Hyperhomocysteinemia plays an important role in endothelial dysfunction,1 a prominent factor in the pathogenesis of cardiovascular diseases.2 Elevated plasma total homocysteine (p-tHcy) levels are associated with a high likelihood of coronary artery disease (CAD), independent of other risk factors for atherosclerosis.3 Homocysteine (Hcy) levels >15 μmol/L are common in subjects with cardiovascular disease.4 High Hcy concentrations inhibit the proliferation of vascular endothelial cells,5 decrease the antioxidant activity of superoxide dismutase on endothelial cell surfaces, and impair endothelial function.6 They are also implicated in the pathophysiology of essential hypertension.7 Hyperhomocysteinemia exerts its deleterious vascular effects through production of free-reactive oxygen species that cause oxidative stress2 as a result of antioxidant/prooxidant imbalance and probably also by inhibition of intracellular glutathione peroxidase.8
Synthesis and trans-sulfuration of Hcy involve biochemical pathways that lead to the formation of other aminothiols, such as glutathione (GSH) and cysteine (Cys), and determine the thiol redox pattern. An imbalance of the redox pattern may favor establishment of oxidative stress. In fact GSH, together with vitamin E and ascorbic acid, constitute the nonenzymatic antioxidant system that plays an important role in contrasting oxidative stress and maintaining the physiological redox state. Conversely, altered Cys levels were considered a risk factor for vascular disease.9
The effects of folate supplementation on p-tHcy levels have been studied in different patient populations,10 and reductions in p-tHcy proportional to pretreatment levels were consistently found. Folate supplementation lowers p-tHcy in CAD,11,12 in dialysis patients,13 and improves endothelial function in CAD,11,12 in familial hypercholesterolemia14 and in acute hyperhomocysteinemia.10
In recent years, several studies have focused on the hypothesis that folate may also have an intrinsic antioxidant property via an electron-donating effect of the 5-amino group of the reduced form of the pterin core of 5-methyltetrahydrofolate (5-MTHF);15 it may further contribute to the improvement of endothelial function, independent of Hcy lowering, by interacting with the antioxidant system.10–12,14 Indeed, a reduction of nitrate and a restoration of the reduced form of tetrahydrobiopterin, an essential cofactor of nitric oxide synthase, were observed following folate therapy.16
The interplay between thiols and antioxidant vitamins during folate treatment are currently undetermined. The aim of this study was to investigate in hyperhomocysteinemic subjects the effects of 5-MTHF on the thiol redox pattern, antioxidant vitamins, and lipid peroxidation, as assessed by free malondialdehyde (MDA) levels, the reactive nonconjugated MDA form.
MATERIALS AND METHODS
Seventy-two subjects with moderate (fasting levels of p-tHcy 11–30 μmol/L) to medium (p-tHcy 31–100 μmol/L) hyperhomocysteinemia17 were enrolled in the study. Exclusion criteria were a previous history of immunological or neoplastic diseases or chronic renal failure. Three patients, who were taking calcium channel blockers, omitted these medications for at least 48 hours before the study. All previous treatments were maintained unaltered during the 3-month period. No vitamin supplements were allowed for 2 weeks before and during the study. None of the subjects was taking oral contraceptives. Twenty-seven patients were mild smokers, but all of them refrained from smoking 1 day before the study. Each patient filled out a questionnaire on his or her own medical history and lifestyle and was instructed to maintain strictly the same lifestyle and diet to avoid undesired influences of diet changes during the study.
The hyperhomocysteinemic subjects enrolled in the study were either outpatients at the Thrombosis and Haemostasis Clinic Unit of Niguarda Cà Granda Hospital in follow-up for previous thrombotic events or had been referred to our laboratory because of the casual finding of elevated p-tHcy levels.
After an overnight fast, blood was sampled from an antecubital vein for measurement of reduced and total aminothiols (plasma and blood Hcy, Cys, GSH, and cysteinylglycine [Cys–Gly]), vitamin E, ascorbic acid, MDA, vitamin B12, folate, glucose, creatinine, γ-glutamyl transpeptidase, fibrinogen, total proteins, albumin, total cholesterol, triglycerides, and LDL- and HDL-cholesterol. The same determinations were obtained 3 months after enrollment.
Forty-eight subjects were assigned to oral 5-MTHF for 3 months, at a daily dose of 15 mg (Prefolic, Zambon, Milan, Italy), while 24 formed the no-treatment group. Genotyping for detection of methylenetetrahydrofolate reductase (MTHFR 677C→T polymorphism) and cystathionine β-synthase (CβS 833 T→C polymorphism) genotypes was used in all cases. The study protocol was approved by Local Ethics Committee. All of the subjects gave written informed consent to participate in the study.
Preparation and analyses for thiols and vitamins were performed immediately after sample collection. Blood-reduced GSH (b-rGSH) was determined by prompt acidification of whole blood according to the method previously described.18 Levels of plasma-reduced GSH (p-rGSH) are low (1%–2%), so b-rGSH concentrations may come close to GSH content inside the cellular fraction of blood (red, white blood cells, platelets).
p-rGSH and total forms of thiols and blood-total GSH (b-tGSH) were determined according to methods validated in our laboratory.19 MDA levels were determined in −80°C stored plasma by a reference method based on the gas chromatography-mass spectrometry technique with isotope dilution.20Plasma vitamin E and ascorbic acid levels were analyzed by isocratic high-performance liquid chromatography (HPLC) separation, as previously described.21
Thiol separation was performed by HPLC (ProStar, Varian, Surrey, UK), as previously described.19 The calibration curves were linear over the range of 1.87–1000 μmol/L, the correlation coefficient ranged between 0.96 and 1. Intra- and interassay precision, determined as coefficient of variation, ranged between 0.60% and 6.14% and between 2.10% and 7%, respectively. Analytical recoveries, obtained by adding to the biological fluid amounts of analyte, were in the range of 95% to 100%.
Within-assay precisions of vitamin E and ascorbic acid were 9.01% for ascorbic acid and 2.77% for vitamin E, while between-assay precision was 8.98% for ascorbic acid and 3.20% for vitamin E. Analytical recoveries, obtained by adding to the biological fluid amounts of analyte, ranged from 80% to 100% for both assays.
Vitamin B12 and folate were measured using a Roche/Hitachi 917 Analyzer (Roche Diagnostic GmbH, Mannheim, Germany) by competitive immunoassay using direct chemiluminescence (Roche Diagnostic GmbH), whereas glucose, creatinine, γ-glutamyl transpeptidase, fibrinogen, total cholesterol, and triglycerides were determined using standard laboratory methods using a Modular Analytics E170 Analyzer (Roche Diagnostic GmbH) by colorimetric enzymatic assay (Roche Diagnostic GmbH). HDL-cholesterol was measured after precipitation with dextran sulfate-magnesium and LDL-cholesterol was calculated using Friedewald method.
DNA was extracted from aliquots of blood cellular fraction and stored at −80°C. The MTHFR 677C→T genotype was identified by polymerase chain reaction amplification, followed by HinfI restriction digestion, and CβS 833 T→C genotype was identified by polymerase chain reaction amplification, followed by BsrI restriction digestion. Digestion products were separated by 3% agarose gel electrophoresis, stained with ethidium bromide. The MTHFR wild-type allele gave a 198-bp fragment, and the insertion variant gave 2 fragments of 175 and 23 bp. The CβS wild-type allele gave a 174-bp fragment, and the insertion variant gave 2 fragments of 132 and 42 bp.22
Results are expressed as median and interquartile range (I–III). Baseline between-group differences were assessed by unpaired Student t test for continuous variables or by the Mann–Whitney U test for non-normally distributed variables and by χ2 or Fisher exact test for categorical variables. ANOVA with repeated measures was used to test changes from baseline to 3 months between groups. P values are given for time course (T), between group (G), and the time-group interaction (5-MTHF therapy effect, I). The relationship among parameters was assessed by simple and multiple linear regression analysis. The associations were presented as linear regression coefficient (β). A 2-tailed P value <0.05 was considered statistically significant. The statistical analyses were carried out with the Statistical Package for the Social Sciences (SPSS, Chicago, IL) release 10.0 for Windows and S-Plus release 6.0 (MathSoft, Seattle, WA).
Demographic, clinical, and baseline biochemical characteristics of the hyperhomocysteinemic population are presented in Table 1. Thirty-two subjects (44.4%) had a family history of CAD, 24 (33.3%) and 23 (31.9%) had previous (>6 months) arterial or venous occlusive vascular events, respectively, whereas 3 subjects had experienced both arterial and venous thrombosis. The 833TT CβS genotype was present in all cases, and only 7 subjects had the 677CC MTHFR genotype (Table 1).
Effect of 5-MTHF Therapy on Redox Status
Baseline redox status parameters, ascorbic acid, vitamin E, and MDA levels were not different between untreated and treated groups (Table 2). After 3 months of 5-MTHF therapy, plasma folate levels increased from 4.4 (3.5–5.2) μg/L at baseline to 230.0 (125.0–430.0) μg/L (P<0.0001). Conversely, folate concentrations were unchanged in the untreated group (4.2 [3.1–5.7] and 4.4 [3.5–5.5] μg/L, respectively). Levels of p-tHcy decreased after 3 months of treatment by 62% (47%–78%) and fell into the normal range (p-tHcy <11 μmol/L) in 37 of the 48 5-MTHF-treated patients (77.1%), whereas no change was observed in untreated subjects. Levels of plasma-reduced Hcy (p-rHcy) also decreased after 3 months only in treated patients, but the p-rHcy:tHcy ratio (p-r:tHcy) was unchanged during treatment. Furthermore, b-tGSH levels were significantly reduced by 5-MTHF treatment. No changes were found for other thiols, vitamin E, ascorbic acid, and MDA levels.
Correlation Between p-tHcy Lowering and Redox State Changes
To assess whether absolute p-tHcy changes from baseline to 3 months (Δp-tHcy=3 months p-tHcy baseline p-tHcy) in 5-MTHF-treated patients were influenced by other variables, we tested the relationship between Δp-tHcy and changes of thiols, vitamins, and MDA by simple linear regression. As Δp-tHcy was correlated to baseline p-tHcy (β=−0.746, P=0.0001), further analyses were adjusted by baseline p-tHcy. Δp-tHcy was associated positively only with changes from baseline to 3 months of p-rCys-Gly (Δp-rCys-Gly; β=0.673, P=0.001): a Δp-rCys-Gly of 4.53 μmol/L corresponded to an effect on Δp-tHcy of 3.05 μmol/L (95% CI 1.31–4.79; P=0.001).
Comparison of 5-MTHF Effect Between Moderate and Medium Hyperhomocysteinemic Subjects
To assess the impact of medium hyperhomocysteinemia on the response to folate treatment, a separate analysis of the redox pattern was performed between medium hyperhomocysteinemic subjects (n=19) and moderate hyperhomocysteinemic subjects (n=29). Baseline redox state parameters, ascorbic acid, vitamin E, and MDA levels were not different except for higher Hcy levels (P=0.0001), as expected, and higher b-rGSH (P=0.01) and b-tGSH levels (P=0.02) in medium hyperhomocysteinemia.
After 5-MTHF treatment, the Δp-tHcy observed in medium hyperhomocysteinemic subjects was greater than the one found in the moderate hyperhomocysteinemic group (Fig. 1). Levels of p-tHcy decreased after 3 months of treatment by 75% (67%–83%) versus 57% (46%–64%) and fell into the normal range in 10/19 (53%) medium hyperhomocysteinemic subjects and in 27/29 (93%) moderate hyperhomocysteinemic subjects. Decrease of p-rHcy after treatment was also greater in medium than in moderate hyperhomocysteinemic subjects (Table 3).
Although b-tGSH levels were significantly reduced in the overall population after 3 months (P=0.01 for time), this reduction was significant only in medium hyperhomocysteinemic subjects after 5-MTHF treatment (P=0.02 for time-group interaction). Moreover, levels of b-rGSH also decreased after treatment but only in medium hyperhomocysteinemia (P=0.04 for time-group interaction). No changes were found for other thiols, vitamin E, ascorbic acid, and MDA levels (Table 3). Δp-tHcy observed in medium hyperhomocysteinemia, adjusted by baseline p-tHcy, was correlated only to Δp-rCys-Gly (P=0.005). In moderate hyperhomocysteinemia, Δp-tHcy was not correlated to changes in other thiols, vitamin E, ascorbic acid, and MDA.
Higher baseline p-tHcy levels were observed in the 34 5-MTHF-treated subjects carrying the variant 677TT MTHFR (50% with medium hyperhomocysteinemia) than in the 14 treated nonhomozygous subjects (wild-type+heterozygous subjects; 29.5 [17.1–41.6] and 16.7 [13.1–25.4] μmol/L, respectively, P=0.02). A significantly greater Δp-tHcy was observed in the former group with respect to that found in nonhomozygous subjects (P=0.045 for time-group interaction; Fig. 2).
5-MTHF Improves the Antioxidant Defense in Hyperhomocysteinemia Through Interaction With GSH Metabolism
The present study demonstrates that high-dose 5-MTHF therapy is able to markedly reduce or revert Hcy level to normal values in subjects with moderate to medium hyperhomocysteinemia; furthermore, a favorable mechanism on the antioxidant defense through interaction with GSH metabolism has been postulated.
We evaluated the effect of 5-MTHF on thiol dynamics in conjunction with ascorbic acid and vitamin E, key molecules of the antioxidant system in hyperhomocysteinemic subjects. A marked decrease in both reduced and total forms of Hcy in hyperhomocysteinemic subjects after high-dose 5-MTHF therapy was observed and, for the first time to our knowledge, a reduction of b-tGSH was found, consistent with an effect on the GSH metabolism. The total form of GSH measured in our study includes the oxidized GSH, all conjugate forms of GSH (among them protein-bound GSH and GSH mixed disulfides with other thiols), produced through oxidative processes or thiol-disulfide exchange reactions, and reduced free GSH. Our data indicate a reduction of GSH oxidative forms following folate treatment. In addition, p-tHcy reduction was associated only with Δp-rCys-Gly, after adjustment by baseline p-tHcy. Because Cys-Gly is the metabolite of GSH catabolism, this relationship points to a link between Hcy decrease and reduction of GSH breakdown after 5-MTHF therapy. Taken together, these findings suggest that 5-MTHF treatment favors the establishment of an antioxidant state in hyperhomocysteinemic subjects through amelioration of GSH turnover.
An intrinsic antioxidant property of 5-MTHF, in addition to its Hcy-lowering ability, has been suggested by many authors.23,24 Our data indicate an effect of folates on redox state, related to GSH metabolism, as supported by the decrease of b-tGSH forms, not associated with changes in reduced GSH. This antioxidant action of folate is evident in subjects with medium hyperhomocysteinemia and suggests a better GSH cycle disposal after 5-MTHF treatment in medium than in moderate hyperhomocysteinemic forms. Our data do not confirm those of Mayer et al,25 who found a significant increase of erythrocyte GSH levels after folate treatment in hyperhomocysteinemic patients with symptomatic atherosclerosis, probably because of methodological differences. In the preanalytical phase, critical for a correct determination of thiol species, we immediately performed whole-blood acidification before sample centrifugation to avoid rapid and spontaneous redistribution of thiol species26 and to allow careful measurement of reduced forms. The acidification after sample centrifugation, as applied by Mayer et al,25 may conversely produce incorrect determination of total and reduced forms of GSH. Furthermore, we used isocratic HPLC, the reference method for thiol determination that is more specific than the colorimetric assay.
In our study population, ascorbic acid and vitamin E levels did not change after 5-MTHF therapy. Baseline MDA was not affected by chronically high Hcy levels because its concentrations were similar in moderate and medium hyperhomocysteinemia. Moreover, 5-MTHF had no effect on MDA levels, as previously reported in CAD patients.24,27 MDA is a widely known marker of lipid peroxidation,21 which increases in many different diseases correlated to oxidative stress. From our data, lipid peroxidation is not affected either by plasma Hcy levels or by 5-MTHF treatment. Alternatively, oxidative damage caused by hyperhomocysteinemia is likely targeted to other cell structures, such as protein or DNA,17 and not detected by markers of lipid peroxidation.
Elevated cellular Hcy concentrations inhibit glutathione peroxidase type 1, an important enzyme of the antioxidant system, for which GSH acts as substrate, and therefore lead to a decrease in bioactive nitric oxide and promote reactive oxygen species formation.28 The baseline intracellular GSH pattern differed between subjects with moderate or medium hyperhomocysteinemia, a finding that may result from a greater inhibition of glutathione peroxidase type 1 activity because of higher Hcy concentrations in the latter group. Lowering of b-tGSH and b-rGSH after 5-MTHF treatment, a feature of subjects with medium hyperhomocysteinemia, suggests a better GSH cycle disposal after therapy in subjects with more elevated Hcy, and is probably the consequence of glutathione peroxidase type 1 activation because overexpression of cellular glutathione peroxidase has in fact counteracted the adverse effects of Hcy on endothelial function.29
In this prospective study, a high dose of 5-MTHF (15 mg/day) for 3 months determined a marked p-tHcy decrease (62%) and normalization of p-tHcy levels in the majority of treated subjects, even in those with medium hyperhomocysteinemia, including subjects carrying the variant 677TT of MTHFR genotype. These findings favorably compare with previous pharmacological studies,10,13,25,30,31 in which lower doses of folic acid or 5-MTHF resulted in lower p-tHcy reduction (range 1.0%–46.3%). 5-MTHF is the active form of folinic derivates, therefore its pharmacological action is not affected by the efficiency of the metabolic conversion pathway32 and results in greater bioavailability of 5-MTHF as compared to folic acid.33
Because a decrease of a few micromoles of p-tHcy determines a significant reduction of risk for CAD and deep venous thrombosis,34 Hcy lowering to the normal range by a high dose of 5-MTHF, as observed in most of our treated subjects, should have a greater impact on the cardiovascular risk associated with hyperhomocysteinemia. The relationship observed between 5-MTHF-induced Hcy lowering and baseline p-tHcy suggests that treatment should be optimized, with folate doses tailored to p-tHcy levels. Low-dose folate supplementation may be insufficient to effectively decrease Hcy in subjects with medium hyperhomocysteinemia.
Dietary changes may alter vitamin status independently of assigned treatment; however, the maintenance of the same lifestyle during the 3-month period in each patient should have prevented the undesired influence of dietary changes on vitamin status.
High-dose 5-MTHF treatment for 3 months ensures marked Hcy lowering to normal values even in subjects with medium hyperhomocysteinemia; therefore, a high dose of 5-MTHF should be the first-line treatment in subjects with elevated plasma Hcy levels. 5-MTHF-induced Hcy lowering was correlated only with changes of p-rCys-Gly and was affected only by baseline p-tHcy. Moreover, 5-MTHF treatment caused a reduction of intracellular total GSH. These findings suggest a favorable interaction of 5-MTHF with GSH metabolism.
Appreciation is expressed to the clinical staff of the Thrombosis and Haemostasis Clinic Unit of Niguarda Cà Granda Hospital for their invaluable help in the recruitment of subjects and to Drs Cristina Patrosso and Silvana Penco for collaboration in genotype characterization. The secretarial assistance of Elisabetta Spagnolo in the preparation of the manuscript is also acknowledged.
1. Chambers JC, McGregor A, Jean-Marie J, et al. Demonstration of rapid onset vascular endothelial dysfunction after hyper-homocysteinemia. An effect reversible with vitamin C therapy. Circulation. 1999;99:1156-1160.
2. Welch GN, Loscalzo J. Homocysteine and atherothrombosis. N Engl J Med. 1998;338:1042–1050.
3. Malinow MR. Plasma Homocyst(e)ine and arterial occlusive diseases: a mini-review. Clin Chem. 1994;40:173–176.
4. Stampfer MJ, Malinow MR, Willett WC, et al. A prospective study of plasma homocyst(e)ine and risk of myocardial infarction in US physicians. JAMA. 1992;268:877–881.
5. Wang H, Yoshizumi M, Lai K, et al. Inhibition of growth and p21ras methylation in vascular endothelial cells by homocysteine but not cysteine. J Biol Chem. 1997;272:25380–25385.
6. Yamamoto M, Hara H, Adachi T. Effect of homocysteine on the binding of extracellular-super-oxide dismutase to the endothelial cell surface. FEBS Lett. 2000;486:159–162.
7. Rodrigo R, Passalacqua W, Araya J, et al. Implications of oxidative stress and homocysteine in the pathophysiology of essential hypertension. J Cardiovasc Pharmacol. 2003;42:453–461.
8. Schnabel R, Lackner KJ, Rupprecht HJ, et al. Glutathione peroxidase-1 and homocysteine for cardiovascular risk prediction. J Am Coll Cardiol. 2005;45:1631–1637.
9. El-Khairy L, Ueland PM, Refsum H, et al. Plasma cysteine as a risk factor for cardiovascular disease. Circulation. 2001;103:2544–2549.
10. Moat SJ, Lang D, McDowell IFW, et al. Folate, homocysteine, endothelial function and cardiovascular disease. J Nutr Biochem. 2004;15:64–79.
11. Liem A, Reynierse-Buitenwerf GH, Zwinderman AH, et al. Secondary prevention with folic acid: effects on clinical outcomes. J Am Coll Cardiol. 2003;41:2105–2113.
12. Doshi SN, Mc Dowell IFW, Moat SJ, et al. Folic acid improves endothelial function in coronary artery disease via mechanisms largely independent of homocysteine lowering. Circulation. 2002;105:22–26.
13. Apeland T, Mansoor MA, Seljeflot I, et al. Homocysteine, malondialdehyde and endothelial markers in dialysis patients during low-dose folinic acid therapy. J Int Med. 2002;252:456–464.
14. Verhaar M, Wever RMF, Pharm D, et al. 5-methyltetrahidrofolate, the active form of folic acid, restores endothelial function in familial hypercholesterolemia. Circulation. 1998;97:237–241.
15. Rezk BM, Haenen GR, van der Vijgh WJ, et al. Tetrahydrofolate and 5-methyltetrahydrofolate are folates with high antioxidant activity. Identification on the antioxidant pharmacophore. FEBS Lett. 2003;555:601–605.
16. Loscalzo J. Folate and nitrate-induced endothelial dysfunction. A simple treatment for a complex pathobiology. Circulation. 2001;104:1086–1088.
17. Medina MA, Urdiales JL, Sánchez-Amores MI. Roles of homocysteine in cell metabolism. Eur J Biochem. 2001;268:3871–3882.
18. De Chiara B, Bigi R, Campolo J, et al. Blood glutathione as a marker of cardiac allograft vasculopathy in heart transplant recipients. Clin Transplant. 2005;19:367–371.
19. Accinni R, Parodi O. Determination of plasma homocysteine. In: Drew AF, ed. Atherosclerosis: Experimental Methods and Protocols. Totowa, NJ: Humana Press; 2001:77–103.
20. Cighetti G, Debiasi S, Paroni R, et al. Free and total malondialdehyde assessment in biological matrices by gas chromatography-mass spectrometry: what is needed for an accurate detection. Anal Biochem. 1999;266:222–229.
21. Hultqvist M, Hegbrant J, Nilsson-Thorell C, et al. Plasma concentration of vitamin C, vitamin E and/or malondialdehyde as markers of oxygen free radical production during hemodialysis. Clin Nephrol. 1997;47:37–46.
22. Kang SS, Wong PW, Susmano A, et al. Thermolabile methylene-tethrahydrofolate reductase: an inherited risk factor for coronary artery disease. Am J Hum Genet. 1991;48:536–545.
23. Doshi SN, McDowell IF, Moat SJ, et al. Folate improves endothelial function in coronary artery disease: an effect mediated by reduction of intracellular super-oxide? Arterioscler Thromb Vasc Biol. 2001;21:1196–1202.
24. Coppola A, D'Angelo A, Fermo I, et al. Reduced in vivo oxidative stress following 5-methyltetrahydrofolate supplementation in patients with early-onset thrombosis and 677TT methyltetrahydrofolate reductase genotype. Br J Haematol. 2005;131:100–108.
25. Mayer O, Filiposký J, Hromádka M, et al. Treatment of hyperhomocysteinemia with folic acid: effects on homocysteine levels, coagulation status, and oxidative stress markers. J Cardiovasc Pharmacol. 2002;39:851–857.
26. Williams RH, Maggiore JA, Reynolds RD, et al. Novel approach for determination of the redox status of homocysteine and other aminothiols in plasma from healthy subjects and patients with ischemic stroke. Clin Chem. 2001;47:1031–1039.
27. Jonasson T, Ohlin AK, Gottsater A, et al. Plasma homocysteine and markers for oxidative stress and inflammation in patients with coronary artery disease–a prospective randomized study of vitamin supplementation. Clin Chem Lab Med. 2005;43:628–634.
28. Upchurch GR, Welch GN, Fabian AJ, et al. Homocyst(e)ine decreases bioavailable nitric oxide by a mechanism involving glutathione peroxidase. J Biol Chem. 1997;272:17012–17017.
29. Weiss N, Zhang YY, Heydrick S, et al. Overexpression of cellular glutathione peroxidase rescues homocyst(e)ine-induced endothelial dysfunction. Proc Natl Acad Sci U S A. 2001;98:12503–12508.
30. Cafolla A, Dragoni F, Girelli G, et al. Effect of folic acid and vitamin C supplementation on folate status and homocysteine level: a randomised controlled trial in Italian smoker-blood donors. Atherosclerosis. 2002;163:105–111.
31. Homocysteine Lowering Trialists' Collaboration. Dose-dependent effects of folic acid on blood concentrations of homocystine: a meta-analysis of the randomized trials. Am J Clin Nutr. 2005;82:806–812.
32. Stover PJ. Physiology of folate and vitamin B12 in health and disease. Nutr Rev. 2004;62:S3–S12.
33. Willems FF, Boers GH, Blom HJ, et al. Pharmacokinetic study on the utilization of 5-methyltetrahydrofolate and folic acid in patients with coronary artery disease. Br J Pharmacol. 2004;141:825–830.
34. Wald DS, Law M, Morris JK. Homocysteine and cardiovascular disease: evidence on casuality from a meta-analysis. Br Med J. 2002;235:1202–1206.
© 2006 Lippincott Williams & Wilkins, Inc.