Association between MSX1 SNPs and nonsyndromic cleft lip with or without cleft palate in the Korean population.

The purpose of this study was to investigate the contribution of MSX1 gene to the risk of nonsyndromic cleft lip with or without cleft palate (NS-CL ± P) in the Korean population. The samples consisted of 142 NS-CL ± P families (9 with cleft lip, 26 with cleft lip and alveolus, and 107 with cleft lip and palate; 76 trios and 66 dyads). Three single nucleotide polymorphisms (SNPs: rs3821949, rs12532, and rs4464513) were tested for association with NS-CL ± P case-parent trios using transmission disequilibrium test (TDT) and conditional logistic regression models (CLRMs). Minor allele frequency, heterozygosity, χ(2) test for Hardy-Weinberg equilibrium, and pairwise linkage disequilibrium (LD) at each SNP were computed. The family- and haplotype-based association test programs were used to perform allelic and genotypic TDTs for individual SNPs and to fabricate sliding windows of haplotypes. Genotypic odds ratios (GORs) were obtained from CLRMs using R software. Although the family-based TDT indicated a meaningful association for rs3821949 (P = 0.028), the haplotype analysis did not reveal any significant association with rs3821949, rs12532, or rs4464513. The A allele at rs3821949 had a significant increased risk of NS-CL ± P (GOR, 1.64; 95% confidence interval,1.03-2.63; P = 0.038, additive model). A positive association is suggested between MSX1 rs3821949 and NS-CL ± P in the Korean population.

INTRODUCTION

Nonsyndromic cleft lip and/or palate (NS-CL/P) is a common congenital craniofacial deformity in humans and is known to be caused by a combination of genes and environmental interactions. The frequency of NS-CL/P is higher in Asian populations (1/500 or higher) than in Caucasian (1/1,000) or African populations (1/2,500) (1-4).

The muscle segment homeobox1 (MSX1) gene at 4p16.1 encodes a DNA-binding sequence and is expressed in spatially-restricted regions of the head during early development. Mutations in this gene have been known to be associated with NS-CL/P, Witkop syndrome, Wolf-Hirschhorn syndrome, and autosomal dominant hypodontia (5-7). In animals, homozygous Msx1-deficient transgenic mice exhibit cleft palate, deficiency of the alveolar bone in the mandible and maxilla, incisor development failure, and arrest of molar development (8, 9). Gong (10) showed that there was misregulation in the expression of the Msx1 gene in embryos of A/WySn mice with cleft palate.

Complete sequencing of MSX1 in humans has revealed several novel mutations, and it is estimated that approximately 2% of NS-CL/P patients carry mutations in this gene (6, 7, 11). There have been several association and linkage studies between MSX1 gene variants and NS-CL/P in humans (12-18). However, in terms of the Korean population, only one article has been published for association and linkage study between MSX1 single nucleotide polymorphisms (SNP) and NS-CL/P.

Park et al. (19) reported significant evidence of linkage in the presence of disequilibrium for 1170G/A of exon 2 and the disease risk decreased with the presence of the A allele (AA genotype: odds ratio, 0.26; 95% confidence interval [CI], 0.10-0.99). However, their study has several limitations as follows: They analyzed only novel SNPs with low occurrence frequency. The samples included cleft lip, cleft lip and palate, and cleft palate only. However, cleft palate only has been regarded as a separated identity etiologically and embryologically from clefts involving the lip with or without the palate (20). In addition, when trio-case with cleft palate only (n = 8) was excluded, the numbers of trio-case with cleft lip with or without palate was relatively small (n = 44; 14 with cleft lip, and 30 with cleft lip and palate).

Therefore, it is needed to determine the association and linkage relationship about tag SNPs in the Korean population and to confine the samples with nonsyndromic cleft lip with or without cleft palate (NS-CL ± P) in order to get meaningful results for epidemiologic studies of cleft patients. The purpose of this study was to investigate the contribution of the MSX1 gene to the risk of NS-CL ± P in the Korean population, whose samples are independent form those of Park et al. (19). We tested three tag single nucleotide polymorphism (SNP) markers in and around the MSX1 gene in Korean NS-CL ± P case-parent families using the transmission disequilibrium test (TDT) and conditional logistic regression models (CLRM).

MATERIALS AND METHODS

Subjects

The sample population consisted of 142 Korean NS-CL ± P families (90 males and 52 females; 9 with cleft lip, 26 with cleft lip and alveolus, and 107 with cleft lip and palate; 76 trios and 66 dyads, Table 1). Five orthodontists performed clinical investigation to diagnose NS-CL ± P. For mutation analysis and case-control studies, peripheral venous blood samples of patients and their parents were collected at either Seoul National University Dental Hospital (SNUDH) or Samsung Medical Center (SMC).

Table 1

Demographic information of gender and cleft type

Each numerical number is the number of actual patient with CL, CLA, and CLP. CL, cleft lip only; CLA, cleft lip and alveolus; CLP, cleft lip and palate.

Extraction of genomic DNA and genotyping

Genomic DNA samples were extracted from peripheral venous blood lymphocytes using a commercial DNA extraction kit (Quiagen Inc., Valencia, CA, USA) and were genotyped using VeraCode Technology® (Illumina Inc., San Diego, CA, USA) at SNP Genetics Inc. (Seoul, Korea).

Selection of SNPs

SNP markers located from 2kb-5' to 2kb-3' of the MSX1 gene were obtained from literature review and the NCBI dbSNP database (http://www.ncbi.nlm.nih.gov/SNP/). Among tag-SNPs covering all SNPs in the MSX1 gene, three SNP markers with minor allele frequency (MAF) greater than 1% in the Japanese population (rs3821949, rs12532, and rs4464513) were selected using the web-based program TAG SNP selection (TagSNP; http://snpinfo.niehs.nih.gov/guide.htm#snptag) (21).

These SNPs achieved high "design scores" (a predictor of usable genotypes provided by Illumina Inc., San Diego, CA, USA). Their heterozygosity was greater than 0.1 in the Japanese population (www.hapmap.org/index.html.en). The genotype call rate and sample call rate were considered acceptable at ≥ 95%. Primers for each SNP were synthesized using Oligator technology (Illumina Inc.).

Statistical analysis

The MAF, heterozygosity, and χ2 test for the Hardy-Weinberg equilibrium (HWE) at each SNP were computed using the genotypes of parents. Pairwise linkage disequilibrium (LD) was computed as both D' and r for all SNPs using the Haploview program (http://www.broad.mit.edu/mpg/haploview/index.php/)(22-24).

The family-based association test program was used to perform allelic and genotypic transmission disequilibrium tests (TDTs) for individual SNPs and the haplotype-based association test program was used to fabricate sliding windows of haplotypes consisting of two and three SNPs (http://www.biostat.harvard.edu/fbat/default.html) (24, 25). The permutation option (26) was used to obtain empirical P values for observed versus expected transmission and to compute the -log10 P value for each SNP/haplotype within the MSX1 gene.

Genotypic odds ratios (GORs) for heterozygotes and homozygotes under additive, dominant, and recessive models were calculated separately for individual SNPs. A matched case-control dataset was generated with each NS-CL ± P case matched to three possible pseudo-control subjects created from the non-transmitted parental allele (24). The GORs were obtained from conditional logistic regression models for matched sets using publicly available subroutines in the R software (www.r-project.org).

Ethics statement

The study protocol was reviewed and approved by the institutional review board at each institution (SNUDH IRB CRI-G07002 and SMC IRB #2007-08-086, respectively). Informed consent was received from each subject before the sampling.

RESULTS

Demographic information for proband gender and cleft type

One hundred forty-two Korean NS-CL ± P families had 76 trios and 66 dyads. The cases included 90 males and 52 females; 9 with cleft lip (CL), 26 with cleft lip and alveolus (CLA), and 107 with cleft lip and palate (CLP) patients (Table 1).

MAF, heterozygosity, HWE, and LD analyses

The MAF for rs4464513 was lowest (0.295), while the MAFs for rs12532 and rs3821949 showed values of 0.359 and 0.484, respectively. None of the three SNPs exhibited any evidence of deviation from HWE (Table 2). Among three SNPs, the values of Pairwise LD (D'/r) for pairs of rs3821949-rs12532, rs12532-rs4464513, and rs3821949-rs4464513 were 0.56/0.16, 0.74/0.21, and 0.97/0.71, respectively.

Table 2

Marker information and transmission disequilibrium test (TDT) results for three single nucleotide polymorphisms (SNPs) in the MSX1 gene in cleft lip with or without cleft palate (CL±P) in 142 CL±P families

*Over-transmitted alleles are in bold type; †Transmission/non-transmission counts from heterozygous parents; ‡Significant P values for individual SNP and global P values for sliding windows of haplotypes of two and three SNPs from TDT analyses; §P < 0.05. MAF, minor allele frequency; HWE, Hardy-Weinberg equilibrium; TDT, transmission disequilibrium test; SNP, single nucleotide polymorphism.

TDT analyses for individual markers and haplotypes

Although the family-based TDT using individual SNPs indicated a significant association for rs3821949 (P = 0.028, Table 2), the haplotype analysis did not reveal any significant association with rs3821949, rs12532, or rs4464513 (Table 2). Because three SNPs used in this study are tag SNPs which are independent of each other, there was no synergic interaction between these SNPs and significance in haplotype analysis could not be exhibited.

Genotypic odds ratios (GOR)

The A allele at rs3821949 had a significant increased risk of NS-CL ± P in an additive model (GOR, 1.64; 95% CI, 1.03-2.63; P = 0.0384, Table 3).

Table 3

Genotypic odds ratios (GORs) for heterozygotes and homozygotes for individual SNPs in 76 NS-CL±P trios

LD represents linkage disequilibrium; CI, confidence interval. *'N' and 'n' refer to the number of subjects carrying the genotype and the number of case/pseudo-control sets generated, respectively; †P values of χ2 tests for the conditional logistic regression model for each SNP; ‡P < 0.05.

DISCUSSION

Single marker analysis and genotypic odds ratio analysis in the present study exhibited that rs3821949 has a meaningful P value (P = 0.028, Table 2) and an increased risk of NS-CL ± P under an additive model (GOR, 1.64; 95% CI, 1.03-2.63; P = 0.038, Table 3). Huang et al. (27) also reported similar results for rs3821949 in the Han Chinese population living in Western China. However, if the number of samples were to be increased, there may be a more significant association in rs3821949 when considering the relatively low frequency (2%) of MSX1 mutations in NS-CL/P patients (6, 7). There is no reported result about association between MSX1 rs3821949 and NS-CL ± P in non-Asian population until now. Therefore, it is needed to confirm the ethnic differences in other populations.

In this study, the A allele at rs3821949 appears to increase the risk of NS-CL ± P, while the G allele is under-represented (Tables 2, 3). Huang et al. (27) suggested that the cleft lip and palate (CLP) patients showed a significant difference in allele frequency of rs3821949 (GG vs GA/AA) compared to cleft lip (CL) and cleft palate only (CP) patients. Therefore, further study is needed to investigate whether the association of SNP in rs3821949 is different according to cleft type with a large number of cases.

Numerous previous studies have investigated the role of the MSX1 gene in the etiology of NS-CL/P in different human populations (12, 27-30). Among the proposed pathogenic mutations, a rare SNP, P147Q, has been a mutation of interest. It has been found in approximately 2% of Vietnamese (7) and 1.2% of Han Chinese (27). However, Tongkobpetch et al. (11) reported that the P147Q mutation could not be pathogenic because there was no association between the P147Q variant and NS-CL/P in the Thai population. Since this study tested only three SNPs whose MAF was greater than 1.0% in the Japanese population (rs3821949, rs12532, and rs4464513) (21), the P147Q variant was not included as a SNP in the MSX1 gene in this study. Therefore, further study of the association of the P147Q variant with NS-CL/P in the Korean population might be needed.

The cause and time of formation of clefts vary; cleft lip and alveolus (CLA) results from fusion failure between the medial nasal process and maxillary process in the primary palate (lip and premaxilla), which takes place during the fourth to the seventh week of gestation; cleft palate (CP) results from fusion failure between the palatal processes in the secondary palate, which develops during the seventh to the twelfth week (20). Therefore, the developmental classification between CLA and CLP is needed for epidemiologic studies of cleft patients. However, considering the number of samples in the present study (cleft in the primary palate [n = 35; 9 with cleft lip and 26 with cleft lip and alveolus] and cleft in the primary and secondary palate [n = 107 with cleft lip and palate], Table 1), it is needed to increase the number of the cleft patients with lip and/or alveolus (CL and CLA) for investigating the possibility of difference in association and linkage between MSX1 SNPs and cleft type.

In conclusion, the results from this study suggest a positive association between MSX1 rs3821949 and NS-CL ± P in the Korean population.