Hobbs, Charlotte A. MD, PhD; Cleves, Mario A. PhD; Karim, Mohammad A. MD, PhD; Zhao, Weizhi MS; MacLeod, Stewart L. PhD
Congenital heart defects are the most common birth defects.1 In the United States, approximately 8 to 10 of 1,000 live births are affected by a congenital heart defect (CHD). The etiology of most nonsyndromic CHDs is unknown; the majority are thought to be caused by a complex interaction of multiple factors including genetic and lifestyle factors that alter folate metabolism.2
Adequate intake of folic acid in the periconceptional period has been shown to be protective against birth defects, including CHDs.3 Genetic polymorphisms in genes that code for key enzymes in the folate pathway may reduce enzyme activity, leading to disruptions in folate metabolism. A number of functional single-nucleotide polymorphisms in critical genes in the folate pathway that affect folate metabolism have been characterized including 677C>T (rs1801133) in the methylenetetrahydrofolate reductase gene (MTHFR, NM005957), 742G>A (rs3733890) in the betaine homocysteine methyltransferase gene (BHMT, NM0011713), and 776C>G (rs1801198) in the transcobalamin II gene (TCII, NM000355).4–6
Both fetal and maternal genetic susceptibilities affect the intrauterine environment during the first 8 weeks of pregnancy, when the primitive heart is forming and developing. Genetic factors, in conjunction with lifestyle factors such as obesity, cigarette smoking, and ethanol intake, may work in combination to increase the risk of CHDs. In this study, we hypothesized that three functional single-nucleotide polymorphisms in maternal folate-related genes will interact with maternal lifestyle factors to increase the risk of CHDs.
MATERIALS AND METHODS
The protocol and provisions for informed consent were reviewed and approved by the institutional review board at the University of Arkansas for Medical Sciences. Those in the case group were mothers of affected infants and were identified and ascertained through the Arkansas Reproductive Health Monitoring System, a statewide birth defects registry. Inclusion criteria for those in the case group were as follows: 1) resident of Arkansas at the time of the completion of the index pregnancy and at the time of enrollment in the study; 2) pregnancy outcome was a liveborn neonate, a stillborn fetus, or elective termination; 3) pregnancy ended between September 1998 and September 2006; 4) a physician diagnosis of a nonsyndromic septal, conotruncal, or right-sided or left-sided obstructive heart defect that had been confirmed by prenatal or postnatal echocardiogram, surgery, autopsy report, or any combination of the three; and 5) the individual had completed participation in the National Birth Defects Prevention Study.7 Cases in which the pregnancy also was affected by a known single gene disorder, chromosomal abnormality, or syndrome were excluded; only nonsyndromic CHDs were included in this study. Further details regarding the National Birth Defects Prevention Study have been previously published.7
Those in the control group were Arkansas mothers who had live births that were unaffected by any birth defect, were randomly chosen from all birth certificates registered at the Arkansas Department of Health with birth dates between December 1998 and April 2007, and had completed participation in the National Birth Defects Prevention Study. Mothers in the case and control groups with type 1 or type 2 diabetes mellitus at the time of conception were excluded from analyses. By design, the National Birth Defects Prevention Study protocol has a case:control ratio of 3:1.
Women in both the case and control groups were interviewed using a structured computer assisted telephone interview that was specifically designed for the National Birth Defects Prevention Study. In this questionnaire, the respondent was asked about smoking and drinking alcoholic beverages before and after conception, and was also asked during which months. Similarly, questions were also asked regarding the use of multivitamins, the educational level of each participant, and household income.
DNA was extracted from maternal buccal cell samples or peripheral blood lymphocytes by use of Pure Gene DNA purification reagents (Gentra Systems, Minneapolis, MN) according to the manufacturer's protocol. Genomic DNA was quantified by use of Applied Biosystems' TaqMan RNaseP Detection Reagents (Applied Biosystems by Life Technologies, Carlsbad, CA) using a standard curve of genomic DNA. The standard curve samples as well as the genomic DNA samples of unknown concentration were subjected to an initial denaturalization at 95°C for 10 minutes followed by 40 polymerase chain reaction (PCR) cycles (95°C for 15 seconds, 60°C for 1 minute) in an ABI PRISM 7900HT real-time PCR instrument (Applied Biosystems). DNA concentrations were calculated from the standard curve using ABI software. Genomic DNA (10 to 15 ng) was used as a template for whole-genome amplification (WGA) by use of the Sigma GenomePlex WGA kit and a protocol provided by the manufacturer. Whole-genome amplification product was quantified as above and 10 ng was used for each genotyping assay. Polymerase chain reaction of polymorphic alleles was accomplished by the use of TaqMan Assays (ABI) on an Applied Biosystems 7900 Sequence Detection instrument for the following polymorphisms: BHMT NM001713 rs3733890; MTHFR NM005957 rs1801133; TCII NM 000355 rs1801198. Polymerase chain reaction conditions were 95°C for 10 minutes followed by 40 cycles of 95°C for 15 seconds, then, 60°C for 1 minute. After PCR, an allelic discrimination read was performed using ABI software (Applied Biosystems).
The four periconceptional lifestyle factors (body mass index [BMI, calculated as weight (kg)/[height (m)]2).], alcohol use, smoking, and folic acid supplementation) were summarized by counts and proportions and were compared using χ2 tests if categorical, and were summarized by means and standard deviations and compared using t tests if continuous. Deviations from Hardy-Weinberg equilibrium in the mothers in the control group stratified by race or ethnicity were tested using an exact Hardy-Weinberg equilibrium test implemented in STATA's GENHW command.8 Odds ratios (ORs) and 95% confidence intervals (CIs) for the association between the occurrence of CHD and allele and genotypes frequencies for each of three single-nucleotide polymorphisms under various genetic models were estimated using linear logistic regression adjusting for maternal race or ethnicity. These associations were further examined using logistic regression models after combining lifestyle factors and genotype and also stratified on lifestyle factor levels. These logistic regression models were adjusted for race or ethnicity and all lifestyle factors.
All eligible women in the case and control groups who met entry criteria were included in the study. With a sample size of 572 women in the case group and 363 in the control group, assuming an additive genetic model, a minor allele frequency of 35%, a lifestyle factor that is present in 25% of the population, conservative ORs of 1.2 for the main effect of the genetic and environmental factor and a two-sided test with α=0.05, the study has over 80% power to detect a minimal OR of 1.85 for a one-way interaction between genotype and lifestyle factor. Statistical analyses were performed using STATA 10.1 software (StataCorp LP, College Station, TX).
As presented in Table 1, there were a total of 572 women in the case group and 363 in the control group included in the current study. The mean maternal age at time of delivery was 26.7 years for those in the case group and 26.4 years for those in the control group (P=.48). Compared with women with normal BMIs, significantly more women in the case group than in the control group were obese (BMI of 30 or more) before pregnancy (26.7% and 21.5%, respectively; P=.02). There were no statistically significant differences between women in the case group and those in the control group for periconceptional alcohol drinking, smoking, or intake of folate supplements. Neither household income nor educational level differed significantly between those in the case and control groups.
No significant deviations from Hardy-Weinberg equilibrium were observed for any single-nucleotide polymorphism within race. There were no significant differences in the distribution of MTHFR 677C>T, BHMT 742G>A, or the TCII 776C>G polymorphisms between women with CHD-affected pregnancies and those with unaffected pregnancies (Table 2). We evaluated allelic, genomic, additive, dominant, and recessive inheritance models; none of the maternal genotypes were associated with case–control status.
Multiple logistic regression models were used to estimate whether the effect of genotypes was modified by lifestyle factors (Tables 3 and 4) or periconceptional folate intake. The reference group for Table 3 was the major allele homozygous genotype (MTHFR 677 CC, BHMT GG, and TCII CC) and the most favorable lifestyle factor. The effect of maternal obesity on the developing heart was increased among those carrying the MTHFR 677 TT genotype and those carrying either one or two copies of the A allele in the BHMT 742 polymorphism. Specifically, obese women who carried two copies of the MTHFR 677 T allele were 4.62 times (95% CI 1.54–13.83) more likely to have a CHD-affected pregnancy than women who were normal weight and had no copies of the T allele (Table 3). Similarly, women who had a BMI of 30 or more and carried one or two copies of the A allele in the BHMT polymorphism were 1.82 times (95% CI 1.09–3.03) more likely to have a CHD-affected pregnancy than women who were normal weight and had no copies of the G allele (data not shown). The maternal TCII polymorphism did not alter the effect of obesity on the risk of CHDs.
To further understand the association between CHDs and the combined effect of genetic polymorphisms and lifestyle factors, we examined the effect of functional polymorphisms within each stratum of lifestyle factors using as reference groups individuals with the same lifestyle factor who were homozygous for the major allele of each functional polymorphism (Table 4). In this way, we were able to estimate the effect of genotype among those with similar lifestyle choices. For example, among women who were obese, those who carried two copies of the T allele in the MTHFR polymorphism were 3.93 times (95% CI 1.24–12.50) more likely to have a CHD-affected pregnancy than women who were obese but carried a CC genotype. Among women who smoked, those who carried one or more copies of the G allele in the TCII polymorphism were 1.81 times (95% CI 1.06–3.11) more likely to have a CHD-affected pregnancy than women who smoked and carried the CC genotype. Similarly, among women who drank alcohol, those who carried the GC or GG genotype in the TCII polymorphism were 1.71 times (95% CI 1.00–2.92) more likely to have a CHD-affected pregnancy than women who drank alcohol but carried the CC genotype. There was no significant evidence that periconceptional folate supplement use modified the association between CHDs and any of the three candidate polymorphisms.
In this case–control study, we evaluated the effect of three folate-related functional polymorphisms combined with maternal periconceptional obesity, smoking, alcohol intake, and folate supplement use on the risk of CHDs. Our intent was to determine whether the association between CHDs and functional polymorphisms in folate-related genes is modified by lifestyle factors that are known to impair folate metabolism. Several findings should be noted.
Maternal genotype was not independently associated with CHD risk for any of the functional polymorphisms. Multiple previous studies have examined the association between CHDs and MTHFR 677C>T.9–12 Our inability to find evidence of an association with the 677C>T polymorphism is in agreement with a recent meta-analysis.13 Among six eligible studies included in the meta-analysis, the overall OR for the association between CHDs and MTHFR 677C>T was 1.2 (95% CI 0.83–1.74). As with previous studies, we were unable to find evidence that the BHMT 742G>A or the TCII 776C>G polymorphisms were associated with CHDs.11,14
Our findings indicate that periconceptional maternal obesity, smoking, and alcohol use combined with functional polymorphisms may increase the risk of CHD-affected pregnancies. The evidence supporting an association between obesity and CHDs has been included in a recently published meta-analysis by Stothard and colleagues.15 Among the 18 studies included in the meta-analysis, the overall OR for the association between CHDs and maternal obesity was 1.30 (95% CI 1.12–1.51). Among our population, the OR associated with maternal obesity, independent of underlying genetic susceptibilities, was 1.48 (95% CI 1.05–2.09), which was consistent with Stothard et al's meta-analysis (data not shown). The effect of maternal obesity was markedly accentuated among women who carried either the MTHFR 677C>T polymorphism or the BHMT 742G>A, with the risk being highest among those who carried the MTHFR TT genotype.
There have been multiple epidemiologic studies examining the association between CHDs and maternal smoking and alcohol intake with inconclusive results.16–18 We found evidence suggesting that the effect of periconceptional smoking and alcohol intake may depend on underlying genetic susceptibilities. Specifically, the TCII 776 CG and GG genotypes when combined with either smoking or alcohol intake increased the risk of CHD-affected pregnancies. This association may be related to lower levels of folate and vitamin B12 in individuals who smoke or drink, coupled with the less efficient transport of B12 into cells and tissue due to the phenotype caused by the TCII 776C>G polymorphism.
The current study has several limitations. Only maternal genetic susceptibilities were considered. The genetics of cardiac development is likely a complex interplay between both maternal and fetal genetic susceptibilities. Sample size limitations prevented us from examining individual cardiac phenotypes. It is possible that maternal genetic susceptibilities and lifestyle factors interrupt specific cell types and developmental stages of cardiogenesis. However, it is equally possible that maternal factors that are present throughout the 8 weeks of cardiogenesis are just as likely to affect all stages of cardiogenesis, and thus the inclusion of all cardiac phenotypes is justified. Our study was limited to the investigation of only three folate-related polymorphisms, and due to sample size limitations we were unable to examine gene-gene interactions. Information regarding lifestyle factors was obtained through maternal report. We cannot rule out misclassification or residual confounding and interactions with other genetic polymorphisms, environmental and lifestyle factors, or a combination of these. We were also unable to include information about family history of CHDs or previously affected pregnancies.
In conclusion, the detrimental effect of maternal obesity on the developing heart appears to be greatest among women who carry the MTHFR 677C>T or BHMT 742G>A polymorphism. Maternal smoking or alcohol use or both in combination with the TCII 776C>G polymorphism may increase the risk of CHDs. The current study provides further evidence that the etiology of nonsyndromic CHDs is complex, requiring both genetic susceptibilities and environmental and lifestyle factors to be investigated. Future studies to examine the association between CHDs and maternal and fetal genetic susceptibilities in folate-related genes in association with maternal lifestyle factors are essential to understanding the effect of one-carbon metabolism on the development of CHDs. An improved understanding of CHD etiology will provide evidence upon which personalized primary prevention programs can be built that target the reduction of CHD occurrence. Risk stratification of reproductive-aged women based on knowledge of the impact of genetic susceptibilities, lifestyle factors, and their interaction may play an important role in minimizing adverse pregnancy outcomes, including heart defects and other structural malformations.
Future research studies investigating the association between CHDs and gene-environment interactions may expand the number of common variants by including haplotype tagging single-nucleotide polymorphisms in candidate pathways, genome-wide association studies, or both. Our study was limited to investigating the association between maternal lifestyle factors and maternal genotypes. It is likely that fetal genotypes will also be very important and further evaluation of gene-gene interactions (epistasis) are also needed.
1. Christianson A, Howson C, Modell B. March of Dimes global report of birth defects: the hidden toll of dying and disabled children. White Plains (NY): March of Dimes Birth Defects Foundation; 2006.
2. Botto LD, Mulinare J, Erickson JD. Do multivitamin or folic acid supplements reduce the risk for congenital heart defects? evidence and gaps. Am J Med Genet A 2003;121A:95–101.
3. Jenkins KJ, Correa A, Feinstein JA, Botto L, Britt AE, Daniels SR, et al. Noninherited risk factors and congenital cardiovascular defects: current knowledge: a scientific statement from the American Heart Association Council on Cardiovascular Disease in the Young: endorsed by the American Academy of Pediatrics. Circulation 2007;115:2995–3014.
4. Afman LA, Lievers KJ, van der Put NM, Trijbels FJ, Blom HJ. Single nucleotide polymorphisms in the transcobalamin gene: relationship with transcobalamin concentrations and risk for neural tube defects. Eur J Hum Genet 2002;10:433–8.
5. 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.
6. Li F, Feng Q, Lee C, Wang S, Pelleymounter LL, Moon I, et al. Human betaine-homocysteine methyltransferase (BHMT) and BHMT2: common gene sequence variation and functional characterization. Mol Genet Metab 2008;94:326–35.
7. Yoon PW, Rasmussen SA, Lynberg MC, Moore CA, Anderka M, Carmichael SL, et al. The National Birth Defects Prevention Study. Public Health Rep 2001;116:32–40.
8. Cleves MA. Hardy-Weinberg equilibrium tests and allele frequency estimation. STATA Technical Bulletin. 1999;48:34–7.
9. Goldmuntz E, Woyciechowski S, Renstrom D, Lupo PJ, Mitchell LE. Variants of folate metabolism genes and the risk of conotruncal cardiac defects. Circ Cardiovasc Genet 2008;1:126–32.
10. Hobbs CA, James SJ, Parsian A, Krakowiak PA, Jernigan S, Greenhaw JJ, et al. Congenital heart defects and genetic variants in the methylenetetrahydrofolate reductase gene. J Med Genet 2006;43:162–6.
11. Shaw GM, Lu W, Zhu H, Yang W, Briggs FB, Carmichael SL, et al. 118 SNPs of folate-related genes and risks of spina bifida and conotruncal heart defects. BMC Med Genet 2009;10:49.
12. van Driel LM, Verkleij-Hagoort AC, de Jonge R, Uitterlinden AG, Steegers EA, van Duijn CM, et al. Two MTHFR polymorphisms, maternal B-vitamin intake, and CHDs. Birth Defects Res A Clin Mol Teratol 2008;82:474–81.
13. van Beynum IM, den Heijer M, Blom HJ, Kapusta L. The MTHFR 677C->T polymorphism and the risk of congenital heart defects: a literature review and meta-analysis. QJM 2007;100:743–53.
14. Verkleij-Hagoort AC, van Driel LM, Lindemans J, Isaacs A, Steegers EA, Helbing WA, et al. Genetic and lifestyle factors related to the periconception vitamin B12 status and congenital heart defects: a Dutch case-control study. Mol Genet Metab 2008;94:112–9.
15. Stothard KJ, Tennant PW, Bell R, Rankin J. Maternal overweight and obesity and the risk of congenital anomalies: a systematic review and meta-analysis. JAMA 2009;301:636–50.
16. Grewal J, Carmichael SL, Ma C, Lammer EJ, Shaw GM. Maternal periconceptional smoking and alcohol consumption and risk for select congenital anomalies. Birth Defects Res A Clin Mol Teratol 2008;82:519–26.
17. Källén K. Maternal smoking and congenital heart defects. Eur J Epidemiol 1999;15:731–7.
18. Malik S, Cleves MA, Honein MA, Romitti PA, Botto LD, Yang S, et al. Maternal smoking and congenital heart defects. Pediatrics 2008;121:e810–6.
© 2010 by The American College of Obstetricians and Gynecologists.