Patterns and Possible Roles of LINE-1 Methylation Changes in Smoke-Exposed Epithelia

Tobacco smoking and reduced methylation of long interspersed element-1 (LINE-1) are crucial in oral carcinogenesis. 5′UTR of human LINE-1 sequence contains several CpG dinucleotides which are methylated in various proportions (0–100%). Methylation levels of many LINE-1s in cancer were reduced, hypomethylated. The hypomethylation of each LINE-1 locus can promote instability of genome and repress expression of a gene located on that same chromosome. This study investigated if cigarette smoking influences LINE-1 methylation of oral mucosal cells. The methylation of human LINE-1 in clinically normal oral mucosa of current smokers was compared to non-smokers. By using the combined bisulphite restriction analysis, each LINE-1 sequence was categorised into 4 patterns depending on the methylation status and location of the two 18-bp successive CpG from 5′ to 3′ including mCmC, uCuC, mCuC and uCmC. Of these, mC and uC represent methylated and unmethylated CpG, respectively. The DNA bisulphite sequence demonstrated that most CpGs of mCmC and uCuC were methylated and unmethylated, respectively. Nevertheless, some CpGs of each mCuC or uCmC allele were methylated. Imaging of the digestion products was used to generate %methylation value. No significant difference in the overall LINE-1 methylation level but the differences in percentages of some methylation patterns were discovered. The %mCmC and %uCuC increased, while the %mCuC decreased in current smokers (p = 0.002, 0.015, and mCuC still persisted in persons who had stopped smoking for over 1 year (p = 0.001). The %mCuC also decreased in the higher pack-year smokers (p = 0.028). Smoking possibly altered mCuC to mCmC and uCuC forms, and changes uCmC to uCuC forms. In conclusion, smoking changes methylation levels of partial methylated LINE-1s and increased the number of hypo- and hypermethylated loci. These hypomethylated LINE-1s may possess carcinogenesis potential. Moreover, LINE-1 methylation patterns may be useful for monitoring oral carcinogenesis in smokers.