Minor Fixes To Lactate Dehydrogenase Reactions

[Devlin-Moyer]

MAR04280 (pyruvate[m] + NADH[m] <-> lactate[m] + NAD⁺[m]) should be removed, because having any significant amount of LDH activity in the mitochondrial matrix would rapidly dissipate the mitochondrial membrane potential and generally significantly disrupt oxidative respiration (previous papers claiming to find LDH in the mitochondrial matrix were probably detecting LDH in the intermembrane space)

MAR01453, MAR11476, and MAR11479 should all have their GPRs changed to "ENSG00000111716 or ENSG00000134333 or ENSG00000166796 or ENSG00000166800 or ENSG00000166816 or ENSG00000171989". At the moment, they all start with "(ENSG00000111716 and ENSG00000134333) or ENSG00000111716 or ENSG00000134333", which, as far as I know, is a nonsensical GPR

[haowang-bioinfo]

please also provide supporting evidence for modifying MAR01453, MAR11476, and MAR11479

[Devlin-Moyer]

Oh sorry I misinterpreted what you meant by that comment earlier; I'm not sure what kind of evidence I could provide for that? Because the problem is that the GPRs currently say "(LDHA and LDHB) or LDHA or LDHB or ...", and, as far as I know, the point of a GPR is to convey which genes are necessary for catalysis, so "LDHA and LDHB" would imply that both genes must be expressed concurrently in order to be capable of catalyzing this reaction, and "LDHA or LDHB" would imply that having either gene expressed is sufficient for this reaction to be catalyzed, and both can't be true simultaneously. But I assume you have a much deeper/more nuanced understanding of GPRs than I do, so let me know if I'm wrong about their purpose

[feiranl]

Thanks for identifying the GPR mistake for MAR01453, MAR11476, and MAR11479. We will change that ASAP.

As for the mitochondria rxn MAR04280 removal, I am not 100% sure about this. Firstly, recently, we have noticed there are some compartment annotation issue with Lactate dehydrogenase(#516). And I double checked that in Uniprot annotation, LDHB and LDH6B were annotated to be in mitochondria.

[Devlin-Moyer]

In the paper I cited in my first message, the section titled "Experimental approaches to study the intracellular lactate shuttle in isolated mitochondria" cites a bunch of papers failing to find evidence of LDH activity in the matrix of human mitochondria at the beginning and concludes with:

Specifically, the presence of high activities of LDH on both sides of the inner membrane, along with transmembrane transporters for pyruvate and lactate, provides no driving force to account for the observed steep redox gradient across the inner membrane. In fact, this intracellular lactate shuttle would create a short-circuit pathway to dissipate the cytosolic-matrix redox gradient generated by the malate-aspartate and glycerol-3-phosphate shuttles, as proposed by Sahlin and colleagues (Sahlin et al. 2002). The reality is that the transmembrane redox gradient is, in fact, observed, and its existence was an important motivating factor to explore the malate-aspartate and glycerol-3-phosphate electron shuttles and their energy dependence. Direct mitochondrial lactate oxidation is thermodynamically inconsistent with well-known observations.

The following section, "Insights from modeling in silico", uses a relatively simple kinetic model to show that

pyruvate and lactate oxidation are mutually exclusive; progressively higher matrix LDH activity would increase direct mitochondrial lactate oxidation, but it would also progressively impair the capacity to oxidize pyruvate to rates far below those routinely measured in laboratories.

Also, the Uniprot page for LDHB specifically says that LDHB is located on the inner mitochondrial membrane, and the associated citation repeatedly says that none of their experiments could determine whether LDHB was catalyzing the reaction in the intermembrane space or in the mitochondrial matrix. The Uniprot page for LDH6B (also known as LDHAL6B) does say it's in the mitochondrial matrix, but it doesn't actually cite any sources to substantiate that. Also, the "function" section cites RHEA, which points back to Uniprot, cites this Reactome entry (which cites this paper that never directly demonstrates catalytic activity or subcellular localization of LDHAL6B in humans and seems to be the only paper about LDHAL6B), and cites papers about LDH genes/isozymes in Selenomonas ruminantium, Lactococcus lactis, Pediococcus acidilactici, and Plasmodium falciparum. So I'm not actually convinced that LDHAL6B can actually catalyze this reaction in the first place, much less in the mitochondrial matrix

[Devlin-Moyer]

Also I just read through #516 and #528 and see that you were going to remove most/all genes from the GPR of MAR04281, the peroxisomal LDH reaction; this paper directly observed both LDHA and LDHB in peroxisomes of human cells.

Idk if these details are important to you, but the situation is that there are two possible transcription stop sites in the human LDHB gene, and the longer transcript contains a peroxisomal targeting sequence. LDHA never has this peroxisomal targeting sequence, but LDHA and LDHB form tetramers that can have both LDHA and LDHB monomers in them, so LDHA bound to LDHB with a peroxisomal targeting sequence will be carried along with it into the peroxisome (again, the above paper directly observed this happening)

So anyway I think you should keep LDHA (ENSG00000134333) and LDHB (ENSG00000111716) in the GPR of MAR04281

[feiranl]

Thanks for identifying this. I will fix this ASAP!

[haowang-bioinfo]

MAR04280 (pyruvate[m] + NADH[m] <-> lactate[m] + NAD+[m]) should be removed, because having any significant amount of LDH activity in the mitochondrial matrix would rapidly dissipate the mitochondrial membrane potential and generally significantly disrupt oxidative respiration (previous papers claiming to find LDH in the mitochondrial matrix were probably detecting LDH in the intermembrane space)

@Devlin-Moyer do you happen to have any FBA data supporting this removal?

[Devlin-Moyer]

Are you asking if I have FBA data that specifically shows that flux through the LDH reaction in the mitochondrial matrix disrupts the mitochondrial membrane potential? If so, I don't, because I don't know how one would represent a membrane potential in FBA. I think you'd need to use one of the thermodynamic FBA approaches and have access to some pretty thorough metabolomics data; it's possible that one of the thermodynamics FBA papers did this, but I kind of doubt that any of them did it on Human1 specifically, since most of the thermodynamic FBA papers I've encountered stick to small (<2,000 reactions) microbial models (presumably because their algorithms are generally much slower than ordinary FBA).
That paper I mentioned in the comment you were replying to did make a simplified kinetic model of the situation that supported those claims.

[haowang-bioinfo]

MAR04280 (pyruvate[m] + NADH[m] <-> lactate[m] + NAD+[m]) should be removed, because having any significant amount of LDH activity in the mitochondrial matrix would rapidly dissipate the mitochondrial membrane potential and generally significantly disrupt oxidative respiration (previous papers claiming to find LDH in the mitochondrial matrix were probably detecting LDH in the intermembrane space)

As described in #599, [m] in Human-GEM doesn't represent mitochondrial matrix alone, but a combination of matrix and intermembrane space at some degree. That said, LDH presence in the intermembrane space would warrant keeping MAR04280, while the removal might be an overkill