- Email: [email protected]
G80A reduced folate carrier SNP influences the absorption and cellular translocation of dietary folate and its association with blood pressure in an elderly population
Life Sciences 79 (2006) 957 – 966 www.elsevier.com/locate/lifescie
G80A reduced folate carrier SNP influences the absorption and cellular translocation of dietary folate and its association with blood pressure in an elderly population Lisa Dufficy a , Nenad Naumovski a , Xiaowei Ng a , Barbara Blades a,b , Zoe Yates a , Cheryl Travers c , Peter Lewis c , Jonathon Sturm b , Martin Veysey b , Paul D. Roach a , Mark D. Lucock a,⁎ a
School of Environmental and Life Sciences, University of Newcastle, PO Box 127, Brush Road, Ourimbah NSW 2258, Australia b Teaching and Research Unit, Northern Sydney Central Coast Health, PO Box 361, Gosford NSW 2250, Australia c Public Health Unit, Northern Sydney Central Coast Health, PO Box 361, Gosford NSW 2250, Australia Received 8 March 2006; accepted 5 May 2006
Abstract The functional consequences of the G80A RFC SNP on the expressed reduced folate carrier protein were evaluated by looking at the relationship between intake of folate, plasma folate and cellular stores of the vitamin. The effect on homocysteine was also examined. Homocysteine is a thiol that is known to be inversely associated with folate, and which is considered to be both thrombo- and athrogenic. At high levels, homocysteine may also interfere with nitric oxide mediated vasodilation, cause oxidative injury to, and proliferation of the vascular endothelium, and alter the elastic properties of the vascular wall, contributing to increased blood pressure. Participants (119; 52 male, 67 female) from a NSW retirement village were assessed. Independent of gender, the assimilation of folate from dietary sources into red cells showed a significant association for GG (r = 0.399; p = 0.022) and GA (r = 0.564; p < 0.0001) subjects, but not homozygous recessive (AA) individuals (r = 0.223; p = 0.236). The same genotype based pattern of significance was shown for the association between dietary folate and plasma folate (GG: r = 0.524; p = 0.002, GA: r = 0.408; p = 0.002). No genotype-related pattern of significance was shown for the association between dietary folate and homocysteine. When examined by gender, some differences were apparent; one-way ANOVA showed that genotype influenced diastolic blood pressure in males (p = 0.019), while only females showed a significant correlation between dietary folate and blood pressure within specific genotypes (Systolic pressure GA: r = −0.372; p = 0.025, carriage of A: r = 0.–0.357; p = 0.011. Diastolic pressure GA: r = −0.355; p = 0.034, carriage of A: r = 0.–0.310; p = 0.029). The G80A RFC SNP had an impact on the absorption and cellular translocation of dietary folate and its association with blood pressure in an elderly population. © 2006 Elsevier Inc. All rights reserved. Keywords: Folate; Homocysteine; Polymorphism; Reduced folate carrier; Hypertension; Blood pressure
Introduction The importance of dietary folate in the context of health and wellbeing is now well established: poor folate status along with common single nucleotide polymorphisms (SNPs) in genes that encode folate dependent proteins are linked to vascular disease, many types of cancer, congenital malformations and complications of pregnancy (Lucock, 2004). These effects stem from several key molecular mechanisms ⁎ Corresponding author. Tel.: +61 2 4348 4109; fax: +61 2 4348 4145. E-mail address: [email protected] (M.D. Lucock). 0024-3205/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.lfs.2006.05.009
that sustain the genomic machinery, and which are sensitive to folate nutritional genetics. Folate sensitive mechanisms include maintenance of genomic CpG methylation patterns for regulated gene expression and de novo biosynthesis of nucleotides to prevent DNA instability (Friso and Choi, 2002; Lucock, 2000). Folate nutritional genetics also influence vasculotoxic plasma homocysteine (Hcy) status and modify risk for vascular disease (Den Heijer et al., 2004). Dietary folate is a complex mixture of similar vitamin coenzymes. The native food form is largely methylfolate (5CH3–H4folate), but also contains some formylfolate and
958
L. Dufficy et al. / Life Sciences 79 (2006) 957–966
increasingly, the synthetic pteroylmonoglutamic acid (PteGlu) form of the vitamin. PteGlu is used as a fortificant in a variety of food types and as a supplement in vitamin preparations (Lucock, 2000, 2004). Native dietary folates such as 5CH3–H4folate are a mixture of folyl-oligo-γ-glutamyl (polyglutamyl) coenzymes. These conjugated forms of the vitamin must be hydrolysed to monoglutamyl forms at the intestinal mucosal surface prior to their absorption from the proximal jejunum. The 5CH3–H4folate form is therefore a major ligand for the cellular folate uptake mechanism which involves the reduced folate carrier (RFC-1). This protein is responsible for the uptake of folate from the jejunum (Nguyen et al., 1997) and the subsequent translocation of this trace nutrient across the membranes of a variety of cells (Moscow et al., 1995; Prasad et al., 1995). Although absorption from the gut lumen is based on interaction between monoglutamyl 5CH3– H4folate and RFC-1, other sites of RFC-1 may have differential specificity for monoglutamyl and polyglutamyl folates — as yet no information is available on this. In this study we examine the functional effect of a SNP at position 80 (exon 2) of RFC-1. This substitution of a guanine for an adenine (G80A RFC) leads to an arginine replacing a histidine in the expressed carrier protein. The functional
consequences of the expressed polymorphic receptor protein have been examined in an elderly population living in a retirement village environment. The objective was to identify whether the G80A RFC SNP might be an important factor in known folate sensitive conditions, particularly those linked to vascular disease, since previous work has shown that the G80A RFC SNP affords protection against thrombosis (Yates and Lucock, 2005). Fig. 1 provides a schematic representation of the relationship between the reduced folate carrier protein and putative folate-mediated influences on blood pressure regulation. In a recent report from the large US Nurses Health studies, Forman and colleagues showed that elevated dietary folate intake was associated with a decreased risk of incident hypertension (Forman et al., 2005). The present study was therefore conceived to examine whether any such relationship might be mediated by the nutritional genetics of the reduced folate carrier and its essential dietary folyl ligand. In particular, we examined the effect of the mutant 80A RFC allele on recumbent systolic and diastolic blood pressure and Hcy disposition. This is the first report that quantifies the influence of G80A RFC on the assimilation of dietary folate as reflected by the association between dietary folate intake and erythrocyte levels.
Fig. 1. Schematic representation of the relationship between the reduced folate carrier protein and putative folate-mediated influences on blood pressure regulation.
L. Dufficy et al. / Life Sciences 79 (2006) 957–966
Materials and methods Study design Subjects One hundred and twenty participants from a New South Wales — Central Coast Retirement Village (65–92 years, 52 males and 67 females) were assessed for prevalence of the G80A RFC SNP, folate intake, and blood levels of folate, vitamin B12 and Hcy. Average recumbent systolic and diastolic blood pressures were also obtained. Ethics approval University of Newcastle Human Research Ethics Committee approval – H-782-0304 and Northern Sydney Central Coast Health Human Research Ethics Committee approval – 04/19 apply.
959
Folate/B12 analysis Plasma and red cell folate along with serum vitamin B12 were measured using chemiluminescent immunoassay (Access Immunoassay System, Beckman Instruments, Inc). The laboratory normal reference range was 370–1050 nmol/L red cell folate; 5–21 nmol/L plasma folate; 125–780 pmol/L serum vitamin B12. PCR analysis of genotype The G80A RFC genotyping was based on the method of Winkelmayer et al. (2003). Briefly, sense and antisense primers with the sequences 5′-AGTGTCACCTTCGTCCC-3′ and 5′TCCCGCGTGAAGTTCTTG-3′ respectively, were used for PCR amplification. In the presence of the G allele, the amplicons were digested with the restriction enzyme CfoI to yield three fragments; 125 bp, 68 bp and 37 bp. The polymorphic (A) allele removes a restriction site leading to two fragments; 162 bp and 68 bp.
Methods Blood collection/sample preparation All blood was collected following a 10 hour fast; 4 mL venous blood was collected into lithium heparin for Hcy analysis. Tubes were stored on ice for no longer than 30 min prior to centrifugation at 2000 ×g for 10 min at + 4 °C. Separated plasma for Hcy determination was then stored at − 80 °C prior to analysis. For genetic analysis, 2 mL blood was collected into EDTA tubes and stored at − 80 °C prior to, and following DNA extraction using a QIAamp (Qiagen) DNA blood mini-kit. Bloods collected into 4 mL EDTA and 4 mL plain tubes were used for folate and vitamin B12 determination, and were sent to Westmead Hospital, Sydney for analysis. HPLC analysis of Hcy Following derivatisation with the fluorogenic reagent SBDF, isocratic HPLC with fluorescence detection was used to measure the plasma total Hcy level according to Araki and Sako, 1987 and Dudman et al., 1996 as modified according to Krijt et al., 2001.
Blood pressure determination Blood pressure measurements were taken on three separate occasions over 6 months and the average of the three was taken as the subject's usual blood pressure. At each of the clinic visits, the subjects' blood pressure was measured while they were in a recumbent position (recumbent blood pressure) after they had been resting for at least 5 min. A standard mercury sphygmomanometer was used and the first (SBP) and fifth (DBP) Korotkoff sounds were recorded to the nearest 2 mm Hg. Two sets of readings were taken and averaged for each visit. Food frequency questionnaire To obtain an estimated daily intake of nutrients, an interviewer administered food frequency questionnaire was used. The questionnaire was extensive in covering 225 food items and all food groups. Subjects were also asked to provide a list of all supplements they were taking, and were asked about these during the food frequency questionnaire interview. To assist in the estimation of serving sizes, a number of photographs were used showing specific amounts of common foods eaten.
Table 1 Mean ± SD values for food/blood metabolites and blood pressure G80A RFC genotype (n)
Red cell folate (nmol/L)
Plasma folate (nmol/L)
Dietary folate (μg/day)
Hcy (μMol/L)
Vitamin B12 (pmol/L)
Average recumbent systolic BP
Average recumbent diastolic BP
a: Mean and SD values by genotype for all subjects GG (33) 792 (321) 21.1 (10.5) GA (55) 774 (331) 24.6 (12.0) AA (30) 756 (270) 23.3 (11.7)
453 (204) 472 (194) 460 (195)
9.08 (2.14) 9.30 (2.57) 8.89 (2.71)
254 (69) 267 (121) 285 (155)
136 (11) 135 (13) 132 (14)
75 (7) 75 (8) 72 (7)
b: Mean and SD values by genotype for female subjects GG (17) 845 (312) 23.2 (11.5) GA (36) 714 (267) 23.5 (11.8) AA (14) 714 (280) 23.5 (10.2)
455 (213) 423 (174) 416 (154)
9.00 (2.14) 8.96 (2.48) 8.07 (2.36)
240 (72) 264 (113) 298 (156)
136 (11) 133 (14) 135 (12)
76 (7) 74 (7) 74 (6)
c: Mean and SD values by genotype for male subjects GG (16) 736 (332) 18.9 (9.2) GA (19) 889 (412) 26.7 (12.3) AA (16) 792 (264) 23.1 (13.2)
452 (201) 564 (200) 499 (222)
9.17 (2.20) 9.93 (2.67) 9.60 (2.86)
269 (66) 273 (139) 273 (158)
137 (11) 137 (12) 128 (16)
74 (7) 77 (8) 69 (7)
Data is organised by genotype for a) all subjects, b) females and c) males.
960
L. Dufficy et al. / Life Sciences 79 (2006) 957–966
Fig. 2. (a) Relationship between dietary folate intake and red cell folate within each G80A RFC genotype. Only GG-wildtype and GA-heterozygous individuals exhibit a significant association between variables. (b) Relationship between dietary folate intake and plasma folate within each G80A RFC genotype. Only GG-wildtype and GA-heterozygous individuals exhibit a significant association between variables. (c) Relationship between dietary folate intake and plasma Hcy within each G80A RFC genotype. No genotype exhibits a significant association between variables.
L. Dufficy et al. / Life Sciences 79 (2006) 957–966
961
Fig. 2 (continued).
The food frequency questionnaires were analysed using Foodworks 2.10.146 (Xyris Software, Brisbane, QLD, Australia). This package uses a number of food databases to cover
the majority of foods consumed by Australians. These include; AusFoods (brands), Aus Nut (base foods) and the New Zealand — Vitamin and Mineral Supplements 1999 databases.
962
L. Dufficy et al. / Life Sciences 79 (2006) 957–966
Fig. 2 (continued).
The average daily intake of folate included the average daily intake from foods and supplements and is referred to as dietary folate.
Statistics All data have been tabulated and presented as the mean value ± SD. B-vitamin related indices and blood pressure
L. Dufficy et al. / Life Sciences 79 (2006) 957–966
963
In order to demonstrate any functional effects of the G80A RFC genotype, the degree and significance of association between key variables has been examined. Fig. 2a shows a strong and significant correlation between dietary folate intake and red cell folate. However, this association only reaches significance for the wildtype (GG) and heterozygous (GA) genotypes (p = 0.022; r = 0.399 and p < 0.0001; r = 0.564 respectively). The homozygous recessive genotype (AA) fails to reach significance for this association (p = 0.223; r = 0.236). An essentially similar pattern is shown for the relationship between dietary folate intake and plasma folate (see Fig. 2b). The association between variables being significant for GG and GA genotypes (p = 0.002; r = 0.524 and p = 0.002; r = 0.408 respectively). However, once again possession of two mutant alleles (AA) leads to no association (p = 0.060; r = 0.347), although clearly p is approaching significance. The relationship between dietary folate and Hcy was also examined (see Fig. 2c). With Hcy, no genotype gave a significant association, despite the well-recognised biochemical relationship between folate and this vasculotoxic thiol. However, when examined independent of gender or genotype a significant association (p = 0.011) was seen between dietary folate and Hcy, although this was nowhere near as significant as with the association between dietary folate and both red cell folate (p < 0.0001) and plasma folate (p < 0.0001). Table 2a–d summarises the Pearson correlations (p and r) for these and a range of other important associations. The analysis
measurements were log10 transformed prior to one-way analysis of variance (ANOVA), which was used to assess the differences in continuous variables between individual G80A RFC genotypes. Significant differences were further examined using the Bonferroni post hoc test. The Pearson product–moment coefficient (r) and significance (p) was used to evaluate the evidence (direction and significance) for any relationship between thiol/B-vitamin/blood pressure indices. Descriptive statistics were calculated on the astute statistical package add on for Microsoft Excel, all other statistical analysis was performed on SPSS. Results Table 1a–c show the mean and SD values for red cell folate, plasma folate, dietary folate, plasma Hcy, serum vitamin B12 and both average recumbent systolic and diastolic blood pressures. Table 1a shows values for all participants organised by genotype; Table 1b and c show the same data for female and male participants, respectively. One-way ANOVA shows that genotype influenced only one parameter — recumbent diastolic blood pressure. The average recumbent diastolic blood pressure for male subjects gave a p value of 0.019. The Bonferroni post hoc test demonstrates that within males, the recumbent diastolic BP was significantly lower in wildtype compared to heterozygous individuals (p = 0.016).
Table 2 Pearson correlation coefficient (r) and significance (p) as measures of the association between intake of food folate (μg/day) and a range of key variables: data is organised to show the effect of genotype, carriage of the mutant allele and gender G80A RFC genotype
Red cell folate (nmol/L)
Plasma folate (nmol/L)
r
r
p
Hcy (μMol/L) p
r
p
Average recumbent systolic BP
Average recumbent diastolic BP
r
r
p
p
a: Pearson correlation coefficient (r) and significance (p) as measures of the association between intake of dietary folate (μg/day) and a range of key variables: all subjects All subjects 0.439 <0.0001 0.423 <0.0001 − 0.232 0.011 − 0.095 0.308 0.026 0.780 b: Pearson correlation coefficient (r) and significance (p) as measures of the association between intake of dietary folate (μg/day) and a range of key variables: GG, GA, AA, carriage of mutant A allele (males and females) GG 0.399 0.022 0.524 0.002 − 0.228 0.202 − 0.008 0.963 0.132 0.464 GA 0.564 <0.0001 0.408 0.002 − 0.228 0.094 − 0.107 0.439 − 0.103 0.453 AA 0.223 0.236 0.347 0.060 − 0.258 0.169 − 0.155 0.414 0.162 0.391 Carriage of A 0.458 <0.0001 0.388 0.0002 − 0.236 0.030 − 0.120 0.275 − 0.011 0.921 c: Pearson correlation coefficient (r) and significance (p) as measures of the association between intake of dietary folate (μg/day) and a range of key variables: GG, GA, AA, carriage of mutant A allele (females) All females 0.406 0.001 0.300 0.014 − 0.186 0.132 − 0.267 0.029 − 0.168 0.173 GG 0.547 0.023 0.640 0.006 − 0.065 0.805 − 0.048 0.854 0.120 0.645 GA 0.414 0.012 0.204 0.233 − 0.283 0.094 − 0.372 0.025 − 0.355 0.034 AA 0.088 0.764 0.015 0.960 − 0.129 0.660 − 0.305 0.290 − 0.143 0.627 Carriage of A 0.329 0.019 0.353 0.010 − 0.239 0.094 − 0.357 0.011 − 0.310 0.029 d: Pearson correlation coefficient (r) and significance (p) as measures of the association between intake of dietary folate (μg/day) and a range of key variables: GG, GA, AA, carriage of mutant A allele (males) All males 0.451 0.001 0.578 <0.0001 − 0.366 0.008 0.084 0.558 0.232 0.101 GG 0.258 0.334 0.397 0.128 − 0.404 0.121 0.037 0.892 0.146 0.589 GA 0.658 0.002 0.691 0.001 − 0.370 0.119 0.217 0.372 0.058 0.815 AA 0.275 0.303 0.521 0.039 − 0.450 0.081 − 0.015 0.956 0.487 0.056 Carriage of A 0.508 0.002 0.609 <0.0001 − 0.394 0.019 0.135 0.439 0.280 0.104
964
L. Dufficy et al. / Life Sciences 79 (2006) 957–966
Table 3 G80A RFC genotype distribution and allele frequency G80A RFC genotype
Number
Percentage (frequency)
GG (wildtype) GA (heterozygous) AA (homozygous recessive) G (wildtype allele) A (mutant allele)
33 55 30 121 115
28.0 46.6 25.4 51.3 (0.51) 48.7 (0.49)
assesses data dependent and independent of genotype and gender. Table 2 shows that when gender is taken into account (independent of genotype), a stronger association is shown between variables (dietary folate vs red cell folate, plasma folate and plasma Hcy) within males than females. When one examines the data with respect to carriage of the mutant A allele, once again males exhibit a particularly strong association between intake of dietary folate and red cell folate, plasma folate and plasma Hcy as compared to females. A significant relationship also exists between intake of dietary folate and both recumbent systolic and diastolic blood pressure in females but not males. However, this is only found in women with the G80A RFC GA genotype and in women when examined for carriage of the mutant A allele. Table 3 shows both the genotype distribution and frequency of alleles within this population. The distribution of genotypes in this Australian population is almost identical to that described recently by the present authors for a UK population (Yates and Lucock, 2005): GG–28.0% vs 30.7%; GA–46.6% vs 44.6%; AA–25.4% vs 24.7%. The UK study provided no information on dietary folate intake or blood pressure, and so could not evaluate the functional consequences of this SNP as effectively as the present study does. Discussion Results suggest that the G80A RFC SNP might indeed be an important factor in known folate sensitive conditions — almost 50% of the population carry the mutant allele. The present results raise two interesting issues: Firstly, the common G80A RFC polymorphism, when present as the homozygous recessive genotype, does impact upon the absorption and cellular translocation of dietary folate as might be predicted on an a priori basis, but until now had not been demonstrated in vivo. Secondly, dietary folate can influence both recumbent systolic and diastolic blood pressure by a mechanism that appears to be modified by G80A RFC genotype. Our findings may not be entirely unexpected, since it has already been shown that elevated dietary folate intake is associated with a decreased risk of incident hypertension in women (Forman et al., 2005). However, the present results are the first to implicate a direct role for G80A RFC in modifying blood pressure; diastolic blood pressure was lower in wildtype (GG) males compared to heterozygote (GA) males (Bonferonni analysis), and the best correlation between blood pressure and dietary folate was in individuals (females) who carried a mutant allele. Taking all data into account, it would seem that the mutant A allele marginally impairs the absorption and
translocation of dietary folate to cellular stores (a process that only occurs during erythropoiesis). Questions that arise as a consequence of this are i) how does dietary folate influence blood pressure? and ii) can other common folate SNPs influence the relationship between dietary folate and blood pressure? It is an interesting ‘generalised’ observation that the lowest recumbent systolic and diastolic blood pressures are associated with the homozygous recessive genotype (Table 1a shows that increasing carriage of the polymorphic allele gives an apparent graded response with respect to lowering mean blood pressure). Larger numbers would be needed to examine whether this trend is in fact real. However, this may have a bearing on a previous study by the present authors in which the mutant A allele offered a significant protective effect against thrombosis in a completely different subject cohort (odds ratio = 0.56 CI; 0.34–0.92) (Yates and Lucock, 2005). In speculating how G80A RFC might influence the relationship between folate and important vascular endpoints like blood pressure and thrombotic events, it is worth considering whether systemic folate could interact with nitric oxide metabolism, which requires structurally similar biopterins for regulated metabolism (Lucock et al., 2002; Moat et al., 2004; Hayden and Tyagi, 2004; Hyndman et al., 2002; Stroes et al., 2000). Indeed, a recent study showed that administration of 5CH3–H4folate directly into the brachial artery profoundly enhanced flow-mediated dilatation. This ‘pharmacologic’ effect is blocked by monomethyl arginine, suggesting that the influence of folate is in fact on nitric oxide metabolism, and therefore could involve a mechanism potentially quite distinct from the well established one that lowers Hcy (Doshi et al., 2003). By contrast, two randomised controlled trials have shown that the lowering of blood pressure observed with folate treatment was associated with a lowering of blood Hcy (Mangoni et al., 2002; van Dijk et al., 2001). There are several mechanisms by which Hcy may contribute to elevated blood pressure. It has been related to impairment of vascular endothelial and smooth muscle cell function and its proposed effects include interference with the vasodilation mediated by endothelium derived nitric oxide (eNO), increasing oxidative stress leading to oxidant injury to the endothelium, stimulating the proliferation of vascular smooth muscle cells and altering the elastic properties of the vascular wall (Hayden and Tyagi, 2004). High Hcy levels may therefore cause endothelial dysfunction which could contribute to alterations in endothelium-dependent vasomotor regulation and in turn to an increased blood pressure. Endothelial dysfunction usually refers to the reduced bioavailability of eNO, the endothelium-derived relaxation factor needed to keep arteries vasodilated and blood pressure normal. It is produced from L-arginine in a two-step reaction by the endothelial enzyme nitric oxide synthase (eNOS). There is evidence to suggest that a high Hcy concentration may interfere with eNO production by increasing the production of radical oxygen species which could result in the oxidation and depletion of tetrahydrobiopterin (BH4), a crucial cofactor in the production of nitric oxide by eNOS (Stanger and Weger, 2003). This is thought to lead to the uncoupling of the enzyme
L. Dufficy et al. / Life Sciences 79 (2006) 957–966
reaction and cause eNOS to become an “out of control” superoxide radical-producing enzyme (Xia et al., 1998) which further exacerbates the oxidant stress and produces the Hcymediated vascular damage (Stroes et al., 2000). Folate itself could also be directly related to a lowering of blood pressure through its own pleiotropic effects on the vasculature, independent of Hcy (Moat et al., 2004; Hayden and Tyagi, 2004; Stroes et al., 2000). It has been postulated to exert direct anti-oxidative effects itself and thus contribute to the restoration of impaired endothelial eNO synthesis (Moat et al., 2004; Hayden and Tyagi, 2004; Stroes et al., 2000). While it is known that a depletion of 5CH3–H4folate, the main circulating metabolite of folate, leads to the uncoupling of eNO synthase from its cofactor BH4, it is not clear how 5CH3–H4folate maintains or restores the coupling (Moat et al., 2004; Hayden and Tyagi, 2004; Stroes et al., 2000). One suggestion is that oxidative stress oxidises BH4 to its inactive metabolite, dihydrobiopterin (BH2), and that 5CH3–H4folate may restore the oxidised BH2 to its requisite reduced BH4 form by donating hydrogens and electrons to BH2 (Hayden and Tyagi, 2004; Stroes et al., 2000). With BH4 restored, the endothelial eNOS reaction could then be recoupled and the production of eNO normalised. Alternatively, because the pteridine ring structure of tetrahydrofolate (H4folate) is very similar to that of BH4, it has been postulated that H4folate may be able to bind directly to the BH4 binding site on eNOS and serve as a replacement cofactor for eNO synthesis when BH4 is low (Hyndman et al., 2002). With H4folate as a cofactor, eNO production is at best likely to be very much less efficient. However, H4folate may be able to stabilise eNOS, prevent its uncoupling and halt superoxide production. This could spare remaining BH4 from oxidation and allow it to participate in eNO synthesis. It is apparent that the functional effect of the G80A RFC SNP is to influence membrane transport of folate. However, the precise details involved in this effect are unclear. The paradox being that while the mutant allele reduces the regulated absorption and translocation of dietary folate into cellular stores, it also seems to yield the lowest blood pressure, and offer protection against a thromboembolic event (Yates and Lucock, 2005). The present study provides little evidence for an obvious G80A RFC-mediated effect between dietary folate and Hcy, and in this respect supports our previous study (Yates and Lucock, 2005). Associations between dietary folate and Hcy are far weaker than they are between dietary folate and blood folates, although with a larger number of subjects, clearer effects on Hcy may be detected. In this respect, gender differences may also become more apparent, with males seemingly more susceptible to an effect of genotype on both blood pressure and Hcy. It is also likely that while folate is recognised as having pleiotropic effects on various pathways, including different routes by which it might influence vascular endpoints, its clinical effect may also stem from an interactive effect of G80A RFC with A1298C MTHFR, A2756G MS, A66G MSR and C677T MTHFR genotypes, as shown recently by Lucock and Yates (in press). The net effect of any such
965
interaction is likely to be complex, and studies are surely warranted to measure the influence of folate and folate genotype on the metabolic disposition of Hcy, glutathione, biopterin and eNO, all of which are likely to be important determinants of vascular disease risk. Conclusions The G80A RFC SNP has been shown to affect the transport of folates across biological membranes, and is therefore likely to modulate folate sensitive clinical conditions. Larger studies are required to ascertain the precise role and extent of impact of this SNP in such complex disorders, although we have now provided sufficient evidence to show an effect on mechanisms related to both blood clotting (Yates and Lucock, 2005) and blood pressure. References Araki, A., Sako, Y., 1987. Determination of free and total homocysteine in human plasma by high-performance liquid chromatography with fluorescence detection. Journal of Chromatography A 422, 43–52. Den Heijer, M., Lewington, S., Clarke, R., 2004. Homocysteine, MTHFR and risk of venous thrombosis: a meta-analysis of published epidemiological studies. Journal of Thrombosis and Haemostasis 3, 292–299. Doshi, S., McDowell, I., Moat, S., Lewis, M., Goodfellow, J., 2003. Folate improves endothelial function in patients with coronary heart disease. Clinical Chemistry and Laboratory Medicine 41, 1505–1512. Dudman, N.P., Guo, X.W., Crooks, R., Xie, L., Silberberg, J.S., 1996. Assay of plasma homocysteine: light sensitivity of the fluorescent 7-benzo-2-oxa-1, 3-diazole-4-sulfonic acid derivative, and use of appropriate calibrators. Clinical Chemistry 42, 2028–2032. Forman, J.P., Rimm, E.B., Stampfer, M.J., Curhan, G.C., 2005. Folate intake and the risk of incident hypertension among US women. Journal of the American Medical Association 293, 320–329. Friso, S., Choi, S.W., 2002. Gene-nutrient interactions and DNA methylation. Journal of Nutrition 132, 2382S–2387S. Hayden, M.R., Tyagi, S.C., 2004. Homocysteine and reactive oxygen species in metabolic syndrome, type 2 diabetes mellitus, and atheroscleropathy: the pleiotropic effects of folate supplementation. Nutrition Journal 3, 4. Hyndman, M.E., Verma, S., Rosenfeld, R.J., Anderson, T.J., Parsons, H.G., 2002. Interaction of 5-methyltetrahydrofolate and tetrahydrobiopterin on endothelial function. American journal of physiology. Heart and circulatory physiology 282, 2167–2172. Krijt, J., Vackova, M., Kozich, V., 2001. Measurement of homocysteine and other aminothiols in plasma: advantages of using tris(2-carboxyethyl) phosphine as reductant compared with tri-n-butylphosphine. Clinical Chemistry 47, 1821–1828. Lucock, M., 2000. Folic acid: nutritional biochemistry, molecular biology, and role in disease processes. Molecular Genetics and Metabolism 71, 121–138. Lucock, M., 2004. Is folic acid the ultimate functional food component for disease prevention? British Medical Journal 328, 211–214. Lucock, M., Yates, Z., in press. Synergy between 677 TT MTHFR genotype and related folate SNPs regulates homocysteine level. Nutrition Research. Lucock, M., Yates, Z., Hall, K., Leeming, R., Rylance, G., MacDonald, A., Green, A., 2002. The impact of phenylketonuria on folate metabolism. Molecular Genetics and Metabolism 76, 305–312. Mangoni, A.A., Sherwood, R.A., Swift, C.G., Jackson, S.H., 2002. Folic acid enhances endothelial function and reduces blood pressure in smokers: a randomized controlled trial. Journal of Internal Medicine 252, 497–503. Moat, S.J., Lang, D., McDowell, I.F.W., Clarke, Z.L., Madhavan, A.K., Lewis, M.J., Goodfellow, J., 2004. Folate, homocysteine, endothelial function and cardiovascular disease. Journal of Nutritional Biochemistry 15, 64–79.
966
L. Dufficy et al. / Life Sciences 79 (2006) 957–966
Moscow, J.A., Gong, M., He, R., Sgagias, M.K., Dixon, K.H., Anzick, S.L., Meltzer, P.S., Cowan, K.H., 1995. Isolation of a gene encoding a human reduced folate carrier (RFC1) and analysis of its expression in transportdeficient, methotrexate-resistant human breast cancer cells. Cancer Research 55, 3790–3794. Nguyen, T.T., Dyer, D.L., Dunning, D.D., Rubin, S.A., Grant, K.E., Said, H.M., 1997. Human intestinal folate transport: cloning, expression, and distribution of complementary RNA. Gastroenterology 112, 783–791. Prasad, P.D., Ramamoorthy, S., Leibach, F.H., Ganapathy, V., 1995. Molecular cloning of the human placental folate transporter. Biochemical and Biophysical Research Communications 206, 681–687. Stanger, O., Weger, M., 2003. Interactions of homocysteine, nitric oxide, folate and radicals in the progressively damaged endothelium. Clinical Chemistry 41, 1444–1454. Stroes, E.S., van Faassen, E.E., Yo, M., Martasek, P., Boer, P., Govers, R., Rabelink, T.J., 2000. Folic acid reverts dysfunction of endothelial nitric oxide synthase. Circirculation Research 86, 1129–1134. van Dijk, R.A., Rauwerda, J.A., Steyn, M., Twisk, J.W., Stehouwer, C.D., 2001. Long-term homocysteine-lowering treatment with folic acid plus pyridoxine
is associated with decreased blood pressure but not with improved brachial artery endothelium-dependent vasodilation or carotid artery stiffness: a 2year, randomized, placebo-controlled trial. Arteriosclerosis, Thrombosis, and Vascular Biology 21, 2072–2079. Winkelmayer, W.C., Eberle, C., Sunder-Plassmann, G., Fodinger, M., 2003. Effects of the glutamate carboxypeptidase II (GCP2 1561C>T) and reduced folate carrier (RFC1 80G>A) allelic variants on folate and total homocysteine levels in kidney transplant patients. Kidney International 63, 2280–2285. Xia, Y., Tsai, A.L., Berka, V., Zweier, J.L., 1998. Superoxide generation from endothelial nitric-oxide synthase: a Ca2+/calmodulin-dependent and tetrahydrobiopterin regulatory process. Journal of Biological Chemistry 273, 25804–25808. Yates, Z., Lucock, M., 2005. G80A reduced folate carrier SNP modulates cellular uptake of folate and affords protection against thrombosis via a non homocysteine related mechanism. Life Sciences 77, 2735–2742.
G80A reduced folate carrier SNP influences the absorption and cellular translocation of dietary folate and its association with blood pressure in an elderly population Lisa Dufficy a , Nenad Naumovski a , Xiaowei Ng a , Barbara Blades a,b , Zoe Yates a , Cheryl Travers c , Peter Lewis c , Jonathon Sturm b , Martin Veysey b , Paul D. Roach a , Mark D. Lucock a,⁎ a
School of Environmental and Life Sciences, University of Newcastle, PO Box 127, Brush Road, Ourimbah NSW 2258, Australia b Teaching and Research Unit, Northern Sydney Central Coast Health, PO Box 361, Gosford NSW 2250, Australia c Public Health Unit, Northern Sydney Central Coast Health, PO Box 361, Gosford NSW 2250, Australia Received 8 March 2006; accepted 5 May 2006
Abstract The functional consequences of the G80A RFC SNP on the expressed reduced folate carrier protein were evaluated by looking at the relationship between intake of folate, plasma folate and cellular stores of the vitamin. The effect on homocysteine was also examined. Homocysteine is a thiol that is known to be inversely associated with folate, and which is considered to be both thrombo- and athrogenic. At high levels, homocysteine may also interfere with nitric oxide mediated vasodilation, cause oxidative injury to, and proliferation of the vascular endothelium, and alter the elastic properties of the vascular wall, contributing to increased blood pressure. Participants (119; 52 male, 67 female) from a NSW retirement village were assessed. Independent of gender, the assimilation of folate from dietary sources into red cells showed a significant association for GG (r = 0.399; p = 0.022) and GA (r = 0.564; p < 0.0001) subjects, but not homozygous recessive (AA) individuals (r = 0.223; p = 0.236). The same genotype based pattern of significance was shown for the association between dietary folate and plasma folate (GG: r = 0.524; p = 0.002, GA: r = 0.408; p = 0.002). No genotype-related pattern of significance was shown for the association between dietary folate and homocysteine. When examined by gender, some differences were apparent; one-way ANOVA showed that genotype influenced diastolic blood pressure in males (p = 0.019), while only females showed a significant correlation between dietary folate and blood pressure within specific genotypes (Systolic pressure GA: r = −0.372; p = 0.025, carriage of A: r = 0.–0.357; p = 0.011. Diastolic pressure GA: r = −0.355; p = 0.034, carriage of A: r = 0.–0.310; p = 0.029). The G80A RFC SNP had an impact on the absorption and cellular translocation of dietary folate and its association with blood pressure in an elderly population. © 2006 Elsevier Inc. All rights reserved. Keywords: Folate; Homocysteine; Polymorphism; Reduced folate carrier; Hypertension; Blood pressure
Introduction The importance of dietary folate in the context of health and wellbeing is now well established: poor folate status along with common single nucleotide polymorphisms (SNPs) in genes that encode folate dependent proteins are linked to vascular disease, many types of cancer, congenital malformations and complications of pregnancy (Lucock, 2004). These effects stem from several key molecular mechanisms ⁎ Corresponding author. Tel.: +61 2 4348 4109; fax: +61 2 4348 4145. E-mail address: [email protected] (M.D. Lucock). 0024-3205/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.lfs.2006.05.009
that sustain the genomic machinery, and which are sensitive to folate nutritional genetics. Folate sensitive mechanisms include maintenance of genomic CpG methylation patterns for regulated gene expression and de novo biosynthesis of nucleotides to prevent DNA instability (Friso and Choi, 2002; Lucock, 2000). Folate nutritional genetics also influence vasculotoxic plasma homocysteine (Hcy) status and modify risk for vascular disease (Den Heijer et al., 2004). Dietary folate is a complex mixture of similar vitamin coenzymes. The native food form is largely methylfolate (5CH3–H4folate), but also contains some formylfolate and
958
L. Dufficy et al. / Life Sciences 79 (2006) 957–966
increasingly, the synthetic pteroylmonoglutamic acid (PteGlu) form of the vitamin. PteGlu is used as a fortificant in a variety of food types and as a supplement in vitamin preparations (Lucock, 2000, 2004). Native dietary folates such as 5CH3–H4folate are a mixture of folyl-oligo-γ-glutamyl (polyglutamyl) coenzymes. These conjugated forms of the vitamin must be hydrolysed to monoglutamyl forms at the intestinal mucosal surface prior to their absorption from the proximal jejunum. The 5CH3–H4folate form is therefore a major ligand for the cellular folate uptake mechanism which involves the reduced folate carrier (RFC-1). This protein is responsible for the uptake of folate from the jejunum (Nguyen et al., 1997) and the subsequent translocation of this trace nutrient across the membranes of a variety of cells (Moscow et al., 1995; Prasad et al., 1995). Although absorption from the gut lumen is based on interaction between monoglutamyl 5CH3– H4folate and RFC-1, other sites of RFC-1 may have differential specificity for monoglutamyl and polyglutamyl folates — as yet no information is available on this. In this study we examine the functional effect of a SNP at position 80 (exon 2) of RFC-1. This substitution of a guanine for an adenine (G80A RFC) leads to an arginine replacing a histidine in the expressed carrier protein. The functional
consequences of the expressed polymorphic receptor protein have been examined in an elderly population living in a retirement village environment. The objective was to identify whether the G80A RFC SNP might be an important factor in known folate sensitive conditions, particularly those linked to vascular disease, since previous work has shown that the G80A RFC SNP affords protection against thrombosis (Yates and Lucock, 2005). Fig. 1 provides a schematic representation of the relationship between the reduced folate carrier protein and putative folate-mediated influences on blood pressure regulation. In a recent report from the large US Nurses Health studies, Forman and colleagues showed that elevated dietary folate intake was associated with a decreased risk of incident hypertension (Forman et al., 2005). The present study was therefore conceived to examine whether any such relationship might be mediated by the nutritional genetics of the reduced folate carrier and its essential dietary folyl ligand. In particular, we examined the effect of the mutant 80A RFC allele on recumbent systolic and diastolic blood pressure and Hcy disposition. This is the first report that quantifies the influence of G80A RFC on the assimilation of dietary folate as reflected by the association between dietary folate intake and erythrocyte levels.
Fig. 1. Schematic representation of the relationship between the reduced folate carrier protein and putative folate-mediated influences on blood pressure regulation.
L. Dufficy et al. / Life Sciences 79 (2006) 957–966
Materials and methods Study design Subjects One hundred and twenty participants from a New South Wales — Central Coast Retirement Village (65–92 years, 52 males and 67 females) were assessed for prevalence of the G80A RFC SNP, folate intake, and blood levels of folate, vitamin B12 and Hcy. Average recumbent systolic and diastolic blood pressures were also obtained. Ethics approval University of Newcastle Human Research Ethics Committee approval – H-782-0304 and Northern Sydney Central Coast Health Human Research Ethics Committee approval – 04/19 apply.
959
Folate/B12 analysis Plasma and red cell folate along with serum vitamin B12 were measured using chemiluminescent immunoassay (Access Immunoassay System, Beckman Instruments, Inc). The laboratory normal reference range was 370–1050 nmol/L red cell folate; 5–21 nmol/L plasma folate; 125–780 pmol/L serum vitamin B12. PCR analysis of genotype The G80A RFC genotyping was based on the method of Winkelmayer et al. (2003). Briefly, sense and antisense primers with the sequences 5′-AGTGTCACCTTCGTCCC-3′ and 5′TCCCGCGTGAAGTTCTTG-3′ respectively, were used for PCR amplification. In the presence of the G allele, the amplicons were digested with the restriction enzyme CfoI to yield three fragments; 125 bp, 68 bp and 37 bp. The polymorphic (A) allele removes a restriction site leading to two fragments; 162 bp and 68 bp.
Methods Blood collection/sample preparation All blood was collected following a 10 hour fast; 4 mL venous blood was collected into lithium heparin for Hcy analysis. Tubes were stored on ice for no longer than 30 min prior to centrifugation at 2000 ×g for 10 min at + 4 °C. Separated plasma for Hcy determination was then stored at − 80 °C prior to analysis. For genetic analysis, 2 mL blood was collected into EDTA tubes and stored at − 80 °C prior to, and following DNA extraction using a QIAamp (Qiagen) DNA blood mini-kit. Bloods collected into 4 mL EDTA and 4 mL plain tubes were used for folate and vitamin B12 determination, and were sent to Westmead Hospital, Sydney for analysis. HPLC analysis of Hcy Following derivatisation with the fluorogenic reagent SBDF, isocratic HPLC with fluorescence detection was used to measure the plasma total Hcy level according to Araki and Sako, 1987 and Dudman et al., 1996 as modified according to Krijt et al., 2001.
Blood pressure determination Blood pressure measurements were taken on three separate occasions over 6 months and the average of the three was taken as the subject's usual blood pressure. At each of the clinic visits, the subjects' blood pressure was measured while they were in a recumbent position (recumbent blood pressure) after they had been resting for at least 5 min. A standard mercury sphygmomanometer was used and the first (SBP) and fifth (DBP) Korotkoff sounds were recorded to the nearest 2 mm Hg. Two sets of readings were taken and averaged for each visit. Food frequency questionnaire To obtain an estimated daily intake of nutrients, an interviewer administered food frequency questionnaire was used. The questionnaire was extensive in covering 225 food items and all food groups. Subjects were also asked to provide a list of all supplements they were taking, and were asked about these during the food frequency questionnaire interview. To assist in the estimation of serving sizes, a number of photographs were used showing specific amounts of common foods eaten.
Table 1 Mean ± SD values for food/blood metabolites and blood pressure G80A RFC genotype (n)
Red cell folate (nmol/L)
Plasma folate (nmol/L)
Dietary folate (μg/day)
Hcy (μMol/L)
Vitamin B12 (pmol/L)
Average recumbent systolic BP
Average recumbent diastolic BP
a: Mean and SD values by genotype for all subjects GG (33) 792 (321) 21.1 (10.5) GA (55) 774 (331) 24.6 (12.0) AA (30) 756 (270) 23.3 (11.7)
453 (204) 472 (194) 460 (195)
9.08 (2.14) 9.30 (2.57) 8.89 (2.71)
254 (69) 267 (121) 285 (155)
136 (11) 135 (13) 132 (14)
75 (7) 75 (8) 72 (7)
b: Mean and SD values by genotype for female subjects GG (17) 845 (312) 23.2 (11.5) GA (36) 714 (267) 23.5 (11.8) AA (14) 714 (280) 23.5 (10.2)
455 (213) 423 (174) 416 (154)
9.00 (2.14) 8.96 (2.48) 8.07 (2.36)
240 (72) 264 (113) 298 (156)
136 (11) 133 (14) 135 (12)
76 (7) 74 (7) 74 (6)
c: Mean and SD values by genotype for male subjects GG (16) 736 (332) 18.9 (9.2) GA (19) 889 (412) 26.7 (12.3) AA (16) 792 (264) 23.1 (13.2)
452 (201) 564 (200) 499 (222)
9.17 (2.20) 9.93 (2.67) 9.60 (2.86)
269 (66) 273 (139) 273 (158)
137 (11) 137 (12) 128 (16)
74 (7) 77 (8) 69 (7)
Data is organised by genotype for a) all subjects, b) females and c) males.
960
L. Dufficy et al. / Life Sciences 79 (2006) 957–966
Fig. 2. (a) Relationship between dietary folate intake and red cell folate within each G80A RFC genotype. Only GG-wildtype and GA-heterozygous individuals exhibit a significant association between variables. (b) Relationship between dietary folate intake and plasma folate within each G80A RFC genotype. Only GG-wildtype and GA-heterozygous individuals exhibit a significant association between variables. (c) Relationship between dietary folate intake and plasma Hcy within each G80A RFC genotype. No genotype exhibits a significant association between variables.
L. Dufficy et al. / Life Sciences 79 (2006) 957–966
961
Fig. 2 (continued).
The food frequency questionnaires were analysed using Foodworks 2.10.146 (Xyris Software, Brisbane, QLD, Australia). This package uses a number of food databases to cover
the majority of foods consumed by Australians. These include; AusFoods (brands), Aus Nut (base foods) and the New Zealand — Vitamin and Mineral Supplements 1999 databases.
962
L. Dufficy et al. / Life Sciences 79 (2006) 957–966
Fig. 2 (continued).
The average daily intake of folate included the average daily intake from foods and supplements and is referred to as dietary folate.
Statistics All data have been tabulated and presented as the mean value ± SD. B-vitamin related indices and blood pressure
L. Dufficy et al. / Life Sciences 79 (2006) 957–966
963
In order to demonstrate any functional effects of the G80A RFC genotype, the degree and significance of association between key variables has been examined. Fig. 2a shows a strong and significant correlation between dietary folate intake and red cell folate. However, this association only reaches significance for the wildtype (GG) and heterozygous (GA) genotypes (p = 0.022; r = 0.399 and p < 0.0001; r = 0.564 respectively). The homozygous recessive genotype (AA) fails to reach significance for this association (p = 0.223; r = 0.236). An essentially similar pattern is shown for the relationship between dietary folate intake and plasma folate (see Fig. 2b). The association between variables being significant for GG and GA genotypes (p = 0.002; r = 0.524 and p = 0.002; r = 0.408 respectively). However, once again possession of two mutant alleles (AA) leads to no association (p = 0.060; r = 0.347), although clearly p is approaching significance. The relationship between dietary folate and Hcy was also examined (see Fig. 2c). With Hcy, no genotype gave a significant association, despite the well-recognised biochemical relationship between folate and this vasculotoxic thiol. However, when examined independent of gender or genotype a significant association (p = 0.011) was seen between dietary folate and Hcy, although this was nowhere near as significant as with the association between dietary folate and both red cell folate (p < 0.0001) and plasma folate (p < 0.0001). Table 2a–d summarises the Pearson correlations (p and r) for these and a range of other important associations. The analysis
measurements were log10 transformed prior to one-way analysis of variance (ANOVA), which was used to assess the differences in continuous variables between individual G80A RFC genotypes. Significant differences were further examined using the Bonferroni post hoc test. The Pearson product–moment coefficient (r) and significance (p) was used to evaluate the evidence (direction and significance) for any relationship between thiol/B-vitamin/blood pressure indices. Descriptive statistics were calculated on the astute statistical package add on for Microsoft Excel, all other statistical analysis was performed on SPSS. Results Table 1a–c show the mean and SD values for red cell folate, plasma folate, dietary folate, plasma Hcy, serum vitamin B12 and both average recumbent systolic and diastolic blood pressures. Table 1a shows values for all participants organised by genotype; Table 1b and c show the same data for female and male participants, respectively. One-way ANOVA shows that genotype influenced only one parameter — recumbent diastolic blood pressure. The average recumbent diastolic blood pressure for male subjects gave a p value of 0.019. The Bonferroni post hoc test demonstrates that within males, the recumbent diastolic BP was significantly lower in wildtype compared to heterozygous individuals (p = 0.016).
Table 2 Pearson correlation coefficient (r) and significance (p) as measures of the association between intake of food folate (μg/day) and a range of key variables: data is organised to show the effect of genotype, carriage of the mutant allele and gender G80A RFC genotype
Red cell folate (nmol/L)
Plasma folate (nmol/L)
r
r
p
Hcy (μMol/L) p
r
p
Average recumbent systolic BP
Average recumbent diastolic BP
r
r
p
p
a: Pearson correlation coefficient (r) and significance (p) as measures of the association between intake of dietary folate (μg/day) and a range of key variables: all subjects All subjects 0.439 <0.0001 0.423 <0.0001 − 0.232 0.011 − 0.095 0.308 0.026 0.780 b: Pearson correlation coefficient (r) and significance (p) as measures of the association between intake of dietary folate (μg/day) and a range of key variables: GG, GA, AA, carriage of mutant A allele (males and females) GG 0.399 0.022 0.524 0.002 − 0.228 0.202 − 0.008 0.963 0.132 0.464 GA 0.564 <0.0001 0.408 0.002 − 0.228 0.094 − 0.107 0.439 − 0.103 0.453 AA 0.223 0.236 0.347 0.060 − 0.258 0.169 − 0.155 0.414 0.162 0.391 Carriage of A 0.458 <0.0001 0.388 0.0002 − 0.236 0.030 − 0.120 0.275 − 0.011 0.921 c: Pearson correlation coefficient (r) and significance (p) as measures of the association between intake of dietary folate (μg/day) and a range of key variables: GG, GA, AA, carriage of mutant A allele (females) All females 0.406 0.001 0.300 0.014 − 0.186 0.132 − 0.267 0.029 − 0.168 0.173 GG 0.547 0.023 0.640 0.006 − 0.065 0.805 − 0.048 0.854 0.120 0.645 GA 0.414 0.012 0.204 0.233 − 0.283 0.094 − 0.372 0.025 − 0.355 0.034 AA 0.088 0.764 0.015 0.960 − 0.129 0.660 − 0.305 0.290 − 0.143 0.627 Carriage of A 0.329 0.019 0.353 0.010 − 0.239 0.094 − 0.357 0.011 − 0.310 0.029 d: Pearson correlation coefficient (r) and significance (p) as measures of the association between intake of dietary folate (μg/day) and a range of key variables: GG, GA, AA, carriage of mutant A allele (males) All males 0.451 0.001 0.578 <0.0001 − 0.366 0.008 0.084 0.558 0.232 0.101 GG 0.258 0.334 0.397 0.128 − 0.404 0.121 0.037 0.892 0.146 0.589 GA 0.658 0.002 0.691 0.001 − 0.370 0.119 0.217 0.372 0.058 0.815 AA 0.275 0.303 0.521 0.039 − 0.450 0.081 − 0.015 0.956 0.487 0.056 Carriage of A 0.508 0.002 0.609 <0.0001 − 0.394 0.019 0.135 0.439 0.280 0.104
964
L. Dufficy et al. / Life Sciences 79 (2006) 957–966
Table 3 G80A RFC genotype distribution and allele frequency G80A RFC genotype
Number
Percentage (frequency)
GG (wildtype) GA (heterozygous) AA (homozygous recessive) G (wildtype allele) A (mutant allele)
33 55 30 121 115
28.0 46.6 25.4 51.3 (0.51) 48.7 (0.49)
assesses data dependent and independent of genotype and gender. Table 2 shows that when gender is taken into account (independent of genotype), a stronger association is shown between variables (dietary folate vs red cell folate, plasma folate and plasma Hcy) within males than females. When one examines the data with respect to carriage of the mutant A allele, once again males exhibit a particularly strong association between intake of dietary folate and red cell folate, plasma folate and plasma Hcy as compared to females. A significant relationship also exists between intake of dietary folate and both recumbent systolic and diastolic blood pressure in females but not males. However, this is only found in women with the G80A RFC GA genotype and in women when examined for carriage of the mutant A allele. Table 3 shows both the genotype distribution and frequency of alleles within this population. The distribution of genotypes in this Australian population is almost identical to that described recently by the present authors for a UK population (Yates and Lucock, 2005): GG–28.0% vs 30.7%; GA–46.6% vs 44.6%; AA–25.4% vs 24.7%. The UK study provided no information on dietary folate intake or blood pressure, and so could not evaluate the functional consequences of this SNP as effectively as the present study does. Discussion Results suggest that the G80A RFC SNP might indeed be an important factor in known folate sensitive conditions — almost 50% of the population carry the mutant allele. The present results raise two interesting issues: Firstly, the common G80A RFC polymorphism, when present as the homozygous recessive genotype, does impact upon the absorption and cellular translocation of dietary folate as might be predicted on an a priori basis, but until now had not been demonstrated in vivo. Secondly, dietary folate can influence both recumbent systolic and diastolic blood pressure by a mechanism that appears to be modified by G80A RFC genotype. Our findings may not be entirely unexpected, since it has already been shown that elevated dietary folate intake is associated with a decreased risk of incident hypertension in women (Forman et al., 2005). However, the present results are the first to implicate a direct role for G80A RFC in modifying blood pressure; diastolic blood pressure was lower in wildtype (GG) males compared to heterozygote (GA) males (Bonferonni analysis), and the best correlation between blood pressure and dietary folate was in individuals (females) who carried a mutant allele. Taking all data into account, it would seem that the mutant A allele marginally impairs the absorption and
translocation of dietary folate to cellular stores (a process that only occurs during erythropoiesis). Questions that arise as a consequence of this are i) how does dietary folate influence blood pressure? and ii) can other common folate SNPs influence the relationship between dietary folate and blood pressure? It is an interesting ‘generalised’ observation that the lowest recumbent systolic and diastolic blood pressures are associated with the homozygous recessive genotype (Table 1a shows that increasing carriage of the polymorphic allele gives an apparent graded response with respect to lowering mean blood pressure). Larger numbers would be needed to examine whether this trend is in fact real. However, this may have a bearing on a previous study by the present authors in which the mutant A allele offered a significant protective effect against thrombosis in a completely different subject cohort (odds ratio = 0.56 CI; 0.34–0.92) (Yates and Lucock, 2005). In speculating how G80A RFC might influence the relationship between folate and important vascular endpoints like blood pressure and thrombotic events, it is worth considering whether systemic folate could interact with nitric oxide metabolism, which requires structurally similar biopterins for regulated metabolism (Lucock et al., 2002; Moat et al., 2004; Hayden and Tyagi, 2004; Hyndman et al., 2002; Stroes et al., 2000). Indeed, a recent study showed that administration of 5CH3–H4folate directly into the brachial artery profoundly enhanced flow-mediated dilatation. This ‘pharmacologic’ effect is blocked by monomethyl arginine, suggesting that the influence of folate is in fact on nitric oxide metabolism, and therefore could involve a mechanism potentially quite distinct from the well established one that lowers Hcy (Doshi et al., 2003). By contrast, two randomised controlled trials have shown that the lowering of blood pressure observed with folate treatment was associated with a lowering of blood Hcy (Mangoni et al., 2002; van Dijk et al., 2001). There are several mechanisms by which Hcy may contribute to elevated blood pressure. It has been related to impairment of vascular endothelial and smooth muscle cell function and its proposed effects include interference with the vasodilation mediated by endothelium derived nitric oxide (eNO), increasing oxidative stress leading to oxidant injury to the endothelium, stimulating the proliferation of vascular smooth muscle cells and altering the elastic properties of the vascular wall (Hayden and Tyagi, 2004). High Hcy levels may therefore cause endothelial dysfunction which could contribute to alterations in endothelium-dependent vasomotor regulation and in turn to an increased blood pressure. Endothelial dysfunction usually refers to the reduced bioavailability of eNO, the endothelium-derived relaxation factor needed to keep arteries vasodilated and blood pressure normal. It is produced from L-arginine in a two-step reaction by the endothelial enzyme nitric oxide synthase (eNOS). There is evidence to suggest that a high Hcy concentration may interfere with eNO production by increasing the production of radical oxygen species which could result in the oxidation and depletion of tetrahydrobiopterin (BH4), a crucial cofactor in the production of nitric oxide by eNOS (Stanger and Weger, 2003). This is thought to lead to the uncoupling of the enzyme
L. Dufficy et al. / Life Sciences 79 (2006) 957–966
reaction and cause eNOS to become an “out of control” superoxide radical-producing enzyme (Xia et al., 1998) which further exacerbates the oxidant stress and produces the Hcymediated vascular damage (Stroes et al., 2000). Folate itself could also be directly related to a lowering of blood pressure through its own pleiotropic effects on the vasculature, independent of Hcy (Moat et al., 2004; Hayden and Tyagi, 2004; Stroes et al., 2000). It has been postulated to exert direct anti-oxidative effects itself and thus contribute to the restoration of impaired endothelial eNO synthesis (Moat et al., 2004; Hayden and Tyagi, 2004; Stroes et al., 2000). While it is known that a depletion of 5CH3–H4folate, the main circulating metabolite of folate, leads to the uncoupling of eNO synthase from its cofactor BH4, it is not clear how 5CH3–H4folate maintains or restores the coupling (Moat et al., 2004; Hayden and Tyagi, 2004; Stroes et al., 2000). One suggestion is that oxidative stress oxidises BH4 to its inactive metabolite, dihydrobiopterin (BH2), and that 5CH3–H4folate may restore the oxidised BH2 to its requisite reduced BH4 form by donating hydrogens and electrons to BH2 (Hayden and Tyagi, 2004; Stroes et al., 2000). With BH4 restored, the endothelial eNOS reaction could then be recoupled and the production of eNO normalised. Alternatively, because the pteridine ring structure of tetrahydrofolate (H4folate) is very similar to that of BH4, it has been postulated that H4folate may be able to bind directly to the BH4 binding site on eNOS and serve as a replacement cofactor for eNO synthesis when BH4 is low (Hyndman et al., 2002). With H4folate as a cofactor, eNO production is at best likely to be very much less efficient. However, H4folate may be able to stabilise eNOS, prevent its uncoupling and halt superoxide production. This could spare remaining BH4 from oxidation and allow it to participate in eNO synthesis. It is apparent that the functional effect of the G80A RFC SNP is to influence membrane transport of folate. However, the precise details involved in this effect are unclear. The paradox being that while the mutant allele reduces the regulated absorption and translocation of dietary folate into cellular stores, it also seems to yield the lowest blood pressure, and offer protection against a thromboembolic event (Yates and Lucock, 2005). The present study provides little evidence for an obvious G80A RFC-mediated effect between dietary folate and Hcy, and in this respect supports our previous study (Yates and Lucock, 2005). Associations between dietary folate and Hcy are far weaker than they are between dietary folate and blood folates, although with a larger number of subjects, clearer effects on Hcy may be detected. In this respect, gender differences may also become more apparent, with males seemingly more susceptible to an effect of genotype on both blood pressure and Hcy. It is also likely that while folate is recognised as having pleiotropic effects on various pathways, including different routes by which it might influence vascular endpoints, its clinical effect may also stem from an interactive effect of G80A RFC with A1298C MTHFR, A2756G MS, A66G MSR and C677T MTHFR genotypes, as shown recently by Lucock and Yates (in press). The net effect of any such
965
interaction is likely to be complex, and studies are surely warranted to measure the influence of folate and folate genotype on the metabolic disposition of Hcy, glutathione, biopterin and eNO, all of which are likely to be important determinants of vascular disease risk. Conclusions The G80A RFC SNP has been shown to affect the transport of folates across biological membranes, and is therefore likely to modulate folate sensitive clinical conditions. Larger studies are required to ascertain the precise role and extent of impact of this SNP in such complex disorders, although we have now provided sufficient evidence to show an effect on mechanisms related to both blood clotting (Yates and Lucock, 2005) and blood pressure. References Araki, A., Sako, Y., 1987. Determination of free and total homocysteine in human plasma by high-performance liquid chromatography with fluorescence detection. Journal of Chromatography A 422, 43–52. Den Heijer, M., Lewington, S., Clarke, R., 2004. Homocysteine, MTHFR and risk of venous thrombosis: a meta-analysis of published epidemiological studies. Journal of Thrombosis and Haemostasis 3, 292–299. Doshi, S., McDowell, I., Moat, S., Lewis, M., Goodfellow, J., 2003. Folate improves endothelial function in patients with coronary heart disease. Clinical Chemistry and Laboratory Medicine 41, 1505–1512. Dudman, N.P., Guo, X.W., Crooks, R., Xie, L., Silberberg, J.S., 1996. Assay of plasma homocysteine: light sensitivity of the fluorescent 7-benzo-2-oxa-1, 3-diazole-4-sulfonic acid derivative, and use of appropriate calibrators. Clinical Chemistry 42, 2028–2032. Forman, J.P., Rimm, E.B., Stampfer, M.J., Curhan, G.C., 2005. Folate intake and the risk of incident hypertension among US women. Journal of the American Medical Association 293, 320–329. Friso, S., Choi, S.W., 2002. Gene-nutrient interactions and DNA methylation. Journal of Nutrition 132, 2382S–2387S. Hayden, M.R., Tyagi, S.C., 2004. Homocysteine and reactive oxygen species in metabolic syndrome, type 2 diabetes mellitus, and atheroscleropathy: the pleiotropic effects of folate supplementation. Nutrition Journal 3, 4. Hyndman, M.E., Verma, S., Rosenfeld, R.J., Anderson, T.J., Parsons, H.G., 2002. Interaction of 5-methyltetrahydrofolate and tetrahydrobiopterin on endothelial function. American journal of physiology. Heart and circulatory physiology 282, 2167–2172. Krijt, J., Vackova, M., Kozich, V., 2001. Measurement of homocysteine and other aminothiols in plasma: advantages of using tris(2-carboxyethyl) phosphine as reductant compared with tri-n-butylphosphine. Clinical Chemistry 47, 1821–1828. Lucock, M., 2000. Folic acid: nutritional biochemistry, molecular biology, and role in disease processes. Molecular Genetics and Metabolism 71, 121–138. Lucock, M., 2004. Is folic acid the ultimate functional food component for disease prevention? British Medical Journal 328, 211–214. Lucock, M., Yates, Z., in press. Synergy between 677 TT MTHFR genotype and related folate SNPs regulates homocysteine level. Nutrition Research. Lucock, M., Yates, Z., Hall, K., Leeming, R., Rylance, G., MacDonald, A., Green, A., 2002. The impact of phenylketonuria on folate metabolism. Molecular Genetics and Metabolism 76, 305–312. Mangoni, A.A., Sherwood, R.A., Swift, C.G., Jackson, S.H., 2002. Folic acid enhances endothelial function and reduces blood pressure in smokers: a randomized controlled trial. Journal of Internal Medicine 252, 497–503. Moat, S.J., Lang, D., McDowell, I.F.W., Clarke, Z.L., Madhavan, A.K., Lewis, M.J., Goodfellow, J., 2004. Folate, homocysteine, endothelial function and cardiovascular disease. Journal of Nutritional Biochemistry 15, 64–79.
966
L. Dufficy et al. / Life Sciences 79 (2006) 957–966
Moscow, J.A., Gong, M., He, R., Sgagias, M.K., Dixon, K.H., Anzick, S.L., Meltzer, P.S., Cowan, K.H., 1995. Isolation of a gene encoding a human reduced folate carrier (RFC1) and analysis of its expression in transportdeficient, methotrexate-resistant human breast cancer cells. Cancer Research 55, 3790–3794. Nguyen, T.T., Dyer, D.L., Dunning, D.D., Rubin, S.A., Grant, K.E., Said, H.M., 1997. Human intestinal folate transport: cloning, expression, and distribution of complementary RNA. Gastroenterology 112, 783–791. Prasad, P.D., Ramamoorthy, S., Leibach, F.H., Ganapathy, V., 1995. Molecular cloning of the human placental folate transporter. Biochemical and Biophysical Research Communications 206, 681–687. Stanger, O., Weger, M., 2003. Interactions of homocysteine, nitric oxide, folate and radicals in the progressively damaged endothelium. Clinical Chemistry 41, 1444–1454. Stroes, E.S., van Faassen, E.E., Yo, M., Martasek, P., Boer, P., Govers, R., Rabelink, T.J., 2000. Folic acid reverts dysfunction of endothelial nitric oxide synthase. Circirculation Research 86, 1129–1134. van Dijk, R.A., Rauwerda, J.A., Steyn, M., Twisk, J.W., Stehouwer, C.D., 2001. Long-term homocysteine-lowering treatment with folic acid plus pyridoxine
is associated with decreased blood pressure but not with improved brachial artery endothelium-dependent vasodilation or carotid artery stiffness: a 2year, randomized, placebo-controlled trial. Arteriosclerosis, Thrombosis, and Vascular Biology 21, 2072–2079. Winkelmayer, W.C., Eberle, C., Sunder-Plassmann, G., Fodinger, M., 2003. Effects of the glutamate carboxypeptidase II (GCP2 1561C>T) and reduced folate carrier (RFC1 80G>A) allelic variants on folate and total homocysteine levels in kidney transplant patients. Kidney International 63, 2280–2285. Xia, Y., Tsai, A.L., Berka, V., Zweier, J.L., 1998. Superoxide generation from endothelial nitric-oxide synthase: a Ca2+/calmodulin-dependent and tetrahydrobiopterin regulatory process. Journal of Biological Chemistry 273, 25804–25808. Yates, Z., Lucock, M., 2005. G80A reduced folate carrier SNP modulates cellular uptake of folate and affords protection against thrombosis via a non homocysteine related mechanism. Life Sciences 77, 2735–2742.