Nutritional factors and gender influence age-related DNA methylation in the human rectal mucosa

Abstract

Aberrant methylation of CpG islands (CGI) occurs in many genes expressed in colonic epithelial cells, and may contribute to the dysregulation of signalling pathways associated with carcinogenesis. This cross-sectional study assessed the relative importance of age, nutritional exposures and other environmental factors in the development of CGI methylation. Rectal biopsies were obtained from 185 individuals (84 male, 101 female) shown to be free of colorectal disease, and for whom measurements of age, body size, nutritional status and blood cell counts were available. We used quantitative DNA methylation analysis combined with multivariate modelling to investigate the relationships between nutritional, anthropometric and metabolic factors and the CGI methylation of 11 genes, together with LINE-1 as an index of global DNA methylation. Age was a consistent predictor of CGI methylation for 9/11 genes but significant positive associations with folate status and negative associations with vitamin D and selenium status were also identified for several genes. There was evidence for positive associations with blood monocyte levels and anthropometric factors for some genes. In general, CGI methylation was higher in males than in females and differential effects of age and other factors on methylation in males and females were identified. In conclusion, levels of age-related CGI methylation in the healthy human rectal mucosa are influenced by gender, the availability of folate, vitamin D and selenium, and perhaps by factors related to systemic inflammation.

Figure 1

Summary descriptions of methylation values for the nine genes and for LINE-1.

Figure 2

Pearson correlation coefficients for the associations between the exposure variables and the methylation of the nine genes, LINE-1 and PCA1 for all subjects (grey), males only (white) and females only (black).

Figure 3

PCA1 methylation in the male (Δ) and female (○) subjects calculated from the CGI methylation values for the nine genes (open symbols) and predicted from the GA model (filled symbols).