IFR is exploring epigenetic changes in cells that line the human gut, which are linked to our vulnerability to developing cancer. Here, Professor Ian Johnson comments on DNA methylation, and on a new study from Nigel Belshaw’s group which suggests that prolonged exposure to increased, supra-nutritional doses of folic acid may cause these epigenetic changes in human cells.

The term “epigenetics” was introduced in 1942 by the biologist C.H. Waddington, to account for what he described as “the causal interactions between genes and their products, which bring the phenotype into being”.  Clearly mechanisms must exist to regulate how genetically identical cells differentiate into the multiplicity of different forms and functions found in complex organisms. But their discovery was beyond the biology of Waddington’s day, and it has taken many decades for the term to become widely recognised and understood.

A modern definition of epigenetics is “the study of heritable changes or differences in gene expression occurring independently of differences in the genetic sequence.”  The instructions that determine such gene expression patterns are encoded by chemical modifications to the DNA backbone, or to the histones proteins around which DNA is wound to form chromatin. By analogy with the genome, the total set of such modifications is now often referred to as the “epigenome”.

The word “heritable” as used here refers to the transmission of epigenetic information through the mitotic divisions of somatic cells. (There is some experimental evidence to suggest that transmission through the germ-line may also occur but that is beyond the scope of this article.)

Amongst the most actively investigated epigenetic marks are the methyl groups covalently bound to many, but not all of the cytosine residues of DNA.  Methylated cytosines occur only in cytosine-guanine dinucleotides (CpGs).  Such CpGs are not scattered randomly through the human genome, but tend to occur more densely in regions known as CpG islands. CpG islands are found within the regulatory promoter sequences of more than half of all human genes.

The CpG islands of actively transcribed genes are typically unmethylated, whereas methylated CpGs are associated with silent non-expressed genes.  This is a normal mechanism of gene regulation, but when the CpG island of what should be an active functioning gene becomes methylated, its expression may be inhibited or completely silenced, leading to abnormal physiological activity in the affected cells. So profound are the consequences of this mechanism that 5-methyl cytosine has been called the “fifth base” of DNA.

Using highly sensitive techniques to quantify DNA methylation, Dr Nigel Belshaw’s group at IFR specialises in the study of epigenetics in the human gut.

Epigenetics in the human gut

Professor Ian Johnson points out that the interior surface of the human large intestine is a particularly rewarding tissue in which to investigate epigenetic mechanisms, both because of its very high rates of cell division and differentiation, and because of the unusual anatomy of the mucosa, which is based on a vast array of discrete and seemingly identical crypts.  Each crypt is a minute glandular pit, lined with columnar epithelial cells that migrate continuously from the crypt base toward the mucosal surface, where they become senescent and are eventually extruded into the lumen of the gut.

Dr Nigel Belshaw

All the migrating and maturing cells are derived from a few rapidly dividing stem cells near the crypt base.  Since epigenetic marks are transmitted through cell division, the DNA methylation signature of each stem cell is reproduced in that fraction of the crypt cell population to which it gives rise.  Nigel Belshaw’s group has explored the methylation patterns generated in this way by quantifying the percentage of methylated cytosine residues in the CpG islands of genes regulating crypt cell proliferation, using DNA extracted from individual human colonic crypts (1).

Multivariate statistical analysis showed that although the methylation patterns of crypts isolated from each individual volunteer differed markedly from one another, on average they were more similar to each other than to crypts from other individuals. Moreover the methylation patterns were related to the age of the individual from whom they were obtained, and crypts from healthy volunteers differed subtly from those obtained from colorectal cancer patients. This is consistent with the hypothesis that aberrant DNA methylation of the colonic mucosa, which is one aspect of what Waddington described as the “epigenetic landscape”, may tend to disrupt normal cell proliferation, and hence contribute to an individual’s eventual vulnerability to cancer.

If the CpG methylation patterns of ageing colonic crypts can ultimately contribute to the development of disease, it is important to understand how those of healthy individuals change with time.  Certain CpG islands in other tissues are known to become more methylated with age,  but it has been shown that the methylation patterns of identical twins tend to diverge, indicating that environmental factors must somehow influence the rate at which aberrant CpG methylation is acquired through life.

What can affect DNA methylation in crypt cells?

Crypt epithelial cells are exposed to a host of chemical signals that might conceivably influence methylation of DNA, including food residues and bacterial metabolites derived from the gut lumen, and nutrients, endocrine factors and pharmacological agents delivered via the blood vessels and extracellular compartments of the mucosa.  One obvious mechanism through which food-borne factors might influence DNA methylation involves the methylation reaction itself, which is mediated by the methyl transferase enzymes (DNMT) 1, 3A and 3B, and which relies on the availability of the intracellular methyl donor S-adenosyl methionine (SAM), which is generated by reactions requiring a supply of methyl groups derived from 5-methyltetrahydrofolic acid. This compound is derived from a variety of food-borne compounds, collectively termed the folates, and classed amongst the water soluble vitamins of the B-group.

Sub-optimal folate intakes do occur in the UK population, and in women of child-bearing age they are known to be associated with an increased risk of neural tube defects.  Conversely, since folate supplements are readily available, and some countries enforce mandatory supplementation of cereal foods with folic acid (the parent compound of the folate group) some human populations can acquire relatively high folate stores. There is therefore much interest in exploring the biochemical effects of sub- or supra-optimal levels of folate provision.

Does folate affect methylation?

In a recent study (2) Michelle Charles, a research student working with Nigel Belshaw, explored the effects of varying levels of 5-methylTHF and folic acid on the methylation of CpG islands in two human cell types,WI-38 fibroblasts and FHC, an essentially normal colon epithelial cell line, cultured in vitro.  When the folates were present at levels similar to those occurring in normal human tissues, there was no effect on CGI methylation. However at the supra-physiological concentrations often found in commercial cell culture media, the methylation levels of two target genes, ESR1 and p16, increased significantly. This effect was associated with a decline in the intracellular concentration ratio of SAM to S-adenosylhomocysteine (SAH), which is the end-product of DNA methylation reactions.

These observations raise the possibility that prolonged exposure to increased, supra-nutritional doses of folic acid, perhaps through the consumption of vitamin supplements or fortified foods over an extended period, might contribute to aberrant DNA methylation in humans. Moreover, as Nigel Belshaw points out, hypermethylation of CpG islands has often been reported in cultured cells used as models for human ageing.  Such studies routinely employ standard commercial cell-culture media. These new observations therefore raise the possibility that the aberrant methylation patterns observed in some of these experiments may have been induced by what amounts to physiological stress associated with inappropriately high levels of folic acid in the culture media. Another possibility is that other cultured tissues, including perhaps embryos undergoing in vitro fertilization, might acquire aberrant methylation patterns if the folate levels in the culture media are not carefully optimised. Various studies to investigate these intriguing possibilities are in progress at IFR.

(1)    Belshaw NJ, Pal N, Tapp HS, Dainty JR, Lewis MP, Williams MR, Lund EK, Johnson IT. Patterns of DNA methylation in individual colonic crypts reveal aging and cancer-related field defects in the morphologically normal mucosa. Carcinogenesis. 2010 31(6):1158-63.

(2)    Charles MA, Johnson IT, Belshaw NJ (2012) Supra-physiological folic acid concentrations induce aberrant DNA methylation in normal human cells in vitro.