Genetic Study of 12 SNPs involved in 11 Folate Metabolism Genes and Neural Tube Defects in Suzhou Children

Objective: Neural tube defect (NTD) incidence could be effectively reduced by folic acid supplementation before and during pregnancy. We studied single nucleotide polymorphisms (SNPs) involved in folate metabolism to explore genetic susceptibility to NTD. We studied the association between 12 SNPs involved in 11 folate metabolism genes and NTDs. Methods: We enrolled 76 children with NTD and 188 control children. We genotyped 12 folate metabolism SNPs including CBS-C699T, DHFR-c594+59del19, GSTO1-C428T, MTHFD-G1958A, MTHFR-C677T, MTHFR-A1298C, MTR-A2756G, MTRR-A66G, NFE2L2-ins1+C11108T, RFC1-G80A, TCN2-C776T, and TYMS-1494del6 using SNaPShot genotyping technology and confirmed by Sanger sequencing. Results: One SNP, TYMS-1494del6, and one compound wide-type genotype of RFC1-G80A, MTHFR-A1298C and TCN2-C776T might decrease NTD risk, and three compound mutation genotypes of MTHFD-G1958A, MTHFRC677T, and MTR-A2756G; MTHFD-G1958A, MTR-A2756G, and RFC1-G80A; and RFC1-G80A, MTHFR-A1298C, and TCN2-C776T might increase NTD risk. The TT genotype of TYMS-1494del6 (P<0.001) and the AT+TT genotype of TYMS-1494del6 (P=0.009) were significant genotypes. Fourteen in 188 control babies carried the compound wide-type genotype of RFC1-G80A, MTHFR-A1298C, and TCN2-C776T, but none in 76 NTD babies (P=0.014). The ratios of the two compound mutants for MTHFD-G1958A, MTHFR-C677T, MTR-A2756G, and MTHFD-G1958A, RFC1-G80A, MTR-A2756G in NTD were higher than in control babies (P=0.021) and RFC1G80A, MTHFR-A1298C, and TCN2-C776T (P=0.029). Conclusions: The TT genotype of TYMS-1494del6 and the two wide-type genotypes of RFC1-G80A, MTHFRA1298C, and TCN2-C776T are protective in NTD. Three compound mutation genotypes of MTHFD-G1958A, MTHFR-C677T, MTR-A2756G, MTHFD-G1958A, MTR-A2756G, RFC1-G80A, and RFC1-G80A, MTHFR-A1298C, TCN2-C776T might increase susceptibility to NTD.


Background
Neural tube defects (NTDs) are a group of serious birth defects that affect the developing nervous system and include anencephaly, spina bifida, and encephalocele [1]. NTDs are the second most common of birth defects, which are preventable and are significant causes of infant death and childhood disability [1,2]. It is estimated that there are more than 300,000 NTDs worldwide each year, many of these occurring in low-resource countries [3]. Anencephaly and spina bifida are the most common NTDs and occur with about equal frequency, whereas encephalocele is seen less frequently. In anencephaly, there is partial or complete absence of the skull bones usually with a remnant of the brain, which is almost always fatal before or shortly after birth [4]. Spina bifida with meningomyelocele results from failure of the vertebral arches to close over an open neural tube defect, exposing the spinal cord and nerves [4]. Although most cases of spina bifida are open, 10% to 15% are closed or covered by skin. Spina bifida is compatible with survival, although in most cases individuals have moderate to severe disabilities and, in some cases, may have mental retardation [2,5]. Researchers generally agreed that NTD occurrence is caused by genetic and environmental factors, though its etiology is not fully understood. Folic acid effectively reduced NTD, and this is supported by evidence from recent systematic reviews on folic acid for the prevention of NTDs in both high-and low-income countries [6,7]. Results of randomized controlled trials and several observational

Journal of Molecular and Genetic Medicine
studies showed that 50% to 85% of NTDs can be prevented if women consume a folic acid-containing supplement before and during the early weeks of pregnancy in addition to dietary folate [1].
Although folic acid supplementation effectively reduces NTD risk, its mechanism is not fully clear. According to the folic acid metabolic pathways [8], two effects of folic acid supplementation protect people from NTDs: (1) it ensures adequate DNA synthesis, methylation, and chromosome repair [9]; and (2) it eliminates the toxic effects of homocysteine to the fetus through the cysteine metabolic pathway [10]. Epidemiologic evidence linking genetic susceptibility in folate metabolism to birth defects is most extensive for NTDs. Genes involved in folate metabolism are heavily investigated for NTD genetic risk, including 5, 10-methylenetetrahydrofolate reductase (MTHFR), Cystathionine beta-synthase (CBS), and Methyltetrahydrofolatehomocysteine methyltransferase (MTR). Also, single nucleotide polymorphisms (SNPs) located in MTHFR, CBS, and MTR have been investigated for their association with NTDs, such as MTHFR-C677T, -A1298C, CBS-T833C, and MS-A2756G. However, findings on these genetic variants have not been consistent [11][12][13][14][15]. Also, few reports exist on other genes involved in the folate metabolic pathway. In this report, we detected 12 SNPs distributed in 11 folate metabolism genes using SNaPShot genotyping technology and analyzed the association between SNPs involved in folate metabolism and NTDs.

Data sources
We enrolled 264 subjects in this study with their informed consent, including 76 cases of NTD babies and 188 normal control babies born in the Suzhou Maternal-child Medical Center from 2008 to 2011. The case group included 76 infants diagnosed with NTDs in the Nanjing Medical University Affiliated Suzhou Hospital and Suzhou University Affiliated Children Hospital from 2008 to 2011, and the information of gender, age and diagnosis was shown in Table 1. The control group included 188 healthy infants born in the Nanjing Medical University Affiliated Suzhou Hospital during the same period. They were randomly selected. All babies were Han Chinese, and we collected their umbilical cord blood when they were born.

Genotyping method
We detected 12 SNPs distributed in 11 folate metabolic enzyme genes in one reaction using SNaPShot genotyping technology. We modified SNaPShot protocol [16] and confirmed it using Sanger sequencing established in our laboratory. The primers (with patents) were synthesized by the Genearray Company in China.

Statistical analysis
We analyzed the genotypes frequencies between the two groups using the X 2 -test, and Hardy-Weinberg genetic equilibrium were tested by non-parametric test. We used binary logistic regression analysis to analyze the association between SNPs and NTD, and to remove the influence of other factors, Values of P<0.05 were considered statistically significant.

Compound wide-type genotypes
No compound wild-type genotypes showed a significant difference in NTD and control babies except the compound wild-type genotype RFC1-G80A, MTHFR-A1298C, and TCN2-C776T (Table 5). Fourteen control babies carried the compound mutation genotype of the total 188 control babies, but no NTD babies carried this compound mutation. This indicated that the compound wild-type genotype RFC1-G80A, MTHFR-A1298C, and TCN2-C776T is a protective factor of NTDs, P=0.014. Additionally, two compound wild-type genotypes including MTHFD-G1958A, MTHFR-C677T, and MTR-A2756G, and MTHFD-G1958A, MTR-A2756G, and RFC1-G80A may be protective NTD factors, with OR=0.73 and 0.61, respectively, although they showed no significant differences given our small sample size.

Groups
Compound wide-type genotypes OR 95% CI P

Discussion
There were several interesting opinions revealed in this study. We found the TT and AT of TYMS-1494del6 were the protective NTD factors, with OR=0.39 and 0.56 respectively. Furthermore, TYMS-1494del6 was also a protective NTD factor as it removed the effects of other SNPs, RR=0.445, P<0.001. TYMS was the key enzyme in the de novo synthesis of 29-deoxyuridine -59-monophosphate (dTMP), the essential precursor of DNA biosynthesis and the repairing process [17]. The most common deletion polymorphism of TYMS was 1494del6 in intron-1, which might have played a more associative role in regulating gene expression than the promoter [17]. At present, two studies were carried out on the association between the TYMS gene SNPs and NTDs, and the results indicated that several TYMS gene SNPs could have been NTD risk factors [18,19]. However, no study reported the association between TYMS-1494del6 and NTDs, and our results indicated it was an interesting candidate SNP for NTDs.
We found another candidate SNP for NTDs was TCN2-C776T. When it removed the effects from other SNPs, TCN2-C776T could be an NTD risk factor, with RR=1.924, P=0.003. TCN's main function was transporting VB12 into cells, critical for balancing folate metabolism. The most common polymorphism was TCN2-C776T, which encoded for arginine in place of proline at codon 259 of the amino acid sequence [20]. Several studies reported TCN-C776G was associated with NTD [21][22][23][24]. To a certain extent, our results supported the above conclusion.
The association of SNPs localized at MTHFR gene with NTD risk was the research focus. We investigated the distribution of MTHFR-C677T and -A1298C in NTDs and could not conclude that they were NTD genetic risk factors, which contrasted with most of the other researchers' conclusions [15,25]. The cause of the contradictory results may include that since NTDs are caused by the combined effects of many genes, different genetic backgrounds and nutrition may result in different genetic NTD susceptibility. MTHFD1-G1958A was a common polymorphism, resulting in a substitution of alanine for glycine at codon 653, located at the domain 10formylTHF enzyme synthase. The purified Arg653Gln enzyme had normal substrate affinity, but the mutation reduced MTHFD1's metabolic activity  [26,27]. Several studies reported that MTHFD1-G1958A could be a maternal NTD risk factor [28,29], but not a children factor. However, De Marco et al. [30] found an increased risk for heterozygous 1958G/A and homozygous 1958A/A genotypes in children. Our results also indicated a negative association between MTHFD1-G1958A and NTDs, consistent with conclusions from studies by Parle-McDermott and Brody.
The most common polymorphism in the MTR gene was MTR-A2756G, which decreased MTR activity and increased the cellular homocysteine level [31,32]. To date, numerous studies have reported that an association exists between the MTR-A2756G polymorphism and maternal NTD risk. However, their results remain inconsistent [33]. The latest meta-analysis by Ouyang indicated that the MTR-A2756G polymorphism was significantly associated with maternal NTD risk in Caucasians [34] but not associated with NTD risks in Caucasian children [34]. According to our studies, MTR-A2756G was not associated with NTD risks in Chinese children.
MTRR is a member of the electron transferase family with three characteristic binding sites, including FMN, FAD, and NADPH. Disturbances in catalytic activity can lead to high homocysteine levels, which could be an NTD risk factor [22]. The most common polymorphism in the MTRR gene was A66G substitution, leading to isoleucine changing to methionine in amino acid 22. Therefore, the association between MTRR-A66G and NTDs has been extensively researched. Pietrzyk et al. [35] postulated that MTRR-A66G should be regarded as an independent risk factor for spina bifida. Also, Dunlevy found that the 66GG genotype frequency is significantly higher in people with NTDs and their mothers than in the control group [36]. However, most studies, including the meta-analysis, indicated MTRR-A66G was not associated with NTD risk [34], In Chinese children, MTRR-A66G heterozygous and homozygous genotype were not shown a association with NTDs.
CBS catalyzes homocysteine and serine to synthesize irreversible cystathionine. Disturbances in this process could lead to an increased cellular homocysteine level. CBS is a cytoplasmic protein and has a catalytic core with a conserved amino acid sequence. In addition, 60 percent of about 150 mutations could lead to decreased protein catalytic activity. There are three polymorphisms in CBS, including CBS-844ins68bp, CBS-T833C, and CBS-G919A, reported as genetic factors for CHDs [37] and NTDs [18]. In our study, we did not find that a novel SNP CBS-C699T was the NTD risk factor.
The most common polymorphism in the RFC1 gene is a highfrequency G to A SNP at position 80 that results in a change of arginine-27 to histidine-27 in exon 2. This makes 5-formyl tetrahydrofolate cofactor transport 2-fold lower than the wildtype, because the arginine (CGG) at residue 27 could be critical for targeting and integrating protein to the plasma membrane. Many studies showed that women carrying the GA and AA genotypes had higher RBC folate concentrations relative to women carrying the RFC1-80GG genotype, and this was an independent risk factor for their children with CHD [16]. However, our results did not support the above conclusion. The possible reason may have been that folate metabolism is a complicated metabolic process in which many enzymes play important roles of maintaining metabolism. Due to the different genetic backgrounds in our study, the major gene defects leading to elevated homocysteine were different.
DHFR's continuous reduction of DHF to tetrahydrofolates (THFs) played an important role in folate metabolism. The most common polymorphism in the DHFR gene was a 19bp deletion within intron I, a well known site of regulatory sequences for some genes. Three studies reported the association between DHFR-c594+59del19 and NTDs but did not reach a consistent conclusion [38][39][40]. In this study, we did not find that NTDs were correlated with DHFR-c594+59del19.
We first reported the association between two SNPs, including NFE2L2-ins1C11108T and GST01-C428T, and NTDs. NFE2L2 was a transcription factor regulating antioxidant enzyme protein expression at the transcriptional level, and plays an important role in the process of HCY metabolism. GSTO1 played an important role in the metabolic pathway from homocysteine to glutathione, which is referred as the transsulfuration pathway. Mostly polymorphisms of GSTM1, GSTT1, and GSTP1 were revealed to be associated with increased susceptibility to several cancers and other diseases [41,42]. In this study, NFE2L2-ins1C11108T and GST01-C428T were not considered NTD risk factors.

Conclusion
The vast majority of NTDs resulted from the interaction of multiple genes and the environment. The effect of one SNP or gene tended to be too "minor" to be found. Also, combined SNPs effects on NTDs may be more obvious than simple superposition of SNPs. However, few studies focused on the association between NTDs and the compound mutations of several significant SNPs located on enzyme genes involved in folate metabolism. We further analyzed the combined effects of SNPs in multiple folate metabolism enzyme genes on NTDs. We found that three compound mutation genotypes, including MTHFD-G1958A, MTHFR-C677T, and MTR-A2756G; MTHFD-G1958A, MTR-A2756G, and RFC1-G80A; and RFC1-G80A, MTHFR-A1298C and TCN2-C776T could be considered risk factors for NTDs. Recently, the two compound mutation genotypes of MTHFD-G1958A, MTHFR-C677T, and MTR-A2756G, and MTHFD-G1958A, MTR-A2756G, and RFC1-G80A were also reported as CHD risk factors by Wang B, et al. which also indicated the compound mutation genotype should be considered an independent factor. The compound wide-type genotypes of MTHFD-G1958A, MTHFR-C677T, and MTR-A2756G, and MTHFD-G1958A, MTR-A2756G, and RFC1-G80A could be protective factors for NTD, with OR<1. However, the differences between the two compound wide-type genotypes did not reach statistical significance, P>0.05, which may be due to our small sample size. The compound mutation genotype of RFC1-G80A, MTHFR-A1298C, and TCN2-C776T was another genetic factor for NTDs, OR=2.07, P=0.029. Furthermore, we found evidence that the compound wide-type genotype of RFC1-G80A, MTHFR-A1298C, and TCN2-C776T was a protective factor for NTDs. We found that 14 control babies carried the compound wide-type genotype of RFC1-G80A, MTHFR-A1298C, and TCN2-C776T, but not in NTD babies.
Limited by the sample size and SNPs detected in our studies, we could not analyze and screen all compound genotypes. Therefore, further studies with large sample sizes and more SNPs related to folate metabolism are needed to research NTDs. High-throughput genotyping technologies have superiority in detecting SNPs, such as Gene Chip Technique, Genome-wide Association Analysis, and Second-generation Sequencing Technology, which may be the breakthroughs for exploring NTD genetic risk factors.