Towards full Quantum Mechanics based Protein-Ligand Binding Affinities

Stephan Ehrlich, Andreas H. Göller, and Stefan Grimme (2017)

Contributed by Jan Jensen

Erlich et al. presents absolute binding free energies for activated serine protease factor X (FXa) and tyrosine-protein kinase 2 predicted using DFT. Here I'll focus on FXa. The calculations are based on truncated model systems consisting of ca 1000 atoms. The geometries are optimised using HF-3c/C-PCM and select constraints, the RRHO free energy correction with DFTB3-D3, the electronic energy with PBE-3c, and the solvation free energy with COSMO-RS and PBE0/def-SVP. The energy terms are simply added together to give a total free energy and the binding free energy is simply the change in free energy upon binding without any additional corrections.

The MAD is similar to that found for host-guest complexes but there are clearly some outliers. The authors ascribe L19 and L27 to errors in the structures due to HF-3c artefacts, while L23 is ascribed to the movement of a crystal water molecule and L10 is the only charged ligand where the error in the solvation free energy is likely higher. The error is below 1.5 kcal/mol for 14 of the 25 ligands.

Clearly there is room for improvement but I do think the results are quite encouraging. A MM-PB(GB)SA study in which five different solvation models are tested for the same ligands found maximum $r$ values of 0.28 and 0.60 using ensemble averaged and energy minimised structures respectively. Furthermore, study determined the relative binding free energies using thermodynamic integration, which is generally considered the current gold standard in the drug design, for five ligand pairs (see table, energies in kcal/mol). Given that there is only five points any statistical analysis of the accuracy would be suspect, but I don't think TI can be said to outperform DFT.

	DFT	TI	exp
5->18	4.0	-0.4	1.1
5->12	2.1	0.4	0.4
5->21	7.5	1.3	4.4
5->17	4.1	-0.2	-0.3
5->24	4.1	0.4	3.6

The real question is whether the DFT results can be systematically improved and the main sticking point here will ultimately be the solvation free energy, especially for charged ligands. The continuum model ultimately relies on a fit to experimental data so there is some degree of empiricism that is hard to remove. In principle it can be done by adding explicit water molecules but then the question is how to deal with the sampling in a cost effective way.

This work is licensed under a Creative Commons Attribution 4.0

Preparation of an ion with the highest calculated proton affinity: ortho-diethynylbenzene dianion

Poad, B. L. J.; Reed, N. D.; Hansen, C. S.; Trevitt, A. J.; Blanksby, S. J.; Mackay, E. G.; Sherburn, M. S.; Chan, B.; Radom, L., Chem. Sci. 2016, 7, 6245-6250
Contributed by Steven Bacharach
Reposted from Computational Organic Chemistry with permission

The new benchmark has been set for superbases. The previous record holder was LiO^–, with a computed proton affinity of 424.9 kcal mol^-1. A new study by Poad, et al., examines the dianions of the three isomeric phenyldiacetylides: 1o, 1m, and 1p.¹ Their computed proton affinities (G4(MP2)-6X) are 440.6, 427.0, and 425.6 kcal mol^-1, respectively. The optimized geometries of these dianions are shown in Figure 1.

1o

1m

1p

Figure 1. Optimized geometries of 1o, 1m, and 1p.

The authors also prepared these bases inside a mass spectrometer. All three deprotonate water, but do not deprotonate methane, though that might be a kinetic issue.

The authors speculate that 1o will be hard to beat as a base since loss of an electron is always a concern with small dianions.

References

1) Poad, B. L. J.; Reed, N. D.; Hansen, C. S.; Trevitt, A. J.; Blanksby, S. J.; Mackay, E. G.; Sherburn, M. S.; Chan, B.; Radom, L., "Preparation of an ion with the highest calculated proton affinity: ortho-diethynylbenzene dianion." Chem. Sci. 2016, 7, 6245-6250, DOI: 10.1039/C6SC01726F.

InChIs

1o: InChI=1S/C10H4/c1-3-9-7-5-6-8-10(9)4-2/h5-8H/q-2
InChIKey=RVSCTJNIQWGMPY-UHFFFAOYSA-N

1m: InChI=1S/C10H4/c1-3-9-6-5-7-10(4-2)8-9/h5-8H/q-2
InChIKey=ATCNFGGQUWGWOE-UHFFFAOYSA-N

1p: InChI=1S/C10H4/c1-3-9-5-7-10(4-2)8-6-9/h5-8H/q-2
InChIKey=GGQMWKMAMDPRPA-UHFFFAOYSA-N

'
This work is licensed under a Creative Commons Attribution-NoDerivs 3.0 Unported License.

Conformer-specific hydrogen atom tunnelling in trifluoromethylhydroxycarbene

Mardyukov, A.; Quanz, H.; Schreiner, P. R., Nat. Chem. 2017, 9, 71–76
Contributed by Steven Bacharach
Reposted from Computational Organic Chemistry with permission

The Schreiner group has again reported an amazing experimental and computational study demonstrating a fascinating quantum mechanical tunneling effect, this time for the trifluoromethylhydroxycarbene (CF₃COH) 2.¹ (I have made on a number of posts discussing a series of important studies in this field by Schreiner.) Carbene 2 is formed, in analogy to many other hydroxycarbenes, by flash vapor pyrolysis of the appropriate oxoacid 1 and capturing the products on a noble gas matrix.

Carbene 2t is observed by IR spectroscopy, and its structure is identified by comparison with the computed CCSD(T)/cc-pVTZ frequencies. When 2t is subjected to 465 nm light, the signals for 2t disappear within 30s, and two new species are observed. The first species is the cis conformer 2c, confirmed by comparison with its computed CCSD(T)/cc-pVTZ frequencies. This cis conformer remains even with continued photolysis. The other product is determined to be trifluoroacetaldehyde 3. Perhaps most interesting is that 2t will convert to 3 in the absence of light at temperatures between 3 and 30 K, with a half-life of about 144 h. There is little rate difference at these temperatures. These results are quite indicative of quantum mechanical tunneling.

To aid in confirming tunneling, they computed the potential energy surface at CCSD(T)/cc-pVTZ. The trans isomer is 0.8 kcal mol^-1 lower in energy that the cis isomer, and this is much smaller than for other hydroxycarbenes they have examined. The rotational barrier TS1 between the two isomer is quite large, 26.4 kcal mol^-1, precluding their interchange by classical means at matrix temperatures. The barrier for conversion of 2t to 3 (TS2) is also quite large, 30.7 kcal mol^-1, and insurmountable at 10K by classical means. No transition state connecting 2c to 3 could be located. These geometries and energies are shown in Figure 1.

2c 0.8

TS1 26.4

2t 0.0

TS2 30.7

3 -49.7

Figure 1. Optimized geometries at CCSD(T)/cc-pVTZ. Relative energies (kcal mol^-1) of each species are listed as well.

WKB computations at M06-2X/6-311++G(d,p) predict a half-life of 172 h, in nice agreement with experiment. The computed half-life for deuterated 2t is 10⁶ years, and the experiment on the deuterated analogue revealed no formation of deuterated 3.

The novel component of this study is that tunneling is conformationally selective. The CF₃ group stabilizes the cis form probably through some weak H^…F interaction, so that the cis isomer can be observed, but no tunneling is observed from this isomer. Only the trans isomer has the migrating hydrogen atom properly arranged for a short hop over to the carbon, allowing the tunneling process to take place.

References

1) Mardyukov, A.; Quanz, H.; Schreiner, P. R., "Conformer-specific hydrogen atom tunnelling in trifluoromethylhydroxycarbene." Nat. Chem. 2017, 9, 71–76, DOI: 10.1038/nchem.2609.

InChIs

1: =1S/C3HF3O3/c4-3(5,6)1(7)2(8)9/h(H,8,9)
InChIKey=GVDJEHMDNREMFA-UHFFFAOYSA-N

2: InChI=1S/C2HF3O/c3-2(4,5)1-6/h6H
InChIKey=FVJVNIREIXAWKU-UHFFFAOYSA-N

3: InChI=1S/C2HF3O/c3-2(4,5)1-6/h1H
InChIKey=JVTSHOJDBRTPHD-UHFFFAOYSA-N

'
This work is licensed under a Creative Commons Attribution-NoDerivs 3.0 Unported License.