DNA methylation is one of the primary ways your cells silence genes. When small chemical tags called methyl groups attach to DNA near a gene’s starting point, they effectively switch that gene off by preventing the cellular machinery from reading it. This process is essential for normal development, but when it goes wrong, it can contribute to diseases like cancer.
How Methylation Blocks a Gene
Your DNA contains stretches called CpG islands, regions where the nucleotides cytosine and guanine appear frequently next to each other. These islands sit near the start of many genes. To qualify as a CpG island, a region needs a GC content above 50%, a length greater than 200 base pairs, and a higher-than-expected density of CpG pairs. When methyl groups attach to the cytosines in these islands, the gene downstream gets shut down.
This silencing happens through two routes. First, methylation can directly block transcription factors, the proteins that normally land on DNA and kick-start gene reading, from physically attaching. The methyl group changes the shape of the DNA surface just enough that these proteins can no longer dock. Second, methylated DNA attracts a different set of proteins that actively repress the gene. One well-studied example is a protein called MeCP2, which binds specifically to methylated CpG sites and then recruits enzymes that strip chemical tags off nearby histone proteins (the spools around which DNA is wound). This causes the DNA to coil more tightly, packing it into a dense, inaccessible form called heterochromatin. Once DNA is packed this tightly, the gene-reading machinery simply cannot reach it.
Where Your Body Uses Gene Silencing
Methylation-based silencing is not a glitch. It is a core part of how your body develops and functions. Every cell in your body carries the same DNA, but a liver cell looks and behaves nothing like a brain cell because each cell type has a different pattern of methylated and unmethylated genes. Methylation helps lock in these cell identities by permanently turning off genes a given cell type does not need.
One striking example is genomic imprinting. You inherit two copies of most genes, one from each parent, and both are usually active. But for a small set of genes, one parental copy is deliberately silenced through methylation. The repressed copy is methylated while the active copy is not. These imprints are established during the formation of sperm and egg cells, maintained throughout development, then erased and re-established in the next generation’s germ cells. Imprinted genes tend to cluster together in regions controlled by imprinting control centers, which carry methylation patterns specific to maternal or paternal origin.
Another major example is X-chromosome inactivation. Females carry two X chromosomes, and to prevent a double dose of X-linked gene products, one copy is largely shut down in every cell. DNA methylation is one of the key mechanisms that locks this inactivation in place, particularly at CpG islands across the silenced X chromosome.
When Silencing Goes Wrong in Cancer
In healthy cells, CpG islands near gene-starting regions are typically unmethylated, keeping those genes available for use. Cancer cells break this pattern. They often show heavy methylation (hypermethylation) at the promoters of tumor suppressor genes, the very genes whose job is to slow cell growth or trigger the death of damaged cells. With these guardians silenced, cells can grow unchecked.
Several well-known tumor suppressors are commonly shut down this way. The retinoblastoma gene (Rb1), which controls cell division, is frequently hypermethylated in various cancers. So is E-cadherin, a protein that keeps cells stuck together (its loss helps tumors spread). GSTP1, which helps detoxify harmful chemicals, is silenced in many prostate cancers. Importantly, these genes are not mutated or deleted. Their DNA sequence is perfectly intact. They are simply switched off by methylation, making this a purely epigenetic form of cancer progression.
Methylation Is Reversible
Unlike mutations in the DNA code itself, methylation can be undone. Your cells contain a family of enzymes called TET proteins (TET1, TET2, and TET3) that chemically modify methylated cytosine through a series of oxidation steps. First, TET converts methylated cytosine into hydroxymethylcytosine. On most of its target sites, the enzyme stops here. But at certain regulatory regions where TET activity is high, the oxidation continues further, producing two additional modified forms. These heavily oxidized bases are then recognized and snipped out by a DNA repair enzyme, which replaces them with a clean, unmodified cytosine. The net result is complete removal of the methyl mark, and the gene can become active again.
This “passive” versus “active” distinction matters. Passive demethylation happens gradually when cells divide and the methylation marks simply are not copied onto new DNA strands. Active demethylation, driven by TET enzymes and the repair pathway, can happen regardless of cell division, allowing more rapid and targeted gene reactivation.
Hydroxymethylcytosine, the first product of TET activity, is not just an intermediate on the path to demethylation. It appears to have its own regulatory role. In embryonic stem cells, TET1 binds to CpG-rich regions to prevent unwanted methylation and converts existing methyl marks to hydroxymethylcytosine, helping maintain the flexible gene expression patterns these cells need as they prepare to specialize into different cell types.
Gene Body Methylation: The Exception
The relationship between methylation and gene silencing has an important exception. Everything described above applies to methylation at the promoter, the region near the gene’s starting point. Methylation within the body of a gene (the sequence that actually encodes the protein) has the opposite correlation: it is positively associated with gene expression. Genes that are being actively read tend to be methylated along their length.
The precise reason for this is still being worked out, but experiments have shown that when gene body methylation is stripped away, the gene’s expression drops. Restoring that methylation brings expression back up. This has therapeutic implications. Some cancers feature genes that have become overactive partly through changes in gene body methylation, and drugs that alter methylation patterns could potentially normalize their activity.
Drugs That Reverse Gene Silencing
Because aberrant methylation drives disease, drugs that strip methyl marks off DNA have entered clinical use. Two hypomethylating agents, azacitidine (approved in 2004) and decitabine (approved in 2006), work by inhibiting the enzymes that maintain DNA methylation. They were among the first epigenetic drugs approved by the FDA, initially for myelodysplastic syndrome, a group of bone marrow disorders that can progress to leukemia. By reactivating tumor suppressor genes that cancer cells had silenced, these drugs can slow disease progression without targeting the DNA sequence itself.
The reversibility of methylation is precisely what makes it such an attractive drug target. A mutated gene requires gene therapy or workarounds, but a methylated gene just needs its chemical tags removed to start working again. This principle continues to drive development of newer epigenetic therapies across multiple cancer types.