Bioactive food compounds have different effects on global DNA methylation, an epigenetic mechanism associated with chromosomal stability and genome function. Since the diet is characterized by a mixture of foods, we aimed to identify dietary patterns in women, and to evaluate their association with long interspersed nuclear elements (LINE-1) methylation, a surrogate marker of global DNA methylation. We conducted an observational cross-sectional study of 349 women from Southern Italy, with no history of severe diseases. Dietary patterns were derived by food frequency questionnaire and principal component analysis. LINE-1 methylation of leukocyte DNA was assessed by pyrosequencing. We observed that intake of wholemeal bread, cereals, fish, fruit, raw and cooked vegetables, legumes, soup, potatoes, fries, rice, and pizza positively correlated with LINE-1 methylation levels. By contrast, vegetable oil negatively correlated with LINE-1 methylation levels. Next, we demonstrated that adherence to a prudent dietary pattern—characterized by high intake of potatoes, cooked and raw vegetables, legumes, soup and fish—was positively associated with LINE-1 methylation. In particular, women in the 3rd tertile exhibited higher LINE-1 methylation level than those in the 1st tertile (median = 66.7 %5mC; IQR = 4.67 %5mC vs. median = 63.1 %5mC; IQR = 12.3 %5mC; p < 0.001). Linear regression confirmed that women in the 3rd tertile had higher LINE-1 methylation than those in the 1st tertile (β = 0.022; SE = 0.003; p < 0.001), after adjusting for age, educational level, employment status, smoking status, use of folic acid supplement, total energy intake and body mass index. By contrast, no differences in LINE-1 methylation across tertiles of adherence to the Western dietary pattern were evident. Interestingly, women who exclusively adhered to the prudent dietary pattern had a higher average LINE-1 methylation level than those who exclusively or preferably adhered to the Western dietary pattern (β = 0.030; SE = 0.004; p < 0.001; β = 0.023; SE = 0.004; p < 0.001; respectively), or those with no preference for a specific dietary pattern (β = 0.013; SE = 0.004; p = 0.002). Our study suggested a remarkable link between diet and DNA methylation; however, further mechanistic studies should be encouraged to understand the causal relationship between dietary intake and DNA methylation.
A cross-sectional study of global DNA methylation and risk of colorectal adenoma
