Influence of 5,10-methylenetetrahydrofolate reductase polymorphism on whole-blood folate concentrations measured by LC-MS/MS, microbiologic assay, and Bio-Rad radioassay.
To cover a wide range of folate concentrations, we purchased EDTA anticoagulated whole blood samples from 2 blood banks. These specimens included 96 and 75 samples from US and European blood donors, respectively, collected in spring 2006. We also purchased non-anticoagulated whole blood samples from the US blood bank to prepare matching serum samples. Whole blood samples were stored at -70[degrees]C before hemolysis with 1% (wt/vol) L-ascorbic acid, pH2.7 (1:11 dilution). The same hemolysate, incubated at 37[degrees]C for 4 h toensure complete folate deconjugation (8) and then frozen at -70[degrees]C, was used for LC-MS/MS folate pattern analysis and for total folate (TFOL) analysis by microbiologic assay. A hemolysate without incubation was used for the BR; the manufacturer states that no incubation is needed if the he molysate is frozen and assayed later. MTHFR C677T (rs1801133) polymorphism was genotyped using the MGB Eclipse[TM] probe system (Nanogen) (12).
We analyzed the data with SAS (version 9; SAS Institute) and Analyze-it for Microsoft Excel software. The effect of genotype (C/C, C/T, and T/T) and data set (US and Europe) on TFOL concentrations was studied by 2-factor ANOVA for each method separately. For descriptive statistics, each data set was analyzed separately. In addition to individual folate species, we also calculated the ratio of 5C[H.sub.3]THF to TFOL and of reduced folates other than 5C[H.sub.3]THF [non-5C[H.sub.3]THF; sum of 5-formyltetrahydrofolic acid (5CHOTHF), 5,10methenyltetrahydrofolic acid (5,10CH=THF), and tetrahydrofolic acid (THF)] to TFOL, as described previously (1-5). To compare the methods, we used the sum of folate species determined by LC-MS/MS and TFOL determined by microbiologic assay or BR. Method regression equations are presented only for the combined set because we found no differences between the 2 data sets other than the concentration ranges. Methods were compared by least-squares regression, after log-transformation of the data, to account for nongaussian distribution. As in our method comparison paper for serum folate (13), we developed multiple linear regression models that used adummybinary variable (0, 1), IND, to account for intercept differences, and an interaction variable (IND x [log.sub.10] microbiologic assay) to account for differences in slope over 2 discrete concentration intervals (cutoff of 370 nmol/L for microbiologic assay representing the median concentration). The fit of the models was evaluated by comparing the sum of squared residuals (SSR) to the predicted residual sum of squares (PRESS). We tested the recoveries of the microbiologic assay and BR methods by adding each of the 5 folate calibrators (Merck Eprova) at 10 nmol/L [in 1% (wt/vol) L-ascorbic acid, pH 2.7, for microbiologic assay and in protein diluent for BR] to a whole-blood hemolysate pool.
The concentrations of folate species in whole blood were higher in the US than in the European sample set, as was the median (range) TFOL concentration by LCMS/ MS [378 (228-820) nmol/L, US set (n = 96); 250 (122-582) nmol/L, European set (n = 75)] (Table 1). The relative whole-blood folate pattern [median (range)] was similar for individuals with C/C (n = 73) and C/T (n = 66) genotype, regardless of the sample set: 88% (71%91%) and 86% (50%-91%), respectively, for 5C[H.sub.3]THF vs 12% (9%-29%) and 14% (9%-51%) for non5C[H.sub.3]THF (see also Fig. 1 in Data Supplement). Individuals with T/T (n = 32) genotype displayed 58% (22%87%) 5C[H.sub.3]THF vs 42% (13%-78%) non-5C[H.sub.3]THF forms, with 5,10CH=THF being the largest contributor to the non-5C[H.sub.3]THF portion. We did not find differences in serum folate pattern or TFOL by genotype (see Table 1 in the Data Supplement).
We found a significant effect of data set (P <0.001) on TFOL concentrations determined by LC-MS/MS or microbiologic assay, but no effect of genotype and no genotype x data set interaction [see Fig. 2 in the Data Supplement for scatter plots by genotype (panel A) and by data set (panel C)]. The LC-MS/MS and microbiologic assays showed a correlation of r = 0.94, but LC-MS/MS produced slightly lower values, resulting in a significant negative concentration-dependent difference of -10% (95% CI, -12% to -8%; 2 SD limits of agreement, -32% to 20% based on [log.sub.10]-transformed Bland Altman analysis). The multiple linear regression equation was as follows:
[log.sub.10] LC-MS/MS = 0.2959 + (0.8697 x [log.sub.10] microbiologic assay) - (0.1582 x IND) + (0.0534 x IND x [log.sub.10] microbiologic assay),
with IND=0 for microbiologic assay results [less than or equal to]370 nmol/L and IND = 1 for microbiologic assay results >370 nmol/L (Fig. 1A). The SSR (0.5406) agreed with the PRESS (0.5679).
The BR assay produced lower results than the other 2 assays. Because of a significant genotype x data set interaction (P = 0.0041), the 2-factor ANOVA for this assay was followed by 1-factor ANOVA, looking at the effect of genotype in each data set separately. In both data sets, genotype had a significant effect on TFOL concentrations determined by BR (US, P = 0.0424, C/T different from T/T; Europe, P = 0.0081, C/C different from T/T) [see Fig. 2 inthe Data Supplement for scatter plots by genotype (panel B) and by data set (panel D)]. We therefore analyzed the data separately for C/C + C/T and T/T genotype. We also present data based on all genotypes. The BR and microbiologic assays showed a correlation of r = 0.87 for all genotypes, r = 0.94 for C/C + C/T, and r = 0.88 for T/T, but BR produced lower values, resulting in a significant negative concentration dependent difference: -45% for all genotypes (95% CI, -46%to -43%; 2SDlimits of agreement, -62% to -20% based on [log.sub.10]-transformed Bland Altman analysis); -48% for C/C + C/T (95% CI, -46% to -43%; 2 SD limits of agreement, -62% to -20%); -31% for T/T (95% CI, -35% to -26%; 2 SD limits of agreement, -52% to 0%). The multiple linear regression equations were as follows:
For all genotypes,
[log.sub.10] BR = 0.3479 + (0.7606 x [log.sub.10]
microbiologic assay) - (0.0875 x IND)
+ (0.0357 x IND = [log.sub.10] microbiologic assay),
SSR = 0.8953, PRESS = 0.9343 (Fig. 1B).
For C/C + C/T,
[log.sub.10] BR = 0.3040 + (0.7695 x [log.sub.10]
microbiologic assay) - (0.1389 x IND)
+ (0.0551 x IND = [log.sub.10] microbiologic assay),
SSR = 0.3879, PRESS = 0.4120 (Fig. 1C).
[log.sub.10] BR = -0.3160 + (1.0837 x [log.sub.10]
microbiologic assay) - (0.6236 x IND)
+ (0.1932 x IND x [log.sub.10] microbiologic assay),
SSR = 0.1457, PRESS = 0.1862 (Fig. 1D),
with IND = 0 for microbiologic assay results [less than or equal to]370 nmol/L and IND = 1 for microbiologic assay results >370 nmol/L.
Our recovery experiments with whole blood samples measured by microbiologic assay showed satisfactory recovery [mean (SD), n = 3 d] for 5CH3 THF [97% (11%)], 5CHOTHF [124% (7%)], and 5,10CH = THF [107% (13%)] and underrecovery for THF [46.4% (8%)]. These results compared well with our previous report of recoveries in serum samples (13). The BR showed satisfactory recovery (n = 2 days) for 5,10CH = THF [115% (10%)], but underrecovered 5C[H.sub.3]THF [51% (4%)] and 5CHOTHF [18% (0.1%)] and overrecovered THF [152% (19%)]. The underrecovery of 5C[H.sub.3]THF and 5CHOTHF, also seen in serum samples (13), could be a primary cause for the lower results obtained by BR. The shift in folate pattern in favor of non-5C[H.sub.3]THF in the T/T genotype appears to account for the higher responses of the BR for this genotype.
[FIGURE 1 OMITTED]
The current study investigated the influence of MTHFR C677T polymorphism on whole blood folate pattern and response by 3 different assays. As reported previously (1-5), our results confirm that individuals with the T/T genotype showed an accumulation of non-5C[H.sub.3]THF forms (approximately 30%-40%) in their whole blood. The novelty of this study is that it analyzed whole blood from a larger number of individuals with T/T genotype (n = 32) and that it considered data from the US, where folate fortification has been in place for almost a decade, and from Europe, where the folate status of the population is lower due to the absence of widespread fortification. Interestingly, the TFOL concentration by either LC-MS/MS or microbiologic assay was not affected by the genotype in either sample set. Previous reports showed lower erythrocyte folate concentrations in individuals with T/T vs C/C genotype, only when folate status was low (below the median of the studied population or excluding patients taking vitamins) (2, 13). Thus, our finding indicates that the folate status of the individuals in the European sample set might be adequate. We confirmed previous findings of high interindividual variability in non-5C[H.sub.3]THF accumulation in individuals with the T/T genotype (from <20% to 80% of TFOL) (4-5). We observed less interindividual variability in individuals with the C/C or C/T genotype for the European set than for the US set. Owing to the lack of a genotype effect for the LC-MS/MS and microbiologic assay, we were able to provide one regression equation that was applicable to a wide concentration range and that covers populations with and without folate fortification. However, we had to use separate equations for the BR assay depending on genotype. The equation for all genotypes will provide only a rough estimate in situations in which genotype information is not available.
Grant/funding Support: None declared.
Financial Disclosures: none declared.
Acknowledgments: We thank Drs. Anja Ruschel-Mitscherlich and Arnulf Hubbuch (Roche Diagnostic, Munich, Germany) for providing the whole blood samples from the European blood bank. We thank Margaret Gallagher and Stanimila Nikolova (CDC, Atlanta, Georgia) for oversight and technical assistance with the MTHFR genotyping. We thank Irene Williams and DonnaLaVoie(CDC,Atlanta, Georgia) and Neelima Paladugula (Battelle, Columbus, Ohio) for their technical assistance with the BR and LC-MS/MS methods.
(1.) Bagley PJ, Selhub J. A common mutation in the methylenetetrahydrofolate reductase gene is associated with an accumulation of formylated tetrahydrofolates in red blood cells. Proc Natl Acad Sci U S A 1998;95:13217-20.
(2.) Friso S, Choi S-W, Girelli D, Mason JB, Dolnikowski GG, Bagley PJ, et al. A common mutation in the 5,10-methylenetetrahydrofolate reductase gene affects genomic DNA methylation through an interaction with folate status. Proc Natl Acad Sci U S A 2002;99:5606-11.
(3.) Fazili Z, Pfeiffer CM. Measurement of folates in serum and conventionally prepared whole blood lysates: application of an automated 96-well plate isotope-dilution tandem mass spectrometry method. Clin Chem 2004;50: 2378- 81.
(4.) Davis SR, Quinlivan EP, Shelnutt KP, Maneval DR, Ghandour H, Capdevila A, et al. The methylenetetrahydrofolate reductase 677C>T polymorphism and dietary folate restriction affect plasma one-carbon metabolites and red blood cell folate concentrations and distribution in women. J Nutr 2005;135:1040-4.
(5.) Smulders YM, Smith DEC, Kok RK, Teerlink T, Gellekink H, Vaes WHJ, et al. Red blood cell folate vitamer distribution in healthy subjects is determined by the methylenetetrahydrofolate reductase C677T polymorphism and by the total folate status. J Nutr Biochem 2007;18:693-9.
(6.) Ueland PM, Hustad S, Schneede J, Refsum H, Vollset SE. Biological and clinical implications of the MTHFR C677T polymorphism. Trends Pharmacol Sci 2001;22:195-201.
(7.) Molloy AM, Mills JL, Kirke PN, Whitehead AS, Weir DG, Scott JM. Whole-blood folate values in subjects with different methylenetetrahydrofolate reductase genotypes: differences between the radio-assay and microbiological assays. Clin Chem 1998;44:186-8.
(8.) Pfeiffer CM, Fazili Z, McCoy L, Zhang M, Gunter EW. Determination of folate vitamers in human serum by stable-isotope-dilution tandem mass spectrometry and comparison with radioassay and microbiologic assay. Clin Chem 2004;50: 423-32.
(9.) Fazili Z, Pfeiffer CM, Zhang M, Jain R. Erythrocyte folate extraction and quantitative determination by liquid chromatography-tandem mass spectrometry: comparison of results with microbiologic assay. Clin Chem 2005;51:2318-25.
(10.) O'Broin S, Kelleher B. Microbiological assay on microtitre plates of folate in serum and red cells. J Clin Pathol 1992;45:344-7.
(11.) Molloy A, Scott JM. Microbiological assay for serum, plasma, and red cell folate using cryopreserved, microtiterplate method. Methods Enzymol 1997;281:43-53.
(12.) Frosst P, Blom HJ, Milos R, Goyette P, Sheppard CA, Matthews RG, et al. A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase. Nat Genet 1995;10:111-3.
(13.) Fazili Z, Pfeiffer CM, Zhang M. Comparison of serum folate species analyzed by liquid chromatography-tandem mass spectrometry with total folate measured by microbiologic assay and BioRad radioassay. Clin Chem 2007;53: 781- 4.
(14.) Icke GC, Dennis M, Sjollema S, Nicol DJ, Eikelboom JW. Red cell N5-methyltetrahydrofolate concentrations and C677T methylenetetrahydrofolate reductase genotype in patients with stroke. J Clin Pathol 2004;57:54-7.
 Nonstandard abbreviations: LC-MS/MS, liquid chromatography-tandem mass spectrometry; BR, Bio-Rad radioassay; 5C[H.sub.3]THF, 5-methyltetrahydrofolic acid; THF, tetrahydrofolic acid; TFOL, total folate; 5CHOTHF, 5-formyltetrahydrofolic acid; 5,10CH=THF, 5,10-methenyltetrahydrofolic acid; SSR, sum of squared residuals; PRESS, predicted residual sum of squares; FA, folic acid.
Zia Fazili, Christine M. Pfeiffer, * Mindy Zhang, Ram B. Jain, and Deborah Koontz
Division of Laboratory Sciences, National Center for Environmental Health, Centers for Disease Control and Prevention, Atlanta, Georgia; * address correspondence to this author at: Division of Laboratory Sciences, National Center for Environmental Health, Centers for Disease Control and Prevention, 4770 Buford Hwy, NE, Mail Stop F55, Atlanta, GA 30345. e-mail CPfeiffer@cdc.gov.
Table 1. Descriptive statistics for whole-blood folate concentrations measured by LC-MS/MS, microbiologic assay, and BR methods. (a) Folate species by LC-MS/MS Genotype n 5[CH.sub.3]THF 5CHOTHF US sample set All study participants 96 Concentration, nmol/L 304 (94.7-703) 41.4 (22.7-93.9) Ratio to TFOL, % 86 (22-92) (b) MTHFR 677 C/C 42 Concentration, nmol/L 317 (180-600) 38.6 (22.7-73.8) Ratio to TFOL, % 87 (71-91) MTHFR 677 C/T 40 Concentration, nmol/L 321 (175-703) 42.5 (25.7-93.9) Ratio to TFOL, % 86 (49-89) MTHFR 677 T/T 14 Concentration, nmol/L 248 (94.7-414) 49.8 (27.1-88.7) Ratio to TFOL, % 63 (22-87) European sample set All study participants 75 Concentration, nmol/L 207 (30.2-462) 29.2 (13.1-68.9) Ratio to TFOL, % 85 (23-91) MTHFR 677 C/C 31 Concentration, nmol/L 239 (120-462) 30.8 (16.5-64.4) Ratio to TFOL, % 88 (80-91) MTHFR 677 C/T 26 Concentration, nmol/L 205 (109-385) 28.8 (13.0-46.2) Ratio to TFOL, % 86 (78-91) MTHFR 677 T/T 18 Concentration, nmol/L 150 (30.2-334) 34.9 (18.5-68.9) Ratio to TFOL, % 57 (23-83) Folate species by LC-MS/MS Genotype n THF 5,10CH THF US sample set All study participants 96 Concentration, nmol/L 0 (0-142) 10.1 (0-212) Ratio to TFOL, % 14 (8.5-79) (c) MTHFR 677 C/C 42 Concentration, nmol/L 0 (0-68.4) 8.79 (0-23.5) Ratio to TFOL, % 13 (8.5-29) MTHFR 677 C/T 40 Concentration, nmol/L 0 (0-142) 10.8 (0-49.8) Ratio to TFOL, % 14 (11-51) MTHFR 677 T/T 14 Concentration, nmol/L 0 (0-68.7) 73.5 (9.00-212) Ratio to TFOL, % 37 (13-78) European sample set All study participants 75 Concentration, nmol/L 0 (0-23.5) 9.65 (0-167) Ratio to TFOL, % 15 (8.8-77) MTHFR 677 C/C 31 Concentration, nmol/L 0 (0-16.7) 7.68 (0-16.5) Ratio to TFOL, % 13 (8.8-20) MTHFR 677 C/T 26 Concentration, nmol/L 0 (0-17.8) 8.16 (0-18.4) Ratio to TFOL, % 14 (9.0-22) MTHFR 677 T/T 18 Concentration, nmol/L 0 (0-23.5) 69.3 (10.3-167) Ratio to TFOL, % 43 (17-77) Total folate Microbiologic Genotype n LC-MS/MS assay US sample set All study participants 96 Concentration, nmol/L 378 (228-820) 449 (240-1086) Ratio to TFOL, % MTHFR 677 C/C 42 Concentration, nmol/L 365 (243-698) 383 (263-915) Ratio to TFOL, % MTHFR 677 C/T 40 Concentration, nmol/L 383 (251-820) 448 (272-1086) Ratio to TFOL, % MTHFR 677 T/T 14 Concentration, nmol/L 368 (228-598) 412 (240-596) Ratio to TFOL, % European sample set All study participants 75 Concentration, nmol/L 250 (122-582) 242 (120-671) Ratio to TFOL, % MTHFR 677 C/C 31 Concentration, nmol/L 275 (136-543) 242 (158-522) Ratio to TFOL, % MTHFR 677 C/T 26 Concentration, nmol/L 238 (123-441) 236 (120-461) Ratio to TFOL, % MTHFR 677 T/T 18 Concentration, nmol/L 238 (122-582) 270 (138-671) Ratio to TFOL, % Total folate Genotype n BR US sample set All study participants 96 Concentration, nmol/L 230 (116-505) Ratio to TFOL, % MTHFR 677 C/C 42 Concentration, nmol/L 202 (121-382) Ratio to TFOL, % MTHFR 677 C/T 40 Concentration, nmol/L 220 (116-437) Ratio to TFOL, % MTHFR 677 T/T 14 Concentration, nmol/L 265 (171-505) Ratio to TFOL, % European sample set All study participants 75 Concentration, nmol/L 155 (66.6-434) Ratio to TFOL, % MTHFR 677 C/C 31 Concentration, nmol/L 155 (76.7-289) Ratio to TFOL, % MTHFR 677 C/T 26 Concentration, nmol/L 141 (66.6-229) Ratio to TFOL, % MTHFR 677 T/T 18 Concentration, nmol/L 203 (98.3-434) Ratio to TFOL, % Data are median (range). (a) Zero is recorded instead of the limit-of-detection (LOD) values, which are as follows: THF, 0.29 nmol/L hemolysate; 5,10CH THF, 0.55 nmol/L hemolysate. (b) Ratio between 5[CH.sub.3]THF and TFOL by LC-MS/MS. (c) Ratio between non-5[CH.sub.3]THF (5CHOTHF + THF + 5,10CH = THF) and TFOL by LC-MS/MS.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Brief Communications|
|Author:||Fazili, Zia; Pfeiffer, Christine M.; Zhang, Mindy; Jain, Ram B.; Koontz, Deborah|
|Date:||Jan 1, 2008|
|Previous Article:||Evaluation of a fully automated serum assay for total N-terminal propeptide of type I collagen in postmenopausal osteoporosis.|
|Next Article:||Adrenomedullin blood concentrations in infants subjected to cardiopulmonary bypass: correlation with monitoring parameters and prediction of poor...|