Methylenetetrahydrofolate reductase (MTHFR) polymorphism A1298C (Glu429Ala) predicts decline in renal function over time in the African-American Study of Kidney Disease and Hypertension (AASK) Trial and Veterans Affairs Hypertension Cohort (VAHC)

Abstract

Background: Hyperhomocysteinemia is associated with increased venous thrombosis and cardiovascular disease (CVD). Mutations in the human methylenetetrahydrofolate reductase (MTHFR) gene have been associated with increased homocysteine levels and risks of CVD in various populations including those with kidney disease. Here, we evaluated the influence of MTHFR variants on progressive loss of kidney function.

Methods: We analyzed 821 subjects with hypertensive nephrosclerosis from the longitudinal National Institute of Diabetes and Digestive and Kidney Diseases African-American Study of Kidney Disease and Hypertension (AASK) Trial to determine whether decline in glomerular filtration rate (GFR) over ∼4.2 years was predicted by common genetic variation within MTHFR at non-synonymous positions C677T (Ala222Val) and A1298C (Glu429Ala) or by MTHFR haplotypes. The effect on GFR decline was then supported by a study of 1333 subjects from the San Diego Veterans Affairs Hypertension Cohort (VAHC), followed over ∼4.5 years. Linear effect models were utilized to determine both genotype [single-nucleotide polymorphism (SNP)] and genotype (SNP)-by-time interactions.

Results: In AASK, the polymorphism at A1298C predicted the rate of GFR decline: A1298/A1298 major allele homozygosity resulted in a less pronounced decline of GFR, with a significant SNP-by-time interaction. An independent follow-up study in the San Diego VAHC subjects supports that A1298/A1298 homozygotes have the greatest estimated GFR throughout the study. Haplotype analysis with C677T yielded concurring results.

Conclusion: We conclude that the MTHFR-coding polymorphism at A1298C is associated with renal decline in African-Americans with hypertensive nephrosclerosis and is supported by a veteran cohort with a primary care diagnosis of hypertension. Further investigation is needed to confirm such findings and to determine what molecular mechanism may contribute to this association.

Fig. 1.

MTHFR polymorphism A1298C and haplotype with C677T predict decline in renal function over time in the AASK Trial. GFR decline (mL/min/1.73 m²/year) is plotted on the Y-axis and time in weeks from start of study is plotted on the X-axis. (a) SNP (genotype). The influence of MTHFR polymorphism A1298C on longitudinal GFR decline is shown. Adjusted GFR values from the joint analysis model, which combines longitudinal and time to event analysis are plotted using an adaptive regression cubic spline with 95% confidence intervals (dashed lines) for fitted curves. AASK subjects that were homozygous for the wild-type allele (A1298/A1298, light gray line) had the slowest decline in renal function (measured by iothalamate clearances over ∼4.2 years). Heterozygotes (dark gray line) and homozygotes for the minor allele (1298C/1298C, black line) had greater declines, showing the significant SNP-by-time interaction, P = 0.023 (main SNP effect P = 0.65), with adjustment for age, sex, drug and BP goal randomization groups. Below the graph indicates the number of GFR measurements that are included by week at each time point. (b) Haplotype. The influence of MTHFR haplotype 677C/1298C on longitudinal GFR decline are plotted for the AASK Study. Adjusted GFR values from the linear mixed effects model are plotted using an adaptive regression cubic spline with 95% confidence intervals (dashed lines) for fitted curves. Haplotype 677C/1298C had a haplotype-by-time interaction on GFR (by iothalamate clearance) as determined by linear mixed model analysis, such that those with at least one copy (dark gray line) had accelerated decline (P = 0.013; main haplotype effect P = 0.36) compared to those with no copies of the haplotype (solid black line).

Fig. 2.

MTHFR polymorphism A1298C and haplotype with C677T was associated with renal function in the San Diego VAHC. GFR decline (mL/min/1.73 m²/year as estimated by the MDRD equation) is plotted on the Y-axis and time in weeks from start of study is plotted on the X-axis. (a) SNP (genotype). The influence of MTHFR polymorphism A1298C on longitudinal GFR decline is plotted for the VAHC Study. Adjusted GFR values from the joint model which combines longitudinal and time to event analysis are plotted using an adaptive regression cubic spline with 95% confidence intervals (dashed lines) for fitted curves. The mean curve for A1298/A1298 (solid black line), A1298/1298C (dark gray line) and 1298C/1298C (light gray line) are plotted. Subjects who were homozygous for the wild-type allele (A1298/A1298, light gray line) had the greatest eGFR. The difference in eGFR was statistically significant P = 0.0041 for the main SNP effect (SNP-by-time interaction, P = 0.58), adjusted for age, sex and ethnicity. Adjusting for possible confounding due to ethnic admixture using a principal components analysis yielded similar results. Below the graph indicates the number of subjects per week at each time point. (b) Haplotype. The influence of MTHFR haplotype 677C/1298C on longitudinal eGFR decline is plotted for the VAHC Study. Adjusted eGFR values from the linear mixed effects model are plotted using an adaptive regression cubic spline with 95% confidence intervals (dashed lines) for fitted curves. Subjects with at least one copy of the haplotype 677C/1298C (dark gray line) had worse renal function as determined by linear mixed model analysis compared to those with 0 copies (solid black line), main haplotype effect P = 0.047 adjusted for sex, ethnicity and baseline age (haplotype-by-time interaction P = 0.25).