Polymorphic variants in VAX1 and the risk of nonsyndromic cleft lip with or without cleft palate in a population from northern China.

BACKGROUND: Nonsyndromic cleft lip with or without cleft palate (NSCL/P) is one of the most common craniofacial birth defects, and the etiology of NSCL/P involves both genetic and environmental factors. Genome-wide association study (GWAS) identified a novel susceptibility locus of ventral anterior homeobox 1 (VAX1) in patients with NSCL/P. However, the association of single nucleotide polymorphisms (SNPs) of VAX1 with NSCL/P is inconclusive due to the differences in the racial and ethnic populations. The aim of this study was to replicate the association between VAX1 and NSCL/P in a northern Chinese Han population.
METHODS: Our study included 186 patients with NSCL/P and 223 healthy individuals from northern China. Five SNPs (rs4752028, rs10787760, rs7078160, rs6585429, and rs1871345) on VAX1 were genotyped using the SNaPshot method.
RESULTS: Recessive genetic model analysis revealed that homozygous genotype CC of VAX1 rs4752028 was associated with an increased risk of NSCL/P (odds ratio = 1.89, 95% confidence interval = 1.12-3.19, P = 0.017), but the results were not significant after the Bonferroni correction for multiple comparisons. The allele and genotype frequencies of rs10787760, rs7078160, rs6585429, and rs1871345 and the allele frequencies of rs4752028 showed no significant differences between cases and controls. Haplotype and SNP-SNP interaction analyses did not detect any significant association of VAX1 with the occurrence of NSCL/P.
CONCLUSION: VAX1 rs4752028 was weakly associated with NSCL/P development in the studied northern Chinese Han population.

Introduction

Nonsyndromic cleft lip with or without cleft palate (NSCL/P), which results from an impaired facial process growth and fusion during embryogenesis, is one of the most common craniofacial birth defects in humans.[ The average incidence of NSCL/P is 1 to 2 of 1000 births globally.[ Moreover, its prevalence varies widely among different geographical regions and ethnic groups.[ The highest incidence of NSCL/P is observed in the Asians and Native Americans, and the lowest is observed in the Africans. The high prevalence of NSCL/P (1.42/1000) is reported in the Chinese population.[ Newborns with NSCL/P may have speech and feeding problems, poor nutrition, psychiatric diseases, and infection of the middle ear. Although these defects can be partly corrected by a series of surgical interventions and multi-disciplinary treatments, they still bring long-term burdens to the individual, family, and society.[

The etiology of NSCL/P is complex and is associated with both the genetic and environment factors.[ Although the specific genetic and environmental risk factors associated with NSCL/P remain unclear, the linkage and association analysis and the genome-wide scanning have provided significant evidence for the potential candidate genes in the development of NSCL/P, such as MSX1,[IRF6,[PVRL1,[SUMO1,[ and FGF.[

Recently, 2 genome-wide association analyses have confirmed a susceptibility candidate gene that may be involved in NSCL/P, VAX1 at locus 10q25.3,[ and it was later identified on meta-analysis by Ludwig et al.[ These studies have revealed that single nucleotide polymorphisms (SNPs) in or near VAX1 were involved in the risk of NSCL/P, and the analysis of participants of European and Asian origin from multiple populations provided a significant evidence for the susceptibility gene.[ A previous study performed using mouse model suggested that lack of functionally active VAX1 results in craniofacial deformity, including cleft palate.[ However, several similar investigations were conducted in the Mesoamericans, Central Africans, Southeast Asians, and southern Chinese population, which showed inconsistent results.[ To the best of our knowledge, no GWAS examining the association between SNPs of VAX1 and the risk of NSCL/P in a northern Chinese population has been reported. Therefore, the aim of our current study was to investigate whether the 5 SNPs (rs4752028, rs10787760, rs7078160, rs6585429, and rs1871345) of VAX1 were associated with the susceptibility of NSCL/P in a northern Chinese Han population.

Methods

Subjects

The current case–control study was approved by the Institutional Ethics Committee of the Harbin Medical University and was a hospital-based study. Diagnosis of the case group (patients with NSCL/P) was done through clinical investigations by 2 experienced dentists to assess individual phenotypic features and cases were identified through medical records; the cases with other major congenital anomaly and syndromes were excluded from the study. The study consisted of 186 cases (101 female subjects and 85 male subjects) who visited the department of Oral and Maxillofacial Surgery, Harbin Medical University Affiliated Stomatological Hospital during the period from March 2006 to April 2011. During the same period, 223 healthy controls (118 female subjects and 105 male subjects), who had no history of congenital malformation or familial history of orofacial clefting, were also selected from the same hospital. A signed informed consent was obtained from each patient, volunteer, or their guardians. About 1-mL peripheral venous blood was withdrawn from each participant.

Polymorphism selection

To investigate the role of VAX1 gene polymorphisms on NSCL/P risk in a northern Chinese population. We selected VAX1 tag SNPs (rs10787760, rs6585429, and rs1871345), which were selected from CHB (Beijing Han population of China) with a minor allele frequency (MAF) >0.05 in the HapMap Project. In addition, on the basis of the genome-wide association studies (GWAS) of Mangold et al,[ we selected other 2 SNPs (rs7078160 and rs4752028) of VAX1.

Genotyping

DNA was extracted from the peripheral venous blood samples from each participant using the QIAamp DNA Blood Kit (Valencia, CA), according to the manufacturer's protocol. Genotyping for VAX1 (rs4752028, rs10787760, rs7078160, rs6585429, and rs1871345) polymorphisms was performed using SNaPshot technology. Primers were designed using the Primer 3 software (http://frodo.wi.mit.edu/). PCRs consisting of 10 to 50 ng DNA, 1 × HotStarTaq buffer (Invitrogen, Carlsbad, CA), 3 mM MgCl2, 300 μM of each dNTP, 0.08 μM of each primer, and one unit of HotStarTaq polymerase (Invitrogen, Carlsbad, CA) were set up in a 20 μl reaction volume. A touchdown PCR program was used with the following conditions: initial denaturation at 95°C for 10 minutes, followed by 20 cycles at 94°C for 20 seconds, annealing at 55°C for 40 seconds, and extension at 72°C for 40 seconds; the annealing temperature was decreased by 0.5°C per cycle. The extension reaction contained 1 × ABI PRISM SNaPshot Multiplex Ready Reaction Mix (Applied Biosystems, Foster City, CA), 0.5 μM of each primer and 1 μl of each PCR product and was carried out according to the manufacturer's instructions (Applied Biosystems). Further, the PCR products were purified and scanned using 3730 Genetic Analyzer (Life Technologies Corporation, Vancouver, British Columbia, Canada).

Statistical Analyses

Hardy–Weinberg equilibrium (HWE) of the genotype distributions of cases and controls was examined by using Chi-square (χ2) test. The differences in genotype and allele frequencies of the tested SNPs between cases and control groups were evaluated using standard χ2 and Fisher tests. The association between SNPs and risk of NSCL/P was evaluated by calculating the odds ratios (ORs) and 95% confidence intervals (CIs). Bonferroni correction of P < 0.01 (0.05/5) was used to note the statistical significance and solve the problem of multiple comparisons. Statistical analyses were performed using PLINK (a free open-source whole genome association analysis toolset), and R. Linkage disequilibrium (LD) was evaluated using Haplotype 4.2 software[ depending on D′ and r values. SNP–SNP interactions in VAX1 were evaluated using the R package of Multifactor Dimensionality Reduction (MDR).[ A result with P value of <0.05 was considered as a statistically significant result.

Results

All the SNPs in the cases and control groups were observed to be consistent with the Hardy–Weinberg equilibrium (P > 0.05) (Table 1). The allele and genotype frequencies of rs10787760, rs7078160, rs6585429, and rs1871345 and the allele frequencies of rs4752028 in the NSCL/P cases were not significantly different from those in the controls (P > 0.05) as summarized in Table 2. Further, analyses of the dominant and recessive genetic models revealed that VAX1 rs4752028 was differently distributed between the cases and control groups (P = 0.017). The results of the recessive genetic model showed that the homozygous genotype CC was associated with an increased risk of NSCL/P (OR = 1.89, 95% CI = 1.12–3.19), combining the TT and CT genotype of VAX1 rs4752028. However, the results did not show significance after the Bonferroni correction for multiple comparisons was applied (corrected P = 0.05/5 = 0.01). The LD pattern among these 5 SNPs is depicted in Fig. 1. A haplotype block was constructed in this region with the D′ and r values (Table 3). In the haplotype analysis, the haplotype distributions between cases and control groups were compared and it was noted that no haplotype was associated with the risk of NSCL/P (P > 0.05; Table 4). The results of MDR analyses of SNP–SNP interactions are summarized in Table 5 and Fig. 2. The 1-locus based model of rs4752028 revealed the highest cross-validation consistency (CVC) of 10/10 and testing balanced accuracy (TA) of 0.53, but it not reached statistical significance (P = 0.127). However, no SNP–SNP interactions were found to be associated with the risk of NSCL/P.

Table 1

The 5 studied VAX1 SNPs.

Table 2

Frequencies and ORs of genotypes and alleles of VAX1 SNPs in controls and cases.

Figure 1

Linkage disequilibrium (LD) blocks for the VAX1 haplotype analysis.

Table 3

Pairwise linkage disequilibrium measures for VAX1.

Table 4

Haplotype association between SNPs rs10787760, rs6585429, and rs1871345.

Table 5

Interaction models by MDR analysis.

Figure 2

SNP1: rs4752028. Interaction analysis between SNPs in NSCL/P using MDR. The dark and light bars in each cell are the combined genotypes of cases and controls, respectively. Each cell color represents the risk degree used—gray-high risk, light-low risk.

Discussion

The current study investigated the association of VAX1 with the risk of NSCL/P in a northern Chinese Han population. In this study, we successfully genotyped 5 SNPs of VAX1, and the allele and genotype frequencies of rs10787760, rs7078160, rs6585429, and rs1871345 and the allele frequencies of rs4752028 showed no significant differences between cases and controls. We found that the homozygous genotype CC of rs4752028 was associated with an increased risk of NSCL/P (OR = 1.89, 95% CI = 1.12–3.19, P = 0.017, using a recessive model). However, the association disappeared after the Bonferroni correction that controlling for multiple comparisons.

These results seem to be a little disappointing; however, given the complicated heterogeneous nature of NSCL/P and a number of other confounding factors, this is an expected result.[Studies from animal models suggested that VAX1 has played a crucial role in the process during craniofacial development. VAX1 was widely expressed in the craniofacial structures in rats, while VAX1-knockout mice exhibit phenotype of cleft palate. In humans, the VAX1 mutation could result in an uncharacterized syndrome with bilateral lip, which was one of the clinical features of the patients.[ Several studies have analyzed VAX1 polymorphisms in NSCL/P and showed controversial results. Some previous studies have identified that VAX1 rs7078160 was associated with the risk of NSCL/P in the Estonians and Mesoamericans.[ However, these studies have not found association of VAX1 rs7078160 with NSCL/P in populations from Poland,[ Brazil,[ and southern China.[ Consistent with results of the study conducted by Pan et al[ in a southern Chinese population, we did not found an association between rs7078160 and NSCL/P. Moreover, the MAF of VAX1 rs7078160 in our cases was similar to which in Pan's research in the southern Chinese population (0.47 vs 0.48).[ These results suggested that rs7078160 may not be associated with the susceptibility of NSCL/P in Chinese population. Recently, de Aquino et al[ conducted a study and revealed NSCL/P risk was associated with rs10787760, rs6585429, and rs1871345 of VAX1 in Brazilian population. In their research, none of the alleles and genotypes showed statistical significance between cases and controls. They found that the frequency of VAX1 GAC haplotype was higher in patients than that in the controls, though the differences were not significant after the Bonferroni correction. However, inconsistent with findings from de Aquino et al,[ haplotype analyses of rs10787760, rs6585429, and rs1871345 in our study showed no significant differences between the cases and controls. The inconsistent results may be due to the genetic heterogeneity among various populations and different causative variants in different haplotypic backgrounds.

In the current study, we replicated the result reported in the study by Mangold et al.[ The association of VAX1 rs4752028 with NSCL/P risk was weak in our study, and the reasons for the discrepancy may be attributed to the existing allelic heterogeneity at this locus in various populations. The MAF for VAX1 rs4752028 in our controls was 0.43, which is distinctly different from that of 0.16 in the European population. Nevertheless, the MAF of VAX1 rs4752028 in our controls was similar to those observed in HapMap CHB populations (0.43 vs 0.35). However, the interpretation of our results has some limitations; after the Bonferroni correction for multiple comparisons, the association was found to be nonsignificant. Fundamentally, multiple markers (comparisons) were used for testing multiple corrections, correcting for spurious associations, and it may be stringent to our research and could lead to a loss of the significant finding. Moreover, our work was a hospital-based case–control study and the selection bias was unavoidable, where the subjects may not be a representative of the general population.

The MDR approach used to explore gene–gene interactions for orofacial clefting has been confirmed.[ Our study failed to find SNP–SNP interactions of VAX1. The discrepancy may be impacted by the sample size, which was insufficient to detect a modest effect of the tested SNP variants.

In conclusion, our study has demonstrated that SNP rs4752028 was involved with the risk of NSCL/P in a northern Chinese Han population, although weak, which to some extent, revealed an association between VAX1 and the risk of NSCL/P. Discrepancy in results may be due to a complex genetic background and environmental exposure among different populations. Therefore, further studies are required to confirm the current data in a larger sample and with various ethnic groups and to determine the association between VAX1 and the risk of NSCL/P.