What MTHFR A1298C and C677T Change, and What They Don't

There is no symptom list for MTHFR. The common variants (C677T, rs1801133, and A1298C, rs1801131) are inherited by a large fraction of the population, and carrying one or two copies does not produce a recognizable clinical picture. What the genotype changes is a measurable biochemical parameter: the efficiency with which 5,10-methylenetetrahydrofolate is reduced to 5-methyltetrahydrofolate, the form of folate that donates a methyl group to homocysteine. So the right question is not “what are my symptoms” but “what is my plasma total homocysteine, my folate status, and my B12 status, and does the genotype change how I should read those numbers.” Fatigue, anxiety, and brain fog are too common and too nonspecific to be attributed to a variant that roughly a third of people carry.

What the two common variants do to the enzyme

C677T is a missense change, Ala222Val, in the catalytic domain. The valine form destabilizes the binding of flavin adenine dinucleotide (FAD), the riboflavin-derived cofactor, so the enzyme is thermolabile and loses activity faster under stress. In vitro, lymphocyte extracts from 677TT individuals typically show on the order of 30% of wild-type activity, and heterozygotes roughly 60–70%. Critically, that loss is partly rescued by folate and riboflavin: FAD binding is stabilized when 5-methylTHF is bound, which is why the homocysteine effect of 677TT is seen mostly in people with low folate status and largely disappears at high folate status.

A1298C is Glu429Ala in the regulatory domain. Its effect on enzyme activity is much smaller, and in most population studies A1298C homozygotes do not show elevated homocysteine. Compound heterozygosity (677CT plus 1298AC) has been reported to reduce activity more than either heterozygote alone, but the biochemical shift is modest. Systematic variant-effect mapping across MTHFR missense variants confirms this picture: variant effects are conditional on folate and riboflavin availability, and many variants that look damaging under low-cofactor conditions behave near-normally when cofactor supply is adequate ¹. An earlier screen of rare MTHFR missense alleles found a similar phenomenon, with several reduced-activity variants restored by supplemental folate in a complementation assay ². The general clinical review literature reaches the same conclusion: the common polymorphisms have wide associations reported and small, environment-dependent effect sizes ³.

Getting your genotype out of your own data

If you have whole-genome sequencing, the two variants are single nucleotide positions on chromosome 1p36.22. MTHFR is transcribed from the minus strand, which is the source of most confusion about which allele is “the mutation.” On GRCh38, rs1801133 sits at chr1:11796321. Its reference is G and its alternate is A on the plus strand, corresponding to 677C>T on the transcript. The second variant, rs1801131, sits at chr1:11794419. It corresponds to 1298A>C, with T as the reference and G as the alternate. Confirm both against dbSNP before you trust a script, and check which build your VCF uses, since the GRCh37 coordinates differ by about 60 kb.

bcftools view -r chr1:11794419,chr1:11796321 sample.g.vcf.gz \
  | bcftools query -f '%CHROM\t%POS\t%ID\t%REF\t%ALT\t[%GT\t%DP\t%GQ]\n'

Look at depth and genotype quality as well as the genotype string. The string alone is not enough. At 30x mean coverage these positions are unremarkable and usually well covered, but a DP under about 10 or a GQ under 20 is worth re-checking. If you are working from a consumer array export rather than sequencing, both SNPs are directly genotyped on common chips, so no imputation is involved. Array text files, however, report alleles on a platform-specific strand convention. Verify by checking allele frequencies: the 677T allele runs around 30% in European ancestries and the 1298C allele around 30% as well, so a result implying a 2% allele is almost certainly a strand flip.

One more thing the raw data will show you: MTHFR has multiple transcript variants and alternative first exons, so if you are mapping RNA-seq reads or interpreting an expression value, specify which transcript you mean ⁴. Isoform-level quantification with a tool like salmon or kallisto against a GENCODE transcriptome is more informative here than a single gene-level count.

The measurements that carry signal

Plasma total homocysteine (tHcy) is the readout. It integrates the inputs below, which makes it the correct single number to watch and also means an abnormal value is not self-explanatory.

• MTHFR activity
• folate
• B12
• riboflavin
• B6
• renal function
• thyroid status

Pre-analytics dominate the result. Homocysteine continues to be exported from red cells into plasma after draw, raising measured values on the order of 10% per hour at room temperature in uncentrifuged EDTA blood. Draw fasting into EDTA, keep the tube on ice, and centrifuge and separate plasma within 30 to 60 minutes. A “high homocysteine” that came from a tube that sat on a bench for three hours is an artifact, and this is the most common reason a self-ordered panel produces a scary number.

Pair tHcy with serum folate, red cell folate (which reflects the last few months rather than the last meal), serum B12, and methylmalonic acid (MMA). MMA is the discriminator: it rises in B12 deficiency and not in folate deficiency, so elevated tHcy with normal MMA points toward the folate and MTHFR axis, while both elevated points toward B12. If you want to distinguish genotype effect from nutritional state, this four-marker panel does most of the work. Interpretation of any abnormal result belongs with a physician, particularly in pregnancy, with a personal or family history of venous thrombosis, or with neurologic findings.

When MTHFR variants are genuinely serious

Severe MTHFR deficiency is a different entity from the common polymorphisms. It is caused by rare biallelic loss-of-function or severely hypomorphic variants, presents with markedly elevated homocysteine and low or low-normal methionine, and can appear in infancy with encephalopathy or later with gait disturbance, psychiatric features, or thrombosis. The mutation-update literature catalogs well over a hundred distinct disease-causing alleles with residual activity typically below 20% ⁵, and characterization of individual homocystinuria-associated mutations shows the mechanism ranges from truncation to disrupted FAD binding ⁶. This is the case where a variant in your VCF genuinely matters, and it is why an annotation pipeline that only checks the two famous SNPs is inadequate. Annotate the whole gene: run VEP or snpEff over the MTHFR region, filter for coding and splice-region consequences, and check gnomAD allele frequency and ClinVar classification for anything rare. If you find a rare biallelic finding, that is a genetics clinic conversation. It is not a self-interpretation exercise.

Questions people also ask

Should you take B12 if you have MTHFR? Genotype alone does not answer that. Serum B12 plus MMA, with homocysteine, tells you whether B12 status is adequate. If MMA is elevated, that is a finding to bring to a clinician, who will also want to exclude absorption problems and renal impairment before anything else.

What vitamins should you avoid with MTHFR? The claim that people with MTHFR variants should avoid folic acid is not supported by the enzyme biology. The variants reduce the conversion of methylene-THF to methyl-THF, a step downstream of folic acid reduction, and folate-responsiveness of reduced-activity variants has been demonstrated directly in functional assays ²¹. Folate intake during pregnancy is a clinical matter to settle with an obstetrician.

Can MTHFR mutations cause anxiety? Associations between the common polymorphisms and psychiatric outcomes have been reported repeatedly, with small effect sizes and inconsistent replication across the broader literature ³. A 677TT genotype is not an explanation for an individual’s anxiety, and treating it as one tends to displace evaluation of causes that are treatable.

How does MTHFR deficiency affect B12 levels? Reduced MTHFR activity lowers the supply of 5-methylTHF needed by methionine synthase, which can trap B12 in unusable form functionally without lowering serum B12 measurably. That is why MMA and homocysteine together read the pathway better than serum B12 alone.

Does MTHFR genotype belong in a broader panel? Genomic panels that report methylation-pathway variants alongside many other genes have been described as clinical decision support rather than as standalone answers ⁷, and reviews of nutrigenomic frameworks stress the same limitation: single-variant interpretation without biochemical measurement is speculative ⁸.

Footnotes

1. Jochen Weile, Nishka Kishore, Song Sun, et al. Shifting landscapes of human MTHFR missense-variant effects. The American Journal of Human Genetics, 2021. https://doi.org/10.1016/j.ajhg.2021.05.009 ↩ ↩²

2. Nicholas J. Marini, Jennifer Gin, Janet Ziegle, et al. The prevalence of folate-remedial MTHFR enzyme variants in humans. Proceedings of the National Academy of Sciences, 2008. https://doi.org/10.1073/pnas.0802813105 ↩ ↩²

3. Antoni F. Araszkiewicz, Krzysztof Jańczak, Paweł Wójcik, et al. MTHFR Gene Polymorphisms: A Single Gene with Wide-Ranging Clinical Implications—A Review. Genes, 2025. https://doi.org/10.3390/genes16040441 ↩ ↩²

4. Anja Homberger, Michael Linnebank, Carsten Winter, et al. Genomic structure and transcript variants of the human methylenetetrahydrofolate reductase gene. European Journal of Human Genetics, 2000. https://doi.org/10.1038/sj.ejhg.5200522 ↩

5. D. Sean Froese, Martina Huemer, Terttu Suormala, et al. Mutation Update and Review of Severe Methylenetetrahydrofolate Reductase Deficiency. Human Mutation, 2016. https://doi.org/10.1002/humu.22970 ↩

6. Sahar Sibani, Benedicte Christensen, Erin O’Ferrall, et al. Characterization of six novel mutations in the methylenetetrahydrofolate reductase (MTHFR) gene in patients with homocystinuria. Human Mutation, 2000. https://doi.org/10.1002/(sici)1098-1004(200003)15:3<280::aid-humu9>3.0.co;2-i ↩

7. Sharon Hausman-Cohen, Carol Bilich, Sandeep Kapoor, et al. Genomics as a Clinical Decision Support Tool for Identifying and Addressing Modifiable Causes of Cognitive Decline and Improving Outcomes: Proof of Concept Support for This Personalized Medicine Strategy. Frontiers in Aging Neuroscience, 2022. https://doi.org/10.3389/fnagi.2022.862362 ↩

8. Jung Min Cho, Joo Won Shin. Nutrigenomics for Autism Spectrum Disorder: A Multidisciplinary Framework for Personalized Nutrition in Special Education. Clinical Nutrition Research, 2025. https://doi.org/10.7762/cnr.2025.14.4.329 ↩