Hidden alternative structures of proline isomerase essential for catalysis

Abstract

A long-standing challenge is to understand at the atomic level how protein dynamics contribute to enzyme catalysis. X-ray crystallography can provide snapshots of conformational substates sampled during enzymatic reactions, while NMR relaxation methods reveal the rates of interconversion between substates and the corresponding relative populations. However, these current methods cannot simultaneously reveal the detailed atomic structures of the rare states and rationalize the finding that intrinsic motions in the free enzyme occur on a timescale similar to the catalytic turnover rate. Here we introduce dual strategies of ambient-temperature X-ray crystallographic data collection and automated electron-density sampling to structurally unravel interconverting substates of the human proline isomerase, cyclophilin A (CYPA, also known as PPIA). A conservative mutation outside the active site was designed to stabilize features of the previously hidden minor conformation. This mutation not only inverts the equilibrium between the substates, but also causes large, parallel reductions in the conformational interconversion rates and the catalytic rate. These studies introduce crystallographic approaches to define functional minor protein conformations and, in combination with NMR analysis of the enzyme dynamics in solution, show how collective motions directly contribute to the catalytic power of an enzyme.

Figure 1. Room-temperature X-ray crystallography and Ringer analysis detect conformational substates in CypA

a, Local maxima above the 0.3σ threshold (yellow line) in Ringer plots reveal alternate side-chain conformations in room-temperature (red line) but not cryogenic (blue line) electron density for Ser99, Leu98, Met61 and Arg55. b, Electron-density maps calculated using room-temperature X-ray data define the alternate conformers of Leu98, Ser99 and Phe113. 2Fo-Fc electron density (blue mesh; 1σ); positive (green) and negative (red) Fo-Fc difference density (3σ). c, 2Fo-Fc composite simulated-annealing omit electron density maps (1.0σ (dark blue) and 0.3σ (light blue)) show a unique conformation for Phe113 in the 1.2-Å-resolution cryogenic structure (blue) and distinct major (red) and minor (orange) conformers in the 1.39-Å-resolution room-temperature structure. Electron density around the main chain and the surrounding residues was omitted for clarity. d, Steric collisions across the network of major (red) and minor (orange) conformers of Arg55, Met61, Phe113 and Ser99 explain how side-chain motions link the active site to remote buried residues.

Figure 2. The structure of the Ser99Thr mutant resembles the minor conformer of wild-type CypA

a, _χ1 Ringer plot (0.3σ threshold is shown as yellow line) of the Ser99Thr mutant (dashed green) and room-temperature, wild-type Ser99 CypA structure (red) show that Thr99 occupies both positions populated by the Ser99-OH_γ group. The angular offset between the major peaks reflects a backbone shift. b, The 2Fo-Fc simulated-annealing omit electron density map of the Ser99Thr CypA mutant (1.0σ (dark blue) and 0.3σ (light blue)) shows apparently unique conformations for Thr99 and Phe113. The structure confirmed the prediction that rotation of Phe113 to the “out” position is coupled to rotation of the Ser99 hydroxyl to the minor rotamer. c, Phe113 and Met61 in Ser99Thr CypA (green, right) are detected exclusively in the position of the minor state of the wild-type enzyme (orange, left).

Figure 3. The Ser99Thr mutation shifts the equilibrium toward the minor wild-type conformation and slows motions in the dynamic network in free CypA

a, Significant ¹H-¹⁵N chemical-shift differences between Ser99Thr and wild-type CypA (red) propagate through group I residues (Arg55, Phe113, and Ser99 shown as black sticks). b, Linear amide chemical shift changes (arrows) between wild-type (black), Lys82Ala (red) and Ser99Thr (blue) CypA reflect the inversion of the major/minor equilibrium due to the Ser99Thr mutation. c, Residues undergoing slow (red) or fast (blue) motions on the NMR time scale in Ser99Thr (right) coincide with previously identified group I (red) and group II (blue) residues in wild-type (left) CypA (amides in grey are prolines or overlapped peaks). d, Temperature dependence of CPMG ¹⁵N NMR relaxation data for group I (left) and group II (right) in Ser99Thr CypA reveal that the mutation impedes group I conformational dynamics (R_EX ~ k₁ and R_EX increases with temperature). In contrast, group II residues are unaffected by the mutation and display the opposite temperature dependence characteristic of fast motions on the NMR time scale. Dispersion curves were normalized to R₂⁰ at 30 °C.

Figure 4. Impeded motions in the dynamic network severely reduce the catalytic power of a chemically competent enzyme

a, Mutations affecting the enzyme dynamics (Ser99Thr) or the chemical step (Arg55Lys) each drastically reduce k_cat/K_M by reducing the bidirectional isomerization step on the enzyme (k_cat^isom) and not substrate affinity (K_D) of CypA. b, ¹H-¹H NOE-exchange spectra at 0.2 s mixing time showing isomerization of the peptide A¹A²P³F⁴ (1 mM) by catalytic amounts of wild-type (black), Ser99Thr (green) and Arg55Lys (blue) CypA. Assignments and dashed lines connecting exchange peaks are included for wild-type. Much higher concentrations of the Ser99Thr and Arg55Lys variants are needed relative to wild-type CypA to obtain similar exchange peaks, reflecting severely reduced catalytic activity. c, The cis-and trans- peaks coalesce (asterisk) for wild-type CypA at the same enzyme concentration as the mutant forms due to its much greater activity. The only remaining off-diagonal peak is a P³α-A²α NOE characteristic for a cis-prolyl peptide bond.