There are thousands of different cell types in our body. Whether their function is to carry oxygen, conduct nerve signals or produce hormones, every single cell uses the same genetic information. Our bodies achieve their incredible functional diversity in part through epigenetics, the regulation of gene expression via alterations to the genome that don't change the sequence of DNA.
"Epigenetics is really the software of our cells," said Professor Manel Esteller, a geneticist at the University of Barcelona and cancer epigenetic group leader at the Josep Carreras Leukaemia Research Institute.
In the same way that a single desktop computer can run many programs, epigenetics allows our cells' DNA hardware to produce many different outputs.
Epigenetic changes influence the course and treatment of a range of diseases, but their impact is most evident in cancer. A recent review, which identified 91 links between epigenetic enzyme alterations and pathology, found that 80% of these connections were related to cancer. "Cancer is a disease of cell differentiation and cellular behavior," said Peter Jones, chief scientific officer at the Van Andel Institute and a veteran epigenetics researcher. Epigenetic markers, like methylation, play an essential role in directing the decisions that cells take as they differentiate. When mutations occur in the genes controlling this process, cancer often follows. While the mutations themselves are difficult to fix, the epigenetic changes can be reversed.
"There's been a tremendous activity in developing drugs that target these epigenetics," Jones added.
When a methyl group binds to a gene body or promoter, it obstructs access for transcription factors that encourage gene expression. The first hint that epigenetics might have a role in cancer came in 1983, when Professor Andrew Paul Feinberg and Professor Bert Vogelstein identified that tumor cells have less methylation than normal cells. Further study demonstrated that as cancer cells shift from benign to invasive states, the degree of methylation in their genomes further withers away.
Imagine that the production of a specialized cell in our body is controlled by a series of faders, as on an audio mixer. To produce each cell type's unique sound, epigenetic markers move these faders to carefully prepared positions. In cancer, the loss of these markers means that every fader is set to maximum. Cancer drowns out cells' unique melodies in favor of a wall of sound that produces uncontrollable growth at any cost.
In cancer, the cell has some failsafe mechanisms to suppress this spread, called tumor suppressor genes. But in contrast to the overall picture of demethylation across the genome, hotspots of hypermethylation form around tumor suppressor genes. These hotspots prevent the out-of-control cell metabolism from accessing the growth-limiting mechanisms encoded by tumor suppressor genes.
Jones points out that while some cases of the childhood cancer retinoblastoma are caused by mutations in the tumor suppressor RB1 gene, others are caused by excessive methylation of RB1's gene promoter. "Even though the person doesn't have a mutation, they don't have a functional gene because it's been shut off," he said.
A web of proteins governs methylation. These proteins include writers, readers and erasers of epigenetic marks. Mutations or disruptions to any of these enzymes can lead to pro-cancer effects.
DNMT3A is a key protein involved in methylation. Both overexpression and silencing of this protein can lead to cancer. Excessive methylation by DNMT3A is implicated in the blood cancer myelodysplastic syndrome (MDS).A drug Jones pioneered called azacitidine improves MDS outcomes by reducing methylation throughout the genome and reactivating tumor suppressor genes. On the other hand, shutting off DNMT3A entirely, in combination with other pro-cancer mutations, can induce rapid blood cancer phenotypes.
The development of epigenetic therapeutics hasn't been straightforward. Epigenetic regulation is complex, and the way that cancer alters epigenetic markers can vary even between cells in the same tumor. There are over 40 approved and experimental cancer therapies that target epigenetic mechanisms. These include methylation inhibitors, like azacitidine, and promoters of another type of mark called acetylation. These latter drugs, called HDAC inhibitors (HDAC is a protein that suppresses acetylation), include vorinostat and romidepsin - both approved for the treatment of cutaneous T cell lymphoma.
The complex landscape of epigenetic modification within tumors often reduces the efficacy of epigenetic monotherapies. Instead, many treatments are given as combination therapies. Azacytidine and vorinostat, a dual DNMT inhibitor-HDAC inhibitor treatment, is more effective in treating blood cancer than monotherapies. Epigenetic therapies can also be used in combination with other branches of cancer therapy - for example, immunomodulatory therapies or gamma-secretase inhibitors.
Despite these successes, Jones said that the field is still awaiting a breakthrough as significant as the one cancer immunotherapies achieved with the release of checkpoint inhibitor therapies. One limitation of epigenetic cancer therapies is poor efficacy in the treatment of solid tumors, like lung cancer. Esteller said that such treatments struggle to reach all cells in these tumors. At higher levels, epigenetic therapies like azacytidine can be toxic, so simply adding more of the drug isn't an option.
Esteller said that the therapies are more useful in cancers like leukemia and lymphoma because there is less chromosomal damage. In solid tumors, the DNA "hardware" is damaged, so bug fixes to the epigenetic "software" are less effective, he explained. Researchers have found more success by combining epigenetic therapies with other treatments, like immunotherapies.
Beyond the development of cancer treatments, epigenetics has the potential to advance cancer diagnostics. The distinctive pattern of methylation observed in cancer cells - overall demethylation paired with hotspots of elevation around specific genes - can be used for predictive markers of cancer. Tumors and other cancer cells release genetic material into the bloodstream, where it floats as cell-free DNA (cfDNA). This circulating DNA can be measured for mutations as well as methylation patterns through techniques collectively termed liquid biopsy. In head-to-head diagnostic comparisons between targeted gene, whole genome sequencing and bisulfite analysis, which measures genome-wide methylation, the latter proved to be the most sensitive technique. Esteller explains that while cancers affecting different tissues can share similar mutations, there is a broader range of potential epigenetic markers. This diversity means that epigenetic diagnostic panels can look at dozens of biomarkers to identify the specific type of cancer present.
The field is growing rapidly. Esteller said the advent of single-cell technologies will allow researchers to identify epigenetic heterogeneity in cancer patients or even within single tumors, which will improve treatment targeting. As for diagnostics, Jones sees epigenetic analysis becoming a routine aspect of cancer prevention in the not-too-distant future: "When you go for your annual blood test, you're going to get a screen done on your circulating DNA," he said.