| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Hormone Laboratory (P.I.H., S.H.), Haukeland University Hospital, Bergen 5021, Norway; LOCUS for Homocysteine and Related Vitamins (P.M.U., J.S.), Institute of Medicine, University of Bergen, 5021 Bergen, Norway; LOCUS for Homocysteine and Related Vitamins (S.E.V.), Department of Public Health and Primary Health Care, University of Bergen, 5021 Bergen, Norway; Cancer Registry of Norway (T.G.), Institute of Population-Based Cancer Research, Montebello, 0310 Oslo, Norway
Address all correspondence and requests for reprints to: Per Magne Ueland, Section for Pharmacology, Institute of Medicine, Armauer Hanssens hus, University of Bergen, 5021 Bergen, Norway. E-mail: per.ueland{at}ikb.uib.no.
| Abstract |
|---|
|
|
|---|
Objective: We investigated how betaine affected total homocysteine (tHcy) concentration within the frame of variable B-vitamin status and according to the methylenetetrahydrofolate reductase (MTHFR) 677C->T genotype.
Design/Setting/Patients: This is a population-based study with a cross-sectional design. It includes 10,601 healthy men and women aged 5064 yr.
Outcome Measures: Plasma samples were analyzed for tHcy, betaine, choline, dimethylglycine, riboflavin, and vitamin B6, whereas folate and vitamin B12 were analyzed in serum.
Results: Betaine was a strong determinant of plasma tHcy in subjects with low serum folate and the MTHFR TT genotype. The association was further strengthened at low levels in the circulation of the other B-vitamins (B2, B6, and B12). Thus, in subjects with the combination of serum folate in the lowest quartile, low vitamin B2, B6, and B12 status, and the MTHFR TT genotype, the difference in tHcy (mean, 95% confidence interval) across extreme plasma betaine quartiles was 8.8 (1.316.2) µmol/liter.
Conclusion: Betaine may thus be an important one-carbon source, particularly in MTHFR 677 TT subjects with inadequate B-vitamin status.
| Introduction |
|---|
|
|
|---|
Betaine, at doses up to 9 g/d, has for years been used as a tHcy-lowering agent in patients with homocystinuria (4). It also reduces fasting tHcy (5, 6, 7, 8, 9, 10) and reduces the increase in tHcy induced by high doses of methionine, i.e. postmethionine load (PML) tHcy in healthy subjects (7, 8). We recently demonstrated that plasma betaine is a strong predictor of PML tHcy (11). Betaine was a weaker predictor of fasting tHcy, but the association was strengthened under conditions of low serum folate (11), suggesting that betaine is an alternative source of one-carbon units in folate-depleted subjects (12). Betaine provides one-carbon units by serving as a substrate for betaine-Hcy methyltransferase (BHMT), which catalyzes the remethylation of Hcy to methionine in the liver and kidney (13).
In most studies of healthy subjects, folate status is a more powerful predictor of plasma tHcy than vitamin B12, but overt B12 deficiency is associated with severe hyperhomocysteinemia (1). Vitamin B6 has minor (14) or no (15) effect on fasting tHcy, but may exert an effect on PML tHcy (16). Recent studies have added riboflavin (vitamin B2) to the spectrum of B-vitamins affecting the tHcy concentration (17). The relationship between B-vitamins and plasma concentration of tHcy is probably explained by their roles as substrates or cofactors for enzymes involved in Hcy remethylation or transsulfuration (18). Methylenetetrahydrofolate reductase (MTHFR) catalyzes the formation of 5-methyltetrahydrofolate (5-MTHF), which is the prevailing folate form in plasma (19) and the form involved in the conversion of Hcy to methionine (20). The homozygous TT genotype of the common MTHFR 677 C->T polymorphism, which has a prevalence of approximately 10% in most Caucasian populations (21), is associated with impaired catalytic properties of the enzyme (22, 23), and is an important genetic determinant of plasma tHcy, particularly under conditions of low folate (24) or riboflavin (17) status.
Plasma tHcy is a responsive indicator of B-vitamin and MTHFR status. This is explained by metabolic reactions dependent on folate, vitamin B12, riboflavin, or vitamin B6, which converge on Hcy. In this study we used plasma tHcy as a probe to investigate how betaine affects one-carbon metabolism within the frame of variable B-vitamin status. We investigated a large population of 10,601 subjects. The study had sufficient statistical power to obtain novel data on the influence of betaine on one-carbon metabolism, with emphasis on the effect modification by B-vitamin status and the MTHFR 677 C->T polymorphism.
| Subjects and Methods |
|---|
|
|
|---|
A total of 10,601 samples from subjects enrolled in the Norwegian Colorectal Cancer Prevention (NORCCAP) study were analyzed. NORCCAP is a prospective randomized study investigating sigmoidoscopy combined with fecal occult blood testing as screening modalities for colorectal cancer in a middle aged (range, 5064 yr) population of both genders (25). Inclusion in the study took place from 19992001 at three hospitals in the city of Oslo and in Telemark County in southern Norway. A number of lifestyle variables were recorded, and blood samples were collected (26). The study was approved by the Regional Ethics Committee, and The Data Inspectorate approved the study protocol. Written informed consent was obtained from all participants.
Blood collection and biochemical analysis
At inclusion, blood samples were collected into EDTA Vacutainer tubes and tubes without additive. Serum was allowed to clot for 1 h at room temperature, whereas EDTA samples were immediately put on ice. Samples were centrifuged at 1100 x g for 10 min; serum and plasma were then separated and frozen at 20 C and stored at 80 C until analysis. Folate (27) and vitamin B12 (28) were determined in serum, and tHcy (29), vitamin B6 (30), riboflavin (the nonphosphorylated form) (30), choline, betaine, and dimethylglycine (DMG) (31) were determined in plasma by published methods. Vitamin B6 exists in three main forms in plasma, i.e. pyridoxal 5'-phosphate (PLP), pyridoxal, and pyridoxic acid. PLP is the cofactor form and is measured in the present study (30). The vitamins and metabolites were analyzed in different matrices (serum vs. EDTA plasma) because EDTA interferes with the folate assay (by inhibiting the growth of Lactobacillus casei) and decreases the preanalytical variability for some metabolites, particularly for choline. MTHFR 677 C->T genotyping was performed by using real-time PCR with 5V exonuclease (Taqman) probes (32).
Statistics
Summary measures include medians with 5th to 95th percentiles. Partial Spearman rank coefficients, adjusted for age, sex, and study center, were used to evaluate associations between variables. The relationship between betaine and tHcy was studied according to folate levels and MTHFR 677 genotype by a Gaussian generalized additive regression model (GAM) (33), as implemented in R (34). The relationships between plasma tHcy and various predictors were also assessed in multiple linear regression models, which were adjusted for age, sex, creatinine, study center, betaine, folate, riboflavin, vitamin B6, vitamin B12, and MTHFR 677 C->T genotype. The vitamin levels were categorized into quartiles for the whole study population. Biochemical variables were represented in the regression models as indicator variables denoting membership to a quartile. The regression coefficient thus estimated the difference in mean tHcy between a reference quartile and the other quartiles. The upper quartile was chosen as reference quartile for betaine, folate, riboflavin, vitamin B6, and vitamin B12, whereas the lower quartile was chosen as the reference for creatinine. Mean tHcy levels across categories for each factor were tested for linear trend. We estimated the strength of the betaine-tHcy relationship in separate groups according to levels of folate (high quartile, Q4; intermediate, Q2 + Q3; low, Q1), MTHFR 677 C->T genotype, and vitamin B26-12 status (high, intermediate, or low). Vitamin B26-12 status was based on a vitamin B26-12 index score. This index was calculated as the sum of quartile scores (03) for riboflavin, vitamin B12, and vitamin B6. An index of 02 was designated low, 35 was intermediate, and 69 was high vitamin B26-12 status. In linear regression analyses with tHcy as the dependent variable, test for trends in the metabolite regression coefficients across the MTHFR 677 C->T genotype (three levels) or folate status (three levels) or the combined B2-B6-B12 status variable (three levels) were obtained from appropriate product terms in the regression model with main effects retained in the model. In these models, we used a linear categorized representation (equally spaced integer scores) of the metabolites. For example, a test for trend in the betaine coefficients across MTHFR genotypes was obtained from the product term in a model with betaine (linear; four levels) + MTHFR (linear; three levels) + betaine*MTHFR + adjustment variables.
SPSS (SPSS Inc., Chicago, IL.) version 11.0 was used for all analyses except the GAM regression analyses, for which we used the statistical package R (34).
| Results |
|---|
|
|
|---|
Plasma and serum samples from 10,601 healthy subjects (age range, 5064 yr; 50.8% female) were analyzed. Sex and age distributions were similar within the CC (51.4%), CT (40.6%), and TT (8.0%) MTHFR genotypes. Men had higher levels of tHcy, betaine, choline, DMG, and creatinine, but lower levels of vitamin B12, folate, and riboflavin than women (P < 0.001, data not shown). Subjects with the TT genotype had approximately 10% lower betaine levels, 30% lower folate levels, and 15% higher levels of tHcy, compared with CC subjects (Table 1
).
|
Adjusted Spearman correlation coefficients between concentrations of betaine, choline, DMG, and B-vitamins are shown in Table 2
. Betaine, choline, and DMG correlated strongly, but strong positive relations were also found between several B-vitamins, as well as between betaine and B-vitamins.
|
Determinants of plasma betaine were assessed in a multiple regression model including study center, age, sex, creatinine, choline, folate, riboflavin, vitamins B6 and B12, and MTHFR 677 C->T genotype. Plasma choline (standardized ß-coefficient = 0.37), sex (ß = 0.36), and serum folate (ß = 0.12) were the strongest predictors of betaine level (P < 0.001, adjusted R2 = 0.28). Subgroup analyses showed that these three variables were also the strongest determinants of plasma betaine in each of the three MTHFR genotypes (data not shown).
Determinants of plasma tHcy
The relationships between plasma tHcy and concentrations of various B-vitamins and creatinine were assessed in a multiple regression model, which also included age, sex, and study center (Table 3
). Betaine, all B-vitamins, and creatinine were significant predictors of tHcy, and the associations with betaine, folate, and riboflavin in particular were strongest in the TT group (Table 3
). The estimated difference in mean plasma tHcy between subjects in the lowest compared with the highest quartile of betaine concentrations were 3.2 µmol/liter in the TT, 1.2 µmol/liter in the CT, and 1.1 µmol/liter in the CC groups.
|
We constructed dose-response curves for the relationship between plasma betaine and tHcy in strata defined by quartiles of serum folate and by MTHFR 677 C->T genotypes. The curves were obtained using a Gaussian generalized additive model (GAM) (33) and were adjusted for age, sex, creatinine, and study center (Fig. 1
). An inverse dose-response relationship between betaine and tHcy was found, which was strong in subjects with the TT genotype and folate in the lowest quartile. A moderate negative betaine-tHcy association was observed in subjects in the lowest folate quartile combined with the CT or CC genotypes, whereas in subgroups with serum folate in quartiles 24, the betaine-tHcy relation was weak or essentially absent, irrespective of genotype.
|
To estimate the effect of B-vitamins other than folate, we constructed a B26-12 index score, which is the sum of quartile scores (03) for concentrations of riboflavin, vitamin B12, and vitamin B6. A B26-12 index score was used to avoid small subgroups defined by combinations with each of the B-vitamins. In a multiple regression analyses (adjusting for age, sex, study center, and creatinine), this index score was a stronger predictor of plasma tHcy (ß = 0.20) than quartiles of each B-vitamin or summary scores of each combination of two B-vitamins (ß < 0.18) (result not shown).
We estimated the simultaneous effect of MTHFR genotype, folate, and other B-vitamins (i.e. B26-12 index score) on the betaine-tHcy relationship by multiple linear regression. The model contained these three variables and was additionally adjusted for age, sex, creatinine, and study center. The effect parameter was the difference in mean tHcy between the highest and lowest betaine quartiles (Fig. 2
and Table 4
).
|
|
| Discussion |
|---|
|
|
|---|
Study design
The large study population allowed precise estimates of betaine-tHcy associations in the entire population, as well as in subgroups. For example, of the 850 subjects with the TT genotype, 397 had low serum folate and 144 individuals (1.4% of the study population) had both low serum folate and low vitamin B26-12 status. Furthermore, age and ethnicity can influence tHcy levels, and the homogeneity of the study population reduces confounding from these variables.
Concentrations of tHcy, betaine, and folate
The median concentration of plasma betaine was 35.4 µmol/liter, which is in agreement with previous reports from studies of healthy individuals (11, 31, 36). Median tHcy was approximately 15% higher and folate was 30% lower in subjects with the TT genotype, compared with the CC genotype, as has consistently been demonstrated in other studies (24, 37). Betaine concentrations were related to MTHFR genotype and showed a small decrease according to the number of T alleles. This may reflect that betaine and 5-MTHF are fungible sources of methyl groups (38), and impaired formation of 5-MTHF in subjects with the MTHFR 677 TT genotype may lead to increased betaine consumption through the BHMT pathway.
Folate and other B-vitamins
B-vitamins, including folate, vitamin B12, riboflavin, and vitamin B6, serve as cofactors in several pathways of one-carbon metabolism (18). In addition, the metabolic functions of these B-vitamins are interconnected, because some enzymes involved in their activation require another B-vitamin as a cofactor (39). Such a functional network, and the fact that intakes of different B-vitamins correlate (40), may explain the positive relationships between the concentrations of B-vitamins reported in the present study (Table 2
) and by others (40). This is the background for constructing a composite variable for the concentrations of riboflavin, vitamin B6 and vitamin B12, i.e. the vitamin B26-12 index.
We observed an inverse relationship between plasma betaine and tHcy, which extended across the whole concentration range of betaine (Fig. 1
). The tHcy difference between subjects in extreme betaine quartiles was strongest at low serum folate and was particularly pronounced in the subgroup with a low vitamin B26-12 status (Fig. 2
and Table 4
). These observations agree with the idea that Hcy is directed to the BHMT pathway under conditions of impaired folate-dependent remethylation (41). Impaired remethylation may occur if methionine synthase (MS) function is inhibited directly, by insufficient dietary supply of vitamin B12 or folate, or indirectly, by low MTHFR activity. Conceivably, high concentrations of the cofactors (methylcobalamin for MS or flavin adenine dinucleotide for MTHFR) may boost low enzyme activities caused by insufficient concentrations of substrates (5-MTHF and 5,10-methylenetetrahydrofolate, respectively) and vice versa. Such cooperative mechanisms may explain the interaction between folate and other B-vitamins observed in the present study.
MTHFR 677 C->T genotype
The strong association between tHcy and betaine in subjects with the combination of low B-vitamin status and the MTHFR TT genotype may reflect up-regulation of the betaine-dependent Hcy remethylation under conditions of impaired activity of 5-MTHF-dependent MS. 5-MTHF formation is inhibited when MTHFR activity is low (20), and in vitro studies on recombinant human MTHFR show that the variant associated with the 677 C->T transition has a propensity to lose its cofactor, flavin adenine dinucleotide, particularly at low folate (42). This might explain why riboflavin, folate, and MTHFR genotype are cooperative determinants of the tHcy-betaine relationship.
Conclusion
Plasma betaine is a strong determinant of plasma tHcy, predominantly in subjects with low levels of folate and other B-vitamins and the MTHFR 677TT genotype. This observation strongly supports the idea that betaine is an important source of one-carbon units, in particular when folate status is compromised, whether caused by inadequate nutrition or genetic factors. This points to the role of betaine as a one-carbon source but also highlights the importance of betaine measurement in studies on folate and Hcy status in relation to chronic diseases.
| Footnotes |
|---|
Disclosure Summary: P.I.H., S.H., S.E.V., T.G., and J.S. have nothing to declare. P.M.U. reports having received consulting fees from Nycomed and is a member of the steering board of the nonprofit Foundation to Promote Research into Functional Vitamin B12 Deficiency. A PTC application (62924 [52365]) for a patent entitled "Determination of folate in fresh and stored serum or plasma as paraaminobenzoylglutamate" was filed on March 3, 2005; P.M.U. is listed as one of the inventors.
First Published Online February 6, 2007
Abbreviations: BHMT, Betaine-Hcy methyltransferase; DMG, dimethylglycine; GAM, generalized additive model; Hcy, homocysteine; MS, methionine synthase; 5-MTHF, 5-methyltetrahydrofolate; MTHFR, methylenetetrahydrofolate reductase; PLP, pyridoxal 5'-phosphate; PML, postmethionine load; tHcy, total Hcy.
Received July 10, 2006.
Accepted January 25, 2007.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
Q.-H. Yang, L. D Botto, M. Gallagher, J. Friedman, C. L Sanders, D. Koontz, S. Nikolova, J D. Erickson, and K. Steinberg Prevalence and effects of gene-gene and gene-nutrient interactions on serum folate and serum total homocysteine concentrations in the United States: findings from the third National Health and Nutrition Examination Survey DNA Bank Am. J. Clinical Nutrition, July 1, 2008; 88(1): 232 - 246. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. M. Ueland and S. Hustad Homocysteine and Folate Status in an Era of Folic Acid Fortification: Balancing Benefits, Risks, and B-vitamins Clin. Chem., May 1, 2008; 54(5): 779 - 781. [Full Text] [PDF] |
||||
![]() |
S. V. Konstantinova, G. S. Tell, S. E. Vollset, O. Nygard, O. Bleie, and P. M. Ueland Divergent Associations of Plasma Choline and Betaine with Components of Metabolic Syndrome in Middle Age and Elderly Men and Women J. Nutr., May 1, 2008; 138(5): 914 - 920. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. M. Wallace, M. P Bonham, J. Strain, E. M Duffy, P. J Robson, M. Ward, H. McNulty, P. W Davidson, G. J Myers, C. F Shamlaye, et al. Homocysteine concentration, related B vitamins, and betaine in pregnant women recruited to the Seychelles Child Development Study Am. J. Clinical Nutrition, February 1, 2008; 87(2): 391 - 397. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Solis, K. Veenema, A. A. Ivanov, S. Tran, R. Li, W. Wang, D. J. Moriarty, C. V. Maletz, and M. A. Caudill Folate Intake at RDA Levels Is Inadequate for Mexican American Men with the Methylenetetrahydrofolate Reductase 677TT Genotype J. Nutr., January 1, 2008; 138(1): 67 - 72. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |