International Journal of Cardiology 105 (2005) 15 18

www.elsevier.com/locate/ijcard

Lack of evidence of association between MTHFR C677T polymorphism

and congenital heart disease in a TDT study design
Alexandre C. PereiraT, Jose Xavier-Neto, Sonia M. Mesquita, Gloria F.A. Mota,
Antonio Augusto Lopes, Jose Eduardo Krieger
Laboratory of Genetics and Molecular Cardiology, Heart Institute, Sao Paulo University Medical School, Sao Paulo, Brazil
Ambulatory of Pediatric Cardiology, Heart Insitute, Sao Paulo University Medical School, Sao Paulo, Brazil

Received 19 July 2004; received in revised form 22 September 2004; accepted 2 October 2004
Available online 26 February 2005

Abstract

Introduction: Hyperhomocysteinemia is frequently associated with congenital defects of the heart and neural tube. A common missense
mutation in the MTHFR gene (C to T substitution at position 677 changing valine to alanine) produces a variant with reduced enzymatic
action, resulting in higher plasma levels of homocysteine. The aim of this study is to investigate whether MTHFR C677T functional genetic
variant is associated with an increased risk of congenital heart disease (CHD) development using a family-based case-control design and the
Transmission Disequilibrium Test (TDT) approach.
Methods: We selected 91 consecutive patients with congenital heart disease for the study. From these patients we were able to obtain samples
on 147 parents. The C677T polymorphism at the MTHFR gene was determined from each participant.
Results: A statistically significant association was disclosed in univariate analysis using a family-based case-control design ( pb0.0001
assuming an additive genetic model, pb0.0001 assuming a dominant genetic model, and p=0.01 assuming a recessive genetic model). This
association was explained by an increased frequency of the T allele in patients as compared to their fathers. However, by using a TDT
approach a highly non-significant result was obtained and no association could be defined between this locus and congenital heart disease.
Conclusions: We did not find sufficient evidence for an association between MTHFR C677T genotype and congenital heart disease in our
study group. Previous reports on such association may be due to population genetic structure.
D 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Genetics; Congenital heart disease; Homocysteine; MTHFR

1. Introduction An abnormality in homocysteine metabolism, appa-

Hyperhomocysteinemia is frequently associated with
congenital defects of the heart and neural tube and is a
suspected factor in atherosclerosis and neoplasia [1,2]. In
addition to association reports, a causal link has been
suggested by showing a crucial role of perturbed homo-
cysteine/methionine metabolism in the etiology of congen-
ital cardiovascular malformation, neural tube defects, and
cardiovascular disease [3].
T Corresponding author. Laboratory of Genetics and Molecular Cardio-
logy, Heart Institute, Sao Paulo University Medical School, Sao Paulo,
Brazil.
E-mail address: lbmpereira@incor.usp.br (A.C. Pereira).
0167-5273/$ - see front matter D 2005 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.ijcard.2004.10.049
rently related to methionine synthase, is present in many
women who give birth to children with neural tube
defects [4]. Conversely, folic acid (FA) therapy has been
shown to reduce the incidence of neural tube, cranio-
facial, and cardiovascular defects and low birth weight
[3].
Interestingly, despite several epidemiological and exper-
imental evidences linking homocysteine/folate pathways
and congenital heart malformations, approximately 50% of
cystathionine beta-synthase deficient expectant mothers
with high plasma homocysteine levels (100 mM) have
given birth to infants with no apparent anomalies, indicating
that there are individuals who are less sensitive to the
toxicity of homocysteine than others [5].
16 A.C. Pereira et al. / International Journal of Cardiology 105 (2005) 1518

Several known conditions can cause hyperhomocysteine-
mia: genetic disorders, folic acid deficiency, vitamin B12
deficiency, vitamin B6 deficiency, renal failure, hypothyr-
oidism, increasing age and smoking. In particular, a common
missense mutation (C to T substitution at position 677
changing a valine to alanine) produces a thermolabile variant
of MTHFR with reduced enzymatic action resulting in higher
plasma levels of homocysteine. Individuals with the homo-
zygous T/T genotype have mildly elevated plasma homo-
cysteine levels and lower folate levels than individuals who
have either the heterozygous C/T or wild-type C/C genotype.
We recently showed that the Brazilian population has a
high frequency of mild folate deficiency [6]. In this context,
MTHFR genotypes may show an increased penetrance
regarding congenital heart disease (CHD) incidence, as has
been shown for hyperhomocysteinemia [6]. In addition, we
and others have suggested that homocysteine reduction after
folate supplementation is predicted to be partially modulated
by MTHFR genotype [6,7]. Indeed, the possibility that
genotype-specific periconceptional use of folate supplemen-
tation reduces the rate of specific structural CHD has been
proposed as a potential clinical use of this genetic marker.
This hypothesis has been tested by some groups with
discordant findings [11,14,15].
Genetic association testing is typically carried out as a
case-control comparison in which marker frequencies are
compared in samples of affected individuals and unaffected
controls. Classically, case-control studies can be confounded
by population stratification. Genetic association studies are
no exception to this difficulty, and various strategies have
been devised to control for stratification. An example is the
Transmission Disequilibrium Test (TDT), which tests for
association by defining whether the genotype transmitted by
a parent to an affected offspring is concordant, or discordant,
with the expected frequency based on Mendelian segregation.
This ensures that case and control genotypes come from the
same population, reducing the potential for stratification [8].
The aim of this study is to investigate a possible
association between the MTHFR C677T functional genetic
variant and an increased risk of CHD development using both
a family-based case-control design and the TDT approach.
and their parents were also invited to participate in the study.
We were able to obtain DNA samples from 147 parents
from the selected patients.
Venous blood was obtained for genomic DNA extraction
from each participant of the study.
The protocol was approved by the Institutional Review
Board of the University of Sao Paulo Medical School and all
participants read and signed an approved informed consent.
2.2. Genetic analysis
DNA was extracted from peripheral blood leukocytes by
a standard salting-out procedure.
Analysis of the C677T polymorphism at the MTHFR
gene was carried out by polymerase chain reaction (PCR).
The C677T polymorphism was determined as previously
described [9]. Two primers for analysis of the CT change
are: 5VTGA AGG AGA AGG TGT CTG CGG GA-3V and
5VAGG ACG GTG CGG TGA GAG TG-3V. These primers
generate a fragment of 198 bp. The substitution creates a
HinfI recognition sequence that digests the 198-bp fragment
into 175- and 23-bp fragments.
2.3. Statistical analysis
Allele and genotype frequencies were calculated through
gene counting. HardyWeinberg equilibrium was tested
with the chi-square test. The association between the
MTHFR C677T polymorphism and disease was carried
out first by comparing genotype frequencies from affected
siblings and their parents through the chi-square test.
Finally, the same association was also tested in a more
robust manner by examining the transmission of markers
from parents to affected offspring using the TRANSMIT
algorithm [10]. p values less than 0.05 on a 2-sided test
were considered significant.
3. Results
We have genotyped 238 individuals, 91 being patients
with congenital heart disease and 147 parents of these

2. Materials and methods

Table 1
2.1. Study population Number and relative frequency of anatomical diagnosis
Number %
This is an observational cross-sectional study designed to Atrium septal defect (ASD) 2 2.2
ascertain different genetic variables associated with con- Ventricular septal defect (VSD) 5 5.5
genital heart disease. Study participants were recruited from ASD + VSD 1 1.1
Aortic coarctation 1 1.1
a tertiary care center specialized in the care of patients with
Atriumventricular septal defect 2 2.2
congenital heart disease. Aortic stenosis 1 1.1
We have selected 91 consecutive patients with congenital Ebstein anomaly 20 22.0
heart disease seen at the Heart Institute Ambulatory of Pulmonary stenosis 1 1.1
Pediatric Cardiology, Sao Paulo, Brazil. Patients were Patent ductus arteriosus 1 1.1
Tetralogy of Fallot 57 62.6
selected based on the diagnosis of congenital heart disease
 A.C. Pereira et al. / International Journal of Cardiology 105 (2005) 1518
Table 2
Alele and genotype frequency in patients and parents
Patient Mother
Alele
C67 69.0 65.3
T67 31.0 34.7
Genotype
CC 47.8 43.5
CT 42.4 43.5
TT 9.8 12.9
patients. The number and relative frequency of the
anatomical diagnosis are shown in Table 1.
Allele and genotype frequencies are shown in Table 2.
Allele and genotype frequencies were in HardyWeinberg
equilibrium. Interestingly, by comparing father genotype
frequencies with patient genotype frequencies, a statistically
significant association was disclosed ( pb0.0001 assuming
an additive genetic model, pb0.0001 assuming a dominant
genetic model, and p=0.01 assuming a recessive genetic
model). This association was explained by an increased
frequency of the T allele in patients as compared to their
fathers.
The possibility that an increased frequency of the T
allele, associated with increased levels of homocysteine and
lower levels of folate in the Brazilian population [6], is
associated with congenital heart disease in our population is
tempting and biologically plausible. Nevertheless, we have
previously shown that allele and genotype frequencies in
this genetic locus are highly dependent on ethnicity in the
Brazilian population. Therefore, a positive result in uni-
variate analysis, even in a family-based case-control study,
could be the result of population stratification and an
unequal ethnic distribution.
We have decided to better explore this potential bias
through the use of a Transmission Disequilibrium Test.
Through the use of the TRANSMIT algorithm we analyzed
the 285 persons in the study following a case-parents
design. Following this design, there were 91 nuclear
families. The resultant global chi-squared test, on 1 degree
of freedom was 0.11235, a non-significant result.
4. Discussion
Several reports have described a positive association
between homocysteine metabolism and neural tube defects
(NTD) [1]. Because neural crest cells contribute to the
outflow septation of the heart, one might assume that the
same hampered homocysteine metabolism may play the
same role in CHD [5].
Recently, Scanlon et al. reported that total maternal folate
intake of z245 Ag was inversely related to risk of cardiac
outflow tract defects among those with transposition of the
great arteries, but positively related among those with
normally related vessels. This difference disappeared when
17
maternal intake of supplemental folic acid of z400 Ag
compared with b400 Ag was considered, excluding dietary
Father intake [11].
Indeed, several lines of evidence have been explored to
77.3 explain the cellular pathways linking homocysteine to
22.7
development defects. For instance, it has been suggested
that homocysteine-induced congenital defects are due to the
62.5 specific ability of homocysteine to inhibit conversion of
29.7 retinal to retinoic acid [12]. Another possible mechanism
7.8 already explored is related to a maternal-to-fetal folate-
transport defect or an inherent fetal biochemical disorder
that is neutralized by supplementation. In this scenario, it
has been shown that mice lacking the folic acid-binding
protein Folbp1 are defective in early embryonic develop-
ment [13].
In addition, several reports have tried to associate
polymorphisms in the 5,10-methylenetetrahydrofolate
reductase gene (MTHFR) to folate-dependent neural tube
defects. Results, however, have been discordant regarding
both association and methodological aspects.
Junker et al. were the first to search for a possible
association between congenital heart disease incidence and
MTHFR C677T polymorphism. Through the use of a case-
control design they found a significant association between
the presence of the TT genotype and pulmonary valve
stenosis, hypoplastic left heart syndrome, coarctation of the
aorta, aortic valve stenosis or subaortic stenosis [14].
Storti et al. genotyped 103 Italian mothers with
conotruncal defects offspring, 200 control mothers, 103
affected children and their fathers in the only study to date
that has used family control data to assort a possible
association between this genetic marker and congenital heart
disease. Like our data, no increased risk was observed for
the prevalence of the 677TT genotype by itself in affected
children and in their mothers [15].
At least two possibilities may explain the discordant
findings between these studies. Firstly, population genetic
structure in this locus has been reported in several
populations worldwide, including ours [6]. The existence
of population genetic structure may lead to false positive
genetic association studies due to unbalanced distribution
between cases and controls. The use of the TRANSMIT
algorithm, through its family-based TDT design, is a nice
alternative to this caveat since it can adjust for population
genetic structure, especially in admixed populations. In
addition, it is also possible that this genetic variant may be
only associated with a particular type of anatomical
diagnoses. In fact, different studies have used patients with
very different sets of anatomical diagnosis under the
congenital heart disease diagnose. For instance, according
to Junker et al. a significant association was found in
patients with pulmonic valve stenosis, hypoplastic left heart
syndrome, coarctation of the aorta, and aortic valve stenosis,
but these diagnoses are the very minority of our patients.
In conclusion, we did not find any association between
MTHFR C677T genotype and congenital heart disease in
18 A.C. Pereira et al. / International Journal of Cardiology 105 (2005) 1518
our study group. Previous reports on such association may
be due to population genetic structure or different
anatomical diagnoses used in different studies. Future
association reports on this locus should not only use study
designs that correct for population genetic structure, but
also a study population with a homogeneous set of
anatomical diagnoses.
References
[1] Molloy AM, Weir DG, Scott JM. Homocysteine, folate enzymes and
neural tube defects. Haematologica 1999;84:53 6.
[2] Scott CH, Sutton MS. Homocysteine: evidence for a causal relation-
ship with cardiovascular disease. Cardiol Rev 1999;7:101 7.
[3] Burgoon JM, Selhub J, Nadeau M, Sadler TW. Investigation of the
effects of folate deficiency on embryonic development through the
establishment of a folate deficient mouse model. Teratology 2002;
65(5):219 27.
[4] Mills JL, McPartlin JM, Kirke PN, Lee YJ, Conley MR, Weir DG,
et al. Homocysteine metabolism in pregnancies complicated by
neural-tube defects. Lancet 1995;345(8943):149 51.
[5] Kapusta L, Haagmans ML, Steegers EA, Cuypers MH, Blom HJ,
Eskes TK. Congenital heart defects and maternal derangement of
homocysteine metabolism. J Pediatr 1999;135(6):773 4.
[6] Pereira AC, Schettert IT, Morandini Filho AA, Guerra-Shinohara EM,
Krieger JE. Methylenetetrahydrofolate reductase (MTHFR) c677t
gene variant modulates the homocysteine folate correlation in a mild
folate-deficient population. Clin Chim Acta 2004;340(1-2):99 105.
[7] Fohr IP, Prinz-Langenohl R, Bronstrup A, Bohlmann AM, Nau H,
Berthold HK, et al. 5,10-Methylenetetrahydrofolate reductase geno-
type determines the plasma homocysteine-lowering effect of supple-
mentation with 5-methyltetrahydrofolate or folic acid in healthy young
women. Am J Clin Nutr 2002;75:275 82.
[8] Jorde LB. Linkage disequilibrium and the search for complex disease
genes. Genome Res 2000;10(10):1435 44.
[9] 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.
[10] Clayton D. A generalization of the transmission/disequilibrium test for
uncertain-haplotype transmission. Am J Hum Genet 1999;65(4):
1170 7.
[11] Scanlon KS, Ferencz C, Loffredo CA, Wilson PD, Corea-Villasenor
A, Khoury MJ, et al, Baltimore-Washington Infant Study Group.
Preconceptional folate intake and malformations of the cardiac
outflow tract. Epidemiology 1998;9(1):95 8.
[12] Limpach A, Dalton M, Miles R, Gadson P. Homocysteine inhibits
retinoic acid synthesis: a mechanism for homocysteine-induced
congenital defects. Exp Cell Res 2000;260(1):166 74.
[13] Piedrahita JA, Oetama B, Bennett GD, van Waes J, Kamen BA,
Richardson J, et al. Mice lacking the folic acid-binding protein Folbp1
are defective in early embryonic development. Nat Genet
1999;23(2):228 32.
[14] Junker R, Kotthoff S, Vielhaber H, Halimeh S, Kosch A, Koch HG,
et al. Infant methylenetetrahydrofolate reductase 677TT genotype
is a risk factor for congenital heart disease. Cardiovasc Res
2001;51(2):251 4.
[15] Storti S, Vittorini S, Lascone MR, Sacchelli M, Collavoli A, Ripoli A,
et al. Association between 5,10-methylenetetrahydrofolate reductase
C677T and A1298C polymorphisms and conotruncal heart defects.
Clin Chem Lab Med 2003;41(3):276 80.