Impact of the common MTHFR 677C→T polymorphism on blood pressure in adulthood and role of riboflavin in modifying the genetic risk of hypertension: evidence from the JINGO project

Hypertension is the leading risk factor contributing to all-cause death in every region in the world, estimated to affect 1.13 billion people globally and account for over 9 million deaths annually, predominantly from cardiovascular disease (CVD) [1‐3]. The relationship of blood pressure (BP) with disease is age-specific and most pronounced in adults 40–69 years, where the risk of CVD is estimated to double for each 20 mmHg rise in systolic BP [4]. Recent reports have highlighted the importance of targeting lifestyle and treatment strategies at the individual level in order to improve cardiovascular health [1, 5], and genome-wide association studies (GWAS) have identified specific genes linked with BP which could lead to personalised treatments for hypertension based on genetic characteristics. The earliest of such studies tested 2.5 million single nucleotide polymorphisms (SNPs) and identified eight genetic loci associated with BP, including a region near the gene encoding the folate-metabolising enzyme methylenetetrahydrofolate reductase MTHFR [6], findings confirmed by subsequent GWAS [7].

Of greater relevance to health, clinical studies have linked this gene with BP, with meta-analyses of case-control studies showing that the 677C→T polymorphism in MTHFR is associated with an increased risk of hypertension by 36–87% [8‐10]. Previously, the role of this polymorphism in CVD has been studied extensively in relation to the well-recognised phenotype, elevated homocysteine, whilst the relationship with BP is relatively under-investigated. The variant MTHFR 677TT genotype, which affects 10% of adults worldwide [11], is however reported to increase the risk of CVD (especially stroke) by up to 40%, albeit with a large geographical variation in the extent of excess risk, consistent with a gene-environment interaction [12‐14]. In this regard, only folate was previously considered, but emerging evidence suggests that riboflavin—required in the form flavin adenine dinucleotide (FAD) as a cofactor for MTHFR—may be a key modifying factor linking this polymorphism with CVD via a novel and genotype-specific effect on BP [14]. In three small randomised controlled trials, we previously demonstrated lowering of systolic BP by 6 to 13 mmHg in response to riboflavin when targeted at hypertensive patients with the variant MTHFR 677TT genotype [15‐17].

Our study shows that from young adulthood to 70 years, the MTHFR 677TT genotype predisposes an individual to a systolic BP typical of an adult several years older without this genetic variant. Although this polymorphism was previously linked with BP, this is the first study to examine the genetic risk of hypertension throughout adulthood and to identify the potential for riboflavin to modify the phenotype in affected adults at a younger age and before the onset of hypertension. The observed effect of MTHFR and its modulation by riboflavin in relation to hypertension risk were found to be independent of homocysteine, the typically reported phenotype linking this polymorphism with CVD.

We observed a pattern in the current study (irrespective of MTHFR genotype), whereby systolic BP increased into older age whereas diastolic BP increased until about 60 years and then declined, as previously reported [21, 22]. The results however showed that adults with the variant MTHFR 677TT genotype have higher systolic and diastolic BP compared to others of the same age with CC or CT genotypes. The BP phenotype was not evident above 70 years, presumably as a result of the confounding effect of other age-related determinants of BP. The reason we focussed on the period up to 70 years is because this is a time during which the relationship of BP with disease is most pronounced, with a reported doubling in the risk of CVD for each 20 mmHg rise in systolic BP [4]. The MTHFR 677TT genotype was associated with a 42% increased risk of hypertension in adults up to 70 years, after adjustment for antihypertensive drug use and other significant factors, namely, age, male sex, BMI, alcohol and blood cholesterol, whereas plasma homocysteine was not independently associated with hypertension risk. The extent of excess hypertension owing to this polymorphism is in good agreement with previous estimates from clinical studies, with reported odds ratios in meta-analyses ranging 1.36 (95% CI, 1.20–1.53) to 1.87 (1.31 to 2.68), for worldwide and Chinese populations, respectively [8, 10]. Our findings however show that from young adulthood, this polymorphism contributes to higher BP, suggesting that affected adults could potentially develop hypertension at an earlier age than those without this genetic risk.

Of particular relevance to cardiovascular medicine is the finding that response to routine BP treatment appears to be suboptimal in adults with the MTHFR 677TT genotype. Overall, 49% of participants 18–70 years in this study were under current treatment with antihypertensive drugs, a rate of treatment similar to that reported for adults 20–80 years in England (51%) and considerably less than in adults 20–80 years in the US (74%) or Canada (80%) [21]. In the current study, in line with our previous observations [17], BP control was poorer in the TT genotype, with only 30% of treated adults with the TT compared to 37% in CT and 45% in CC genotypes, achieving BP control. Similarly, reported BP control rates for all treated adults are 37% in England [23] and higher in North America, at 54% in the US [5] and 65% in Canada [23]. Irrespective of prevailing rates of treatment or BP control, however, our findings suggest that within a given population, adults with the TT genotype compared to others without this gene variant will be less likely to achieve target BP with routine treatment, but neither the patient nor the physician will be aware of this. The economic implications of suboptimal BP control are considerable, with the direct costs of hypertension estimated in 2009 at $370 billion annually, representing 10% of healthcare expenditures worldwide [24].

Uniquely, this study enabled the genetic risk of hypertension owing to this polymorphism to be considered in relation to riboflavin (the MTHFR cofactor). Unlike other B vitamins (e.g. folate and vitamin B12), riboflavin biomarkers are rarely measured in human studies and no previous cohort study to have investigated this polymorphism has considered riboflavin [25]. We estimated a 3-fold increased risk of hypertension when the variant TT genotype occurred in combination with deficient riboflavin status (relative to the CC genotype and normal riboflavin status), whereas better riboflavin status was associated with reducing the excess hypertension risk, and normal riboflavin status with no genetic risk. In contrast, among adults with CT or CC genotypes, riboflavin status did not influence the risk of hypertension, evidence that riboflavin has a genotype-specific role in BP. The finding that riboflavin has the potential to modify BP in adults affected by this polymorphism is entirely consistent with our earlier studies in hypertensive patients, which showed a lowering of systolic BP by 6 to 13 mmHg in response to riboflavin supplementation specifically in the TT genotype [15‐17], resulting in a marked increase in blood-pressure control from 32 to 57% (pre versus post riboflavin intervention for 16 weeks), despite no change in antihypertensive treatment over the intervention period [17]. Here we show the potential of riboflavin to modify BP in genetically at-risk adults at an earlier age and the data suggest that the onset of hypertension could be delayed through intervention with riboflavin. Ideally, such intervention would occur prior to commencing antihypertensive treatment and along with lifestyle interventions as per ESC/ESH guidelines for hypertension management [1], especially considering that riboflavin is safe with no known adverse effects even at doses of 100-fold higher than typical dietary intakes [26]. Alternatively, riboflavin could be co-administered with an antihypertensive drug as a novel combination therapy targeted at patients with this genetic risk factor. The potential to prevent or treat hypertension in sub-populations worldwide could be considerable, given that this genotype affects 10% of people globally, ranging 4–26% in Europeans (increasing north to south), 20% in Northern China, to as high as 32% in Mexico [11].

The impact of this polymorphism on BP throughout adulthood and the potential modifying effect of riboflavin are important findings, considering that this polymorphism is linked with an increased risk of stroke [12‐14], and recent evidence shows that living longer in better cardiovascular health during mid-life is associated with lower risk of disease and mortality later in life [27]. Control of BP is highly effective in reducing cardiovascular mortality [5, 23, 24], with each 2 mmHg lower systolic BP associated with a 10% lower risk of stroke [4]. Furthermore, powerful evidence, from the SPRINT trial testing the effects of intensive versus standard blood-pressure control [28] and from meta-analyses of large-scale BP lowering trials [29], highlights significant benefits for cardiovascular risk especially among middle-aged adults [30] of BP-lowering to values below hypertension cut-points. Because of concerns that intensive treatment of BP could also pose certain risks [31], however, there have been calls for newer approaches, including novel combination therapies and non-pharmacological solutions [32]. Our results indicate that the most effective timeframe to target adults with this genetic variant will be up to 70 years, via supplementation with riboflavin to potentially offer an effective low-cost strategy to lower BP. Of note, sub-optimal riboflavin status may be more widespread than is generally recognised, but is largely undocumented as riboflavin biomarkers are rarely measured in human studies [25]. The UK is one of the very few countries worldwide to include a riboflavin biomarker in its population-wide diet and nutrition survey, and recent data shows that over 50% of healthy British adults have low or deficient riboflavin status [33], in close agreement with the current results in Irish adults.

The biological mechanism explaining MTHFR-BP relationship shown here is unknown, but likely involves the potent vasodilator nitric oxide (NO) [34]. Vascular tissue concentrations of 5-methyltetrahydrofolate (the product of the MTHFR reaction) were associated with NO bioavailability and improved endothelial function in patients undergoing coronary artery bypass graft surgery and were found to be lower in those patients with the TT genotype [35, 36]. The current results, considered with our earlier trials [15‐17], indicate that the biologic perturbation leading to higher BP in the TT genotype is modifiable with riboflavin. Molecular studies show that the decreased enzyme activity in the TT genotype is owing to loss of the riboflavin (FAD) cofactor from the active site [37], but riboflavin intervention can restore MTHFR activity in vivo [38]. Restoring MTHFR in vascular tissue could in turn lower BP specifically in individuals with the TT genotype. Mechanistic studies are required, but at this time, the evidence does not support a direct role for homocysteine in BP. Although elevated homocysteine is the characteristic phenotype linked with this polymorphism (and is responsive to riboflavin in the TT genotype [38]), intervention trials to lower homocysteine have shown no corresponding BP response [39], indicating that homocysteine is not causatively related to hypertension. The current results suggest that this polymorphism is linked with CVD via BP independently of homocysteine, and given its importance for clinical outcomes, BP may be the much more relevant target to prevent CVD in those affected by the variant genotype.

A strength of this study is its large sample of adults 18 to 90 years stratified for the relevant polymorphism using data from two cohorts sampled under a common project initiative, from participating centres in Northern Ireland (UK) and the Republic of Ireland (representing two distinct health systems), using standardised methodologies and centralised laboratory analysis to investigate outcomes that were formulated before data collection. Furthermore, the current analysis was based on an a priori hypothesis (linking this polymorphism and riboflavin with BP) whereas other studies of genetic risk factors in relation to disease risk factors are typically opportunistic studies using available data. Thus uniquely, our study provides biomarker data for riboflavin, rarely measured in nutritional studies, and used here to enable the impact of riboflavin on the MTHFR-BP relationship from young adulthood to be demonstrated. The major limitation of this study is its cross-sectional (rather than a longitudinal) design; nonetheless, the study findings in relation to the genotype-specific effect of riboflavin are reinforced by our earlier trials [15‐17] showing significant BP-lowering in response to intervention with riboflavin in CVD patients identified with the relevant genotype.