Background

Neural tube defects (NTDs) are complex congenital malformations of the central nervous system. Anencephaly and spina bifida are the most common and severe forms of NTDs. The birth prevalence of NTDs varies from approximately 0.8/1,000 births in many areas of the US to 3.5/1,000 in Mexico [1, 2]. Epidemiologic studies suggest that both genetic and environmental factors contribute to NTD etiologies. Although most factors appear to explain very little of the population burden of NTDs, maternal nutritional factors do appear to substantially contribute to the complex etiologies of NTDs. Foremost among these factors has been the role of periconceptional intake of folic acid in reducing recurrence and occurrence risks of women for NTD-affected pregnancies [3–11].

Nutrients and nutrition-related factors other than folic acid have been observed to influence NTD risks. For example, increased intakes of methionine, zinc, vitamin C, and choline have been associated with reduced NTD risk [12–15]. With respect to choline, it was recently observed that increased periconceptional intakes of diets with choline were associated with reduced risks of NTD-affected pregnancies that were independent of maternal folate intakes.[15] This observation provided evidence to suggest that deficiencies in methyl donors may be associated with NTD risk, that is to say, a less than optimal methyl-donor supply and DNA methylation status has been a suggested area for research efforts for certain birth defects [16]. Choline, like folate, is a methyl donor in the methylation of homocysteine to methionine [17, 18].

Choline is utilized for the de novo synthesis of phosphatidylcholine (PC) and sphingomyelin through the cytidine diphosphocholine (CDP-choline) pathway. There are three reactions in this pathway. The first reaction is catalyzed by the enzyme choline kinase (CHK; ATP:choline phosphotransferase, EC 2.7.1.32), which phosphorylates choline by donating an ATP [19]. The second reaction involves phosphocholine (P-Cho) cytidylyl transferase (CCT), which catalyzes the formation of CDP-Choline from P-Cho and CTP [20]. The final reaction uses choline phosphotransferase (CPT), which catalyzes the condensation reaction of CDP-Choline with diacylglycerol [21]. Phosphatidylcholine (PC) and sphingomyelin are required for maintaining cell membranes and play important roles in regulation of cell growth, differentiation, and death through the production of diacylglycerol (DAG) and ceramide (CER), which are cell signaling molecules [22, 23].

In gastrulation- and neurulation-stage mouse embryos, choline was elucidated to be used primarily for PC synthesis favoring the CDP-choline pathway, although some betaine and acetylcholine was also generated [24]. Using the choline uptake inhibitor 2-dimethlyaminoethanol (DMAE) and an inhibitor of PC synthesis, 1-O-octadecyl-2-O-methly-rac-glycerol-3-phosphocholine (ET-18-OCH3), Fisher and co-workers observed an increase in cell death and both craniofacial and NTDs in neurulation stage mouse embryos grown in culture [24].

In humans, choline kinase has two isoforms, CHKα and CHKβ, with the α as a dominant isoform. The CHKA gene encoding choline kinase α is located at chromosome 11q13.2. Our study focused on the CHKA gene. PCYT1A and PCYT1B encode CCTα and CCTβ, respectively. PCYT1A located at chromosome 3q29, while PCYT1B is located at chromosome Xp22.11 [25, 26]. In this study we focused on PCYT1A gene.

Given that periconceptional intake of choline has been associated with decreased risk of NTD-affected pregnancies [15], we investigated CHKA and PCYT1A genotypes on risk of spina bifida. We also investigated these genotypes in combination with lowered maternal intake of choline as risk factors of spina bifida.

Methods

Study population

Data investigated were derived from a case-control study that previously described a risk reduction in NTDs associated with maternal periconceptional intake of choline. In brief, these data were derived from the California Birth Defects Monitoring Program, a population-based active surveillance system for collecting information on infants and fetuses with congenital malformations [27]. Births occurring in selected California counties in the period 1989–1991 were eligible for the original case-controlled interview study. For the current investigation, we identified 103 spina bifida infants whose newborn screening blood specimen could be obtained and whose mothers' choline intake was estimated. As controls, we identified 338 non-malformed control infants whose newborn screening blood specimen could be obtained and whose mothers' choline intake was estimated. Among the 103 cases, 36% were non-Hispanic whites, 51% were Hispanics, and 13% were of other race/ethnic background. Among the 338 controls, 56% were non-Hispanic whites, 25% were Hispanics, and 19% were of another race/ethnic background. All samples were obtained with the approval from the state of California Health and Welfare Agency Committee for the Protection of Human Subjects. Genomic DNA used for genotyping was collected from newborn screening blood spots and extracted according to the Puregene Genomic DNA Extraction kit (Gentra, Minneapolis, MN, USA) protocol.

Genotyping procedure

Two intronic CHKA SNPs, hCV1562388 (A>C) and hCV1562393 (C>G) as well as two intronic PCYT1A SNPs, rs939883 (T>A) and rs3772109 (T>C) were selected as tagging SNPs using SNPBrowser software (v2.0) (Applied Biosystems, Foster City, CA, USA). hCV1562388 and hCV1562393 cover a 10 kbp genomic region of the CHKA gene; rs939883 and rs3772109 cover a 40 kbp genomic region of the PCYT1A gene. Samples were genotyped using a fluorescence-based allelic discrimination assay on an ABI PRISM® 7900HT sequence detection system (Applied Biosystems, Forster City, CA, USA), following the manufacturer's protocol. These intronic SNPs were selected based on the assumption that they might be in linkage disequilibrium (LD) with disease-causing variation. Primers and fluorescent dye labeled probes were purchased from ABI as Assay-on-Demand reagents. The Assays-on-Demand SNP genotyping consisted of a 20 × mix of unlabeled PCR primers and TaqMan®probe labeled with FAM™ and VIC™ fluorescent dyes. The FAM™ dye is linked to the 5' end of one allele in the probe while the VIC™ dye is linked to the 5' end of the other allele in the probe. These dyes are used for allelic discrimination of each SNP.

Allelic discrimination PCR reactions were performed on 384-well plates. Each reaction contained 2.5μL TaqMan Universal PCR Master Mix, No Amp Erase® UNG (2 ×), 0.25 μL of 20 × Assay-on-Demand™ SNP Genotyping Assay Mix, 2.25μL gDNA (1–20 ng) diluted in dH2O making up a total volume of 5 μL per reaction. The thermocycling conditions started with a denaturation step at 95°C for 10 min, followed by 45 cycles of denaturation at 92°C for 15 sec, annealing and extension at 60°C for 1 minute. Results were read and interpreted blind as to case/control status, and each assay was performed in duplicate.

Statistical analysis

Deviation from Hardy-Weinberg Equilibrium among control infants was evaluated by a chi-square test. Odds ratios (ORs) and associated 95% confidence intervals (95% CIs) were used to measure associations between infant CHKA, PCTY1A genotypes or compound genotypes, and spina bifida risk. For genotype comparisons, homozygous wild-type infants served as the reference group to which heterozygotes and variant homozygotes were compared. Choline intake values were considered according to quartile cutoffs. For quartile analyses, odds ratios and 95% CI were computed to estimate risk using the lowest quartile as the reference. All statistical analyses for this study were performed using SAS software v9.1 (SAS Institute Inc, Cary, NC, USA). Samples failed the genotyping assay were excluded for statistic analyses.

Results

Table 1 shows the previously observed association in this dataset between choline intake and, specifically, spina bifida risk. That is, ORs indicated that maternal intakes of choline in the periconceptional period were associated with reduced risk.

Table 1 Effect estimates (odds ratio) for spina bifida-affected pregnancies associated with maternal choline intake during the periconceptional period, California 1989–1991. OR: odds ratio; CI: confidence interval.
Full size table

Genotyping results of all SNPs were in Hardy-Weinberg Equilibrium (HWE) among controls (χ2 test: P > 0.05). Among non-Hispanic white and Hispanic white, the minor allele frequencies (MAF) were 0.39 and 0.27 for hCV1562388, 0.20 and 0.12 for hCV1562393, 0.32 and 0.33 for rs939883, 0.38 and 0.47 for rs3772109, respectively. Linkage disequilibrium (LD) were evaluated by D' and r2 using the Haploview program. For CHKA gene, hCV1562388 and hCV1562393 are in complete LD in study population (D' = 0.91, r2 = 0.065). For PCYT1A gene, D' for rs939883 and rs3772109 was 0.81 and r2 was 0.29.

Table 2 shows 'gene-only' effects associated with CHKA SNPs. These data showed a reduced risk of spina bifida for individuals with either one or more C alleles for the SNP hCV1562388 (A>C) but not with SNP hCV1562393(C>G). Infants with AA genotype for PCYT1A SNP rs939883 showed a nearly twofold increased risk of spina bifida relative to those with the TT genotype, however, it is not statistically significant and may be caused by chance.

Table 2 Effect estimates (odds ratios) for spina bifida-affected pregnancies associated with CHKA SNPs hCV1562388 (A>C) and hCV1562393(C>G), PCYT1A SNPs rs939883 (T>A) and rs3772109 (T>C), California 1989–1991. OR: odds ratio; aOR: adjusted odds ratio by maternal ethnicity; CI: confidence interval.
Full size table

Table 3 shows results of analyses that investigated gene-nutrient effects, that is to say, combined effects on risk of spina bifida between maternal choline intake and homozygous genotypes. We did not observe evidence of a gene-nutrient interaction between CHKA SNPs and maternal periconceptional choline intake. The increased risk for PCYT1A observed in gene-only analyses did not appear to be further influenced by maternal choline intake.

Table 3 Effect estimates (odds ratios) for spina bifida-affected pregnancies associated with maternal choline intake, CHKA SNP hCV1562388 (A>C), hCV1562393(C>G), PCYT1A SNPs rs939883 (T>A) and rs3772109 (T>C) genotypes. Breslow-Day test: P > 0.05. Lower 25%: total choline intake ≤ 289.93 mg/day. Higher 75 %: total choline intake > 513.24 mg/day.
Full size table

Discussion

This study investigated an underlying genetic explanation for a previously identified association between choline intake and spina bifida risk [15]. In the current study, we investigated intronic gene variants of two enzymes involved in the metabolism of dietary choline via the CDP-choline pathway. We believe this is the first study to evaluate DNA sequence variants in the human CHKA and PCYT1A genes for a possible association with NTD risk. Reduced risks of spina bifida were found for CHKA SNP hCV1562388, and increased risks were found for SNP rs939883. These risks, however, were not modified by maternal periconceptional intake levels of dietary choline. Thus, our study showed gene-only effects but did not observe gene-nutrient interaction effects associated with choline intake. The results indicate that dietary choline and choline metabolism genes may affect spina bifida risk independently or through some other unknown mechanisms. This interpretation should be taken cautiously owing to limited statistical power; if gene-only effects are true, a lack of gene-nutrient interaction effects may be due to small sample sizes and limited statistical power.

The functional impacts of the CHKA SNP hCV1562388 (A>C), CHKA SNP hCV1562393(C>G), PCYT1A SNP rs939883 and SNP rs3772109, or other sequence variations associated with these tagging SNPs are unknown with respect to choline, PC, or homocysteine concentrations. Unlike the enzymes cystathionine beta-synthase (CBS), methionine synthase reductase (MTR), and 5,10-methylenetetrahydrofolate reductase (MTHFR), that are directly involved in folate-homocysteine metabolism, the CHK metabolic profile may be modulated, as its effect is transmitted through the choline pathway [24].

This study had limitations that lessen our ability to make solid inferences from the results. As noted, the study had limited sample size; therefore, precision was low for many of the estimated effects, particularly those involving gene-nutrient interactions. Our study, explored only four tagging SNPs in two genes, limiting the ability to detect possible genetic modifiers related to choline metabolism.

Conclusion

Despite its limitations, this study provided initial data indicating a potential association between CHKA and PCYT1A gene variants and spina bifida risk. Future studies of additional SNPs within the CHKA and PCYT1A genes should be investigated as potential predictors of spina bifida risks.