic ADP ellipsoids for Ca atoms at the 85% probability level are colored by magnitude, from blue (5 Å 2 ) to red (30 Å 2 ).Formation of the thioimidate intermediate (labeled; pink highlight) during catalysis elevates ADP magnitudes in the active site, the IJ linker, and helix H. Larger ADP magnitudes are indicated with larger ellipsoid volumes and warmer colors.(C,D) Rosenfield difference analysis (62) indicates no major changes in ADP directional preferences upon intermediate formation.Secondary structural elements in ICH are shown on the left to the Rosenfield matrices, with helices as red rectangles and strands as blue arrows.Rosenfield analysis determines the difference between the projections of two atoms' ADP ellipsoids onto the line joining them.Rigid body-like correlated motion of a pair of atoms would result in low difference values (lighter colors).The Rosenfield matrix is computed by averaging mainchain difference ADP projection values in 30 bins (~7 residues per bin).

Table S3 . Crystallographic refinement statistics 1
Anisotropy is defined as the ratio of the smallest to largest eigenvalue of the ADP tensor. 2 Defined as the number of close contacts (atom overlap > 0.4 Å) per 1000 atoms.