A pair of bases in the DNA, say, A-T, have a tautomerized form A*-T* (resulting from switching the sides of both protons along the hydrogen bridges).

I have studied how, by means of DFT computations, one can compute the energy difference between the normal and the anomalous forms. But, another point of interest is the stability of the tautomer.

I have read several times that those anomalous tautomers "decay" to the normal A-T form very quickly (much quicker than the time it lasts a round of DNA replication), which is interesting because it implies that tautomerization is not likely to be a source of spontaneous mutations.

What I would like to learn is how can one compute this expected life-time of the anomalous tautomer

I have an idea (which I don't know if it could be rigth or if maybe is completely worthless), which involves making a Born-Oppenheimer separation for the two protons (''after'' the DFT computation of the electronic ground state) and then study the non-adiabatic transitions between the different vibronic states of the couple of protons. I am happy to provide details, if someone whises so.

But perhaps "my method" is too messy, or naive (or both) and / or there are other standar approaches to compute approximately the expected mean life of anomalous tautomers such as the mentioned A*-T*.

I would be very grateful if someone could apport some insigth and/or references into how are these expected life-times calculated typically in computational chemistry!

  • $\begingroup$ I am by no means an expert, but molecular dynamics simulations are sometimes used to study lifetimes. A lot of force fields exist, both for unreactive and reactive systems. Seeing as how tautomerism involves breaking and reforming of bonds, perhaps the ReaxFF could be fitted, validated, and then used? $\endgroup$ – Yoda May 10 '16 at 22:09
  • 3
    $\begingroup$ find TS for tautomerizarion -> compute rate constant using TS theory -> compute half-life for tautomer $\endgroup$ – Jan Jensen May 11 '16 at 5:49
  • $\begingroup$ I had not heard of the ReaxFF, thanks for the note. @Jan Jense: could you elaborate further your answer, if you have time to spare? $\endgroup$ – user28429 May 11 '16 at 9:55
  • 4
    $\begingroup$ Have a look at molecularmodelingbasics.blogspot.dk/2009/08/… $\endgroup$ – Jan Jensen May 11 '16 at 10:19
  • 1
    $\begingroup$ The protons might tunnel, so kinetics vs. barrier height might have a different relationship. $\endgroup$ – Karsten Theis Jan 4 at 22:23

TLDR: With a quantum/semi-classical transition state theory.

I order to determine the lifetime of the tautomer, you need to know the rate constant for the decay from the tautomer to the conventional dimer. This is a kinetics question rather than a thermodynamic question so you won't be able to answer this by comparing single point DFT energies. The rate will also vary with temperature.

A complicating factor is that the nuclei moving are protons in hydrogen bonds. Hydrogen bonds are frequently double well potentials with two minima, one near each end of the bond - in this case the tautomer and conventional dimer. In these case the physical distance between the minima is small and the proton may tunnel through the potential barrier between the two minima - hence nuclear quantum effects are likely to be important, at least at low temperatures.

Methods that can calculate nuclear quantum effects can be very computational expensive and may require you to work in a reduced dimensional represention - only considering the two protons quantum mechanically and potentially simplifying the surrounding structure.

Two potential methods that could calculate a rate taking account of nuclear quantum effects are semiclassical transtition state theory, or ring-polymer molecular dynamics which has been applied within pockets of larger systems such as proteins.

  • $\begingroup$ For more information about his class of reactions - often referred to as proton-coupled electron transfer - see this review. $\endgroup$ – user213305 Mar 6 at 22:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy