On chromosome 1p13, two genes sit in close proximity: TSPAN2 (tetraspanin-2, a membrane scaffold protein) and NGF (nerve growth factor, a neurotrophin critical for the development and sensitization of pain-sensing neurons). A regulatory variant between them — rs2078371 — carries one of the largest single-SNP effect sizes for migraine susceptibility ever identified, ranking 3rd in effect magnitude among 123 loci in the largest migraine genome-wide association study conducted to date¹¹ 3rd in effect magnitude among 123 loci in the largest migraine genome-wide association study conducted to date
Hautakangas et al. 2022 — 102,084 migraine cases, 771,257 controls, Nature Genetics.

The Mechanism

The C allele at rs2078371 sits in the intergenic region between TSPAN2 and NGF and is annotated as a downstream regulatory variant near LINC01765, a long non-coding RNA in this locus. The precise molecular mechanism remains under investigation, but the most compelling biological model points to NGF (nerve growth factor)²² NGF (nerve growth factor)
a secreted neurotrophin that binds TrkA receptors on small-diameter trigeminal sensory neurons, upregulating TRPV1, TRPA1, and Nav1.8 — the ion channels that amplify pain signals during migraine. When NGF is overexpressed or its signalling is dysregulated, trigeminal afferents become hyperexcitable — a state known as peripheral sensitization, in which normally sub-threshold stimuli (light, sound, head movement) trigger pain.

TSPAN2 is a tetraspanin scaffold protein expressed in oligodendrocytes and in endothelial cells. It organizes cell-surface signalling complexes and may influence the local vascular and neuronal environment of the meninges, the pain-sensitive membranes surrounding the brain. Whether the locus effect is mediated through NGF expression, through TSPAN2 function in meningeal cells, or through the LINC01765 non-coding RNA regulating one or both genes is not yet resolved. The GWAS signal pointing to rs2078371 as a tag variant does not identify the causal allele or the gene it acts on — it marks a chromosomal region where one or more regulatory variants alter migraine susceptibility through this cluster of interacting genes.

The biological interpretation is strongly supported by the clinical observation that the effect is larger in migraine without aura (MO)³³ migraine without aura (MO)
the most common migraine subtype, affecting ~75% of migraine sufferers; it lacks the visual/neurological aura that precedes migraine with aura and is thought to have a larger peripheral sensitization component than in migraine with aura. Peripheral sensitization — the NGF-mediated mechanism — is disproportionately implicated in MO pathophysiology, in contrast with the cortical spreading depression that drives aura.

The Evidence

The genetic evidence is robust across three independent large-scale GWAS:

Hautakangas et al. 2022⁴⁴ Hautakangas et al. 2022
Genome-wide analysis of 102,084 migraine cases identifies 123 risk loci and subtype-specific risk alleles. Nature Genetics. is the most definitive study. Among 123 migraine loci identified across more than 873,000 individuals, rs2078371-C has OR=1.112 and P=6×10⁻⁴² — the 3rd strongest effect size in the entire study. The subtype analysis found the signal was driven primarily by migraine without aura, where OR was larger, directly implicating the peripheral/trigeminal sensitization pathway.

Anttila et al. 2013⁵⁵ Anttila et al. 2013
Genome-wide meta-analysis identifies new susceptibility loci for migraine. Nature Genetics. established the TSPAN2 region at chromosome 1p13 as one of five newly discovered migraine loci in 23,285 cases, providing the first genome-wide-significant evidence for this chromosomal region. Gormley et al. 2016⁶⁶ Gormley et al. 2016
Meta-analysis of 375,000 individuals identifies 38 susceptibility loci for migraine. Nature Genetics. confirmed the signal in a much larger meta-analysis, cementing the locus as one of the most reliably replicated in migraine genetics.

At the gene level, Coskun et al. 2016⁷⁷ Coskun et al. 2016
Association of BDNF and NGF gene polymorphisms with susceptibility to migraine. Neuropsychiatr Dis Treat. showed that NGF coding variants (rs6330 T allele, OR~1.6 vs controls) are themselves associated with migraine risk, providing direct evidence that NGF protein variation influences migraine biology and making the NGF gene a plausible causal mediator of the GWAS signal at rs2078371.

Practical Actions

Migraine management for C allele carriers is dominated by reducing triggers that amplify peripheral sensitization — the mechanism the locus implicates. Three interventions have specific relevance:

Anti-NGF pharmacology is the most mechanistically targeted approach. Anti-NGF monoclonal antibodies (tanezumab, fasinumab) have been studied extensively for pain conditions driven by peripheral sensitization and have been tested in migraine prevention contexts. While none are yet approved specifically for migraine, the biological logic is direct: if elevated NGF activity at this locus drives trigeminal hyperexcitability, dampening NGF signalling addresses the root mechanism. CGRP-pathway treatments (erenumab, fremanezumab, galcanezumab — all approved for migraine prevention) act downstream of the same trigeminal sensitization cascade, making them especially relevant for the MO-predominant mechanism this locus represents.

Trigger mapping and early abortive treatment are behavioural interventions with strong evidence specifically for MO patients with a high peripheral sensitization component — the subtype this variant most strongly predicts. Cutaneous allodynia (scalp tenderness during migraine) is a clinical marker of peripheral sensitization; carriers experiencing it should treat attacks early with triptans before central sensitization develops, since triptans lose efficacy once central sensitization is established.

Interactions

The rs6330 variant in the NGF coding region is a related SNP in the same gene. Carriers of both rs2078371-C (GWAS regulatory signal) and rs6330-T (NGF coding variant associated with migraine with aura) represent compound risk at the same biological locus, though these are independent associations. The combination has not been formally studied in joint analysis.

rs11172113 (near SLC2A9) and rs10166942 (near TRPM8) are other migraine GWAS loci with distinct mechanisms (ion channel and metabolic pathways) that may interact additively with TSPAN2/NGF locus risk. Compound actions across multiple migraine-risk loci should be considered when users carry multiple high-risk genotypes.

The pancreatic beta cell is a specialist — it exists for one purpose: sensing blood glucose and releasing exactly the right amount of insulin. Building and maintaining that specialization requires a master transcription factor called PAX4 (paired box gene 4¹¹ paired box gene 4
PAX4 controls beta-cell development and differentiation during embryogenesis and throughout life). When PAX4 function is compromised, the beta cell loses its identity, its capacity to produce insulin, and its ability to suppress glucagon — the hormone that keeps glucose rising between meals.

The rs2233580 variant encodes a single amino acid change in the PAX4 protein: arginine to histidine at position 192 (Arg192His). This is an East and Southeast Asian–specific variant, virtually absent in European and African populations. Among East Asians it reaches a minor allele frequency of roughly 9–10%, making it one of the most clinically significant population-specific diabetes risk variants known.

The Mechanism

PAX4 normally acts as a transcriptional repressor²² transcriptional repressor
a protein that binds to DNA and turns off neighboring genes. It binds the promoters of the insulin gene and the glucagon gene, keeping glucagon expression suppressed in beta cells while supporting insulin transcription. The Arg192His change sits in the homeodomain DNA-binding region of the protein and measurably weakens this repressor function. Luciferase reporter assays show that PAX4 R192H has significantly reduced ability to repress both the insulin and glucagon promoters compared to wild-type PAX4.

The downstream consequences are visible at the cellular level. In EndoC-βH1 beta cells, PAX4 knockdown impairs insulin secretion and reduces total insulin content. In CRISPR-derived hiPSC islet models³³ CRISPR-derived hiPSC islet models
human induced pluripotent stem cells differentiated into islet-like structures in the lab, deletion or mutation of PAX4 causes de-repression of alpha-cell gene expression (particularly glucagon), an increase in polyhormonal "confused" cells, and reduced beta-cell identity markers. These phenotypes were fully rescued by CRISPR-mediated gene correction — confirming that Arg192His is the causal variant, not a passenger mutation.

The Evidence

The first genome-wide significant association came from a Chinese exome-chip study: 7,189 T2D cases and 10,813 controls in Hong Kong and Guangzhou, combined p = 3.74×10⁻¹⁵⁴⁴ 7,189 T2D cases and 10,813 controls in Hong Kong and Guangzhou, combined p = 3.74×10⁻¹⁵
Guo et al. Exome-chip association analysis reveals an Asian-specific missense variant in PAX4 associated with type 2 diabetes in Chinese individuals. Diabetologia, 2017.

A large study of early-onset T2D in Chinese found that 21.4% of 2,886 patients with early-onset T2D carried at least one Arg192His allele⁵⁵ 21.4% of 2,886 patients with early-onset T2D carried at least one Arg192His allele
Ma et al. Missense variants in PAX4 are associated with early-onset diabetes in Chinese. Diabetes Therapy, 2021. The combined odds ratio was 1.88 (95% CI 1.37–2.60). Carriers showed higher HbA1c, lower 2-hour C-peptide levels, and a dose-dependent shift toward younger age of diagnosis.

Functional confirmation came from a 2023 Nature Communications study using isogenic hiPSC lines carrying the Arg192His allele: participants heterozygous or homozygous for p.His192 showed decreased acute insulin response to glucose and reduced HOMA-B beta-cell function⁶⁶ participants heterozygous or homozygous for p.His192 showed decreased acute insulin response to glucose and reduced HOMA-B beta-cell function
Hastoy et al. PAX4 loss of function increases diabetes risk by altering human pancreatic endocrine cell development. Nature Communications, 2023.

A Singapore cohort study among East Asians confirmed the association with younger age of T2D onset and reduced C-peptide⁷⁷ association with younger age of T2D onset and reduced C-peptide
Ang et al. PAX4 R192H is associated with younger onset of type 2 diabetes in East Asians in Singapore. Diabetes Research and Clinical Practice, 2019.

One pharmacogenomic study found that patients with the CC genotype (Arg/Arg) achieved significantly better fasting glucose control on sulfonylurea monotherapy than CT or TT carriers⁸⁸ CC genotype (Arg/Arg) achieved significantly better fasting glucose control on sulfonylurea monotherapy than CT or TT carriers
Chen et al. Effects of TCF7L2 rs7903146 and PAX4 rs2233580 on hypoglycemic agent therapeutic efficacy. Heliyon, 2024.

Practical Actions

Carriers of one or two T alleles have measurably reduced beta-cell function. The core implication is impaired glucose-stimulated insulin secretion. Dietary patterns that minimize postprandial glucose spikes directly reduce the demand placed on already-compromised beta cells. Monitoring for early metabolic changes (fasting glucose, HbA1c, fasting insulin) is warranted, especially in East Asian individuals who carry the variant — since the risk is population-concentrated, it is most clinically relevant for this group.

Carriers may have reduced benefit from sulfonylureas (which act by stimulating beta-cell insulin release) compared to biguanides (which improve insulin sensitivity), though this pharmacogenomic finding is preliminary and should be discussed with a prescribing physician.

Interactions

PAX4 rs2233580 (R192H) and rs3824004 (R192S) affect the same amino acid position 192 in the PAX4 homeodomain. Among East Asians, compound heterozygous carriers of one R192H and one R192S allele show dramatically higher T2D risk (OR ~3.23 vs OR ~1.78 for single-variant carriers), as documented in Ma et al. 2021 (PMID 33216280). This compound effect is worth flagging in genetic counseling for East Asian patients with early-onset T2D.

Both PAX4 rs2233580 and TCF7L2 rs7903146 affect beta-cell insulin secretion through distinct mechanisms (transcription factor function vs. Wnt/incretin pathway), and their combination in a patient informs differential response to first-line oral diabetes medications.

SLC47A1 encodes MATE1 (multidrug and toxin extrusion protein 1), a transporter that pumps organic cations — including metformin — from renal tubular cells and hepatocytes back into the urine or bile. While OCT1 and OCT2 pull metformin into cells from the bloodstream, MATE1 pushes it out the other side. Together, these transporters determine the net cellular concentration of metformin and, therefore, how powerfully the drug activates its molecular target.

The rs2289669 G>A variant sits in an intron of SLC47A1 on chromosome 17p11.2. The A allele is associated with reduced MATE1 efflux activity, meaning the renal tubular cells hold onto metformin longer before excreting it. This delayed clearance translates into prolonged tissue exposure and, in multiple studies of Chinese and Korean populations, a measurably stronger glucose-lowering response. The effect is not independent of OCT1 genotype at rs622342 — the two transporters are functionally coupled, and their combined genotype determines the full pharmacokinetic picture.

MATE1 is expressed on the apical (urine-facing) membrane of renal proximal tubular cells and on the canalicular membrane of hepatocytes. It functions as an antiporter: it exports organic cations like metformin in exchange for protons flowing inward. By exporting metformin into the renal filtrate and bile, MATE1 completes the cycle that started when OCT2 (in the kidney) and OCT1 (in the liver) imported the drug.

The A allele at rs2289669 reduces MATE1 transporter function, though the exact molecular mechanism is not definitively established. Genomic editing experiments showed that the variant itself does not alter SLC47A1 transcript expression or splicing — it may tag a causal functional variant in linkage disequilibrium, or it may exert regulatory effects not detectable in hepatocellular model systems. Functionally, regardless of mechanism, A allele carriers show reduced renal secretion of metformin: in a controlled Korean study, ranitidine (a competitive MATE1 inhibitor) unmasked a 15.2% lower metformin renal clearance in GA+AA carriers versus GG homozygotes.

The clinical consequence is paradoxical: reduced MATE1 function means more metformin stays in tissues longer, which activates AMPK more durably and suppresses hepatic gluconeogenesis more effectively. In Chinese patients with newly diagnosed type 2 diabetes on metformin monotherapy, the AA genotype produced approximately double the HbA1c reduction of GG carriers. The A allele also appears to amplify the metformin-associated rise in basal GLP-1, suggesting a secondary mechanism beyond direct AMPK activation.

The variant was first linked to metformin response in the Rotterdam Study. Becker et al. 2010¹¹ Becker et al. 2010
Interaction between polymorphisms in the OCT1 and MATE1 transporter and metformin response. Pharmacogenet Genomics 2010 found that rs2289669 alone had no significant main effect on HbA1c change in 98 incident metformin users, but showed a statistically significant multiplicative interaction with OCT1 rs622342 (p=0.015). In OCT1 CC homozygotes, the MATE1 A allele was associated with a 0.68% greater HbA1c reduction — the combined impairment of both influx (CC) and efflux (A allele) creating the most dramatic attenuation of transport, and paradoxically the strongest drug retention.

In Chinese patients, associations were stronger and more direct. He et al. 2015²² He et al. 2015
SLC47A1 rs2289669 G>A variants enhance the glucose-lowering effect of metformin via delaying its excretion. Diabetes Obes Metab 2015 conducted a randomized trial in 220 newly diagnosed type 2 diabetes patients with one-year follow-up. AA homozygotes achieved -2.32% HbA1c reduction versus -1.07% in GG carriers — a roughly two-fold difference — attributable to significantly lower urinary metformin excretion (p<0.01). Liang et al. 2017³³ Liang et al. 2017
Differential increments of basal GLP-1 concentration among SLC47A1 rs2289669 genotypes. Diabet Med 2017 replicated the glycemic association in 291 patients and added a mechanistic layer: A allele carriers showed significantly greater metformin-induced rises in basal GLP-1, and when GLP-1 changes were statistically controlled, the rs2289669 effect on HbA1c disappeared — suggesting GLP-1 secretion mediates the genotype-response relationship.

The pharmacokinetic dimension was confirmed in a controlled Korean study. Cho and Chung 2016⁴⁴ Cho and Chung 2016
The MATE1 rs2289669 polymorphism affects renal clearance of metformin following ranitidine treatment. Pharmacogenomics J 2016 showed that 26 healthy subjects with the GG genotype had 15.2% higher renal clearance of metformin after ranitidine administration than GA+AA carriers — demonstrating that co-administration of MATE1 inhibitors (ranitidine, cimetidine, trimethoprim) amplifies the effect of the A allele genotype.

A large meta-analysis by the MetGen consortium found no significant association of rs2289669 with metformin response. Kalamajski et al. 2022⁵⁵ Kalamajski et al. 2022
Genomic editing of metformin efficacy-associated genetic variants in SLC47A1. Eur J Hum Genet 2022 also showed that CRISPR editing of rs2289669 did not alter SLC47A1 expression or splicing. This raises the possibility that the variant tags a causal locus in LD rather than being directly functional, and that effect sizes in smaller studies were inflated. The evidence level is therefore moderate rather than strong or established.

For GG homozygotes, MATE1 function is normal, and metformin clearance is efficient. If type 2 diabetes is well-controlled on standard doses, no pharmacogenomic adjustment is needed. However, MATE1 inhibitor drugs (ranitidine, cimetidine, trimethoprim) can significantly raise metformin plasma levels, and this risk is magnified in GG carriers who rely on normal MATE1 function to maintain clearance.

For AA homozygotes, the reduced efflux means metformin concentrations are higher and more persistent. This confers a pharmacodynamic advantage — better glucose control at standard doses — but also greater susceptibility to metformin accumulation if renal function declines, or if MATE1-inhibiting drugs are co-administered. Monitoring renal function and being cautious with MATE1 inhibitors is especially important for this genotype.

This variant is the final element of the metformin pharmacogenomics panel alongside OCT1 rs622342 (SLC22A1) and ATM rs11212617. OCT1 controls metformin influx into hepatocytes; MATE1 controls efflux. The Rotterdam Study demonstrated a significant multiplicative interaction: the rs2289669 effect on glucose lowering is amplified several-fold in carriers of the OCT1 rs622342 CC genotype compared to AA carriers. The Chinese studies, conducted primarily in East Asian populations where the A allele is more uniformly represented, may partly explain the stronger main effects observed. Clinically, assessing all three variants together provides the most complete picture of predicted metformin response.

Melanopsin is the photopigment that makes you a social animal of time. Unlike the rod and cone photoreceptors that handle vision, melanopsin sits in a specialized subset of retinal ganglion cells — intrinsically photosensitive retinal ganglion cells (ipRGCs)¹¹ intrinsically photosensitive retinal ganglion cells (ipRGCs)
These cells project to the suprachiasmatic nucleus (SCN), the brain's master clock — and has one job: tell your brain what time of day it is based on ambient light. When melanopsin detects morning blue light, it suppresses melatonin, entrains your circadian clock, boosts alertness, and stabilizes mood. The OPN4 Pro10Leu variant (rs2675703) subtly reduces the amplitude of this melanopsin signal, making the clock's light input slightly quieter.

The Mechanism

At position 10 of the melanopsin protein, a proline (Pro) is swapped for leucine (Leu). This P10L substitution is located in the N-terminal intracellular domain of the G protein-coupled receptor. Rodgers et al. 2018²² Rodgers et al. 2018
Human Molecular Genetics, in-vivo rescue study using viral delivery to mouse ipRGCs tested the Pro10Leu variant directly using a mouse melanopsin-knockout model and found that ipRGCs expressing the human P10L variant showed reduced response amplitude compared to wild-type human melanopsin — even though the light sensitivity threshold (EC50) was unchanged. In behavioral tests, circadian photoentrainment and pupillary light responses were functionally rescued, suggesting the variant is not severely disabling. However, the reduced amplitude means the signal generated per photon is smaller, which matters most under dim or brief light exposures — exactly the conditions of winter mornings at higher latitudes.

The Evidence

The key clinical association comes from Roecklein et al. 2009³³ Roecklein et al. 2009
Journal of Affective Disorders, n=130 SAD / 90 controls. SAD participants had a significantly higher frequency of the homozygous T/T genotype: all 7 individuals (5% of the full sample) who carried TT were in the SAD group, yielding an odds ratio of 5.6 for SAD membership. The study also examined haplotypes across multiple OPN4 variants, but rs2675703 P10L was among the most significant individual associations. Importantly, recruitment methods differed between groups, and the result has not been independently replicated for SAD specifically.

The 2012 chronotype study by Roecklein et al.⁴⁴ Roecklein et al.
Chronobiology International, n=234 non-psychiatric community adults found that TT carriers — unlike CC and CT carriers — showed a significant interaction between daylength and sleep onset: TT individuals went to bed later on long days and earlier on short days, while CC/CT carriers showed no such seasonal shift. This suggests the variant doesn't uniformly disrupt sleep but rather changes how sleep timing responds to seasonal light cues.

A pilot study in Mexican individuals (Gutiérrez-Amavizca et al. 2021⁵⁵ Gutiérrez-Amavizca et al. 2021
IJERPH, n=29 insomnia / 98 controls) found a striking association between the T allele and chronic insomnia (dominant model OR=9.37, p=1×10⁻⁴). The confidence intervals are wide (8.18–335.66 for dominant model), reflecting the small sample; this should be considered a preliminary signal requiring replication.

The 2013 pupil study (Roecklein et al.⁶⁶ Roecklein et al.
Psychiatry Research, n=30) found that post-illumination pupil response (PIPR) was reduced in SAD patients overall, but the reduction tracked with the I394T variant (rs11150800), not P10L. This dissociates the two OPN4 variants: I394T primarily affects light sensitivity threshold, while P10L affects signal amplitude.

A 2025 systematic review (Lucio-Enríquez et al.⁷⁷ Lucio-Enríquez et al.
Frontiers in Neuroscience) synthesizing 9 OPN4 studies confirmed that P10L is associated with SAD, chronotype variability, and chronic insomnia — while noting that the overall body of evidence remains small and that insomnia patients with this variant also showed higher rates of obesity or overweight, potentially reflecting circadian disruption downstream effects.

Practical Actions

The melanopsin pathway is unusually amenable to behavioral intervention precisely because its substrate — light — is something you can control. For TT carriers, the implication of reduced melanopsin signal amplitude is that the circadian system may need a stronger or more precisely timed light input to achieve the same photoentrainment signal. Morning bright light exposure (10,000 lux for 20–30 minutes within an hour of waking) is the most evidence-based way to boost the ipRGC signal, and this is the same intervention used clinically for SAD. Evening blue-light blocking (using amber lenses or screen filters after sunset) reduces competitive melanopsin activation that would otherwise delay the melatonin onset.

CT heterozygotes have a less clear phenotype — most studies grouped CC and CT together as the non-TT reference — but some evidence suggests mild seasonal sleep timing variability. Awareness of this tendency allows proactive adjustments around the autumn-winter clock change.

Interactions

The other functionally studied OPN4 variant, rs11150800 (Ile394Thr / I394T), has a distinct functional profile: it reduces melanopsin light sensitivity (raises the EC50 threshold) and was specifically associated with reduced pupillary light response in SAD patients in the 2013 study. Since P10L affects amplitude and I394T affects sensitivity, individuals carrying both risk alleles may have compounded impairment of the ipRGC light signaling pathway — though no study has examined the compound genotype directly.

Proposed compound interaction — OPN4 P10L + I394T: Individuals carrying both TT (rs2675703) and the I394T risk genotype (rs11150800) may have doubly impaired melanopsin signaling (reduced amplitude AND reduced sensitivity), potentially requiring more intensive circadian light interventions than either variant alone. No published study addresses this combination, so any combined recommendation is mechanistic extrapolation.

PPARG¹¹ PPARG
Peroxisome Proliferator-Activated Receptor Gamma; the master transcription factor for adipocyte differentiation and insulin sensitivity, also the molecular target of thiazolidinedione diabetes drugs encodes PPARγ, the protein that determines how vigorously the body creates new fat cells and how efficiently existing adipocytes respond to insulin. rs2959272 sits in an intronic region of the gene at chromosome 3, position 12,401,334 (GRCh38). It was among six PPARG variants that reached statistical significance for predicting individual variation in weight loss during a structured calorie restriction trial — and separately, the G allele has been associated with elevated plasma renin activity, pointing to a PPARγ-mediated link between adipose tissue and the renin-angiotensin system.

The Mechanism

As an intronic variant, rs2959272 does not change any amino acid in the PPARγ protein. Its functional effect, if direct, would operate through regulatory mechanisms: altering a transcription factor binding site, affecting local chromatin accessibility, or acting as a linkage disequilibrium²² linkage disequilibrium
LD means two variants are inherited together so frequently that one can serve as a proxy marker for another's functional effect tag for nearby functional variants in the PPARG locus. The rs2959272 T/G SNP is also notable as an ambiguous variant — the complement of T is A and of G is C, meaning published papers that used minus-strand notation (reporting C/A alleles) refer to the same biological variant in complementary encoding.

During caloric restriction, PPARγ activity is actively repressed by SIRT1 — a key part of the body's fat-mobilization programme. Picard et al.³³ Picard et al.
Picard F et al. Sirt1 promotes fat mobilization in white adipocytes by repressing PPAR-gamma. Nature, 2004 demonstrated that SIRT1 binds to and represses PPARγ-controlled genes through NCoR and SMRT co-repressors, triggering lipolysis and fat loss from white adipocytes. Variants in intronic regions of PPARG that alter the gene's baseline expression level or its sensitivity to this SIRT1-mediated repression could therefore modulate how much fat is mobilised per unit of caloric deficit — explaining why different PPARG genotypes predict different weight loss trajectories.

A secondary mechanism involves the renin-angiotensin system. PPARγ agonists are known to suppress renin synthesis in juxtaglomerular cells, and conversely, variants that alter PPARγ tone in adipose tissue may influence plasma renin activity (PRA) — a pathway that in turn feeds back on blood pressure regulation and fluid balance during weight loss.

The Evidence

The primary association with weight loss variability comes from a study by Matsuo et al.⁴⁴ study by Matsuo et al.
Matsuo T et al. PPARG genotype accounts for part of individual variation in body weight reduction in response to calorie restriction. Obesity (Silver Spring), 2009, which genotyped eight PPARG variants in 95 middle-aged Japanese women (BMI ≥25 kg/m²) undergoing a 14-week structured calorie restriction program targeting 1,200 kcal/day with weekly dietary lectures. Overall weight loss averaged 7.7 ± 3.1 kg (11.3% reduction). Among all eight SNPs tested, six — including rs2959272 — were significantly associated with the degree of weight reduction. The strongest individual predictor was rs1175544 (P=0.004, accounting for 7% of variance), but rs2959272 co-reached significance in the same cohort. No association was found between these SNPs and changes in coronary heart disease risk factors accompanying weight loss, isolating the signal to body weight trajectory rather than broader metabolic remodelling.

A candidate gene study by Underwood et al.⁵⁵ candidate gene study by Underwood et al.
Underwood PC et al. The relationship between peroxisome proliferator-activated receptor-gamma and renin: a human genetics study. J Clin Endocrinol Metab, 2010 examined rs2959272 in 395 Caucasian-American and 55 African-American hypertensive subjects on a low-sodium diet protocol. Homozygous G-allele carriers (reported as "CC" in that paper's minus-strand notation) showed significantly higher plasma renin activity at baseline in both Caucasians (P=0.016) and African-Americans (P=0.027), with combined Fisher's P=0.002. This finding positions rs2959272 as a functional marker in the PPARγ–renin axis and supports a biological role, though it does not directly quantify the weight loss magnitude difference by genotype.

The overall evidence level is emerging: the calorie restriction signal comes from a single study of 95 participants in one population (Japanese women), and the renin finding, while replicated across two ethnic groups, does not directly address fat mobilisation. Large, multi-ethnic calorie restriction trials with independent replication are needed.

Practical Actions

For TG and GG carriers, the practical implication is that individual weight loss during calorie restriction may differ meaningfully from population averages — the Matsuo data show that PPARG genotype collectively accounts for a substantial portion of observed inter-individual variation. Structured tracking of actual weight trajectory versus expected trajectory (0.5–1 kg per week on a standard 500 kcal/day deficit) can detect early divergence and prompt adjustment of the dietary approach.

The renin-axis finding suggests that GG carriers maintain higher renin tone, which may affect fluid retention and apparent weight loss in the early weeks of calorie restriction. Attention to sodium intake during any weight loss program is particularly relevant for carriers of the G allele.

Interactions

rs2959272 belongs to the broader PPARG intronic locus that includes rs1175544, rs1175540, and rs709158 — all co-significant in the Matsuo 2009 calorie restriction study. These variants are in partial linkage disequilibrium and their combined haplotype context may predict calorie restriction response more accurately than any single SNP alone. The well-established PPARG Pro12Ala variant (rs1801282) directly modifies the PPARγ protein and determines baseline insulin sensitivity; it acts independently of the intronic haplotype but the two together define the full PPARG functional profile. No compound action has yet been characterised for the specific rs2959272 + rs1801282 combination.

APOA5 (Apolipoprotein A5) plays a key role in regulating triglyceride levels. Discovered in 2001¹¹ Discovered in 2001
Pennacchio et al. An apolipoprotein influencing triglycerides in humans and mice revealed by comparative sequencing. Science, 2001 through comparative sequencing of the APOA1/C3/A4 gene cluster, APOA5 was found to strongly influence plasma triglyceride concentrations in both humans and mice.

The Mechanism

The S19W variant (rs3135506) causes a serine-to-tryptophan substitution at position 19 (p.Ser19Trp) in the signal peptide²² The signal peptide is a short amino-acid sequence that directs a newly made protein for secretion out of the cell of the APOA5 protein. This disrupts the signal peptide function, reducing APOA5 secretion into the bloodstream by approximately 50%. Since APOA5 normally lowers triglycerides by stimulating lipoprotein lipase activity and inhibiting VLDL production, reduced secretion leads to higher triglyceride levels.

The Evidence

The ICARIA study³³ ICARIA study
Loria et al. Additive effects of LPL, APOA5 and APOE variant combinations on triglyceride levels and hypertriglyceridemia. BMC Med Genet, 2010 demonstrated that APOA5 S19W carriers have an independent, additive triglyceride-raising effect. Carriers of the rare allele show significantly higher levels of large VLDLs (+133%) and small dense LDLs (+34%), creating a proatherogenic lipid profile.

Guardiola et al.⁴⁴ Guardiola et al.
Guardiola et al. APOA5 variants predispose hyperlipidemic patients to atherogenic dyslipidemia and subclinical atherosclerosis. Atherosclerosis, 2015 confirmed that this variant predisposes carriers to atherogenic dyslipidemia and subclinical atherosclerosis — measurable thickening of artery walls even before symptoms appear.

High triglycerides are an independent risk factor for cardiovascular disease and pancreatitis⁵⁵ Very high triglycerides (above roughly 500 mg/dL) can trigger acute pancreatitis, a serious inflammation of the pancreas.

Practical Implications

The C allele is found in about 6% of Europeans and up to 14% of Hispanics, but is rare (<2%) in East Asian and African populations. Dietary interventions — especially limiting refined carbohydrates and increasing omega-3 intake — are the primary management strategy.

Interactions

Triglyceride-raising effects are additive when combined with variants in LPL and APOE genes. If you also carry APOE E4 (rs429358), your overall cardiovascular risk is compounded.

SIRT1 (Sirtuin 1) is often called the master regulator of aging biology. This NAD+-dependent protein deacetylase¹¹ NAD+-dependent protein deacetylase
an enzyme that removes acetyl groups from proteins in a reaction that consumes NAD+, coupling cellular energy status to gene regulation coordinates DNA repair, inflammation suppression, mitochondrial biogenesis, insulin sensitivity, and stress resistance across virtually every tissue. rs3758391 lies approximately 2 kb upstream of the SIRT1 transcription start site — in the promoter region that determines how much SIRT1 protein the cell makes.

The Mechanism

The T allele at rs3758391 is associated with higher SIRT1 mRNA and protein expression, while the common C allele is associated with a more moderate expression level. The AMD study by Kaikaryte et al.²² The AMD study by Kaikaryte et al.
Kaikaryte K et al. SIRT1: Genetic Variants and Serum Levels in Age-Related Macular Degeneration. Life (Basel). 2022 found that individuals carrying at least one T allele had significantly elevated serum SIRT1 levels compared with CC homozygotes. This might seem paradoxical given that SIRT1 is generally considered protective in aging — but the relationship between SIRT1 expression level and disease risk is not linear. Chronically elevated SIRT1 can disrupt normal cell-cycle checkpoints in proliferating cells, contributing to cancer susceptibility, while the same elevated SIRT1 may benefit post-mitotic tissues like neurons.

The rs3758391 promoter variant is part of a broader regulatory haplotype that includes rs3818292 (intron 1) and rs7895833 (intron 5). The G-T-G combination across these three sites modulates SIRT1 expression in a tissue- and age-specific manner.

The Evidence

Cognitive aging: The Leiden 85-plus study³³ The Leiden 85-plus study
Kuningas M et al. SIRT1 gene, age-related diseases, and mortality: the Leiden 85-plus study. J Gerontol A Biol Sci Med Sci. 2007 followed 1,245 very elderly participants and found that T allele carriers showed better cognitive functioning and a trend toward lower cardiovascular mortality. A Han Chinese aging cohort⁴⁴ Han Chinese aging cohort
Zhang WG et al. SIRT1 variants are associated with aging in a healthy Han Chinese population. Clin Chim Acta. 2010 of 482 individuals (246 aged 60–91 years) found the C allele OR=1.453 (p=0.026) and CC genotype OR=3.042 (p=0.027) were more prevalent in older participants, interpreted as the C allele correlating with reaching old age — suggesting C carriers have lower cumulative disease burden over decades.

Metabolic protection: A large Iranian study by Naseri et al. 2024⁵⁵ Naseri et al. 2024
Naseri R et al. Protective role of SIRT1 (rs3758391 T>C) polymorphism against T2DM and its complications: Influence on GPx activity. Health Sci Rep. 2024 (n=398) found the C allele protective against type 2 diabetes, diabetic neuropathy, and diabetic retinopathy. CC and TC genotypes showed significantly higher glutathione peroxidase (GPx) activity than TT carriers, suggesting the C allele promotes antioxidant capacity. A Bangladeshi case-control study with meta-analysis by Ahmed et al. 2025⁶⁶ Ahmed et al. 2025
Ahmed R et al. Association of SIRT1 rs3758391 Polymorphism With T2DM in Bangladeshi Population. Health Sci Rep. 2025 found TT vs CC OR=3.88 (95% CI 1.34–11.25) for T2DM, with the T allele carrying 1.67-fold higher odds (95% CI 1.07–2.60).

Cancer risk: In a Chinese cohort of 206 DLBCL lymphoma patients, Kan et al. 2018⁷⁷ Kan et al. 2018
Kan Y et al. SIRT1 rs3758391 polymorphism and risk of diffuse large B cell lymphoma in a Chinese population. Cancer Cell Int. 2018 found TT carriers had OR=3.518 (95% CI 1.68–7.39) for disease and the TT genotype was an independent poor prognostic factor (HR=1.981, p=0.006). A Greek study by Papantzimas et al. 2026⁸⁸ Papantzimas et al. 2026
Papantzimas I et al. Association between SIRT1 rs3758391 genetic variant and susceptibility to pancreatic and gastric cancer. Ann Gastroenterol. 2026 found TC genotype protective against pancreatic cancer (OR=0.35, p=0.004) and gastric cancer (OR=0.26, p=0.006), with C allele enriched in healthy controls (OR=0.39, p<0.001).

Survival: A large Russian cohort of 3,312 individuals by Erdman et al. 2025⁹⁹ Erdman et al. 2025
Erdman V et al. Genetic predictors of longevity and survival in cellular homeostasis genes. Gene. 2025 found TT genotype associated with improved survival in diabetes patients (HR=0.40, p=0.006) and multimorbidity (HR=0.48, p=0.025) — suggesting TT carriers who do develop chronic disease fare better once ill, perhaps due to elevated SIRT1-mediated stress resistance.

Contradictions in the literature reflect SIRT1's tissue- and context-specific biology. The T allele's higher SIRT1 expression may benefit neurons and stressed tissues (cognitive protection, survival under disease) while heightening cancer susceptibility through disrupted cell-cycle control. The C allele's lower expression appears to reduce metabolic disease risk and improve antioxidant tone over decades.

Practical Implications

For TT carriers, supporting SIRT1 function through upstream inputs — adequate NAD+ availability, caloric sufficiency, and polyphenol intake — remains important, while cancer screening vigilance gains additional relevance. For CC carriers, the genetic architecture already provides metabolic protection; attention to maintaining the antioxidant advantages (GPx activity) through adequate selenium and zinc intake (cofactors for GPx enzymes) reinforces what the genotype already provides.

Interactions

rs3758391 participates in the rs3818292-rs3758391-rs7895833 three-variant SIRT1 haplotype. The G-T-G combination across these sites was associated with increased odds of exudative age-related macular degeneration in the Lithuanian cohort. rs3758391 also appears in haplotype combinations associated with depression comorbidity in T2DM (PMID 39612426). The full range of SIRT1 regulatory effects requires considering all three regulatory variants together, since each modulates expression through different promoter and intronic elements. The companion SNP rs7895833 is already profiled in the GeneOps database.

The angiotensin-converting enzyme¹¹ angiotensin-converting enzyme
ACE cleaves angiotensin I into angiotensin II (a potent vasoconstrictor) and inactivates bradykinin (a vasodilator). It sits at the centre of the renin-angiotensin-aldosterone system (RAAS) governing blood pressure, fluid balance, and vascular tone gene is home to one of the most studied variants in the history of exercise genetics. The ACE insertion/deletion (I/D) polymorphism — a 287-base-pair Alu repeat sequence in intron 16 — determines circulating and tissue ACE activity in a dose-dependent manner: the D allele drives ACE levels up, the I allele keeps them down. Because standard short-read sequencing arrays cannot reliably genotype this structural variant directly, rs4341 serves as its practical proxy. rs4341 is an intronic C/G SNP in near-complete linkage disequilibrium with the I/D locus: the C allele tags the insertion, and the G allele tags the deletion.

This SNP has been studied in elite mountaineers, endurance runners, rowers, triathletes, sprint athletes, and strength trainers across dozens of populations. The picture that emerges is not a single "athletic gene" but a genuine biological dial — turned one way for aerobic efficiency and endurance, turned the other for power output and strength adaptation.

The Mechanism

ACE sits at a critical enzymatic crossroads. When the D allele is present — particularly in homozygous form (GG at rs4341) — ACE activity is elevated in both serum and skeletal muscle tissue. The consequence is higher circulating angiotensin II²² higher circulating angiotensin II
Angiotensin II is a potent vasoconstrictor that also promotes protein synthesis and cardiac hypertrophy via AT1 receptor signalling, making it anabolic for skeletal muscle under conditions of resistance training, which promotes vasoconstriction and skeletal muscle protein synthesis, and accelerated bradykinin degradation³³ accelerated bradykinin degradation
Bradykinin is a vasodilator that also activates nitric oxide synthase and promotes glucose uptake; its half-life is sharply reduced by elevated ACE activity, which reduces vasodilation and nitric oxide-mediated signalling.

The I allele (C at rs4341) does the opposite: lower ACE activity means less angiotensin II production and prolonged bradykinin half-life. Bradykinin signals through B2 receptors⁴⁴ B2 receptors
B2 bradykinin receptors activate phospholipase C and nitric oxide synthase, improving glucose uptake in skeletal muscle and promoting vasodilation during sustained exercise to promote nitric oxide synthesis, vasodilation, improved glucose uptake in working muscle, and greater mitochondrial efficiency — all traits that support sustained aerobic output.

The Evidence

The foundational evidence comes from decades of athlete cohort studies, capped by two major meta-analyses. A 2022 meta-analysis of 26 studies⁵⁵ 2022 meta-analysis of 26 studies
Ipekoglu G et al. A meta-analysis on the association of ACE and PPARA gene variants and endurance athletic status. J Sports Med Phys Fitness, 2022 covering 2,979 endurance athletes and 10,048 controls found the II genotype (CC at rs4341) significantly enriched among endurance athletes at OR=1.48. A complementary systematic review and meta-analysis of ACE and ACTN3 studies⁶⁶ systematic review and meta-analysis of ACE and ACTN3 studies
Ma F et al. The association of sport performance with ACE and ACTN3 genetic polymorphisms: a systematic review and meta-analysis. PLoS One, 2013 found II genotype at OR=1.35 (95% CI 1.17–1.55) for endurance athlete status, while D allele carriers showed advantages in strength- and power-based events.

The altitude evidence is particularly striking. A study of 141 mountaineers⁷⁷ study of 141 mountaineers
Thompson J et al. Angiotensin-converting enzyme genotype and successful ascent to extreme high altitude. High Alt Med Biol, 2007 attempting peaks above 8,000 metres found the I allele strongly enriched in those who successfully summited — the II group averaged maximum altitudes of 8,559 m versus 8,079 m for DD, a difference of nearly 500 vertical metres. This likely reflects greater aerobic efficiency under hypoxia rather than VO2max per se.

A training study of 58 army recruits⁸⁸ training study of 58 army recruits
Woods DR et al. Endurance enhancement related to the human ACE I-D polymorphism is not due to differences in the cardiorespiratory response to training. Eur J Appl Physiol, 2002 homozygous for either allele found that II subjects showed significantly greater reductions in submaximal oxygen consumption at 80W after training, suggesting the I allele's advantage lies in metabolic efficiency — doing the same work for less oxygen — rather than a simple increase in peak VO2max.

For the D allele, evidence points toward strength adaptation. A 12-year review of ACE exercise genetics⁹⁹ 12-year review of ACE exercise genetics
Puthucheary Z et al. The ACE gene and human performance: 12 years on. Sports Medicine, 2011 noted that the D allele is associated with greater left ventricular mass increases in response to endurance training, larger strength gains in resistance training programs, and enrichment among elite swimmers and short-distance sprinters in several national athlete cohorts.

Practical Implications

This SNP does not determine athletic destiny — elite endurance athletes and elite power athletes carry every genotype. But it does represent a genuine biological tendency that can inform training priorities:

• If you carry CC (II): your aerobic machinery is biased toward efficiency. Sustained efforts — long runs, cycling, rowing, altitude sports — align well with your physiology. Your training response may favour volume over intensity, and altitude camps may bring above-average adaptation.
• If you carry GG (DD): your RAAS is tuned for higher outputs of angiotensin II. Resistance training tends to produce larger strength gains, and explosive, power-based activities play to your physiological tendencies. Cardiovascular monitoring is worth discussing with a physician, since the D allele is linked to higher cardiovascular disease risk in non-athletic contexts.
• If you carry CG (ID): you have the intermediate phenotype — one of each allele, with intermediate ACE activity. Most people carry this genotype. You have genuine versatility without a strong pull in either direction.

Interactions

The ACE I/D has been studied in combination with ACTN3 R577X (rs1815739)¹⁰¹⁰ ACTN3 R577X (rs1815739) in multiple athlete cohorts. The combination of ACE II (CC) with ACTN3 XX (TT) appears to compound endurance advantages, while ACE DD (GG) with ACTN3 RR (CC) compounds power/sprint tendencies. These are observational associations without interventional confirmation but represent the best-studied two-locus interaction in exercise genetics.

ACE activity also interacts with AGTR1 A1166C (rs5186)¹¹¹¹ AGTR1 A1166C (rs5186) — the angiotensin II type 1 receptor variant. Individuals with both elevated ACE activity (D allele) and a more responsive AT1 receptor (C allele) may have amplified angiotensin II signalling, relevant to cardiovascular risk assessment and potentially to training-induced cardiac remodelling.

The MCT1 A1470T (rs1049434)¹²¹² MCT1 A1470T (rs1049434) variant in the lactate transporter gene is a functionally independent but thematically related fitness SNP — lactate clearance complements aerobic capacity in determining sustained high-intensity performance.

The small arteries and capillaries that feed heart muscle — the coronary microcirculation¹¹ coronary microcirculation
vessels less than 500 µm in diameter, responsible for ~70% of total coronary resistance — are increasingly recognized as a distinct site of cardiovascular disease. Unlike the large coronary arteries that atherosclerosis typically affects, microvascular dysfunction operates through impaired vasodilation, excessive vasoconstriction, and structural rarefaction of small vessels. The rs4855559 variant in MYH15 emerged from a systematic genetic screen of 643 patients as a male-specific determinant of this underdiagnosed condition.

The Mechanism

MYH15 encodes myosin heavy chain 15²² myosin heavy chain 15
one of the unconventional class II myosin heavy chains, expressed primarily in extraocular muscles, muscle spindles, and at lower levels in vascular and cardiac tissue. The rs4855559 variant sits within intron 36 of MYH15³³ intron 36 of MYH15
a non-coding region on chromosome 3q13.13, plus-strand position 108,396,189 in GRCh38. As an intronic variant, it does not alter the amino acid sequence of the myosin protein. The likely mechanism is regulatory — altered splicing efficiency, intronic enhancer disruption, or expression-level effects⁴⁴ regulatory — altered splicing efficiency, intronic enhancer disruption, or expression-level effects
the T allele is in linkage disequilibrium with other MYH15 variants previously associated with myocardial infarction and maladaptive cardiac remodeling.

Why myosin heavy chain biology would influence coronary microvascular tone is not yet fully characterized. Candidate mechanisms include altered contractile properties of vascular smooth muscle cells lining small coronary arteries⁵⁵ vascular smooth muscle cells lining small coronary arteries
smooth muscle myosin isoform composition affects vasoconstrictor tone and myogenic reactivity, and indirect effects on cardiomyocyte contractility that alter microvascular perfusion pressure. The pronounced sex-specificity of the association — present in men, absent in women — suggests interaction with androgen-dependent cardiovascular regulatory pathways, though this remains speculative.

The Evidence

Yoshino et al. (2014)⁶⁶ Yoshino et al. (2014)
Single nucleotide polymorphisms associated with abnormal coronary microvascular function. Coron Artery Dis 2014;25(4):281-9 conducted the largest candidate-gene study of coronary microvascular dysfunction to date: 643 patients referred for cardiac catheterization without significant obstructive coronary disease. Coronary flow reserve (CFR) was measured by intracoronary adenosine injection⁷⁷ Coronary flow reserve (CFR) was measured by intracoronary adenosine injection
CFR below 2.5 defined abnormal microvascular function, a widely used clinical threshold. Of 184 patients with abnormal CFR (36 men, 148 women), rs4855559-T was associated with OR 2.27 for abnormal CFR in men (p=0.0029), with a highly significant sex interaction (p=0.0008). In women, the association was null (OR 1.22, p=0.22).

A companion MYH15 variant, rs7630352⁸⁸ rs7630352
a second intronic variant in moderate linkage disequilibrium with rs4855559, showed an even larger effect in men (OR 2.60, p=0.0006). The convergence of two independent MYH15 variants on the same male-specific phenotype strengthens the biological signal.

A 2024 systematic review of genetic determinants of coronary microvascular dysfunction⁹⁹ 2024 systematic review of genetic determinants of coronary microvascular dysfunction
Stein et al., J Am Heart Assoc 2024; surveying 30 SNPs across 22 genes from the available literature confirmed that MYH15 variants are among the most consistently replicated loci and noted their prior association with myocardial infarction and maladaptive remodeling. Separately, a large genome-wide association study (n>28,000 European-ancestry participants) by Purves et al. (2019)¹⁰¹⁰ Purves et al. (2019)
A major role for common genetic variation in anxiety disorders. Mol Psychiatry 2020 found rs4855559-T genome-wide significantly associated with lifetime anxiety disorder (beta=0.12 decrease on a continuous anxiety scale, p=4×10⁻⁸). Whether this reflects shared autonomic regulation between cardiac and anxiety pathways, or independent pleiotropic effects, is unknown.

Evidence is classified as emerging: the CFR association derives from a single candidate-gene study with a modest sample of men (n=36 with CMD), and has not yet been replicated in an independent cohort with similar methodology. The anxiety GWAS is genome-wide significant but identifies correlation, not causation.

Practical Actions

Men carrying one or two T alleles at rs4855559 have approximately double the odds of measurable coronary microvascular dysfunction. Because CMD frequently presents as angina with no obstructive coronary disease — a pattern more common in women but underdiagnosed in men — awareness of this genetic predisposition should lower the threshold for physiological testing if symptoms arise. Coronary flow reserve can be assessed non-invasively using cardiac PET or echocardiography with pharmacological stress¹¹¹¹ Coronary flow reserve can be assessed non-invasively using cardiac PET or echocardiography with pharmacological stress
these tests quantify microvascular vasodilatory capacity without catheterization.

Microvascular dysfunction responds to interventions targeting endothelial health. Nitrate-rich vegetables (beetroot, arugula, spinach) raise plasma nitrite and support NO-mediated vasodilation¹²¹² Nitrate-rich vegetables (beetroot, arugula, spinach) raise plasma nitrite and support NO-mediated vasodilation
particularly relevant when endothelial NOS function is impaired in CMD. Phosphodiesterase inhibitors and ACE inhibitors have shown benefit in small CMD trials. Given the overlap between CMD and anxiety neurobiology implicated by the GWAS data, autonomic modulation strategies may also be relevant.

Interactions

The rs4855559 association concentrates in men and interacts multiplicatively with sex (interaction p=0.0008). This marks rs4855559 as one of the few cardiovascular risk variants with a robust, formally tested sex-by-genotype interaction. A second MYH15 intronic variant, rs7630352, appears to act in the same direction with a larger effect size (OR 2.60 in men); individuals carrying T alleles at both variants may face compounded microvascular risk, though this has not been formally tested in a compound genotype analysis.

Every amino acid you eat is eventually broken down. When your body catabolizes histidine, the final two steps of the pathway are handled by a single bifunctional enzyme — formimidoyltransferase cyclodeaminase (FTCD)¹¹ formimidoyltransferase cyclodeaminase (FTCD)
encoded on chromosome 21q22.3; enzyme is most abundantly expressed in liver. What makes this enzyme unusual is where its product goes: rather than simply releasing a waste metabolite, FTCD feeds a one-carbon unit directly into the [folate pool | the reservoir of tetrahydrofolate (THF) derivatives that carry methyl groups for DNA synthesis, methylation reactions, and homocysteine remethylation]. The p.Val101Met missense variant reduces this enzyme's efficiency, quietly throttling the histidine-to-folate pipeline.

The Mechanism

Histidine catabolism produces an intermediate called [formiminoglutamate (FIGLU) | formiminoglutamate — elevated FIGLU in urine has long been used as a clinical marker of folate deficiency; the FIGLU loading test challenges subjects with excess histidine to stress the folate pool]. FTCD's formiminotransferase domain transfers the formimino group from FIGLU onto tetrahydrofolate (THF), generating 5-formimino-THF. The enzyme's cyclodeaminase domain then strips ammonia from this intermediate, yielding [5,10-methenyl-THF | a one-carbon carrier that can be reduced to 5,10-methylene-THF or converted to 5-formyl-THF; both feed directly into folate-dependent reactions including DNA synthesis and methionine regeneration]. This one-carbon unit joins the central folate pool and ultimately supports SAM synthesis via the methionine cycle. The p.Val101Met substitution sits [between β-sheet 4 and α-helix 4 of the formiminotransferase N-subdomain | structural domain required for substrate binding and octamer assembly], a region important for both substrate binding and the protein's characteristic octameric quaternary structure. Computational predictors (SIFT score 1.0, PolyPhen-2 score 0.029) classify the substitution as tolerated; however, the genetic association evidence is unusually strong given the modest prediction scores — suggesting the variant subtly impairs enzyme efficiency at physiological substrate concentrations rather than abolishing activity outright.

The Evidence

The association was established by Pierce et al. 2019²² Pierce et al. 2019
Exome-wide association study of arsenic metabolism phenotypes in 1,660 Bangladeshi adults from the HEALS cohort. The T allele of rs61735836 showed genome-wide significant associations with all three urinary arsenic metabolites: increased inorganic arsenic (iAs%; P = 8×10⁻¹³), increased monomethylarsenic (MMA%; P = 2×10⁻¹⁶), and decreased dimethylarsenic (DMA%; P = 6×10⁻²³). This pattern indicates impaired [sequential methylation of arsenic | inorganic arsenic is methylated by AS3MT using SAM as methyl donor; DMA is the fully methylated, excretable form; low DMA% signals reduced methylation capacity] — exactly what reduced SAM availability from impaired FTCD activity would predict. Carriers of the T allele also showed increased arsenic-induced skin lesion risk (OR = 1.35; P = 1×10⁻⁵), making this a clinically consequential variant in high-arsenic environments. The biological logic connecting FTCD to arsenic metabolism runs through the folate and methionine cycles. A review of nutrition and one-carbon metabolism³³ review of nutrition and one-carbon metabolism
Abuawad et al., 2021 describes how FTCD-derived one-carbon units feed into the folate cycle, which then transfers methyl groups to the methionine cycle to regenerate SAM — the universal methyl donor used by AS3MT (and over 200 other methyltransferases). Reduced FTCD efficiency means fewer one-carbon units entering the folate pool, less SAM synthesized, and therefore impaired methylation capacity across the board. Multiple randomized controlled trials have confirmed that the folate-arsenic methylation link is causally upstream. A double-blind RCT in Bangladesh⁴⁴ double-blind RCT in Bangladesh
Gamble et al. 2006, n=200, 12 weeks showed that folic acid supplementation in folate-deficient adults significantly increased arsenic methylation, with greater DMA% and lower blood arsenic in the supplemented group. A more recent RCT of folic acid plus creatine⁵⁵ RCT of folic acid plus creatine
Bozack et al. 2019 found a 14% increase in blood DMA and a 0.19-unit improvement in the secondary methylation index at 12 weeks. These trials show that boosting the folate pool — which is precisely what FTCD normally does via histidine catabolism — improves methylation capacity. Carriers of the T allele who have reduced FTCD efficiency stand to benefit most from strategies that replenish the folate pool through alternative routes.

Practical Actions

For T allele carriers, the priority is compensating for the reduced input of one-carbon units from histidine catabolism. Because FTCD channels units into THF (not directly into the methyl-THF branch), supplementing with methylfolate (5-MTHF) provides pre-formed methyl groups that bypass the need for the FTCD step. Ensuring adequate vitamin B12 maintains methionine synthase activity, which recycles homocysteine and regenerates THF. Choline and betaine provide an alternative (folate-independent) remethylation route for homocysteine via BHMT, reducing pressure on the folate-dependent pathway. The arsenic association is most directly relevant in populations with high inorganic arsenic exposure (well water in South Asia, parts of South America, or western United States). For people in low-arsenic environments, the functional consequence of the variant is subtler but still present: reduced one-carbon input into the folate pool means that dietary folate demands are slightly higher than for people with efficient FTCD.

Interactions

FTCD feeds one-carbon units into the THF pool that MTHFR (rs1801133, rs1801131) then converts to 5-methylTHF for homocysteine remethylation. A person carrying both FTCD T allele (reduced input into THF pool) and MTHFR 677T (reduced conversion to 5-methylTHF) faces impairment at consecutive steps in the one-carbon pathway — less substrate entering the folate pool, and less efficient conversion of what does enter. The combined effect on methylation capacity would be greater than either variant alone. SHMT1 C1420T (rs1979277) also operates in this pathway, interconverting serine and glycine while transferring one-carbon units to THF. Carriers of variants in FTCD, MTHFR, and SHMT1 collectively represent persons with reduced throughput at multiple nodes of one-carbon metabolism. SLC19A1 (RFC1, rs1051266) governs cellular folate transport. Poor folate uptake by SLC19A1 variants compounds FTCD-related inefficiency by limiting the THF available for FTCD's product to integrate with.