The percentage of males and females among the VSD cases was 46.3% and 53.7% respectively, whereas the percentage of males and females among the control subjects was 44.8% and 55.2%, respectively, in this study. The mean age of case subjects was 4.51 ± 2.39 whereas the mean age of controls was 5.43 ± 0.51. There was a significant difference of mean group of age between cases and controls (p < 0.05).

Genotypes of C677T polymorphism were determined by incubating the PCR product with Hinf1 restriction enzyme. The fragments were resolved on 4% agarose gel. Figure 1 shows the PCR product and PCR-RFLP of MTHFR C677T gene polymorphism. The genotype and allele frequencies of C677T polymorphism of MTHFR are shown in Table 2. The Chi-square test did not show any significant differences between genotype and allele frequencies of C677T polymorphism among cases and controls (p > 0.05) and (p > 0.05), respectively. The percentage of CC and CT genotypes among cases was 51.2% and 48.8%, respectively, compared to 56.8% and 43.2%, respectively among control subjects. The derived allele frequencies of C and T alleles were 75.6% and 24.4%, respectively, among cases and 78.4% and 21.6%, respectively, among controls.

Lane M illustrates the 100bp DNA ladder. Lanes 1 and 2 show the PCR products (198bp), Lanes 3 and 4 show the heterozygote fragments (198 and 1475 bp), Lanes 5 and 6 shows the wild type fragment (198 bp) and Lane N represents the negative test control.

Genotype and allele frequencies of C677T polymorphism of MTHFR were compared between cases and controls. There were no significant differences between genotype and allele frequencies of C677T polymorphism among cases and controls (p > 0.05). The percentage of CC and CT genotypes among cases was 51.2% and 48.8%, respectively, compared to 56.8% and 43.2%, respectively, among control subjects. The derived allele frequencies of C and T alleles were 75.6% and 24.4%, respectively, among cases and 78.4% and 21.6%, respectively, among controls. As there was no TT genotype in cases and controls, we failed to show an association between MTHFR C677T gene polymorphism and the occurrence of VSD. The results obtained from a study done by Zhu et al. on a Chinese population suggested that the MTHFR C677T locus variation is associated with the occurrence of atrial septal defect (ASD) and patent ductus arteriosus (PDA), and carriers of mutant homozygote TT and allele T had a high risk of these two types of CHDs [41]. Junker et al. studied the relationship of MTHFR C677T gene polymorphism with CHDs, and the results showed that TT homozygoty was significantly related to congenital heart structure teratogenesis, especially with pulmonary artery stenosis, left heart dysplasia syndrome and coarctation of aorta [35]. Subsequent studies [36–39] failed to show significant results, which could indicate an association between CHD and the MTHFR gene polymorphism, which is in consistence to our study but in contrast to the findings of Bennouar et al. in a Moroccan population [42].

All genotypes were determined using PCR-RFLP using respective restriction enzymes. In order to genotype C524T polymorphism of MTRR, the amplified PCR product (as shown in Figure 2) was digested with Xho1 restriction enzyme. The resolved fragments on 4% agarose gel showed 247 and 62 bp in C524C homozygote, 309, 247, and 62 bp in C524T heterozygote and 309bp in T524T homozygote. Genotypes of A66G polymorphism were determined by digestion of the PCR product with Nde I restriction enzyme. This digestion produced fragments of 126, 25 bp for MTRR A66A homozygote, 151, 126 and 25 bp in MTRR A66G heterozygote, and 151 bp in MTRR G66G homozygote. Figure 3 shows the PCR product and RFLP analysis of MTRR A66G polymorphism respectively.

In Figure 2, Lane M illustrates the 100 bp DNA ladder. Lanes 1 and 2 show the PCR products (309 bp), Lanes 3 and 4 show the wild type fragments (247 bp), Lanes 5 and 6 show the mutant type fragment (309 bp), Lanes 7 and 8 show the heterozygote fragments (309 and 247 bp) and Lane N represents the negative test control.

In Figure 3, Lane M illustrates the 100 bp DNA ladder. Lanes 1–3 show the PCR products (151 bp), Lane 4,5 shows the wild type fragments (126 bp), Lane 6,7 shows the mutant type fragment (151 bp), Lane 8,9 shows the heterozygote fragments (151 and 126 bp) and Lane N represents the negative test control.

For MTRR C524T polymorphism, the percentage of genotypes C524C, C524T and T524T among cases was 43.1%, 40.7%, and 16.3%, respectively, whereas the frequencies among controls were 52.8%, 43.2% and 4%, respectively (Table 3). There was a significant difference between cases and controls (p < 0.05). Derived allele frequencies of C allele and T allele were 63.4% and 36.6% among cases and 74.4% and 25.6% among controls. There was a significant difference between the allele frequency among cases and controls (p < 0.05). For the A66G polymorphism of MTRR, the genotype and frequencies differed significantly between cases and controls (p < 0.05). The genotypes of A66A, A66G and G66G were 33.3%, 43.9%, and 22.8%, respectively, in cases compared to 49.6%, 42.4% and 8%, respectively, among control subjects. The allele frequencies as shown in Table 3 also showed a significance difference between cases and controls (p < 0.05). C677T genotype frequencies were in agreement with those predicted by the Hardy Weinberg distribution, whereas, MTRR gene polymorphisms did not meet the Hardy Weinberg assumption in neither cases nor controls.

MTRR is a central regulatory enzyme in the metabolic pathway of homocysteine/folate. The malfunction of this enzyme has been considered as a modifiable risk factor for heart and vascular disease [43]. In this case-control study, significant differences were observed for A66G and C524T polymorphisms of MTRR gene between cases and controls (p < 0.05), which is in accordance with some studies [44–47]. We found a modest association between the A66G and C524T alleles of the MTRR gene and CHDs in the Iranian subjects. In contrast, a difference in homocysteine concentration between the 66 MTRR genotypes in CAD and ischemic cerebrovascular disease was not detected in some studies [14,16]. The conflicting results found in the associated studies were due to many confounding factors, such as study design, sample size bias, gen-environment interactions, population heterogeneity, mismatched phenotypes and population stratification. Collectively, the findings from this study provide adequate evidence that the risks of VSD are influenced by the variants of the three polymorphisms of MTHFR and MTRR genes in Iranian subjects. It is important to consider the need for future studies of folate-pathway genes other than those found in the current study, and it is also potentially complex to study the gene-gene and gene-environment interactions and VSD in the Iranian population.

For further confirmation, random samples were selected and repeated with the same PCR conditions to confirm the genotyping results. Those samples were purified and DNA sequencing was performed with the Applied Biosystems 3730xl DNA Sequencer provided by Medigene Sdn. Bhd, Selangor, Malaysia.

In this study, a number of limitations have to be considered. The present study has provided only a genetic association for MTHFR and MTRR gene polymorphisms among Iranian VSD patients as compared to control subjects. However, we failed to analyze the MTHFR and MTRR mRNA levels in both VSD patients and control subjects. Apart from the MTHFR and MTRR gene polymorphisms other polymorphisms such as TBX5, NKX2.5 and GATA4 need to be analyzed to determine the association of the other candidate genes with VSD and other defects. Replication studies with a larger number of samples and assessing the genotypic and allelic frequencies of mothers are also needed in order to understand the etiological factors.

Ethical approval was obtained from the Shaheed Beheshti University of Medical Sciences and also from the Faculty of Medicine and Health Sciences, Universiti Putra Malaysia. A written informed consent was obtained from the parents of the children with CHD. Similarly, consent forms were obtained from all of the control subjects enrolled in this study.

A total of 150 pediatric patients were approached, and 123 pediatric patients were recruited as the other subjects were excluded due to the inconsistent results and extreme values. All the patients were recruited from the pediatric infectious research center (PIRC), Mofid children hospital, Tehran, Iran. All the subjects were visited during July 2010 to Jun 2011 in the hospital from the different parts of the districts. The diagnosis of non-syndromic CHD and classification of the type of the cardiac defect had been done by a pediatric cardiologist based on the clinical and echocardiography findings with or without the diagnostic cardiac catheterization findings and surgical notes. Whereas for the control subjects, a total of 125 healthy control subjects with no history of congenital heart disease were recruited in for this study.

Buccal cell samples were collected on a sterile cytobrush (Qiagen Inc., Chatsworth, CA, USA) Genomic DNA extraction was carried out using the DNA isolation kit (Qiagen Inc., Chatsworth, CA, USA). The purity of extracted DNA was quantified using Eppendorf UVette® in Biophotometer (Eppendorf, Hamburg, Germany).

To determine the genotypes of MTHFR and MTRR genes, genomic DNA was amplified first by the respective primers using the polymerase chain reaction (PCR) technique. The PCR amplification for all the respective polymorphisms was performed in a total volume of a 25 μL reaction mixture consisting of 10 pmol of each primer and the Mastermix (i-DNA Biotechnology (M) Sdn Bhd, Kuchai lama, Kuala Lumpur, Malaysia) and the template DNA. A negative control containing no genomic DNA and a positive control of known genotype were always included in the set of reactions. All the PCR cycling conditions were carried out on an iCycler machine (BioRad Laboratories, Hercules, CA, USA). The amplified PCR products for all the three gene polymorphisms were separated at 2%–4% agarose gel (Bioline, London, UK). The agarose gel was stained in ethidium bromide and visualized using Alpha Imager (Alpha Innotech, San Leandro, CA, USA). The PCR products of the respective genes were digested with 2–4 units of the respective restriction enzymes (Thermo Fisher Scientific, Inc, provided by Research Instruments Sdn Bhd, Petaling Jaya, Malaysia) with 10× Fast Digest Green Buffer in a final volume of 30 μL reaction mixture. Table 4 shows the primers used for the RFLP method, restriction endonucleases and the digested restricted fragment size products. Identical results were obtained when genotyping was performed for 10% of the samples on two separate occasions.

Purified PCR products were sent to Research Biolabs Malaysia to confirm the nucleotide sequence. The sequencing results were aligned with the respective reference gene sequence from the NCBI-GeneBank sequences using the sequence alignment MEGA4 software [48].

Data analysis was done by SPSS version 18.00 (SPSS Inc, South Wacker Drive, Chicago, IL, USA). Alleles and genotype distribution were tested for deviation from the Hardy-Weinberg by a Chi Square test. The odds ratio (OR) and its 95% confidence intervals (CI) were used to illustrate the association, with p < 0.05 considered in all tests to be statistically significant.