Human Nutrition and Metabolism - National Taiwan University

10 downloads 0 Views 129KB Size Report
vegetarianism, this study was designed to examine the effect of Taiwanese ... lacto-vegetarian diet had mildly elevated fasting plasma homocysteine levels ...

Human Nutrition and Metabolism

Plasma Homocysteine Levels in Taiwanese Vegetarians Are Higher than Those of Omnivores1 Chien-Jung Hung, Po-Chao Huang,* Shao-Chun Lu,* Yi-Hwei Li,† Hsien-Bin Huang,** Bi-Fong Lin,‡ Sue-Joan Chang†† and Hsu-Fang Chou2 Department of Biochemistry and †Department of Public Health College of Medicine, Tzu-Chi University, Hualien; *Department of Biochemistry, College of Medicine, National Taiwan University, Taipei; **Institute of Molecular Biology, National Chung Cheng University, Chia-Yi; ‡Department of Agricultural Chemistry, National Taiwan University, Taipei; and ††Department of Biology, National Cheng Kung University, Tainan, Taiwan


vegetarian diet


vitamin B-12



The circulating level of total homocysteine is tightly regulated by the enzymes and B-vitamins involved in methionine metabolism. Genetic or nutritional disorders of methionine metabolism may result in abnormal homocysteine levels. Severe genetic defects of cystathionine ␤-synthase, methionine synthase or MTHFR are known to cause severe hyperhomocysteinemia in children (2,3). Moderate deficiencies of these key enzymes, or a deficiency of folic acid, vitamin B-12 or vitamin B-6 may result in mildly elevated plasma homocysteine4 (4). Hyperhomocysteinemia, both mild and severe, has been linked to various forms of cardiovascular abnormalities (5) and is considered to be an independent risk factor for cardiovascular disease. In recent years, vegetarianism has become increasingly popular in Taiwan through growing awareness of associated health benefits, including lower incidence of heart disease. However,

Homocysteine is a thiol-containing amino acid and is formed in mammals from methionine through transmethylation reactions. Via the transsulfuration reaction, homocysteine is further converted to cystathionine and then cysteine by pyridoxal (PL)3-5-phosphate (vitamin B-6)– dependent cystathionine ␤-synthase and cystathionase, respectively. Homocysteine can also be reconverted to methionine through a remethylation reaction catalyzed by methionine synthase. This reaction requires vitamin B-12 (methylcobalamin) as a coenzyme and 5-methyltetrahydrofolate as a methyl donor. 5-Methyltetrahydrofolate is formed from 5,10-methylenetetrahydrofolate by methylenetetrahydrofolate reductase (MTHFR) (1).

1 Supported by National Health Research Institute, Taiwan (DOH 85, 86-HR527 to P.C.H.) and Tzu-Chi University (TCMRC 8601 to H.F.C.). 2 To whom correspondence should be addressed. E-mail: [email protected] 3 Abbreviations used: MTHFR, methylenetetrahydrofolate reductase; PL, pyridoxal; RDNA, Recommended Daily Nutrient Allowances.

4 The term “plasma homocysteine” used in the text and tables refers to plasma total homocysteine unless otherwise indicated.

0022-3166/02 $3.00 © 2002 American Society for Nutritional Sciences. Manuscript received 21 May 2001. Initial review completed 30 June 2001. Revision accepted 17 October 2001. 152

Downloaded from at Swets 76861023 on June 5, 2009

ABSTRACT Mild hyperhomocysteinemia is an independent risk factor for cardiovascular disease and may result from a deficiency of folate, vitamin B-6 or vitamin B-12. Because vitamin B-12 deficiency is often associated with vegetarianism, this study was designed to examine the effect of Taiwanese vegetarian diets on B-vitamin status and plasma homocysteine levels. Female Buddhist lacto-vegetarians (n ⫽ 45; 31– 45 y) and matched omnivores (n ⫽ 45) recruited in Hualien, Taiwan, were investigated. Taiwanese vegetarians consumed normal amount of folate, but only 21% of Taiwan Recommended Daily Nutrient Allowances (RDNA) values of vitamin B-12. Compared with the omnivores, the vegetarians had significantly higher levels of plasma folate (14.79 ⫾ 7.70 vs. 11.98 ⫾ 8.29 nmol/L), but lower levels of vitamin B-12 (207.7 ⫾ 127.1 vs. 403.5 ⫾ 138.9 pmol/L). Fasting plasma homocysteine levels were significantly higher in vegetarians than in omnivores (mean: 11.20 ⫾ 4.27 vs. 8.64 ⫾ 2.06 ␮mol/L; median: 10.5 vs. 8.5 ␮mol/L). Fasting plasma homocysteine was inversely correlated with plasma folate and vitamin B-12 in the vegetarian group. Multiple regression analysis revealed that plasma folate, vitamin B-12 and creatinine were independent determinants of homocysteine variation and contributed to 38.6% of homocysteine variation in the vegetarians. Compared with the omnivores, vegetarians also had significantly lower serum levels of valine, isoleucine, leucine, lysine, alanine and arginine, but higher levels of glycine. In the vegetarian group, fasting plasma homocysteine correlated negatively with serum threonine, lysine, histidine, arginine and cystine, and these amino acids contributed to 38.7% of homocysteine variation. In conclusion, the Buddhist nuns who consumed a lacto-vegetarian diet had mildly elevated fasting plasma homocysteine levels presumably due to lower levels of plasma vitamin B-12. J. Nutr. 132: 152–158, 2002.


SUBJECTS AND METHODS Subjects. Apparently healthy female vegetarians (31– 45 y) and age-, sex-matched omnivores were recruited from Hualien, a city in eastern Taiwan. These subjects were initially recruited for an investigation of the effect of Taiwanese vegetarian diets on risk factors for heart disease. A description of the recruitment was described in a previous report (15). The majority of the vegetarian subjects were Tzu-Chi Buddhist nuns; the omnivorous control subjects were mostly employees of local hospitals in Hualien. The major excluding criteria were as follows: body weight ⱖ 120% of ideal; a history of chronic disease (renal disease, cancer, diabetes mellitus, heart disease, hypertension); alcohol intake; cigarette smoking; and a regular use of vitamin/mineral supplements. A total of 90 subjects (45 vegetarians, 45 omnivores) were enrolled. Among the vegetarians, 6 were vegan and 39 were lactovegetarian (consuming ⬍ 240 mL of low fat or skim milk per day). The average history of vegetarian practice was 7.9 y, with a minimum of 2 y. The study protocol was approved by the National Health Research Institute, Taiwan, and was explained to the subjects before they gave their informed consent. Dietary assessment. Dietary assessment included a 24-h recall supplemented with a semiquantitative food-frequency questionnaire, which was completed by qualified dietitians. A database for Taiwan food composition (16) was used to calculate the daily energy and nutrient intake. Anthropometric measurements and blood collection. After a 10to 11-h overnight fast, anthropometric measurements were performed on each subject. Blood samples were collected in empty tubes or tubes containing EDTA, potassium oxalate or sodium heparin. Plasma and serum were used for routine blood chemical assays and measurements of vitamins, homocysteine and other amino acid concentrations. Laboratory measurements. Determination of plasma total triglycerides and total, HDL- and LDL-cholesterol was described previously (15). Plasma PL-5-phosphate (vitamin B-6) concentration was determined by HPLC with fluorometric detection (Hitachi, Tokyo, Japan) (17). Serum amino acid levels were measured according to the method of Heinrickson et al. (18) using HPLC with UV detection in the Waters PICO-TAG amino acid analysis system (Waters, Milford, MA). Plasma vitamin B-12 concentration was measured by Microparticle Enzyme Immunoassay, and plasma homocysteine level was determined by Fluorescence Polarization Immunoassay (19). Both assays were performed in Abbott IMX system (Abbott Laboratories, Abbott Park, IL). Plasma folic acid concentration was determined by a Lactobacillus casei microbiological assay as previously reported (20).

Statistical analysis. Data were analyzed using MINITAB Release 13.1 for Windows (Minitab, State College, PA). Results were expressed as means ⫾ SD unless otherwise stated. The AndersonDarling test was performed to determine the normality of the measurements. For normal data, differences between vegetarians and omnivores were compared using Student’s t test. For nonnormal data, such as values of plasma homocysteine, folic acid, PL-5-phosphate, vitamin B-12, creatinine and serum amino acids, their median values were calculated. The Mann-Whitney test, a nonparametric procedure, was conducted to compare the median values between the two dietary groups. Correlations between plasma homocysteine and Bvitamins, and between plasma homocysteine and amino acid levels were analyzed by using Pearson’s correlation and Spearman’s nonparametric correlation. Multiple regression analysis was used to determine the independent predictors of plasma homocysteine concentrations. The level of statistical significance was set at P ⬍ 0.05.

RESULTS The general characteristics of the vegetarian and omnivore subjects were described in our previous report (15). Compared with the omnivores, the vegetarians had similar mean age and height, but lower mean body weight, body mass index and blood pressure. As shown in Table 1, the vegetarians consumed less energy, protein, and fat than did the omnivores, which was consistent with the vegetarians’ lower plasma concentrations of total protein, triglycerides and cholesterol. Although riboflavin and niacin intakes were significantly lower in the vegetarians than in the omnivores, both dietary groups consumed ⬎ 70% of the Taiwan Recommended Daily Nutrient Allowances (RDNA) (21) for thiamin, riboflavin and niacin. Due to the lack of a complete nutrient database for Chinese foods, the estimated daily intake of methionine, folic acid, vitamin B-6 and vitamin B-12 was determined on the basis of data obtained from a semiquantitative food-frequency questionnaire. For vegetarians and omnivores, the approximate daily intake of these nutrients was as follows: methionine (1086 vs. 1313 mg/d), folic acid (329 vs. 156 ␮g/d), vitamin B-6 (1.01 vs. 1.39 mg/d) and vitamin B-12 (0.42 vs. 7.12 ␮g/d), respectively. Compared with the omnivores, the vegetarians consumed more folic acid but less vitamin B-12 at only 21% of Taiwan RDNA. The vegetarians had a significantly higher level of plasma folate but a lower concentration of vitamin B-12 compared with the omnivores (Table 2). The plasma concentrations of TABLE 1 Daily intake of energy, protein, fat and B-vitamins, and plasma protein and lipid levels among vegetarians and omnivores1 Variable Dietary intakes Energy, MJ Protein, g (en%) Fat, g (en%) Thiamin, mg Riboflavin, mg Niacin, mg Plasma concentrations Total protein, g/L Total triglycerides, mmol/L Total cholesterol, mmol/L



6.4 ⫾ 1.7a 47 ⫾ 17c (12) 39 ⫾ 18b (22) 1.0 ⫾ 1.2 0.7 ⫾ 0.4c 10.1 ⫾ 4.1b

7.3 ⫾ 2.0 66 ⫾ 23 (15) 50 ⫾ 19 (26) 0.8 ⫾ 0.5 1.2 ⫾ 0.9 13.1 ⫾ 5.2

74 ⫾ 3.0b 0.74 ⫾ 0.31c 3.89 ⫾ 0.57b

76 ⫾ 4.0 1.11 ⫾ 0.62 4.40 ⫾ 0.91

1 Values are means ⫾ SD, n ⫽ 45; a– c significantly different from omnivores: a P ⬍ 0.05, b P ⬍ 0.01, c P ⬍ 0.001. 2 en%, energy percentage.

Downloaded from at Swets 76861023 on June 5, 2009

an unplanned vegetarian diet may result in deficiencies of iron, calcium, vitamin B-2, B-6 and B-12, nutrients often found in animal products (6). It is reasonable to suspect that even a vegetarian diet with adequate folate intake may result in an abnormal homocysteine level, thus increasing risk for cardiovascular disease. This hypothesis was tested in Chile (7), the United States (8,9) and Australia (10), but it has not yet been evaluated among Taiwanese vegetarians. Unlike Western vegetarian diets, which consist mainly of vegetables, fruits, nuts, seeds, dairy products and eggs, Taiwanese vegetarian diets include few or no dairy products and consist mainly of rice, vegetables, fruits and significant amounts of soy products. A number of studies have reported the effects of soy protein and soy isoflavones in reducing blood lipids (11,12). One possible mechanism for this hypocholesterolemic action of soy is through its effect on changing blood amino acid concentrations (13,14). However, it is not clear whether dietary protein and serum amino acid levels have any effect on plasma homocysteine levels. Therefore, the present study was designed to evaluate the effects of Taiwanese vegetarian diets on B-vitamin status, serum amino acid profile and on plasma homocysteine levels.




TABLE 2 Plasma concentrations of total homocysteine, folate, vitamin B-12, pyridoxal-5-phosphate (PLP) and creatinine in vegetarians and omnivores1 Vegetarian Homocysteine, ␮mol/L Folate, nmol/L Vitamin B-12, pmol/L PLP, nmol/L Creatinine, ␮mol/L

11.20 ⫾ 4.27b (10.5) 14.79 ⫾ 7.70a (12.57) 207.7 ⫾ 127.1b (204) 47.10 ⫾ 18.60 (45.5) 62.71 ⫾ 5.35 (61.81)

Omnivore 8.64 ⫾ 2.06 (8.5) 11.98 ⫾ 8.29 (9.51) 403.5 ⫾ 138.9 (392) 61.60 ⫾ 43.2 (47.0) 59.65 ⫾ 8.41 (61.18)

1 Values are means ⫾ SD (medians), n ⫽ 45; a,b significantly different from omnivores: a P ⬍ 0.05, b P ⬍ 0.0001 (Mann-Whitney Test).

TABLE 3 Serum concentrations of essential (EAA) and nonessential (NEAA) amino acids in vegetarians and omnivores1 EAA/NEAA

Vegetarian (n ⫽ 38)

Omnivore (n ⫽ 45)

␮mol/L Valine Methionine Isoleucine Leucine Lysine Hydroxyproline Glycine Alanine Arginine Cystine

170.0 ⫾ 29.0c (172) 23.4 ⫾ 5.2 (23.3) 44.5 ⫾ 9.2c (44.9) 98.0 ⫾ 20.2b (100) 121.0 ⫾ 21.0c (118) 5.1 ⫾ 1.1c (4.9) 319.0 ⫾ 72.0a (315) 307.0 ⫾ 64.6b (301) 74.9 ⫾ 23.1a (70.3) 57.9 ⫾ 23.1 (51.1)

212.0 ⫾ 46.0 (209) 24.8 ⫾ 5.4 (25.1) 56.6 ⫾ 12.9 (53.9) 117.0 ⫾ 27.6 (114) 158.0 ⫾ 32.0 (159) 10.7 ⫾ 6.9 (8.6) 280.0 ⫾ 82.0 (269) 360.0 ⫾ 92.4 (347) 88.1 ⫾ 22.8 (88.7) 60.5 ⫾ 34.2 (47.9)

a Values are means ⫾ SD (medians); a,b significantly different from omnivores: a P ⬍ 0.05, b P ⬍ 0.01, c P ⬍ 0.0001 (Mann-Whitney Test).

Downloaded from at Swets 76861023 on June 5, 2009

PL-5-phosphate (the coenzyme form of vitamin B-6) and creatinine did not differ between the vegetarians and the omnivores. Interestingly, the vegetarians had significantly higher levels of fasting plasma homocysteine than were found in the omnivores (mean:11.20 ⫾ 4.27 vs. 8.64 ⫾ 2.06 ␮mol/L; median: 10.5 vs. 8.5 ␮mol/L). Table 3 summarizes the serum concentrations of L-amino acids that were found to be affected by the vegetarian diet. Compared with the omnivores, vegetarians had significantly lower levels of valine, isoleucine, leucine, lysine, alanine, and arginine and higher concentrations of glycine. No differences were found in either methionine or cystine levels between these two dietary groups. Some dietary determinants of plasma homocysteine were examined by Spearman’s correlation and multiple regression. Fasting plasma homocysteine in the vegetarians correlated negatively with plasma folate (r ⫽ ⫺0.47, P ⫽ 0.002) and vitamin B-12 (r ⫽ ⫺0.523, P ⬍ 0.001) but did not correlate with plasma PL-5-phosphate (Table 4). The vegetarians also demonstrated a slightly positive correlation between plasma homocysteine and plasma creatinine (r ⫽ 0.237, P ⫽ 0.121). Only the vegetarians demonstrated significant negative correlations of plasma homocysteine with serum levels of threonine, lysine, histidine, arginine and cystine. In the omnivore group, only serum asparagine correlated negatively with plasma homocysteine. No correlation was found between plasma homocysteine and either plasma B-vitamins or plasma creatinine levels in the omnivore group. Thus, diet does affect the correlations of plasma homocysteine with plasma B-vitamins, creatinine and serum amino acids, and homocysteine itself may be influenced by the diet’s interaction effects with these variables. The Spearman correlations between plasma homocysteine and plasma B-vitamins, creatinine, and amino acids were also calculated in subjects combined from the two dietary groups; the results were compared with those obtained in the vegetarian or the omnivore group (Table 4). The degrees of association between plasma homocysteine and plasma folate (r ⫽ ⫺0.124), vitamin B-12 (r ⫽ ⫺0.480), creatinine (r ⫽ 0.199) and some amino acids were diminished in the combined group compared with correlations seen in the vegetarian group, probably due to the confounding and interaction effects caused by the type of diet. Because diet seems to be an important determinant of plasma homocysteine level, the results obtained in all subjects from the combined group may seem misleading. Therefore, we decided to present the results and draw inferences for vegetarians and omnivores independently in the correlation and multiple regression analyses. In addition to Spearman’s correlations, Pearson’s correlation co-

efficients were also calculated (data not shown) to evaluate the relationships between plasma homocysteine and independent variables. The results obtained from these two methods were similar and provided methodological support for the following multiple regression analysis. Multiple regression analysis was used to incorporate the study variables into the statistical model and to assess the strength of the association between plasma homocysteine concentrations and each of the predictors while controlling for the other variables. Plasma levels of folate, vitamin B-12 and creatinine, but not PL-5-phosphate, were significant predictors of plasma homocysteine levels. In the vegetarian group, the coefficient of determination, R2, was calculated to be 0.386, suggesting that these variables contributed 38.6% of homocysteine variation (Table 5). When the level of plasma creatinine was excluded, the coefficient of determination was 0.286 (data not shown), indicating that the plasma level of creatinine contributed ⬃10% of the homocysteine variation. In contrast, neither the plasma B-vitamins nor plasma creatinine was a significant predictor of plasma homocysteine in the omnivore group. Serum amino acids were not combined with plasma Bvitamins and creatinine for the multiple regression analysis to avoid unnecessary, confused information caused by a statistical problem called multicollinearity. This problem occurs when two or more predictors used in regression are correlated to the extent that they convey essentially the same information about the variation of the dependent variable. Consequently, the contribution of each individual predictor becomes less prominent and none can be shown to be statistically significant. The problem of multicollinearity indeed existed when the serum amino acids that were shown to be correlated with plasma homocysteine (Table 4) were analyzed in a separate regression model. The overall R2 was 0.387, suggesting that these amino acids contributed some information to the prediction of plasma homocysteine; however, none of their individual P-values was ⬍ 0.05. Furthermore, when we included serum amino acids with plasma B-vitamins and creatinine in the regression analysis, the power of significance of B-vitamins in predicting plasma homocysteine was weakened because the more predictors included in the regression model, the less power each individual predictor has.



TABLE 4 Correlations of plasma homocysteine concentration with plasma folate, vitamin B-12, pyridoxal-5-phosphate (PLP), creatinine and serum amino acid concentrations in the vegetarians and omnivores1 Vegetarian Variable

Folate Vitamin B-12 PLP Creatinine Amino acid Threonine Methionine Lysine Asparagine Histidine Arginine Cystine

r1 ⫺0.470 ⫺0.523 ⫺0.028 0.237 ⫺0.409 ⫺0.310 ⫺0.376 ⫺0.311 ⫺0.491 ⫺0.550 ⫺0.461



0.002 ⬍0.001 0.855 0.121

⫺0.087 ⫺0.033 0.145 0.047

0.011 0.059 0.020 0.057 0.002 ⬍0.001 0.004

⫺0.017 ⫺0.041 ⫺0.185 ⫺0.323 0.119 ⫺0.221 0.176

(n ⫽ 44)

(n ⫽ 38)

Vegetarian ⫹ Omnivore

Omnivore P


0.572 0.832 0.341 0.761

⫺0.124 ⫺0.480 0.031 0.199

0.912 0.788 0.223 0.030 0.435 0.145 0.249

⫺0.230 ⫺0.187 ⫺0.406 ⫺0.275 ⫺0.112 ⫺0.439 ⫺0.099

(n ⫽ 45)

(n ⫽ 45)

P (n ⫽ 89)

(n ⫽ 83)

0.248 ⬍0.001 0.775 0.062 0.036 0.091 ⬍0.001 0.012 0.315 ⬍0.001 0.375

DISCUSSION The most important finding in this study is that the Taiwanese vegetarians had significantly higher fasting plasma homocysteine levels than the matched omnivores. A mild elevation of plasma homocysteine level has also been reported recently in male ovo-lactovegetarians and vegans in Australia (10) and in male and female lacto- or ovo-lactovegetarians in Chile (7). Although our study and the other two studies (7,10) all reported 30 – 40% greater plasma homocysteine among the vegetarians, their homocysteine values were higher than ours. Mann et al. (10) reported mean homocysteine concentrations of 11.6, 15.8 and 19.2 ␮mol/L for the meat eaters, ovolactovegetarians and vegans, respectively. Mezzano et al. (7) reported geometric mean homocysteine concentrations of 9.55 and 13.5 ␮mol/L for the omnivores and ovo-lactovegetarians, respectively. The mean plasma homocysteine concentration of 8.64 or median of 8.5 ␮mol/L measured in our omnivores is in good agreement with the value of 8.6 ␮mol/L calculated by averaging several mean fasting plasma homocysteine levels of healthy middle-aged women reported previously (22–25). The discrepancy in the mean value of homocysteine concentration observed among the omnivores from different studies may be attributed in part to gender differences. The fasting plasma

TABLE 5 Multiple regression analysis of homocysteine concentrations among vegetarians using plasma vitamins and creatinine as independent predictors1,2 Variable Folate Vitamin B-12 PLP3 Creatinine

Regression coefficient

Standard error


⫺0.266 ⫺0.012 0.059 0.245

0.087 0.004 0.036 0.098

0.004 0.009 0.111 0.017

1 Variables are significant predictors of plasma total homocysteine at P ⬍ 0.05, n ⫽ 44. 2 R2 ⫽ 0.386 3 PLP, pyridoxal-5-phosphate.

homocysteine concentrations were 20% higher in men than in women (25,26) presumably due to greater lean body mass, muscle mass and creatinine production (27) and lower estradiol (28) in men than in women. It will be interesting to determine what causes the mild elevation of fasting plasma homocysteine in vegetarians. Several factors have been shown to affect plasma homocysteine levels, including age, physical determinants, genetic defects, certain disease status, medication and nutritional deficiencies (29). Many of these factors were controlled and can be excluded in our study given our criteria for subject selection. Therefore, only the dietary and nutrition-related factors that are potential predictors of plasma homocysteine are discussed. Several B-vitamins play important regulatory roles in homocysteine metabolism and plasma homocysteine levels in humans. Among them, folate, vitamin B-6 and vitamin B-12 seem to be the most important factors. Because vitamin B-6 and vitamin B-12 are found mainly in animal products, deficiencies of these vitamins are a cause for concern in vegetarian diets. Mann et al. (10) and Mezzano et al. (7) claimed that the impairment of vitamin B-12 status is the main factor in determining elevated homocysteine levels in vegetarians. Indeed, in our vegetarian group, impairment of vitamin B-12 was evident. The mean plasma vitamin B-12 level was at the lower end of the normal range and was 50% lower than that of the omnivore group. In addition, 14 of 45 (31%) vegetarians had plasma vitamin B-12 ⬍ 147.8 pmol/L, a level indicating deficiency. In the present study, not only vitamin B-12, but also folate was negatively correlated with plasma homocysteine, and they were significant determinants of plasma homocysteine in the vegetarian group. Nevertheless, the protective effects of higher plasma folate levels do not successfully counteract the hyperhomocysteinemic effects of low plasma vitamin B-12, thus resulting in an overall increase in homocysteine in the vegetarians. We found no correlation between plasma homocysteine and either plasma folate or vitamin B-12 in omnivores, consistent with previous findings (7). The most likely explanation is that the omnivores had adequate levels of vitamin B-12 and folate, as well as normal and relatively consistent concentrations of plasma homocysteine. The present study demonstrated a lack of correlation be-

Downloaded from at Swets 76861023 on June 5, 2009

1 Spearman’s correlation coefficients were considered significant at P ⬍ 0.05.



levels in the vegetarians, with the exception of methionine, cystine and phenylalanine. On the other hand, serum glycine was 14% higher in the vegetarians than in the omnivores. Thus, the amino acid pattern was slightly unbalanced. However, the plasma albumin levels did not differ between the two groups, and thus the protein nutrition status of the vegetarians was only marginally insufficient. It is not known whether the marginal insufficiency of protein nutrition has any effect on plasma homocysteine level. The strong inverse correlations between plasma homocysteine and serum threonine, lysine, histidine, and arginine in our vegetarian subjects are of interest. The biochemical explanations for the association between plasma homocysteine and these amino acids remain to be determined. Arginine is involved in creatinine synthesis, which couples with the formation of S-adenosylhomocysteine in the liver. Thus, although we would expect to see a positive correlation between serum arginine and plasma homocysteine, they were instead found to be negatively correlated. Glycine stimulates transmethylation of methionine to form S-adenosylhomocysteine (35). Therefore, a higher level of serum glycine in the vegetarian subjects might contribute in part to the elevation of homocysteine levels in the vegetarians. However, no significant association between glycine and homocysteine levels was found in the present study. The lack of correlation between fasting plasma homocysteine and serum amino acids in the omnivores may be explained in part by the fact that they had relatively normal, balanced serum amino acids and normal, consistent plasma homocysteine concentrations. The relatively high coefficient of determination (R2 ⫽ 0.387) observed in the multiple regression analysis strongly suggests that serum L-amino acids are potential predictors of plasma homocysteine in the vegetarian group; however, more studies are required to confirm our observations and identify which amino acids are the independent predictors. The vegetarians ate less fat and less cholesterol and had lower plasma total triglycerides and total cholesterol than the omnivores. The relationships between plasma homocysteine and total cholesterol levels are inconclusive. A positive correlation between plasma homocysteine and total cholesterol levels has been reported (36), whereas no such an association was found in control and hypercholesterolemic patients (37). In the present study, we found no correlation between plasma homocysteine and plasma total cholesterol in either the vegetarians or the omnivores, presumably because their cholesterol levels were all within normal ranges. Caffeine intake is another dietary factor that is positively associated with plasma homocysteine (38). Drinking large quantities of coffee (1 L/d) has been shown to increase fasting plasma homocysteine in healthy subjects (39). In our study, the omnivores drank more coffee (240 vs. 26 mL/d) and more tea (150 vs.106 mL/d) than the vegetarians. However, the effects of caffeine on plasma homocysteine appear to be small at these levels of consumption. Plasma creatinine has been shown to be positively correlated with plasma homocysteine level in men and women (27), a result of both renal function and the relationship between homocysteine production and creatine-creatinine synthesis in the remethylation pathway. In this study, although plasma creatinine was not significantly correlated with plasma homocysteine (P ⫽ 0.121), a somewhat positive association (r ⫽ 0.237) between these two variables was seen in the vegetarians. Plasma creatinine was a significant predictor of plasma homocysteine and increased the regression coefficient of determination (R2) by 10% in our vegetarians.

Downloaded from at Swets 76861023 on June 5, 2009

tween plasma homocysteine and plasma PL-5-phosphate in both dietary groups. This confirms previous reports that vitamin B-6 did not significantly affect levels of fasting plasma homocysteine (30), but that vitamin B-6 could be considered a weak determinant of nonfasting homocysteine concentrations (4,9). Despite the fact that vitamin B-12 and folate are the two important determinants of fasting plasma homocysteine levels, these factors account for only 28.6% of homocysteine variation in the vegetarian group and only 1.3% in the omnivore group. These results suggest that there are other dietary and/or nondietary factors involved in the regulation of homocysteine levels. Riboflavin is also an essential cofactor in both transsulfuration and remethylation pathways of homocysteine metabolism, but the relationship between riboflavin status and homocysteine levels is not yet understood. Although some studies have shown that plasma homocysteine is inversely correlated with both riboflavin intake (31) and plasma riboflavin (32) in healthy subjects, we found no association between plasma homocysteine and riboflavin intake in either vegetarians or omnivores (data not shown). We did not measure plasma riboflavin levels in the present study. Therefore, it is not clear whether plasma riboflavin is a determinant of plasma homocysteine in our subjects, nor is it clear whether a slightly lower intake of riboflavin (70% of RDNA) has any influence on plasma homocysteine. Information regarding the association between niacin status and homocysteine metabolism is limited. Recently, a pharmacologic dosage of niacin, rather than niacin deficiency, has been shown to interfere with methionine metabolism by affecting vitamin B-6 status, leading to hyperhomocysteinemia in rats (33). In our study, the vegetarians consumed ⬃77% of the RDNA for niacin. At this level, the effect of niacin on homocysteine should be very small, if present at all. There is little evidence regarding the association between fasting plasma homocysteine and dietary protein or amino acids. Although Sugiyama et al. (34) observed a significantly lower hepatic level of S-adenosylhomocysteine, the precursor of homocysteine, in rats fed soybean protein compared with rats fed casein protein, Shimakawa et al. (31) showed no correlation between plasma homocysteine and intake of methionine or protein among middle-aged adults. Recently, Jeckel et al. (28) reported a very minor contribution (1%) of plasma protein to the variation in plasma homocysteine in a healthy, middle-aged German population. Although dietary methionine is the precursor of homocysteine and is more abundant in the omnivore diet, there was no difference in serum methionine concentrations between the vegetarian and the omnivore groups. Furthermore, neither group demonstrated a correlation between serum methionine and plasma homocysteine. These results support the previous suggestions that serum methionine and plasma homocysteine are normally maintained at a relatively constant level (10). As shown in Table 3, The omnivores had normal, balanced serum amino acid levels. The vegetarians, on the other hand, had lower serum levels of valine, isoleucine, leucine, lysine and arginine but higher concentrations of glycine. These findings are similar to previous observations made in omnivores who were switched to a plant-based diet for 4 wk (13). The vegetarian subjects of this study consumed 0.92 g protein/kg body weight, mainly from plants (95% of total). Thus, their protein nutritional status was inferior to that of the omnivores, who consumed 1.19 g/kg body weight, of which ⬃47% was of animal origin. This difference in protein intake may contribute to the slightly lower (13–23%) serum essential amino acid


ACKNOWLEDGMENTS The authors are grateful to the Tzu-Chi Buddhist nuns and the control subjects who participated in this study. We also thank KueiHua Lo for her excellent technical assistance, Yin-Huei Wong and Pi-Ru Yu for their dietary assessment and Alice Wang for editorial assistance.

LITERATURE CITED 1. Finkelstein, J. D. (1998) The metabolism of homocysteine: pathways and regulation. Eur. J. Pediatr. 157 (suppl. 2): S40 –S44. 2. Mc Cully, K. S. (1969) Vascular pathology of homocysteinemia: implications for the pathogenesis of arteriosclerosis. Am. J. Pathol. 56: 111–128. 3. Kanwar, Y. S., Manaligod, J. R. & Wong, P.W.K. (1976) Morphologic studies in a patient with homocystinuria due to 5,10-methylenetetrahydrofolate reductase deficiency. Pediatr. Res. 10: 598 – 609. 4. Selhub, J., Jacques, P. F., Wilson, P.W.F., Rush, D. & Rosenberg, I. H.

(1993) Vitamin status and intake as primary determinants of homocysteinemia in the elderly. J. Am. Med. Assoc. 270: 2693–2698. 5. Brattstrom, L. & Wilcken, D.E.L. (2000) Homocysteine and cardiovascular disease: cause or effect ? Am. J. Clin. Nutr. 72: 315–323. 6. Dwyer, J. T. (1991) Nutritional consequences of vegetarianism. Annu. Rev. Nutr. 11: 61–91. 7. Mezzano, D., Mun˜ oz, X., Martinez, C., Cuevas, A., Panes, O., Aranda, E., Guasch, V., Strobel, P., Mun˜ oz, B., Rodriguez, S., Pereira, J. & Leighton, F. (1999) Vegetarians and cardiovascular risk factors: hemostasis, inflammatory markers and plasma homocysteine. Thromb. Haemost. 81: 913–917. 8. Haddad, E. H., Berk, L. S., Kettering, J. D., Hubbard, R. W. & Peters, W. R. (1999) Dietary intake and biochemical, hematologic and immune status of vegans compared with nonvegetarians. Am. J. Clin. Nutr. 70 (suppl.): 586S–593S. 9. DeRose, D. J., Charles-Marcel, Z. L., Jamison, J. M., Muscat, J. E., Braman, M. A., McLane, G. D. & Mullen, J. K. (2000) Vegan diet-based lifestyle program rapidly lowers homocysteine levels. Prev. Med. 30: 225–233. 10. Mann, N. J., Li, D., Sinclair, A. J., Dudman, N.P.B., Guo, X. W., Elsworth, G. R., Wilson, A. K. & Kelly, F. D. (1999) The effect of diet on plasma homocysteine concentrations in healthy male subjects. Eur. J. Clin. Nutr. 53: 895– 899. 11. Anderson, J. W., Johnstone, B. M. & Cook-Newell, M. E. (1995) Metaanalysis of the effects of soy protein intake on serum lipids. N. Engl. J. Med. 333: 276 –282. 12. Anthony, M. S., Clarkson, T. B. & Williams, J. K. (1998) Effects of soy isoflavones on atherosclerosis: potential mechanisms. Am. J. Clin. Nutr. 68: 1390S–1393S. 13. Sanchez, A., Horning., M. C., Shavlik, G. W., Wingeleth, D. C. & Hubbard, R. W. (1985) Changes in levels of cholesterol associated with plasma amino acids in humans fed plant proteins. Nutr. Rep. Int. 32: 1047–1056. 14. Katan, M. B. Vroomen, L.H.M. & Hermus, R.J.J. (1982) Reduction of casein-induced hypercholesterolemia and atherosclerosis in rabbits and rats by dietary glycine, arginine and alanine. Atherosclerosis 43: 381–391. 15. Lu, S. C., Wu, W. H., Lee, C. A., Chou, H. F., Lee, H. R. & Huang, P. C. (2000) LDL of Taiwanese vegetarians are less oxidizable than those of omnivores. J. Nutr. 130: 1591–1596. 16. Department of Health (1998) Taiwan Nutrient Databases. Department of Health, Executive Yuan, Taipei, Taiwan. 17. Kimura, M., Kanehira, K. & Yokoi, K. (1996) Highly sensitive and simple liquid chromatographic determination in plasma of B6 vitamins, especially pyridoxal 5⬘-phosphate. J. Chromatogr. 722: 295–301. 18. Heinrikson, R. L. & Meredith, S. C. (1984) Amino acid analysis by reverse-phase high performance liquid chromatography: precolumn derivatization with phenylisothiocyanate. Anal. Biochem. 136: 65–74. 19. Shipchandler, M. T. & Moore, E. G. (1995) Rapid, fully automated measurement of plasma homocyst(e)ine with the Abbott IMX analyzer. Clin. Chem. 41: 991–994. 20. Horne, D.W. & Patterson, D. (1988) Lactobacillus casei microbiological assay of folic acid derivatives in 96-well microtiter plates. Clin. Chem. 34: 2357– 2359. 21. Department of Health. (1993) Recommended Daily Nutrient Allowances, 5th ed. Department of Health, Executive Yuan, Taipei, Taiwan. 22. Refsum, H., Helland, S. & Ueland, P. M. (1985) Radioenzymic determination of homocysteine in plasma and urine. Clin. Chem. 31: 624 – 628. 23. Andersson, A., Brattstrom, L., Israelsson, B., Isaksson, A, Hamfelt, A. & Hultberg, B. (1992) Plasma homocysteine before and after methionine loading with regard to age, gender, and menopausal status. Eur. J. Clin. Investig. 22: 79 – 87. 24. Kang, S. S., Wong, P.W.K., Cook, H. Y., Norusis, M. & Messer, J. V. (1986) Protein-bound homocyst(e)ine: a possible risk factor for coronary artery disease. J. Clin. Investig. 77: 1482–1486. 25. Nygard, O., Vollset, S. E. & Refsum, H. (1995) Total plasma homocysteine and cardiovascular risk profile. The Hordaland Homocysteine Study. J. Am. Med. Assoc. 274: 1526 –1533. 26. Lussier-Cacan, S., Xhignesse, M., Piolot, A., Selhub, J., Davignon, J. & Genest, J., Jr. (1996) Plasma total homocysteine in healthy subjects: sex specific relation with biological traits. Am. J. Clin. Nutr. 64: 587–593. 27. Brattstrom, L., Lindgren, A., Israelsson, B., Andersson, A. & Hultberg, B. (1994) Homocysteine and cysteine: determinants of plasma levels in middleaged and elderly subjects. J. Intern. Med. 236: 633– 641. 28. Dierkes, J., Jackel, A., Ambrosch, A., Westphal, S., Luley, C. & Boeing, H. (2001) Factors explaining the difference of total homocysteine between men and women in the European investigation into cancer and nutrition Potsdam study. Metabolism 50: 640 – 645. 29. Fonseca, V. Guba, S. C. & Fink, L. M. (1999) Hyperhomocysteinemia and the endocrine system: implications for atherosclerosis and thrombosis. Endocr. Rev. 20: 738 –759. 30. Miller, J. W., Ribaya-Mercado, J. D., Russell, R. M., Shepard, D. C., Morrow, F. D., Cochary, E. F., Sadowski, J. A. Gershoff, S. N. & Selhub, J. (1992) Effect of vitamin B-6 deficiency on fasting plasma homocysteine concentrations. Am. J. Clin. Nutr. 55: 1154 –1160. 31. Shimakawa, T., Nieto, F. J., Malinow, M. R., Chambless, L. E., Schreiner, P. J. & Szklo, M. (1997) Vitamin intake: a possible determinant of plasma homocysteine among middle-aged adults. Ann. Epidemiol. 7: 285–293. 32. Hustad, S., Ueland, P. M., Vollset, S. E., Zhang Y., Bjorke-Monsen, A. L. & Schneede, J. (2000) Riboflavin as a determinant of plasma total homocys-

Downloaded from at Swets 76861023 on June 5, 2009

Nondietary factors such as lifestyle may also affect homocysteine concentrations. DeRose et al. (9) recently reported that self-selected subjects participating in a vegan diet– based lifestyle program, which included moderate physical exercise, stress management, spirituality enhancement sessions, group support and abstention from caffeine, tobacco or alcohol, had significantly reduced plasma homocysteine levels from 8.66 ⫾ 2.7 to 7.53 ⫾ 2.1 ␮mol/L in 1 wk. The authors suggested that multiple factors could have contributed to the homocysteine-lowering effects observed in their study. Although the causal relationship between hyperhomocysteinemia and cardiovascular disease is still under debate, elevated homocysteine level is nonetheless considered an independent graded risk factor for cardiovascular disease (5). Some studies showed that homocysteine may account for 10% of cardiovascular disease risk (40). There is no consensus definition for hyperhomocysteinemia. However, a value of ⬎ 11.1 (41), ⬎ 14.0 (4), ⬎ 15.0 (25) or ⬎ 16.3 ␮mol/L (42) of plasma homocysteine has been suggested to present a higher risk for developing some forms of cardiovascular disease compared with the value of ⬍ 9.0 ␮mol/L of plasma homocysteine (43). Meta-analysis of several retrospective and prospective studies suggests that every 5 ␮mol/L increment in fasting or nonfasting homocysteine above 10 ␮mol/L is associated with a 60 – 80% or 20 –30% increase in cardiovascular risk, respectively (44,45). Boushey et al. (40) also reported that a 5 ␮mol/L homocysteine increment elevated cardiovascular risk by as much as that caused by a cholesterol increase of 0.5 mmol/L (20 mg/dL). In the present study, the vegetarian group showed a 1.2 ␮mol/L increment above 10 ␮mol/L in fasting homocysteine and also showed a 1.3 mmol/L reduction in plasma total cholesterol below 5.2 mmol/L (200mg/dL). Accordingly, it seems reasonable to assume that the harmful effect of the elevated homocysteine levels on the risk for heart disease may be counteracted by the beneficial effect of lower plasma cholesterol levels in the vegetarian subjects. In conclusion, the present study demonstrates significantly higher levels of fasting plasma homocysteine in Buddhist vegetarians than in omnivores residing in Hualien, Taiwan. The mild elevation of plasma homocysteine levels is attributed in part to the net effects of high folate and of low vitamin B-12 on homocysteine levels. Other factors such as creatinine, amino acids and lifestyle may also regulate the homocysteine concentrations. Vegetarian diets have the beneficial health effects of reducing risks of heart disease, but also have the potential health risks caused by an inadequate intake of certain nutrients. It seems appropriate to recommend vitamin B-12 supplementation for our vegetarian subjects to improve the adverse effects of mild elevation of homocysteine levels.




teine: effect modification by the methylenetetrahydrofolate reductase C677T polymorphism. Clin. Chem. 46: 1065–1071. 33. Basu, T. K. & Mann, S. (1997) Vitamin B-6 normalizes the altered sulfur amino acid status of rats fed diets containing pharmacological levels of niacin without reducing niacins hypolipidemic effects. J. Nutr. 127: 117–121. 34. Sugiyama, K., Kanamori, H., Akachi, T. & Yamakawa, A. (1996) Amino acid composition of dietary proteins affects plasma cholesterol concentration through alteration of hepatic phospholipid metabolism in rats fed a cholesterol free diet. J. Nutr. Biochem. 7: 40 – 48. 35. Benevenga, N. J. & Steele, R. D. (1984) Adverse effects of excessive consumption of amino acids. Annu. Rev. Nutr. 4: 157–181. 36. Olszewski, A. J., Szostak, W. B., Bialkowska, M., Rudnicki, S. & McCully, K. S. (1989) Reduction of plasma lipid and homocysteine levels by pyridoxine, folate, cobalamin, choline, riboflavin, and troxerutin in atherosclerosis. Atherosclerosis 75: 1– 6. 37. Laskowska-Klita, T., Szymczak, E. & Radomyska, B. (2001) Serum homocysteine and lipoprotein (a) concentrations in hypercholesterolemic and normocholesterolemic children. Clin. Pediatr. 40: 149 –154. 38. Jacques, P. F., Bostom, A. G., Wilson, P. W., Rich, S., Rosenberg, I. H. & Selhub, J. (2001) Determinants of plasma total homocysteine concentration in the Framingham offspring cohort. Am. J. Clin. Nutr. 73: 499 –500. 39. Urgert, R., van Vliet, T., Zock, P. L. & Katan, M. B. (2000) Heavy coffee

consumption and plasma homocysteine: a randomized controlled trial in healthy volunteers. Am. J. Clin. Nutr. 72: 1107–1110. 40. Boushey, C. J., Beresford, S.A.A., Omenn, G. S. & Motulsky, A. G. (1995) A quantitative assessment of plasma homocysteine as a risk factor for vascular disease: probable benefits of increasing folate intakes. J. Am. Med. Assoc. 274: 1049 –1057. 41. Stampfer, M. J., Malinow, M. R., Willett, W. C., Newcomer, L. M., Upson, B., Ullman, D., Tishler, P. V. & Hennekens, C. H. (1992) A prospective study of plasma homocyst(e)ine and risk of myocardial infarction in US physicians. J. Am. Med. Assoc. 268: 877– 881. 42. Ubbink, J. B., Vermaak, W. J., van der Merwe, A. & Becker, P. J. (1993) Vitamin B-12, vitamin B-6, and folate nutritional status in men with hyperhomocysteinemia. Am. J. Clin. Nutr. 57: 47–53. 43. Nygard, O., Nordrehaug, J. E., Refsum, H., Ueland, P. M., Farstad, M. & Vollset, S. E. (1997) Plasma homocysteine levels and mortality in patients with coronary artery disease. N. Engl. J. Med. 337: 230 –236. 44. Eikelboom, J. W., Lonn, E., Genest, J., Jr., Hankey, G. & Yusuf, S. (1999) Homocyst(e)ine and cardiovascular disease: a critical review of the epidemiologic evidence. Ann. Intern. Med. 131:363–375. 45. Refsum, H., Ueland, P. M., Nygard, O. & Vollset, S. E. (1998) Homocysteine and cardiovascular disease. Annu. Rev. Med. 49: 31– 62.

Downloaded from at Swets 76861023 on June 5, 2009