Association of EPHA3 Gene Polymorphisms with Nonsyndromic Cleft Lip With or Without Cleft Palate

Abstract

Aims:

Nonsyndromic cleft lip with or without palate (NSCL/P) represents a complex condition caused by genetic and environmental factors. The aim of this study was to investigate the relationship between the EPHA3 polymorphisms and NSCL/P.

Materials and Methods:

To investigate the relationship between five EPHA3 single nucleotide polymorphisms (SNPs) and NSCL/P, we selected 180 affected patients and 167 normal controls from the Chinese Han Population. EPHA3 SNPs (rs7650466, rs1398197, rs17801309, rs1054750, and rs7632427) were genotyped using the SNaPshot technique; bioinformatic analyses were performed to determine if any of them were potentially functional SNPs.

Results:

The rs7650466 T allele was associated with the incidence of NSCL/P (OR, 0.211; 95% CI, 0.131-0.338; adjusted p = 4.881 × 10⁻¹⁰) and cleft lip with or without palate (CL/P) (OR, 0.176; 95% CI, 0.104-0.297; adjusted p = 3.617 × 10⁻¹⁰), as well as with protective and dominant effects in both conditions. The rs7650466 T allele could be associated with reduced risk of the malformation. In a bioinformatics analysis, we found potential matching sites (miR-1255a, miR-125a-3p, miR-143, and miR-552) for rs7650466 and preliminarily analyzed its potential function.

Conclusions:

Collectively, our data suggest that the EPHA3 rs7650466 polymorphism confers genetic risk for NSCL/P in the Chinese Han Population. Furthermore, rs7650466 is associated with CL/P incidence in stratified analyses, but not with cleft palate only.

Introduction

Nonsyndromic cleft lip with or without palate (NSCL/P) is one of the most common congenital malformations of the oral and maxillofacial region. Asian and American Indians have the highest NSCL/P prevalence of up to 1/500, which is followed by Europeans with <1/1000. The rate for Africans is the lowest, around 1/2500 (Dixon et al., 2011). NSCL/P congenital malformation often causes many adverse effects, including facial deformity, abnormal swallowing and speech functions, and psychological effects. A series of treatments, such as surgery, speech therapy and, sometimes, psychotherapy, are needed for NSCL/P. Although current medical science is well developed, some NSCL/P treatments are often unsatisfactory. As with other congenital anomalies, reducing the incidence of NSCL/P is desired, which requires knowledge of the causes of this condition. Although the definitive cause of NSCL/P is unclear due to its complex etiology, NSCL/P is accepted to result from an interaction of polygenic and environmental factors. As such, the study of genetic factors associated with NSCL/P is essential for determining the etiology and pathogenesis of this condition.

Cleft lip and palate (CLP) is usually classified as syndromic or nonsyndromic due to the concurrent existence of other congenital malformations or developmental abnormalities. NSCL/P is often divided into cleft lip only (CLO), CLP, and cleft palate only (CPO). CLO and CLP are considered variants of the same malformation and based on their similar epidemiologic features and embryologic timing are grouped to form the group cleft lip with or without cleft palate (CL/P) (Leslie and Marazita, 2013). Studies showed that the EPH receptor (Eph) may be associated with CLP. Ephs are classified into type A and type B, with type B associated with syndromic CL/P, and particularly with craniofrontonasal syndrome (CFNS), which is caused by a mutation in EFNB1. CFNS syndrome is a congenital disease that includes cleft palate, nasal dysplasia, and craniofacial asymmetry (Twigg et al., 2004; Wieland et al., 2004). EFNB1-deficient mice typically exhibit CFNS syndrome, which involves defective secondary palatal growth and decreased primary palatal mesenchymal cell proliferation (Compagni et al., 2003; Davy et al., 2004; Bush and Soriano, 2010). Given the association between Eph type B and CL/P, Eph type A may also be associated with this malformation.

EPHA3 belongs to the type A Eph family (Klein, 2012). EPHA3 is primarily involved in controlling cell adhesion, movement, and contraction as well as regulation of axon guidance and morphogenesis during growth and development (Lawrenson et al., 2002; Smith et al., 2004; Janes et al., 2012, 2014). Animal experiments showed that EPHA3 is expressed in the first and second branchial arches and is highly expressed during the secondary palatine process, which makes EPHA3 a candidate gene for CLP (Agrawal et al., 2014). Other research showed that EPHA3 is strongly expressed in palatal mesenchymal cells during early palatal development (Xavier et al., 2016). A genome-wide association study revealed that rs7632427, located about 3 kB downstream of EPHA3, has a strong regulatory role in the development of NSCL/P (Ludwig et al., 2012). Moreover, rs7632427 was shown to be associated with NSCL/P (Pan et al., 2013).

These studies suggest that EPHA3 plays an important role in the early development of the lip and palate and may be involved in NSCL/P pathogenesis. Therefore, the present study was undertaken to study EPHA3 as a risk factor for NSCL/P within a Chinese cohort.

Materials and Methods

Subjects

A cohort of 180 NSCL/P patients, which included CPO (n = 27; 15%) and CL/P (n = 153; 85%) patients, was identified from individuals who visited the Beijing Stomatological Hospital, Capital Medical University, Beijing, between 2015 and 2017. All included patients had NSCL/P without other congenital malformations. The patients (92 males and 88 females) ranged in age between 3 months and 26 years. A total of 167 sex-matched control samples were identified from the Children's Department of Stomatology, Beijing Stomatological Hospital, Capital Medical University, Beijing. These healthy control samples had no congenital malformations or genetic family history. Control subjects (88 males and 79 females) were between 11 months and 28 years of age. The study protocol was approved by the Ethics Committee of Beijing Stomatological Hospital. Written informed consent for genetic analysis was obtained from all subjects or their guardians.

Single nucleotide polymorphism selection and genotyping assay

Five single nucleotide polymorphisms (SNPs) present in EPHA3 (rs7650466, rs1398197, rs17801309, rs1054750, and rs7632427) were selected based on their presence in the HapMap database for the Chinese Han in Beijing (CHB) population (www.hapmap.org) and the NCBI dbSNP database (www.ncbi.nlm.nih.gov/snp). The minor allele frequency of the selected loci was ≥5% in the Chinese Han Population. Functional SNPs, which may have varying functional effects on protein sequence, transcriptional regulation, RNA splicing, or microRNA (miRNA) binding, were selected for the study. A relevant SNP (rs7632427) from a review of the literature was also selected (Table 1). DNA was extracted from venous blood or saliva using the TIANamp Blood DNA Kit (Tiangen, Beijing, China) or DN39 Saliva Extract Kit (Aidlab, Beijing, China). EPHA3 genotypes of the study subjects were determined using the SNaPshot technique (Applied Biosystems, Warrington, UK), a commonly used approach based on single nucleotide extension of fluorescent labeling. Compared with other techniques, the SNaPshot technique is more robust, accurate, and sensitive. Finally, the DNA sequences were read using GeneMapper^® software version 3.2 (Applied Biosystems).

Table 1.

Information About Five Candidate Single Nucleotide Polymorphisms

SNPs	Variant	MAF	Location
rs7650466	C > T	T = 0.1655/829	3′-UTR
rs17801309	G > A	A = 0.0549/275	Missense
rs7632427	T > C	C = 0.3846/1926	Intergenic
rs1054750	T > C	C = 0.1296/649	Synonymous codon
rs1398197	C > T	A = 0.1126/564	Synonymous codon

SNPs, single nucleotide polymorphisms; MAF, minor allele frequency.

Statistical methods

Each SNP was tested for Hardy-Weinberg equilibrium using the Haploview program version 4.2 (www.broad.mit.edu/mpg/haploview/index.php). Logistic regression analysis was used to test for associations between each SNP allele, genotype, and NSCL/P risk. All of the above statistical calculations were analyzed using Plink software version 1.07 (http://pngu.mgh.harvard.edu/∼purcell/plink/index.shtml). The results were corrected using the Bonferroni correction. A p-value <0.05 was defined as statistically significant. We then performed bioinformatic analyses on all positive SNPs using online software from SNPinfo (https://snpinfo.niehs.nih.gov/cgi-bin/snpinfo/) and miRNASNP (www.bioguo.org/miRNASNP/index.php).

Results

In our study, all five candidate SNPs were in Hardy-Weinberg equilibrium. Logistic regression analysis of alleles revealed that the rs7650466 T allele was associated with NSCL/P incidence (OR, 0.211; 95% CI, 0.131-0.338; p = 9.763 × 10⁻¹¹; adjusted p = 4.881 × 10⁻¹⁰) and was protective against NSCL/P (OR <1, U95 < 1). Compared with individuals carrying the C allele, the reduction in risk for individuals having the rs7650466 T allele was nearly 80%. In the logistic regression of the dominant model, the rs7650466 T allele had a dominant effect on NSCL/P (p = 1.636 × 10⁻¹⁰). The other 4 SNPs showed no statistically significant differences between the NSCL/P and control groups (Table 2).

Table 2.

The Results of Allele Logistic Regression Analysis on the Case and Control Groups

SNPs	Allele	Case (frequency)	Control (frequency)	OR (95% CI)	p^a	p^b	p^c
NSCL/P and control groups
rs7650466	C	328 (0.911)	239 (0.716)	0.211 (0.131-0.338)	9.763 × 10⁻¹¹	4.881 × 10⁻¹⁰	1.636 × 10⁻¹⁰
rs7650466	T	32 (0.089)	95 (0.284)	0.211 (0.131-0.338)	9.763 × 10⁻¹¹	4.881 × 10⁻¹⁰	1.636 × 10⁻¹⁰
rs1398197	C	327 (0.956)	267 (0.967)	1.343 (0.588-3.068)	0.484	1	0.580
rs1398197	T	15 (0.044)	9 (0.033)	1.343 (0.588-3.068)	0.484	1	0.580
rs17801309	G	334 (0.933)	309 (0.942)	1.163 (0.630-2.145)	0.629	1	0.594
rs17801309	A	24 (0.067)	19 (0.058)	1.163 (0.630-2.145)	0.629	1	0.594
rs1054750	T	338 (0.955)	314 (0.952)	0.930 (0.460-1.881)	0.840	1	0.694
rs1054750	C	16 (0.045)	16 (0.048)	0.930 (0.460-1.881)	0.840	1	0.694
rs7632427	T	302 (0.858)	283 (0.852)	0.958 (0.631-1.454)	0.840	1	0.416
rs7632427	C	50 (0.142)	49 (0.148)	0.958 (0.631-1.454)	0.840	1	0.416
CPO and control groups
rs7650466	C	45 (0.833)	239 (0.716)	0.460 (0.207-1.020)	0.056	0.281	0.085
rs7650466	T	9 (0.167)	95 (0.284)	0.460 (0.207-1.020)	0.056	0.281	0.085
rs1398197	C	52 (0.963)	267 (0.967)	1.147 (0.234-5.628)	0.866	1	0.866
rs1398197	T	2 (0.037)	9 (0.033)	1.147 (0.234-5.628)	0.866	1	0.866
rs17801309	G	52 (0.963)	309 (0.942)	0.636 (0.147-2.753)	0.545	1	0.577
rs17801309	A	2 (0.037)	19 (0.058)	0.636 (0.147-2.753)	0.545	1	0.577
rs1054750	T	52 (0.963)	314 (0.952)	0.745 (0.161-3.44)	0.706	1	0.706
rs1054750	C	2 (0.037)	16 (0.048)	0.745 (0.161-3.44)	0.706	1	0.706
rs7632427	T	47 (0.870)	283 (0.852)	0.854 (0.359-2.036)	0.722	1	0.512
rs7632427	C	7 (0.130)	49 (0.148)	0.854 (0.359-2.036)	0.722	1	0.512
CL/P and control groups
rs7650466	C	283 (0.925)	239 (0.716)	0.176 (0.104-0.297)	7.234 × 10⁻¹¹	3.617 × 10⁻¹⁰	6.278 × 10⁻¹¹
rs7650466	T	23 (0.075)	95 (0.284)	0.176 (0.104-0.297)	7.234 × 10⁻¹¹	3.617 × 10⁻¹⁰	6.278 × 10⁻¹¹
rs1398197	C	275 (0.955)	267 (0.967)	1.379 (0.591-3.217)	0.457	1	0.563
rs1398197	T	13 (0.045)	9 (0.033)	1.379 (0.591-3.217)	0.457	1	0.563
rs17801309	G	282 (0.928)	309 (0.942)	1.259 (0.674-2.352)	0.470	1	0.444
rs17801309	A	22 (0.072)	19 (0.058)	1.259 (0.674-2.352)	0.470	1	0.444
rs1054750	T	286 (0.953)	314 (0.952)	0.961 (0.464-1.991)	0.916	1	0.752
rs1054750	C	14 (0.047)	16 (0.048)	0.961 (0.464-1.991)	0.916	1	0.752
rs7632427	T	255 (0.856)	283 (0.852)	0.975 (0.630-1.508)	0.908	1	0.486
rs7632427	C	43 (0.144)	49 (0.148)	0.975 (0.630-1.508)	0.908	1	0.486

p, p value of allele logistic regression analysis.

p, p value after Bonferroni correction.

p, p value of dominant logistic regression analysis.

CPO, cleft palate only; NSCL/P, nonsyndromic cleft lip with or without palate.

In the next stratified analysis, we divided all NSCL/P cases into CPO groups and CL/P groups. None of the 5 SNPs examined was associated with CPO incidence (Table 2). However, the rs7650466 T allele was associated with CL/P incidence (OR, 0.176; 95% CI, 0.104-0.297; p = 7.234 × 10⁻¹¹; adjusted p = 3.617 × 10⁻¹⁰) and had a dominant effect on CL/P (p = 6.278 × 10⁻¹¹; Table 2). Individuals carrying the rs7650466 T allele can have up to an 80% reduction in risk of CL/P. Logistic regression analysis of genotype revealed that there was no significant difference among the 5 SNPs in all groups (p > 0.05).

We next conducted bioinformatics analysis on rs7650466. The potential matching sites of rs7650466 (miR-1255a, miR-125a-3p, miR-143, and miR-552) were identified using SNPinfo software (Fig. 1). Using the miRNA-SNP software analysis, we also found that, under normal circumstances, the rs7650466 C allele was associated with the G allele of the miR-2052 seed region. When T was substituted for the rs7650466 C allele, the binding affinity between the EPHA3 3′-UTR and miR-2052 may decrease, which in turn could alter EPHA3 gene expression levels (Fig. 2).

FIG. 1.

Potential matching sites for rs7650466 using SNPinfo software. miR-1255a, miR-125a-3p, and miR-552 are potential matching sites for the rs7650466 T allele. miR-125a-3p, miR-143, and miR-552 are potential matching sites for the rs7650466 C allele. SNP, single nucleotide polymorphisms.

FIG. 2.

MiRNA-SNP online software forecast results for rs7650466. Normally, the rs7650466 allele of C was associated with the G allele of the miR-2052 seed region. When the rs7650466 allele of C changes to T, the binding ability between the 3′-UTR of the EPHA3 gene and miR-2052 may decrease. miRNA, microRNA.

Discussion

The EPHA3 gene is a member of the Eph receptor family that is located on chromosome 3p11.1. Ephrin A3 regulates cell adhesion, cell movement, and contraction, as well as axon guidance and morphogenesis during growth and development (Lawrenson et al., 2002; Smith et al., 2004; Janes et al., 2012, 2014). Recent evidence suggests that EPHA3 may be a causative gene for NSCL/P (Ludwig et al., 2012; Pan et al., 2013; Agrawal et al., 2014; Xavier et al., 2016). However, several similar studies focused on statistical differences between gene sequences and identified sequence changes that were often located in introns or meaningless areas. These studies also lacked functional analysis of genetic loci and biological mechanisms. In our study, we evaluated associations between functional polymorphic variants of EPHA3 and the risk of NSCL/P in a Chinese Han Population.

The most important finding of the present study is the identification of an association between EPHA3 rs7650466 gene variants and NSCL/P and CL/P. However, the rs7650466 variant was not associated with CPO. Both CL/P and CPO are genetically distinct phenotypes in terms of their inheritance and thus CL/P may have a different genetic background compared with CPO, which could explain the results obtained here. The small number of CPO patients in this population could also explain the lack of association between CPO and rs7650466, which is located in the EPHA3 3′-UTR, a region that is primarily related to miRNA. miRNA is a small noncoding regulatory RNA that serves as a key regulator of around 50% of coding protein genes in the human genome and is involved in nearly all important cellular processes (Ambros, 2004; Bushati and Cohen, 2007; Krol et al., 2010). The miRNA complementary sequence is called the miRNA recognition element (MRE). Most MREs are located in the 3′-UTR of the target messenger RNA. Pairing of miRNA with MRE results in translational inhibition or mRNA instability, which negatively regulates the expression levels of target genes. A single miRNA can correspond to ∼200 transcription products, whereas multiple miRNAs can act on a single target mRNA (Selbach et al., 2008).

In recent years, many studies demonstrated that SNPs located in 3′-UTRs can affect interactions between miRNA and target genes, leading to a series of biological effects. In the current study, SNPinfo online software analysis revealed that the EPHA3 SNP rs7650466 was a potential match site for miR-1255a, miR-125a-3p, miR-143, and miR-552. Using miRNA-SNP software analysis, we also found that the rs7650466 C allele is typically associated with the G allele of the miR-2052 seed region. However, when the rs7650466 C allele changes to T, binding between the EPHA3 3′-UTR and miR-2052 may decrease, which in turn alters EPHA3 expression levels. In our study, the rs7650466 T allele showed a protective effect. Thus, when the C allele mutates to T, the binding capacity of the EPHA3 3′-UTR and miR-2052 may decrease and negatively modulate EPHA3 gene expression, which translates to reduced NSCL/P and CL/P incidence. These bioinformatic analyses provide a solid foundation for further research on the biological effects of functional SNPs in the EPHA3 gene.

Synonymous codons refer to codons that have different base sequences but encode the same amino acids. The frequency of synonymous codons differs such that in the same or different species, different genes typically tend to use one or several specific codons with preference to encode a given amino acid, which is called codon bias. This phenomenon may be related to heterologous gene expression, protein expression level, and gene translation regulation (Powell and Moriyama, 1997; Chiapello et al., 1998; Powell et al., 2003; Liu et al., 2005). In this study, rs1054750 and rs1398197 were synonymous codons, but were associated with NSCL/P incidence.

A missense mutation is a change in a base sequence that produces an amino acid sequence change. Such mutations have varying effects on protein function. In this study, we did not find a correlation between the EPHA3 SNP rs17801309 and NSCL/P onset. Meanwhile, the SNP rs7632427 is located about 3 kB downstream of the EPHA3 gene and was previously suggested to be related to NSCL/P (Ludwig et al., 2012; Pan et al., 2013). However, we found no significant association between rs7632427 and NSCL/P in our study.

Our study has some potential limitations. Although we analyzed 347 samples, this sample size is relatively small for this type of study. In addition, we conducted bioinformatic analyses of rs7650466 in the 3′-UTR, but did not experimentally validate the conclusions. Therefore, we plan to expand the sample size and study the function of rs7650466 in the next stage of our study.

In summary, NSCL/P is a complex disease caused by interaction of multiple genes and environmental factors. Our research contributes to a better understanding of the role of genetic factors in NSCL/P pathogenesis. We discovered that the EPHA3 gene polymorphism rs7650466 was associated with NSCL/P incidence in a Chinese Han population. Furthermore, in stratified analysis, rs7650466 was associated with incidence of CL/P, but not CPO. Larger study populations and functional studies of rs7650466 are needed to confirm our findings.

Footnotes

Acknowledgments

The authors thank all the patients and their families who contributed to this study. This work was supported by the Beijing Municipal Natural Science Foundation (Grant No. 7162074).

Author Disclosure Statement

No competing financial interests exist.

References

Agrawal

, Wang

, Kim

, et al. (2014) Embryonic expression of EphA receptor genes in mice supports their candidacy for involvement in cleft lip and palate. Dev Dyn, 243:1470-1476.

Ambros

(2004) The functions of animal microRNAs. Nature, 431:350-355.

Bush

, Soriano

(2010) Ephrin-B1 forward signaling regulates craniofacial morphogenesis by controlling cell proliferation across Eph-ephrin boundaries. Genes Dev, 24:2068-2080.

Bushati

, Cohen

(2007) microRNA functions. Annu Rev Cell Dev Biol, 23:175-205.

Compagni

, Logan

, Klein

, et al. (2003) Control of skeletal patterning by ephrinB1-EphB interactions. Dev Cell, 5:217-230.

Chiapello

, Lisacek

, Caboche

, et al. (1998) Codon usage and gene function are related in sequences of Arabidopsis thaliana. Gene, 209:GC1-GC38.

Dixon

, Marazita

, Beaty

, et al. (2011) Cleft lip and palate: understanding genetic and environmental influences. Nat Rev Genet, 12:167-178.

Davy

, Aubin

, Soriano

(2004) Ephrin-B1 forward and reverse signaling are required during mouse development. Genes Dev, 18:572-583.

Janes

, Nievergall

, Lackmann

(2012) Concepts and consequences of Eph receptor clustering. Semin Cell Dev Biol, 23:43-50.

10.

Janes

, Slape

, Farnsworth

, et al. (2014) EphA3 biology and cancer. Growth Factors, 32:176-189.

11.

Klein

(2012) Eph/ephrin signalling during development. Development, 139:4105-4109.

12.

Krol

, Loedige

, Filipowicz

(2010) The widespread regulation of microRNA biogenesis, function and decay. Nat Rev Genet, 11:597-610.

13.

Lawrenson

, Wimmer-Kleikamp

, Lock

, et al. (2002) Ephrin-A5 induces rounding, blebbing and de-adhesion of EphA3-expressing 293T and melanoma cells by CrkII and Rho-mediated signaling. J Cell Sci, 115:1059-1072.

14.

Leslie

, Marazita

(2013) Genetics of cleft lip and cleft palate. Am J Med Genet C Semin Med Genet, 163C:246-258.

15.

Ludwig

, Mangold

, Herms

, et al. (2012) Genome-wide meta-analyses of nonsyndromic cleft lip with or without cleft palate identify six new risk loci. Nat Genet, 44:968-971.

16.

Liu

, Dou

, Ji

, et al. (2005) Synonymous codon usage and gene function are strongly related in Oryza sativa. Biosystems, 80:123-131.

17.

Pan

, Han

, Zhang

, et al. (2013) Association and cumulative effects of GWAS-identified genetic variants for nonsyndromic orofacial clefts in a Chinese population. Environ Mol Mutagen, 54:261-267.

18.

Powell

, Sezzi

, Moriyama

, et al. (2003) Analysis of a shift in codon usage in Drosophila. J Mol Evol, 57 Suppl 1:S214-S225.

19.

Powell

, Moriyama

(1997) Evolution of codon usage bias in Drosophila. Proc Natl Acad Sci U S A, 94:7784-7790.

20.

Smith

, Vearing

, Lackmann

, et al. (2004) Dissecting the EphA3/Ephrin-A5 interactions using a novel functional mutagenesis screen. J Biol Chem, 279:9522-9531.

21.

Selbach

, Schwanhäusser

, Thierfelder

, et al. (2008) Widespread changes in protein synthesis induced by microRNAs. Nature, 455:58-63.

22.

Twigg

, Kan

, Babbs

, et al. (2004) Mutations of ephrin-B1 (EFNB1), a marker of tissue boundary formation, cause craniofrontonasal syndrome. Proc Natl Acad Sci U S A, 101:8652-8657.

23.

Wieland

, Jakubiczka

, Muschke

, et al. (2004) Mutations of the ephrin-B1 gene cause craniofrontonasal syndrome. Am J Hum Genet, 74:1209-1215.

24.

Xavier

, Miletich

, Cobourne

(2016) Ephrin ligands and EPH receptors show regionally restricted expression in the developing palate and tongue. Front Physiol, 7:60.