∆G formation of ubiquinol/one

[kevinsh...@u.northwestern.edu]

Hi Elad,

I am trying to use the ComponentContribution class to calculate the values of a reaction involving ubiquinol/one. I have used the following reaction strings to get the thermodynamics at pH = 7, I = 0.1M, pMg = 3 using the search_reaction and standard_dg_prime functions:

1) fumarate + ubiquinol = succinate + ubiquinone --> 24 +/- 7 kJ/mol

2) fumarate + ubiquinol-2 = succinate + ubiquinone-2 --> 74 +/- 7 kJ/mol

3) fumarate + ubiquinol-6 = succinate + ubiquinone-6 -- > 74 +/- 7 kJ/mol

My question is why is 1 so different from 2 and 3? A couple of thoughts I have on this right now.

1) The ubiquinone/ol compounds in reaction 1 have the IDs 167 and 191, both of which I believe are in the training data. Reaction 2 returns IDs 691534 and 10162, which are not in the training data. A difference of 50 kJ/mol seems very large to me and I am not sure where it is coming from. I inspected the compound objects for both of these sets and couldn't quickly find a difference between them.

2) 2 and 3 having the same values is the behavior I would expect, as the change in group vectors is the same between those two reactions. That being said, which of these values is more reliable? For some of my models there may be varying lengths of ubiquinol/one, so a simple solution would be to just use the comopunds that yielded the 24 kJ/mol value for every instance of ubiquinone/ol if that is more accurate. However, I would like to understand the discrepancy in the calculated values here so I can be aware of any potential issues moving forward from here.

Thanks,

Kevin Shebek

[Elad Noor]

Hi Kevin,

Indeed reaction 1 uses direct information about the reduction potential of ubiquinol (stored in this table), while 2-3 use the group contribution method since we don't have specific information about ubiquinol-2 and -6.

Such specialized electron carriers are notoriously difficult to estimate (since they contain a large number of conjugated double bonds), and therefore the direct measurement can deviate a lot from the group contribution estimate.

By the way, maybe you noticed that the InChI structure of "ubiquinol" and "ubiquinol-2" is the same in our database. That is because we need to have a structure in order to estimate pKas, but it is not used for the group contribution in the former case.

I think it is reasonable to assume that all ubiquinols have a similar reduction potential as you suggested. The easiest solution would be to replace all specific ubiquinol/ones with the generic one.

That being said, as far as I know, ubiquinones are typically not present in solution, but rather embedded in the membrane as part of the electron transport chain and therefore applying the "standard" thermodynamic analyses to them is not helpful.

I typically allow their concentration to be totally free (i.e. not necessarily within the bounds of 1 uM to 10 mM). If you do that, the reduction potential becomes meaningless.

Best,

Elad