Inhibition Mechanism of SARS-CoV-2 Main Protease with Ketone-Based Inhibitors Unveiled by Multiscale Simulations

Abstract

<p>We present the results of combined classical and QM/MM simulations for the inhibition of SARS-CoV-2 3CL protease by a recently proposed ketone-based covalent inhibitor, PF-00835231, that is under clinical trial. In the noncovalent complex formed after binding into the active site the carbonyl group of this inhibitor is accommodated into the oxyanion hole formed by the NH main chain groups of residues 143 to 145. The P1-P3 groups of the inhibitor establish similar interaction with the enzyme to those of equivalent groups in the natural peptide substrate, while the hydroxymethyl moiety of the inhibitor partly mimics the interactions established by the P1’ group of the peptide in the active site. Regarding the formation of the covalent complex, the reaction is initiated after the proton transfer from Cys145 to His41. Formation of the covalent hemithioacetal complex takes place by means of the nucleophilic attack of the Sg atom of Cys145 on the electron deficient carbonyl carbon atom and a proton transfer from the catalytic His41 to the carbonyl oxygen atom mediated by the hydroxyl group. Our findings can be used as a guide to propose modifications of the inhibitor in order to increase its affinity by the 3CL protease.</p>

Impact of warhead modulations on the covalent inhibition of SARS-CoV-2 Mpro explored by QM/MM simulations

“…[15] First of all, the enzyme:substrate initial complex is the neutral C145/H41 dyad (E:I) instead of the ion pair dyad C145 -/H41 + (E (+/-) :I). This result is in agreement with studies carried out by us, and others, using different inhibitors and substrates, [14,17,19] but in contrast with the protonation state of the catalytic dyad suggested from the ligand-free SARS-CoV-M pro recently solved by neutron crystallography at pH 6.6. [20] Nevertheless, as already pointed out, questions remain about how pH or the presence of an inhibitor or substrate influence the protonation state of the dyad in SARS-CoV M pro .…”

“…Thus, while some studies report a neutral dyad significantly more stable than the ion pair (by ca. 11 kcal•mol -1 ) and even not being an stable state, [19] others suggest the ion pair is not so destabilized with respect the initial neutral dyad, [14,16] or even slightly more stable than the initial neutral dyad (i.e. our previous study with B1).…”

“…As previously proposed from X-ray diffraction studies, [3] and later supported by computational studies using different approaches, the chemical reaction leading to M pro inactivation requires the imidazole group of H41 to activate the SH group of C145 to generate a highly nucleophilic thiolate (CysS -) that would readily react with the inhibitor. [14][15][16][17][18][19] According to the recent literature, the equilibrium between the neutral dyad and the CysS -/HisH + ion pair appears to be tipped in favour of the neutral pair by ligand binding, but it may depend on the stereoelectronic properties of the ligand itself. Thus, while some studies report a neutral dyad significantly more stable than the ion pair (by ca.…”

Characterization of the non-covalent interaction between the PF-07321332 inhibitor and the SARS-CoV-2 main protease

“…For example, in the classical MD simulations reported in Refs. [ [25] , [26] , [27] ] (extending from a minimum of 0,1 to a maximum of 8 μs), the ligand orientation consistently remained similar (RMSD ′ 2 . 5: 3 Å) to that of the starting pose, prepared using the X-ray with the ligand covalently bound to the cysteine, despite the former was not covalently bound to the protein and the cysteine was protonated.…”

“…[ 25 ] and destabilising in Ref. [ 26 ] One weak point of these QM/MM approaches for the holo forms is that the latter are prepared [ [25] , [26] , [27] ] starting from the X-ray structures where the ligand is already covalently bound to CYS and that were presumably obtained from protein stock pre-incubated with the ligand. The X-ray structure, on the other hand, is not necessarily similar to the actual initial non covalent bonding pose as the transition state involving the substrate and the binding site must be the result of a concerted process with a C145 to H41 PT transfer induced by the ligand non covalent docking, followed by a structural modification involving both the ligand and the nearby residues (including the oxyanion hole [ 28 ]) to arrive at the covalently bonded form seen in the ligand incubated samples.…”