Epigenetic modulation of WNT signalling in the ageing human intestinal epithelium: consequences for tissue homeostasis

The human gut, which is lined with finger-like invaginations called crypts, is continually being renewed. Each crypt is populated by cells derived from a small number of stem cells present at the bases of each crypt. The stem cells usually divide asymmetrically to produce one stem cell and one daughter cell, thus maintaining the stem cell population. The daughter cell divides symmetrically several more times while moving up the side of the crypt until, about halfway up, it stops dividing and differentiates into one of several cell types necessary for normal gut function. These cells become components of the gut surface, where they are programmed to die, and are shed into the faecal stream. The rates of cell division and migration and differentiation are very tightly controlled by the expression of several genes. We have previously shown that several of the genes controlling cell division, migration and differentiation are gradually switched off in the ageing human gut by the addition of methyl groups to the DNA sequences comprising these genes. Since the gut is lined with crypts, where all cells are derived from a small number of stem cells, any change to a stem cell, such as gene methylation, will be passed on to all its progeny and will represent a significant proportion of the population of cells within the crypt. This implies that cells in which these genes are switched off will not be uniformly distributed throughout the gut, but rather they will be localised to specific crypts, resulting in a mosaic pattern of affected crypts that may eventually be at increased risk of disease. We have previously found this to be true in elderly subjects. Here we aim to extend these observations to many more people of different ages, to determine if, as we predict, during ageing more and more of these important genes are switched off by methylation, and that this occurs in increasing numbers of crypts. We also aim to understand what happens in the crypts when these important genes are switched off in the stem cells. Do the cells divide more rapidly? Do they travel further up the crypt while still dividing? Are the cells more or less likely to differentiate into one type of cell or another? Are they more resistant to programmed cell death? The answers to these questions may help us to understand what is going wrong during the ageing process, and may provide insight into strategies for the prevention of age-related disease.

The human gut epithelium is one of the most dynamic tissues in the body, and the maintenance of its integrity is essential for the well-being of the individual throughout life. However, an age-related reduction in crypt number and crypt cell proliferation, and an impaired ability of the epithelium to recover from injury, have all been observed in rodents. The canonical Wnt signalling pathway is the master regulator of intestinal epithelial renewal. Constitutive activation of Wnt signals disrupts intestinal tissue homeostasis in epithelial neoplasia, but preliminary evidence indicates that it also occurs with age in the apparently normal gut. Our pilot studies show that the activity of Wnt signals, and the degree of CpG island methylation of Wnt inhibitor genes, an epigenetic mechanism for gene silencing, both increase with age and vary markedly from crypt-to-crypt in the same individual. We propose that the development of increased Wnt signals along the crypt-axis reflects this age-dependent methylation of Wnt inhibitor genes. To test this hypothesis and understand the consequences for tissue homeostasis we will apply bioimaging and functional genomic approaches to a unique native human colonic crypt culture system. We will measure levels of CpG island methylation of Wnt inhibitor genes, and Wnt signals, in single crypts derived from human volunteers of widely differing age. Age-related variations in levels of DNA methyltransferase activity, and their responses to imposed molecular and cellular stress, will be quantified. We will use RNAi techniques to silence Wnt inhibitors and study the functional consequences for tissue renewal. Finally, we will build a database that can be interrogated with multivariate statistics to derive and test mechanistic hypotheses on the relationship between loss of epithelial tissue homeostasis and levels of CpG island methylation in a panel of Wnt inhibitor genes.