Association of BRCA2 Gene Functional Polymorphisms with Nonsyndromic Cleft Lip With or Without Cleft Palate in a Chinese Population

Abstract

Background:

Nonsyndromic cleft lip with or without cleft palate (NSCL/P) is a complex congenital disease affected by genetic and environmental factors however, the specific pathogenic alleles and regulatory mechanisms remain unclear in many cases. Here, we aimed to study the association between eight potentially functional single nucleotide polymorphisms (SNPs) of the BRCA2 and MGMT genes and NSCL/P in a Chinese population through a case-control study.

Materials and Methods:

To investigate the relationship between potentially functional SNPs of the BRCA2 and MGMT genes and NSCL/P, we selected 200 affected patients and 200 unrelated normal controls in a Chinese population. The BRCA2 gene SNPs (rs11571836, rs144848, rs7334543, rs15869, rs766173 and rs206118) and MGMT gene SNPs (rs12917 and rs7896488) were genotyped using the SNaPshot technique and the resulting data were subjected to statistical and bioinformatic analyses.

Results:

Our study identified for the first time that alleles of the BRCA2 are associated with NSCL/P in a Chinese population and that the s11571836 G allele was protective against NSCL/P. Under four genetic models, rs11571836 had a significant correlation with NSCL/P. Preliminary bioinformatic analyses revealed four potential miRNA matching sites (miR-1244, miR-1323, miR-562, and miR-633) associated with the rs11571836 which is located in the 3′ untranslated region of BRCA2.

Conclusions:

These results support the role of polymorphisms of BRCA2 gene in affecting susceptibility to NSCL/P and its progression, but further research is necessary to elucidate the mechanism by which the BRCA2 gene polymorphisms affect the penetrance of NSCL/P.

Introduction

Cleft lip and/or palate (CL/P) is one of the most common congenital malformations of the maxillofacial region, with a prevalence of ∼1 in 700 live births. Asian and American Indians have the highest incidence rate, while African Americans have the lowest (Dixon et al., 2011). Patients with this condition may have negative effects in a variety of areas, including feeding, speaking, hearing, and social integration.

Therefore, some patients may suffer lifelong adverse effects from the deformity. CL/P is classified as syndromic cleft lip and palate and nonsyndromic cleft lip with or without cleft palate (NSCL/P). As an example, Van der Woude syndrome, which is mostly brought on by mutations in the interferon regulatory factor 6 and grainyhead like transcription factor 3 genes, has a well-studied etiology and pathology (Kondo et al., 2002; Peyrard-Janvid et al., 2014). In contrast, the understanding of the etiology and pathogenesis of NSCL/P, which accounts for ∼70% of CL/P, is very limited.

It is generally accepted that NSCL/P is a genetic disease with a complex etiology caused by genetic and environmental factors. Numerous research has identified a number of genes linked to the prevalence of NSCL/P, although the findings are often unsatisfactory and poorly reproducible. More importantly, previous studies have focused on statistical differences, while neglecting the functional analysis of genetic loci and biological mechanisms.

The process of facial embryonic development involves multiple factors and signaling pathways that together participate in regulating various physiological processes such as cell proliferation, migration, differentiation, and apoptosis. Furthermore, if any of these regulatory components' functions is hampered, it may have an impact on the fusion association of the facial synapse and lead to the emergence of NSCL/P (Ji et al., 2020). Previous studies have found genetic overlap between cancer and NSCL/P and a possible common etiological background (Dunkhase et al., 2016; Wang et al., 2016).

DNA damage repair genes are critical for embryonic development and their abnormalities dramatically increase the risk of cancer. Many animal experiments have demonstrated that mutations in DNA damage repair genes lead to various embryonic developmental abnormalities, including stillbirths, proliferation defects, genetic instability, and developmental malformations of organs, especially craniofacial tissues (Deng and Wang, 2003). The network of genes controlling cellular defense against DNA damage plays a role in the etiology of NSCL/P.

Dysregulation of co-expressed gene networks mainly associated with DNA double strand breaks (DSBs) repair and cell cycle control was found in dental pulp stem cells of NSCL/P patients (Kobayashi et al., 2013). Some adverse environmental factors may lead to oxidative DNA damage, including DSBs, which may increase the risk of developing craniofacial cleft (Feltes et al., 2013). A crucial role for DNA damage response in the formation of the palate has been established by animal research. Furthermore, the breast cancer susceptibility gene (BRCA)1/2-p53-dependent pathway is key in regulating normal palatogenesis in mice (Yamaguchi et al., 2021).

Therefore, DNA damage repair-related genes are closely related to the pathogenesis of NSCL/P, among which, BRCA2 and Methylation of O6-methylguanine-DNA methyltransferase (MGMT) genes are key genes. The BRCA2 gene is required for craniofacial bone development (Kitami et al., 2018) and was found to be weakly associated with NSCL/P in non-Hispanic whites (NHW) and Hispanics (Rodriguez et al., 2018). MGMT is a DNA methyltransferase that repairs damaged DNA by transferring methyl groups at the guanine O6 site to cysteine residues, avoiding gene mutations, cell death, and tumorigenesis caused by alkylating agents (Yu et al., 2020).

10q26.3 distal deletion was associated with cleft palate and the region of deletion contained the MGMT (Plaisancié et al., 2014). Besides, rs7922405, located in the intron of MGMT gene, was associated with NSCL/P (Chen et al., 2018). However, the specific association and mechanism between BRCA2, MGMT gene, and NSCL/P in the Chinese population are still not clear. Therefore, we selected eight functional single nucleotide polymorphisms (SNPs) in BRCA2 and MGMT genes in this case-control study and performed preliminary bioinformatics analysis of the positive results.

Materials and Methods

Subjects

Two hundred NSCL/P patients, 71 males and 129 females, 3 months to 18 years of age, were collected as a case group from Beijing Stomatological Hospital affiliated to Capital Medical University. Among the case group, there were 59 cases of cleft lip only and 141 of cleft lip with cleft palate. The control group was recruited in the Beijing area and consisted of 200 cases, who had no congenital malformations and genetic family history. Among them, there were 80 males and 120 females, ranging from 2 to 25 years old. Besides, the subjects in the case and control groups were unrelated.

Variant selection and genotyping

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of Beijing Stomatological Hospital affiliated to Capital Medical University (No. CMUSH-IRB-KJ-PJ-2020-01). Written informed consent for genetic analysis was obtained from all subjects or their guardians. In the case group, 2 mL of venous blood was drawn, anticoagulated with EDTA, and stored at −80°C. In the control group, 2 mL of saliva was collected by DN-39 saliva collector (Aidlab, Beijing, China) and stored at room temperature. DNA was extracted from venous blood or saliva using a TIANamp blood DNA kit (Tiangen, Beijing, China) or a DN-39 saliva extraction kit (Aidlab). Spectrophotometry and agarose gel electrophoresis were used to determine the quantity and quality of extracted DNA.

The databases selected for functional variants were the HapMap (http://hapmap.ncbi.nlm.nih.gov) and NCBI dbSNP (www.ncbi.nlm.nih.gov/snp) databases. The functional variants for BRCA2 and MGMT genes in Chinese Han population were queried, and the screening criteria were as follows: (1) variants with a minor allele frequency of 0.05 (5%) or greater in Chinese Han population and (2) variants were selected at positions such as frameshift variants, missense variants, 5′ untranslated region (UTR), and 3’UTR. Eight functional variants were finally selected for this study (Table 1).

Table 1.

Information About Eight Candidate Single Nucleotide Polymorphisms

Genes	SNPs	Variant	MAF	Function
BRCA2	rs11571836	A > G	0.32	3′UTR
	rs144848	T > G	0.284	Missense
	rs7334543	A > G	0.073	3′UTR
	rs15869	A > C	0.23	3′UTR
	rs766173	T > G	0.098	Missense
	rs206118	T > C	0.162	5′UTR
MGMT	rs12917	C > T	0.13	Missense
MGMT	rs7896488	G > A	0.238	3′UTR

MAF, minor allele frequency; SNPs, single nucleotide polymorphisms; UTR, untranslated region.

Primers were designed and synthesized using Primer5 software based on gene sequence information. In this study, genotyping was performed by SNaPshot technique according to the manufacturer's protocols (Applied Biosystems, Warrington, United Kingdom). An ABI 3730XL DNA analyzer was used for sequencing and the DNA sequences were read using GeneMapper version 3.2 software (Applied Biosystems). The variant site corresponding to the extension product was identified based on the position of the peak. Then, the type of base added could be known according to the color of the peak, so as to determine the genotype of the sample. After the results were obtained, the positive results were evaluated for preliminary function by bioinformatics softwares.

Data analysis

Haploview version 4.2 program (www.broad.mit.edu/mpg/haploview/index.php) was applied to perform the Hardy-Weinberg Equilibrium (HWE) test. The chi-square (χ²) test was used to calculate whether the frequencies of each allele were statistically different between cases-controls by Plink version 1.07 software. Correlation between five genotypic test models (codominant, dominant, recessive, overdominant, and log additive) and NSCL/P was analyzed by unconditional logistic regression using the SNPStats online program. The optimal genetic model was determined according to the Akaike information criterion (AIC), and the genetic model with the lowest AIC value was the optimal genetic model for the variant. Bonferroni correction was applied to all results. The significance level in all tests was 0.05. Quanto version 1.2.4 software was used to evaluate the statistical power.

The potential function of positive variants was analyzed using bioinformatics tools, including SNPinfo (https://snpinfo.niehs.nih.gov) (Xu and Taylor, 2009), Fathmm (http://fathmm.biocompute.org.uk) (Shihab et al., 2013), PolyPhen2 (http://genetics.bwh.harvard.edu/pph2) (Adzhubei et al., 2010), and MutationTaster2021 (Steinhaus et al., 2021).

Results

A total of 200 NSCL/P cases and 200 controls were recruited to study the relationship between the functional SNPs of BRCA2, MGMT gene, and NSCL/P in a Chinese population. All SNPs included in this study were genotyped in both the case and control groups and belonged to HWE in the control group. The results of the χ² test showed that the frequency of BRCA2 rs11571836 G allele was statistically significant between the case and control groups (odds ratio [OR] = 0.6321, confidence interval [95% CI] = 0.4657-0.8579, adjusted p = 0.025). There was no positive result for the remaining SNPs (Table 2).

Table 2.

Comparison of Allele Frequencies of Eight Single Nucleotide Polymorphisms Between Case Group and Control Group

Genes	SNPs	Allele	Case (Freq)	Control (Freq)	OR (95% CI)	χ²	p^a	p^b
BRCA2	rs11571836	A	275 (0.73)	253 (0.63)	0.6321 (0.4657-0.8579)	8.715	0.003	0.025
	rs11571836	G	101 (0.27)	147 (0.37)	0.6321 (0.4657-0.8579)	8.715	0.003	0.025
	rs144848	T	270 (0.71)	287 (0.72)	1.021 (0.7474-1.396)	0.01771	0.894	1
	rs144848	G	108 (0.29)	113 (0.28)	1.021 (0.7474-1.396)	0.01771	0.894	1
	rs7334543	A	354 (0.94)	367 (0.92)	0.754 (0.4369-1.301)	1.034	0.309	1
	rs7334543	G	24 (0.06)	33 (0.08)	0.754 (0.4369-1.301)	1.034	0.309	1
	rs15869	A	301 (0.8)	298 (0.74)	0.7474 (0.5338-1.046)	2.887	0.089	0.715
	rs15869	C	77 (0.2)	102 (0.26)	0.7474 (0.5338-1.046)	2.887	0.089	0.715
	rs766173	T	335 (0.89)	367 (0.92)	1.427 (0.8858-2.3)	2.154	0.142	1
	rs766173	G	43 (0.11)	33 (0.08)	1.427 (0.8858-2.3)	2.154	0.142	1
	rs206118	T	311 (0.82)	341 (0.85)	1.245 (0.8496-1.825)	1.267	0.26	1
	rs206118	C	67 (0.18)	59 (0.15)	1.245 (0.8496-1.825)	1.267	0.26	1
MGMT	rs12917	C	333 (0.88)	344 (0.86)	0.8301 (0.5453-1.264)	0.7552	0.385	1
	rs12917	T	45 (0.12)	56 (0.14)	0.8301 (0.5453-1.264)	0.7552	0.385	1
	rs7896488	G	277 (0.73)	314 (0.79)	1.363 (0.9785-1.899)	3.365	0.067	0.533
	rs7896488	A	101 (0.27)	84 (0.21)	1.363 (0.9785-1.899)	3.365	0.067	0.533

p Value was calculated by the χ² test.

p Value was calculated by the χ² test after Bonferroni correction. Significant p values are shown in bold.

95% CI, 95% confidence interval; Freq, allele frequency; OR, odds ratio.

Unconditional logistic regression analysis showed that rs11571836 was significantly associated with NSCL/P under codominant, dominant, overdominant, and log additive (codominant model: OR [G/A vs. A/A] = 2.78, 95% CI [G/A vs. A/A] = 1.79-4.33, OR [G/G vs. A/A] = 1.4, 95% CI [G/G vs. A/A] = 0.74-2.66, adjusted p < 0.0001; dominant mode: G/A-G/G vs. A/A, OR = 2.35, 95% CI = 1.56-3.54, adjusted p < 0.0001; overdominant model: G/A vs. A/A-G/G, OR = 2.60, 95% CI = 1.70-3.96, adjusted p < 0.0001; and log-additive model: OR = 1.53, 95% CI = 1.14-2.06, adjusted p = 0.035). The overdominant model was the optimal genetic model of rs11571836 (AIC = 521.1) (Table 3) with a power value of 96.39%.

Table 3.

Association Between Eight Single Nucleotide Polymorphisms and Nonsyndromic Cleft Lip With or Without Cleft Palate Under Different Genetic Models

SNPs	Model	Genotype	Case (Freq.)	Control (Freq.)	OR (95% CI)	p^a	p^b	AIC
rs11571836	Codominant	A/A	111 (59%)	76 (38%)	1	<0.0001	<0.0001	522.1
		G/A	53 (28.2%)	101 (50.5%)	2.78 (1.79-4.33)
		G/G	24 (12.8%)	23 (11.5%)	1.40 (0.74-2.66)
	Dominant	A/A	111 (59%)	76 (38%)	1	<0.0001	<0.0001	524.2
	Dominant	G/A-G/G	77 (41%)	124 (62%)	2.35 (1.56-3.54)	<0.0001	<0.0001	524.2
	Recessive	A/A-G/A	164 (87.2%)	177 (88.5%)	1	0.7	1	541.4
	Recessive	G/G	24 (12.8%)	23 (11.5%)	0.89 (0.48-1.63)	0.7	1	541.4
	Overdominant	A/A-G/G	135 (71.8%)	99 (49.5%)	1	<0.0001	<0.0001	521.1
	Overdominant	G/A	53 (28.2%)	101 (50.5%)	2.60 (1.70-3.96)	<0.0001	<0.0001	521.1
	Log additive	—	—	—	1.53 (1.14-2.06)	0.004	0.035	533.4
rs144848	Codominant	T/T	104 (55%)	99 (49.5%)	1	0.017	0.136	536.8
		G/T	62 (32.8%)	89 (44.5%)	1.51 (0.99-2.31)
		G/G	23 (12.2%)	12 (6%)	0.55 (0.26-1.16)
	Dominant	T/T	104 (55%)	99 (49.5%)	1	0.28	1	541.8
	Dominant	G/T-G/G	85 (45%)	101 (50.5%)	1.25 (0.84-1.86)	0.28	1	541.8
	Recessive	T/T-G/T	166 (87.8%)	188 (94%)	1	0.032	0.256	538.4
	Recessive	G/G	23 (12.2%)	12 (6%)	0.46 (0.22-0.95)	0.032	0.256	538.4
	Overdominant	T/T-G/G	127 (67.2%)	111 (55.5%)	1	0.018	0.144	537.3
	Overdominant	G/T	62 (32.8%)	89 (44.5%)	1.64 (1.09-2.48)	0.018	0.144	537.3
	Log additive	—	—	—	0.98 (0.73-1.34)	0.92	1	542.9
rs7334543	Codominant	A/A	167 (88.4%)	169 (84.5%)	1	0.51	1	543.6
		G/A	20 (10.6%)	29 (14.5%)	1.43 (0.78-2.63)
		G/G	2 (1.1%)	2 (1%)	0.99 (0.14-7.10)
	Dominant	A/A	167 (88.4%)	169 (84.5%)	1	0.27	1	541.7
	Dominant	G/A-G/G	22 (11.6%)	31 (15.5%)	1.39 (0.77-2.50)	0.27	1	541.7
	Recessive	A/A-G/A	187 (98.9%)	198 (99%)	1	0.95	1	543
	Recessive	G/G	2 (1.1%)	2 (1%)	0.94 (0.13-6.77)	0.95	1	543
	Overdominant	A/A-G/G	169 (89.4%)	171 (85.5%)	1	0.24	1	541.6
	Overdominant	G/A	20 (10.6%)	29 (14.5%)	1.43 (0.78-2.63)	0.24	1	541.6
	Log additive	—	—	—	1.30 (0.77-2.21)	0.32	1	542
rs15869	Codominant	A/A	117 (61.9%)	107 (53.5%)	1	0.2	1	541.7
		C/A	67 (35.5%)	84 (42%)	1.37 (0.91-2.08)
		C/C	5 (2.6%)	9 (4.5%)	1.97 (0.64-6.06)
	Dominant	A/A	117 (61.9%)	107 (53.5%)	1	0.093	0.744	540.1
	Dominant	C/A-C/C	72 (38.1%)	93 (46.5%)	1.41 (0.94-2.12)	0.093	0.744	540.1
	Recessive	A/A-C/A	184 (97.3%)	191 (95.5%)	1	0.32	1	542
	Recessive	C/C	5 (2.6%)	9 (4.5%)	1.73 (0.57-5.27)	0.32	1	542
	Overdominant	A/A-C/C	122 (64.5%)	116 (58%)	1	0.18	1	541.2
	Overdominant	C/A	67 (35.5%)	84 (42%)	1.32 (0.88-1.99)	0.18	1	541.2
	Log additive	—	—	—	1.38 (0.97-1.97)	0.073	0.584	539.7
rs766173	Codominant	T/T	152 (80.4%)	169 (84.5%)	1	0.25	1	542.2
		G/T	31 (16.4%)	29 (14.5%)	0.84 (0.48-1.46)
		G/G	6 (3.2%)	2 (1%)	0.30 (0.06-1.51)
	Dominant	T/T	152 (80.4%)	169 (84.5%)	1	0.29	1	541.8
	Dominant	G/T-G/G	37 (19.6%)	31 (15.5%)	0.75 (0.45-1.27)	0.29	1	541.8
	Recessive	T/T-G/T	183 (96.8%)	198 (99%)	1	0.12	0.96	540.6
	Recessive	G/G	6 (3.2%)	2 (1%)	0.31 (0.06-1.55)	0.12	0.96	540.6
	Overdominant	T/T-G/G	158 (83.6%)	171 (85.5%)	1	0.6	1	542.7
	Overdominant	G/T	31 (16.4%)	29 (14.5%)	0.86 (0.50-1.50)	0.6	1	542.7
	Log additive	—	—	—	0.73 (0.46-1.14)	0.17	1	541
rs206118	Codominant	T/T	131 (69.3%)	146 (73%)	1	0.44	1	543.3
		C/T	49 (25.9%)	49 (24.5%)	0.90 (0.57-1.42)
		C/C	9 (4.8%)	5 (2.5%)	0.50 (0.16-1.53)
	Dominant	T/T	131 (69.3%)	146 (73%)	1	0.42	1	542.3
	Dominant	C/T-C/C	58 (30.7%)	54 (27%)	0.84 (0.54-1.30)	0.42	1	542.3
	Recessive	T/T-C/T	180 (95.2%)	195 (97.5%)	1	0.23	1	541.5
	Recessive	C/C	9 (4.8%)	5 (2.5%)	0.51 (0.17-1.56)	0.23	1	541.5
	Overdominant	T/T-C/C	140 (74.1%)	151 (75.5%)	1	0.75	1	542.9
	Overdominant	C/T	49 (25.9%)	49 (24.5%)	0.93 (0.59-1.47)	0.75	1	542.9
	Log additive	—	—	—	0.81 (0.56-1.18)	0.28	1	541.8
rs12917	Codominant	C/C	147 (77.8%)	150 (75%)	1	0.59	1	543.9
		C/T	39 (20.6%)	44 (22%)	1.11 (0.68-1.80)
		T/T	3 (1.6%)	6 (3%)	1.96 (0.48-7.98)
	Dominant	C/C	147 (77.8%)	150 (75%)	1	0.52	1	542.5
	Dominant	C/T-T/T	42 (22.2%)	50 (25%)	1.17 (0.73-1.86)	0.52	1	542.5
	Recessive	C/C-C/T	186 (98.4%)	194 (97%)	1	0.35	1	542.1
	Recessive	T/T	3 (1.6%)	6 (3%)	1.92 (0.47-7.78)	0.35	1	542.1
	Overdominant	C/C-T/T	150 (79.4%)	156 (78%)	1	0.74	1	542.8
	Overdominant	C/T	39 (20.6%)	44 (22%)	1.08 (0.67-1.76)	0.74	1	542.8
	Log additive	—	—	—	1.19 (0.79-1.80)	0.4	1	542.2
rs7896488	Codominant	G/G	101 (53.4%)	123 (61.8%)	1	0.18	1	540.2
		G/A	75 (39.7%)	68 (34.2%)	0.74 (0.49-1.13)
		A/A	13 (6.9%)	8 (4%)	0.51 (0.20-1.27)
	Dominant	G/G	101 (53.4%)	123 (61.8%)	1	0.095	0.76	538.8
	Dominant	G/A-A/A	88 (46.6%)	76 (38.2%)	0.71 (0.47-1.06)	0.095	0.76	538.8
	Recessive	G/G-G/A	176 (93.1%)	191 (96%)	1	0.21	1	540.1
	Recessive	A/A	13 (6.9%)	8 (4%)	0.57 (0.23-1.40)	0.21	1	540.1
	Overdominant	G/G-A/A	114 (60.3%)	131 (65.8%)	1	0.26	1	540.4
	Overdominant	G/A	75 (39.7%)	68 (34.2%)	0.79 (0.52-1.19)	0.26	1	540.4
	Log additive	—	—	—	0.73 (0.52-1.02)	0.064	0.512	538.2

Significant p values are shown in bold. Nominally significant p values are shown in italics.

p Value was calculated by the nonconditional logistic regression.

p Value was calculated by the nonconditional logistic regression after Bonferroni correction.

AIC, Akaike information criterion.

The rs144848 was correlated with NSCL/P under codominant, recessive, and overdominant models (codominant model: OR [G/T vs. T/T] = 1.51, 95% CI [G/T vs. T/T] = 0.99-2.31, OR [G/G vs. T/T] = 0.55, 95% CI [G/G vs. T/T] = 0.26-1.16, p = 0.017; recessive model: G/G vs. T/T-G/T, OR = 0.46, 95% CI = 0.22-0.95, p = 0.032; and overdominant model: G/T vs. T/T-G/G, OR = 1.64, 95% CI = 1.09-2.48, p = 0.018). However, after the Bonferroni correction, the p-value was >0.05. The codominant model was the optimal genetic model for rs144848 (AIC = 536.8) with a power value of 96.9%. Unconditional logistic regression results of the remaining SNPs were negative (Table 3).

In this study, the BRCA2 rs11571836 was found to be associated with NSCL/P, which is located in the 3′UTR. SNPinfo online software predicted potential miRNA binding sites and identified hsa-miR-1244, hsa-miR-1323, hsa-miR-562, and hsa-miR-633 as possible miRNAs bound by the BRCA2 gene (Table 4). Before Bonferroni correction, a missense mutation rs144848 was associated with NSCL/P. In addition, the potential biological effects of this missense mutation on protein function were predicted using Fathmm, PolyPhen2, and MutationTaster2021 software. The result of Fathmm was tolerated, while the result of PolyPhen2 and MutationTaster2021 was benign.

Table 4.

Potential Matching Sites for rs11571836 Using SNPinfo Software

SNPs	Allele	Position	Prediction strand	Forward sequence	miRNA	Score	Energy
rs11571836	A	14	+	GATAATTCCTTTTAGTTTAGCTACTa	hsa-miR-1244	146.00	−13.02
rs11571836	G	14	+	GATAATTCCTTTTGGTTTAGCTACTa	hsa-miR-1244	142.00	−10.94
rs11571836	A	15	+	AGATAATTCCTTTTAGTTTAGc	hsa-miR-1323	150.00	−15.23
rs11571836	G	7	+	cctttTGGTTTAGCTACTat	hsa-miR-562	143.00	−16.07
rs11571836	A	6	+	cTTTTAGTTTAGCTACTATTtt	hsa-miR-633	144.00	−7.98
rs11571836	G	9	+	TTCCTTTTGGTTTAGCTACTATTtt	hsa-miR-633	150.00	−10.52

Discussion

Our study revealed for the first time that BRCA2 was associated with NSCL/P. The rs11571836 G allele was protective against NSCL/P pathogenesis in a Chinese population (OR <1 and U95 < 1). Rs11571836 had a significant correlation with NSCL/P under four genetic models, with the overdominant model being the optimal genetic model. Rs144848 was associated with NSCL/P under three genetic models, and the codominant model was the optimal genetic model. However, rs144848 was not statistically different after the Bonferroni correction. Bioinformatics analysis identified hsa-miR-1244, hsa-miR-1323, hsa-miR-562, and hsa-miR-633 were potential binding miRNAs for rs11571836. Moreover, the missense mutation rs144848 may be a possible harmless mutation on the BRCA2.

With the development of genetics, there is an increasing number of studies on the association of common and complex genetic congenital diseases such as NSCL/P. In particular, the advent of genome-wide association studies has identified dozens of loci associated with the risk of NSCL/P (Avasthi et al., 2021; Beaty et al., 2010; Gowans et al., 2022; Mukhopadhyay et al., 2021), most of which are located in intronic or intergenic regions, but the clinical insights gained from these results are limited. In contrast, functional variants are the most likely to be associated with disease.

Association studies based on functional SNPs are relatively efficient for better understanding their function in complex diseases because they study a group of SNPs that are most likely to cause diseases (Carlton et al., 2006). Therefore, eight functional SNPs of BRCA2 and MGMT genes were selected for our study and the results were subjected to preliminary bioinformatics analysis to lay the foundation for further functional studies.

There are 27 exons in the BRCA2 gene, which is found on chromosome 13q12 (Lancaster et al., 1996). The BRCA2 gene is involved in maintaining genomic stability, especially in the homologous recombination pathway involved in DNA double strand repair. It encodes for the production of proteins that play an important role in regulating the replication of human cells, normal cell growth, and repair of DNA damage to genetic material (D'Alessandro et al., 2018; Roy et al., 2011).

The BRCA2 gene is mainly studied as an oncogene with its mutations associated with a range of malignancies, including breast, ovarian, prostate, and pancreatic cancers (King et al., 2003; Lee et al., 2021). The BRCA2 gene was specifically damaged in epithelial and/or ectodermal mesenchymal cells throughout the formation of the mouse craniofacial structure, resulting in abnormalities in the frontal nasal bone and maxilla (Kitami et al., 2018). BRCA2 played a key role in palatal cleft formation in mouse neural crest-derived ectodermal mesenchymal cells and was widely expressed in palatal epithelial and mesenchymal cells. Neural crest mice lacking BRCA2 in cells exhibit severe cleft palate. The neural crest-specific BRCA1 and BRCA2 double deletion animals showed a more severe cleft palate phenotype, indicating they functioned synergistically in palatogenesis in mice (Yamaguchi et al., 2021).

In NHW and Hispanics, BRCA2 rs206115 was modestly associated with NSCL/P (p = 0.01) and the other BRCA2 SNPs (rs144848 and rs9534342) showed borderline associations with NSCL/P (p = 0.05), but none was statistically significant after Bonferroni correction (Rodriguez et al., 2018). Chinese people made up our research sample, and the findings revealed a connection between rs144848 and NSCL/P. Codominant model was the best genetic model of rs144848, in which the risk of G/T is 1.51 times that of T/T, and G/G is the protective genotype compared with T/T.

However, rs144848 was also not statistically significant after Bonferroni correction. The key finding of this study was to determine the association between the BRCA2 rs11571836 and NSCL/P. The rs11571836 G allele frequency was statistically significant between the case and control groups and the G allele was protective against the incidence of NSCL/P. The overdominant model was the optimal genetic model for rs11571836, in which the risk of NSCL/P of G/A was 2.6 times higher compared with A/A-G/G. Our findings are also the same as those mentioned above, suggesting that cancer and NSCL/P may share a common genetic background (Dunkhase et al., 2016; Wang et al., 2016).

The rs11571836 is located in the 3′UTR of the BRCA2 gene. RNA stability, mRNA translation, and localization are among the activities that are regulated by the 3′UTR, which is situated downstream of the coding sequence (Steri et al., 2018). SNP at 3′UTR of gene can regulate the expression level of gene in both temporal and spatial dimensions, and control the translation process of corresponding protein (Jia et al., 2013; Kuersten and Goodwin, 2003). The 3′UTR is where most of the miRNA and mRNA binding sites are found.

Moreover, miRNAs can pair completely with a segment of target sequence on the 3′UTR of target genes to form a complex that induces gene silencing, inhibits translation, or promotes mRNA degradation, resulting in a decrease in the expression of target genes (Bartel, 2009). We predicted that hsa-miR-1244, hsa-miR-1323, hsa-miR-562, and hsa-miR-633 might be in a targeting relationship with the 3′UTR of BRCA2 gene. Therefore, the BRCA2 gene 3′UTR polymorphic locus rs11571836 may affect the progression of NSCL/P by modifying the targeted binding of corresponding miRNAs to mRNAs and thus regulating the mRNA expression level of BRCA2 gene.

The rs144848 is a missense mutation in BRCA2 gene. The amino acid type and sequence of the polypeptide chain are altered by the missense mutation, which occurs when the codon encoding one amino acid turns into the codon encoding another amino acid following base replacement. Missense mutations affect protein stability, hydrogen bond, kinetics, and activity, may or may not cause disease, and may affect susceptibility to disease and drug therapy (Gong and Blundell, 2010; Stefl et al., 2013).

Around 1.26% of human SNPs occur in protein-coding regions, of which 0.64% are nonsynonymous SNPs (including missense and nonsense mutations) (Gong and Blundell, 2010). Nonsynonymous SNPs are generally prevalent and seem to have no apparent functional significance, according to studies of human nonsynonymous SNPs (Ng et al., 2008). To assess the possible biological effects of mutations on protein function, we employed three types of software; however, all the outcomes were benign.

Although the efficacy of these techniques to predict variant pathogenicity is growing, grouping pathogenic from benign variants according to score does not further our knowledge of the underlying molecular processes in important disorders. At the same time, although a large number of disease-related missense variants can be found in publicly accessible databases, most of them are still of uncertain significance (Iqbal et al., 2020; Landrum et al., 2018). Therefore, it is of great interest to investigate the effect of missense mutations on NSCL/P by functional assay experiments.

A DNA repair protein, which is an effective direct DNA damage repair enzyme, is encoded by the MGMT gene, which has five exons and is located at 10q26. It participates in the cell's resistance to the toxicity of alkylating agents and the induced gene mutation, and prevents mismatch errors during DNA replication (Sharma et al., 2009). Glioblastoma has been the primary focus of research on the MGMT gene (Butler et al., 2020). In the association study between MGMT and NSCL/P, it was found that rs7922405 was associated with NSCL/P in Chinese people (Chen et al., 2018). However, rs7922405 is an intron variant; our study did not include the study of rs7922405, an intron SNP. Furthermore, no association between candidate functional SNPs of MGMT and NSCL/P was found in our study.

Conclusions

Our study identified for the first time the association of rs11571836 and rs144848 in the BRCA2 gene with NSCL/P in a Chinese population, but no statistical difference in rs144848 after Bonferroni correction was found. The results were also subjected to preliminary bioinformatics analysis. A bigger sample size might strengthen the validity of our findings since the number of cases and controls in our research was relatively limited. In addition, more cellular and animal studies are required for this work to investigate the potential role of the BRCA2 gene in the pathogenesis of NSCL/P.

Footnotes

Acknowledgments

We thank the study participants for their support of the study.

Authors' Contributions

S.G.: conceptualization (equal); funding acquisition (lead); investigation (lead); formal analysis (lead); writing—original draft (lead); and writing—review and editing (lead). Z.Z.: investigation (supporting); data curation (equal); and validation (equal). T.G.: investigation (supporting); writing—original draft (supporting); and resources (supporting). Y.X.: investigation (supporting); validation (equal); and resources (supporting). X.Y.: formal analysis (supporting) and data curation (equal). Y.W.: formal analysis (supporting) and data curation (supporting). R.C.: conceptualization (equal); resources (lead); and project administration (lead).

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This study was supported by grants from Discipline Construction Fund of Beijing Stomatological Hospital Affiliated to Capital Medical University (Grant No. 19-09-09).

References

Adzhubei

, Schmidt

, Peshkin

, et al. A method and server for predicting damaging missense mutations. Nat Methods, 2010; 7(4):248-249; doi: 10.1038/nmeth0410-248

Avasthi

, Muthuswamy

, Asim

, et al. Identification of novel genomic variations in susceptibility to nonsyndromic cleft lip and palate patients. Pediatr Rep, 2021; 13(4):650-657; doi: 10.3390/pediatric13040077

Bartel

. MicroRNAs: Target recognition and regulatory functions. Cell, 2009; 136(2):215-233; doi: 10.1016/j.cell.2009.01.002

Beaty

, Murray

, Marazita

, et al. A genome-wide association study of cleft lip with and without cleft palate identifies risk variants near MAFB and ABCA4. Nat Genet, 2010; 42(6):525-529; doi: 10.1038/ng.580

Butler

, Pongor

, Su

, et al. MGMT status as a clinical biomarker in glioblastoma. Trends Cancer, 2020; 6(5):380-391; doi: 10.1016/j.trecan.2020.02.010

Carlton

, Ireland

, Useche

, et al. Functional single nucleotide polymorphism-based association studies. Hum Genomics, 2006; 2(6):391-402; doi: 10.1186/1479-7364-2-6-391

Chen

, Guo

, Shi

, et al. Genetic variants of MGMT, RHPN2, and FAM49A contributed to susceptibility of nonsyndromic orofacial clefts in a Chinese population. J Oral Pathol Med, 2018; 47(8):796-801; doi: 10.1111/jop.12758

D'Alessandro

, Whelan

, Howard

, et al. BRCA2 controls DNA:RNA hybrid level at DSBs by mediating RNase H2 recruitment. Nat Commun, 2018; 9(1):5376; doi: 10.1038/s41467-018-07799-2

Deng

, Wang

. Roles of BRCA1 in DNA damage repair: A link between development and cancer. Hum Mol Genet, 2003; 12 Spec No 1:R113-R123; doi: 10.1093/hmg/ddg082

10.

Dixon

, Marazita

, Beaty

, et al. Cleft lip and palate: Understanding genetic and environmental influences. Nat Rev Genet, 2011; 12(3):167-178; doi: 10.1038/nrg2933

11.

Dunkhase

, Ludwig

, Knapp

, et al. Nonsyndromic cleft lip with or without cleft palate and cancer: Evaluation of a possible common genetic background through the analysis of GWAS data. Genom Data, 2016; 10:22-29; doi: 10.1016/j.gdata.2016.08.017

12.

Feltes

, de Faria Poloni

, Notari

, et al. Toxicological effects of the different substances in tobacco smoke on human embryonic development by a systems chemo-biology approach. PLoS One, 2013; 8(4):e61743; doi: 10.1371/journal.pone.0061743

13.

Gong

, Blundell

. Structural and functional restraints on the occurrence of single amino acid variations in human proteins. PLoS One, 2010; 5(2):e9186; doi: 10.1371/journal.pone.0009186

14.

Gowans

LJJ

, Comnick

, Mossey

, et al. Genome-wide scan for parent-of-origin effects in a sub-Saharan African cohort with nonsyndromic cleft lip and/or cleft palate (CL/P). Cleft Palate Craniofac J, 2022; 59(7):841-851; doi: 10.1177/10556656211036316

15.

Iqbal

, Pérez-Palma

, Jespersen

, et al. Comprehensive characterization of amino acid positions in protein structures reveals molecular effect of missense variants. Proc Natl Acad Sci U S A, 2020; 117(45):28201-28211; doi: 10.1073/pnas.2002660117

16.

, Garland

, Sun

, et al. Cellular and developmental basis of orofacial clefts. Birth Defects Res, 2020; 112(19):1558-1587; doi: 10.1002/bdr2.1768

17.

Jia

, Yao

, Arif

, et al. Regulation and dysregulation of 3′UTR-mediated translational control. Curr Opin Genet Dev, 2013; 23(1):29-34; doi: 10.1016/j.gde.2012.12.004

18.

King

, Marks

, Mandell

, et al. Breast and ovarian cancer risks due to inherited mutations in BRCA1 and BRCA2. Science, 2003; 302(5645):643-646; doi: 10.1126/science.1088759

19.

Kitami

, Kitami

, Kaku

, et al. BRCA1 and BRCA2 tumor suppressors in neural crest cells are essential for craniofacial bone development. PLoS Genet, 2018; 14(5):e1007340; doi: 10.1371/journal.pgen.1007340

20.

Kobayashi

, Alvizi

, Sunaga

, et al. Susceptibility to DNA damage as a molecular mechanism for non-syndromic cleft lip and palate. PLoS One, 2013; 8(6):e65677; doi: 10.1371/journal.pone.0065677

21.

Kondo

, Schutte

, Richardson

, et al. Mutations in IRF6 cause Van der Woude and popliteal pterygium syndromes. Nat Genet, 2002; 32(2):285-289; doi: 10.1038/ng985

22.

Kuersten

, Goodwin

. The power of the 3′ UTR: Translational control and development. Nat Rev Genet, 2003; 4(8):626-637; doi: 10.1038/nrg1125

23.

Lancaster

, Wooster

, Mangion

, et al. BRCA2 mutations in primary breast and ovarian cancers. Nat Genet, 1996; 13(2):238-240; doi: 10.1038/ng0696-238

24.

Landrum

, Lee

, Benson

, et al. ClinVar: improving access to variant interpretations and supporting evidence. Nucleic Acids Res, 2018; 46(D1):D1062-D1067; doi: 10.1093/nar/gkx1153

25.

Lee

, Lee

, Li

. BRCA genes and related cancers: A meta-analysis from epidemiological cohort studies. Medicina (Kaunas), 2021; 57(9):905; doi: 10.3390/medicina57090905

26.

Mukhopadhyay

, Feingold

, Moreno-Uribe

, et al. Genome-wide association study of non-syndromic orofacial clefts in a multiethnic sample of families and controls identifies novel regions. Front Cell Dev Biol, 2021; 9:621482; doi: 10.3389/fcell.2021.621482

27.

, Levy

, Huang

, et al. Genetic variation in an individual human exome. PLoS Genet, 2008; 4(8):e1000160; doi: 10.1371/journal.pgen.1000160

28.

Peyrard-Janvid

, Leslie

, Kousa

, et al. Dominant mutations in GRHL3 cause Van der Woude Syndrome and disrupt oral periderm development. Am J Hum Genet, 2014; 94(1):23-32; doi: 10.1016/j.ajhg.2013.11.009

29.

Plaisancié

, Bouneau

, Cances

, et al. Distal 10q monosomy: New evidence for a neurobehavioral condition?. Eur J Med Genet, 2014; 57(1):47-53; doi: 10.1016/j.ejmg.2013.11.002

30.

Rodriguez

, Maili

, Chiquet

, et al. BRCA1 and BRCA2 gene variants and nonsyndromic cleft lip/palate. Birth Defects Res, 2018; 110(12):1043-1048; doi: 10.1002/bdr2.1346

31.

Roy

, Chun

, Powell

. BRCA1 and BRCA2: Different roles in a common pathway of genome protection. Nat Rev Cancer, 2011; 12(1):68-78; doi: 10.1038/nrc3181

32.

Sharma

, Salehi

, Scheithauer

, et al. Role of MGMT in tumor development, progression, diagnosis, treatment and prognosis. Anticancer Res, 2009; 29(10):3759-3768.

33.

Shihab

, Gough

, Cooper

, et al. Predicting the functional, molecular, and phenotypic consequences of amino acid substitutions using hidden Markov models. Hum Mutat, 2013; 34(1):57-65; doi: 10.1002/humu.22225

34.

Stefl

, Nishi

, Petukh

, et al. Molecular mechanisms of disease-causing missense mutations. J Mol Biol, 2013; 425(21):3919-3936; doi: 10.1016/j.jmb.2013.07.014

35.

Steinhaus

, Proft

, Schuelke

, et al. MutationTaster2021. Nucleic Acids Res, 2021; 49(W1):W446-W451; doi: 10.1093/nar/gkab266

36.

Steri

, Idda

, Whalen

, et al. Genetic variants in mRNA untranslated regions. Wiley Interdiscip Rev RNA, 2018; 9(4):e1474; doi: 10.1002/wrna.1474

37.

Wang

, Qiu

, Shi

, et al. Gene expression profiling analysis contributes to understanding the association between non-syndromic cleft lip and palate, and cancer. Mol Med Rep, 2016; 13(3):2110-2116; doi: 10.3892/mmr.2016.4802

38.

, Taylor

. SNPinfo: integrating GWAS and candidate gene information into functional SNP selection for genetic association studies. Nucleic Acids Res, 2009; 37(Web Server issue):W600-W605; doi: 10.1093/nar/gkp290

39.

Yamaguchi

, Kitami

, Wu

, et al. Alteration of DNA damage response causes cleft palate. Front Physiol, 2021; 12:649492; doi: 10.3389/fphys.2021.649492

40.

, Zhang

, Wei

, et al. O6-Methylguanine-DNA Methyltransferase (MGMT): Challenges and new opportunities in glioma chemotherapy. Front Oncol, 2020; 9:1547; doi: 10.3389/fonc.2019.01547