A genome-wide association analysis: m6A-SNP related to the onset of oral ulcers

Abstract

Oral ulcers are one of the most common inflammatory diseases on oral mucosa that have obvious impacts on patients. Studies have shown that N6-methyladenosine (m6A) RNA transcription modification may be involved in the development of various inflammatory responses, and whether the pathogenesis of oral ulcers is related to m6A is unclear. This study aims to identify how m6A-related single nucleotide polymorphisms (m6A-SNPs) may affect oral ulcers. The UKBB dataset containing 10,599,054 SNPs was obtained from the GWAS database using the keyword "oral ulcer" and compared with the M6AVar database containing 13,703 m6A-SNPs.With 7,490 m6A-SNPs associated with oral ulcers identified, HaploReg and RegulomeDB were used for further functional validation and differential gene analysis was performed using the GEO database dataset GSE37265. A total of 7490 m6A-SNPs were detected in this study, 11 of which were related to oral ulcers (p<5E-08), and all of these SNPs showed eQTL signals. The SNP rs11266744 (p=2.00E-27) may regulate the expression of the local gene CCRL2, thereby participating in the pathogenesis of oral ulcers. In summary, by analyzing genome-wide association studies, this study showed that m6A modification may be involved in the pathogenesis of oral ulcers and CCRL2 may be the targeted gene.

Keywords: CCRL2; GWAS; m6A; oral ulcers; pathogenesis.

Figure 1

Flow chart of research designs and main results.

Figure 2

The Manhattan curve of genome-wide identification of m6A-SNPs associated with oral ulcers. The Manhattan plot showed that among the 7490 m6A-SNPs associated with oral ulcers, 11 SNPs reached the genome-wide significance threshold (p<5E- 8), and 30 SNPs reached the genome-wide prompt threshold (p<5E-5).

Figure 3

The expression levels of selected genes were shown in oral ulcer patients and healthy controls. The expression of genes such as CCRL2, HLA-C, GSDMB, HLA-DPB1 and MAPT were elevated in patients with oral ulcers.

Figure 4

The regional association map of the rs11266744 locus. The SNP rs11266744 was located in the CCRL2 protein-coding region. This region showed very high conservation, transcription level and DNase I hypersensitivity.

Figure 5

The genomic sequence of a representative CCRL2 transcript (ENST00000399036.4) was used to predict the m6A modification peak on the website (http://www.cuilab.cn/sramp), while secondary structure analysis was enabled. (A) The red box indicates the disappearance of the m6A modified peak near rs11266744 after inputting the altered sequence. (B) The local secondary structure around mutation site rs11266744 is shown. rs11266744 (A>C) is located in purple at the predicted m6A modification site (red) with medium confidence. In the secondary structure string, H, M, I, B and P refer to hairpin loop, multiple loop, interior loop, bulged loop, and paired residues, respectively.