Conformational gating, dynamics and allostery in human monoacylglycerol lipase

Inhibition of human Monoacylglycerol Lipase (hMGL) offers a novel approach for treating neurological diseases. The design of inhibitors, targeting active-inactive conformational transitions of the enzyme, can be aided by understanding the interplay between structure and dynamics. Here, we report the effects of mutations within the catalytic triad on structure, conformational gating and dynamics of hMGL by combining kinetics, NMR, and HDX-MS data with metadynamics simulations. We found that point mutations alter delicate conformational equilibria between active and inactive states. HDX-MS reveals regions of the hMGL that become substantially more dynamic upon substitution of catalytic acid Asp-239 by alanine. These regions, located far from the catalytic triad, include not only loops but also rigid α-helixes and β-strands, suggesting their involvement in allosteric regulation as channels for long-range signal transmission. The results identify the existence of a preorganized global communication network comprising of tertiary (residue-residue contacts) and quaternary (rigid-body contacts) networks that mediate robust, rapid intraprotein signal transmission. Catalytic Asp-239 controls hMGL allosteric communications and may be considered as an essential residue for the integration and transmission of information to enzymes’ remote regions, in addition to its well-known role to facilitate Ser-122 activation. Our findings may assist in the identification of new druggable sites in hMGL.

hMGL conformations. In the case of D239A (Fig. S4c), the extreme peak shape heterogeneity and overlapping peaks indicates a state of substantially increased dynamic flexibility. Nevertheless, the peaks for all mutants are distributed enough across the spectra, giving an overall view of the enzyme's behavior during ligand binding. Remarkably, many resonance lines in the bound states with compound 1 appear to sharpen significantly, indicating more homogeneous dynamics in the bound states.

Site-directed Mutagenesis, Expression and Purification of sol-hMGL and Mutants.
The full DNA sequence for designed substitutions was submitted to GenScript (Piscataway, NJ) and synthesized DNA were provided upon full sequencing and cloning in pET-45b(+). Each mutated plasmid was transformed and expressed in BL21 (DE3) E.coli cells. Ten different constructs were generated based on this sol-hMGL template: S122A, S122C, S122T, H269A, D239A, D239N, H121A, L241A, C242A and Y268A.
Recombinant enzyme expression and purification were performed based on a previously reported protocol 2 . 15 N-labeled cells for the catalytic triad mutants were grown in Spectra-9 media for bacterial cell growth (Cambridge Isotopes Labs) and purified. The expression levels and yields were comparable to those of sol-hMGL. SDS-PAGE and size-exclusion chromatography were employed to confirm the purity and oligomeric state of the protein preparations (data not shown). All mutants were high purity and were found to exist in solution as mixtures of monomers and oligomers.
Enzyme Assays. The effects of mutations on enzyme hydrolytic activity were determined by incubation of purified construct with endogenous substrate 2-arachidonoylglycerol (2-AG) and quantifying the product arachidonic acid (AA) by high-performance liquid chromatography (HPLC). The 2-AG substrate was generously supplied by NIH (Bethesda, MD) and AA was purchased from Nu-Chek Prep (Elysian, MN).
The initial velocity data generated from original UV-Vis and HPLC results were fitted to the Michaelis-estimate Vmax, Km, kcat and catalytic efficiency (kcat/Km). Circular Dichroism Spectroscopy. Potential conformational changes in the secondary structure of mutants were monitored in the far-UV region between 190 and 260 nm with enzymes concentrations 10 μM (300 μL) in a quartz cuvette with a path length of 1 mm. The components of buffer (pH 7.4) for CD experiments were 20 mM sodium phosphate, 100 mM NaCl and 2 mM TCEP. Three accumulations of scans were taken and averaged to get the complete spectra.

Nuclear Magnetic Resonance Spectroscopy and Data Analysis.
For 1D 1 H NMR spectra, 3-9-19 WATERGATE 3 pulse sequence (p3919fpgp) with gradients and additional flip back pulse was used for optimal detection of downfield exchangeable proton resonances. Exponential multiplication (broadening factor lb = 20 Hz) was applied for all 1D NMR spectra. 2D 1 H -15 N HSQC NMR spectra were acquired using the standard Bruker pulse sequences supplied with AVANCE 700 spectrometer, as detailed before 2 .
The data were processed and visualized using Topspin 3. Well-Tempered metadynamics simulations. Briefly, the catalytic triad residues and adjacent water molecules were set as the QM region at B3LYP/6-31G (d, p) level, whereas the rest of the protein was set as the MM region. The Qsite calculations were conducted under gas phase, where QM region was optimized up to 700 steps and the MM region remained unchanged. The sol-hMGL as well as each of the mutants was first solvated in SPC water with 10 Å buffer and 0.15 M NaCl. The systems were then equilibrated using the default protein relaxation protocol first with restraints on the solute heavy atoms: 1) 100 ps of Brownian dynamics NVT at 10 K, 2) 12 ps of NVT simulation at 10K, 3) 12 ps of NPT simulation at 10K, and 4) 12 ps of NPT simulation at 300K; followed by 24 ps of NPT simulations at 300K without any restraints. The welltempered metadynamics was carried out for 350 ns at 300K using a Gaussian width of 0.1 Å with the Figure S1. Comparison of the far-UV CD spectra for the catalytic triad mutants of sol-hMGL. The protein concentrations were ~ 10 μM. Figure S2. Thermal denaturation curves for the catalytic triad mutants of sol-hMGL. Enzymes were subjected to a temperature gradient in 20 mM sodium phosphate, 100 mM NaCl, 2 mM TCEP buffer, pH 7.4, and unfolding was followed by monitoring the CD signal at 222 nm. Data points are shown in black circles and the sigmoidal fits used to determine Tm.   sol-hMGL Figure S8. Sequence coverage map for peptic peptides that were identified by MS/MS spectra for D239A mutant. The peptides are presented as bars.
D239A Table S1. Percentage deuterium uptake for the peptides detected in sol-hMGL and two catalytic triad mutants S122A and D239A for D2O immersion of t = 30 s to t = 4 h. The data points are presented as mean ± standard error (1σ), which were computed from data observed in the triplicate experiments. The values are adjusted for back exchange.