Homocysteine, B12 and MTHFR

Methionine synthase takes a methyl group from 5-methyltetrahydrofolate and hands it to homocysteine, making methionine. It cannot work without vitamin B12 [1]. When B12 runs low, homocysteine rises, and folate becomes trapped in its methyl form, so folate cannot do its other jobs either.

That is why homocysteine works as a functional B12 test. Serum B12 measures how much vitamin is in the blood, not whether cells can use it; low serum B12 does not always mean deficiency, and severe functional deficiency can occur with normal serum B12 [1].

Who tends to run low: long-term metformin users, in whom B12 fell by 19% over 4.3 years [7]; vegans and vegetarians without supplements [8]; and older adults. The whole-food plant-based protocol keeps a B12 target for this reason, and HbA1c and metformin users may want B12 on their radar.

B12 has a second job: it is the cofactor for methylmalonyl-CoA mutase in mitochondria. When B12 is low, methylmalonic acid (MMA) accumulates as well as homocysteine [1]. Folate plays no role in that second enzyme, so low folate raises homocysteine without raising MMA.

British haematology guidelines put it this way: serum B12 is the first-line test, plasma MMA is a second-line test to clarify uncertain results, and homocysteine can help as a second-line test but is less specific than MMA [2]. The same guideline warns that no single test is a gold standard, and that strong clinical features of deficiency should be treated without waiting for perfect numbers [2].

A practical reading:

• High homocysteine, high MMA: points to B12 deficiency.
• High homocysteine, normal MMA: points to folate shortfall, MTHFR TT with low folate, kidney or thyroid causes, or lifestyle factors.
• Normal homocysteine and MMA: functional B12 deficiency is unlikely.

Interpretation of these combinations belongs with your clinician.

MTHFR (methylenetetrahydrofolate reductase) converts 5,10-methylenetetrahydrofolate into 5-methyltetrahydrofolate, the main circulating form of folate and the methyl donor for recycling homocysteine [9]. In 1995, researchers identified a common C677T variant, present on about 38% of chromosomes in the population studied, that makes the enzyme heat-sensitive and less active; people with two copies had higher homocysteine [9].

How common TT is varies widely. In the China Kadoorie Biobank, 21% of 156,000 adults carried TT, ranging from 5% to 41% across regions [5].

The size of the effect depends on diet. Across 237 genetic datasets, TT raised homocysteine by 3.12 µmol/L compared with CC in low-folate Asia, but by only 0.13 µmol/L in the Americas, Australia and New Zealand, where food is fortified with folic acid [3]. Riboflavin matters too: the TT enzyme loses its riboflavin cofactor easily, and 1.6 mg of riboflavin a day lowered homocysteine by 22% in TT carriers, with no effect in CC or CT [10].

MTHFR is a natural experiment for homocysteine, so these studies speak to causation:

• Heart disease. In 19 unpublished datasets with 48,175 coronary cases, TT carried an odds ratio of 1.02 (0.98 to 1.07), and 1.01 in unsupplemented low-folate populations. Published studies gave 1.15, and the authors concluded that the gap reflects publication bias [4]. In 156,000 Chinese adults, TT was unrelated to ischaemic heart disease [5].
• Stroke. TT raised stroke odds by 1.68 in low-folate Asia and 1.03 in fortified regions [3]. In China, TT carried 13% higher stroke risk, with a stronger link to intracerebral haemorrhage [5]. Genetic analyses tie homocysteine mainly to small vessel stroke [11].
• Death. A Korean study using MTHFR found no causal effect of homocysteine on all-cause mortality [12].

Earlier claims linking MTHFR to blood clots and pregnancy loss were also tested; meta-analyses did not support a link with venous thromboembolism [6].

The American College of Medical Genetics and Genomics states that MTHFR polymorphism testing has minimal clinical utility and should not be ordered as part of a routine thrombophilia evaluation [6]. The reasons are straightforward:

• Genotype does not predict your level. In folate-replete populations, TT barely changes homocysteine [3]. A homocysteine test measures what the genotype cannot.
• Genotype does not change treatment. If homocysteine is high, the work-up and treatment follow the level and the cause, whatever the genotype [2].
• Genotype does not predict heart disease in unbiased data [4].

Consumer DNA tests often report MTHFR and attach long symptom lists to it. Those lists are not supported by the trial or genetic evidence. If you already know you are TT, a homocysteine test and adequate folate, B12 and riboflavin intake are the evidence-based response [10][13].

Methylfolate (5-methyltetrahydrofolate) skips the MTHFR step, which makes it appealing in theory. The evidence base for lowering homocysteine, however, comes mostly from folic acid: 25 trials pooled for dose effects [13] and outcome trials such as CSPPT, which stratified participants by MTHFR genotype and used 0.8 mg folic acid [14]. Riboflavin specifically lowered homocysteine in TT carriers [10].

In regions with folic acid fortification, TT carriers have homocysteine close to CC carriers [3], which suggests ordinary folic acid intake already compensates for most of the variant's effect.

Choosing between folic acid and methylfolate is a reasonable question for a clinician, especially when B12 status is uncertain, since any folate can mask the anaemia of B12 deficiency.