search

SysBioChalmers / Human-GEM

The generic genome-scale metabolic model of Homo sapiens
https://sysbiochalmers.github.io/Human-GEM-guide/
Creative Commons Attribution 4.0 International
98 stars 41 forks source link

Redox of Alpha-Tocopheryl (Hydroxy)Quinone #611

Open Devlin-Moyer opened 1 year ago

Devlin-Moyer commented 1 year ago

While looking into FAD/ubiquinone reactions in #607, I came across these two reactions that involve ubiquinone and don't have any genes associated with them: MAR03047: alpha-tocopheryl hydroquinone [c] + ubiquinone [c] ⇔ alpha-tocopheryl quinone [c] + ubiquinol [c] MAR06500: alpha-tocopheryl hydroquinone [c] + ubiquinone [m] ⇔ alpha-tocopheryl quinone [c] + ubiquinol [m]

According to this paper, Complexes I and III of the electron transport chain can use alpha-tocopheryl quinone in place of ubiquinone, but ubiquinone/ol and alpha-tocopheryl (hydro)quinone cannot directly oxidize or reduce each other, so MAR06500 should be replaced by new reactions that look like MAR06921 and MAR06918 with ubiquinone/ol replaced with alpha-tocopheryl (hydro)quinone. Since there is not currently a version of either alpha-tocopheryl (hydro)quinone metabolite in [m], and these would be the only mitochondrial reactions they would participate in, it seems like overkill to add [c] to [m] transport reactions for both metabolites, so I'm just suggesting that the new reactions replacing MAR06500 involve both [c] and [m] metabolites, since that would not be functionally different from adding the transport reactions and mitochondrial metabolites.

That same paper also mentioned that the cytochrome b5 reductases (CYB5R1/2/3/4/L) can reduce either either ubiquinone or alpha-tocopheryl quinone using NADH in the cytosol, and I also found this paper that mentioned that CBR1 (ENSG00000159228) can do the same but with NADPH and not NADH, and this paper that found that NQO1 (ENSG00000181019) can use either NADPH or NADH to reduce either ubiquinone or alpha-tocopheryl quinone (both CBR1 and NQO1 are cytosolic according to Uniprot).

MAR01801 and MAR01806 already represent reduction of ubiquinone by NADPH (MAR01801) or NADH (MAR01806) in the cytosol, but are currently only associated with TXNRD1, which Uniprot thinks is only capable of reducing thioredoxin. So I think that the GPRs of MAR01801 and MAR01806 should be replaced with this new list of enzymes that can actually reduce cytosolic ubiquinone, MAR03407 should be edited to represent the reduction of alpha-tocopheryl (hydro)quinone by NADPH, and a new reaction should be created to represent its reduction by NADH:

haowang-bioinfo commented 1 year ago

based on this issue, the redox of alpha-tocopheryl (hydroxy)quinone is associated to OXPHOS complexes. To save the time for evaluation, can you please just directly paste the text here from the references

Devlin-Moyer commented 1 year ago

From this paper: Reduction of alpha-tocopheryl quinone (which this paper calls TQ1) to alpha-tocopheryl hydroxyquinone by Complex I:

The lack of succinate-dependent TQ1 reduction in the presence of antimycin A (vide infra) shows that electron transfer from endogenous UQH2 to exogenous TQ1 does not occur. Therefore, TQ1 is directly reduced by complex I.

Reduction of alpha-tocopheryl quinone to alpha-tocopheryl hydroxyquinone by Complex III:

From the lack of succinate-dependent TQ1 reduction in the presence of antimycin A we conclude, similar to above, that TQ1 interacts directly with the quinone reduction site (Qi pocket) of complex III. Reduction of short-chain UQ homologues and analogues by complex III has been observed before [46]. TQ1 binds to complex III with comparable affinity, but with only 1% reactivity with respect to UQ1 (Table 3).

Also I just realized that I confused Complexes II and III in the original set of suggested changes, so I fixed that (this paper showed that Complex II does not reduce alpha-tocopheryl quinone)

edkerk commented 1 year ago

If TQ1 is directly reduced by complex I, should MAR06500 not also involve translocation of protons?

Devlin-Moyer commented 1 year ago

Oh so looking more closely at the paper I cited when I said that Complex I and Complex III could reduce TQ1, they specified that TQ1 is binding to the same site that ubiquinone usually binds to, so if we wanted to fully represent the mechanism here, MAR06500 should probably be edited to look like MAR06921 but with ubiquinone/ol replaced with alpha-tocopheryl (hydroxy)quinone, and then also add a new reaction that looks like MAR06918 but with ubiquinone/or replaced with alpha-tocopheryl (hydroxy)quinone. I can update my proposed changes & #644 accordingly if you think that sounds reasonable

edkerk commented 1 year ago

Sounds indeed reasonable! Regarding the number of protons being pumped, would that still be the same (4 for complex I and 4 for complex III), or would that depend on the redox potential of TQ1? I'm not sure, and I'm even less sure whether any useful data is available/can be deciphered. In that case, the default of 4 and 4 H+ seems like a reasonable estimate.

Devlin-Moyer commented 1 year ago

Yea I was planning on using the same number of protons as the existing ubiquinone versions, since I think it's already the case that the number of protons pumped per NADH oxidized is somewhat variable when ubiquinone is the final acceptor, and that Complex I does several electron transfer steps between FeS clusters elsewhere in the enzyme that drive at least part of the proton pumping before the final transfer to ubiquinone/alpha-tocopheryl quinone, which presumably have about the same potential differences whether ubiquinone of alpha-tocopheryl quinone is at the end

Devlin-Moyer commented 1 year ago

That inspired me to revisit the other papers I'd initially cited and I found that they all described enzymes that could reduce either ubiquinone or alpha-tocopheryl quinone, but using NAD(P)H and not each other, so I have substantially revised my proposed changes (I think I'm also gonna just close #644 since it is now completely different from these changes)