Dr. Manel Esteller, ICREA research professor, head of the Cancer Epigenetics and Biology Program at the Biomedical Research Institute of Bellvitge (IDIBELL) and professor of genetics at the University of Barcelona
Opinion
The sequence of DNA nucleotide base pairs, the typical object of study in classical genetics, on its own can’t fully explain how our cells work, how complex diseases interfere with that process, or the definition of the species. We need something more. The field of epigenetics provides part of the explanation.
Waddington defined epigenetics in 1939 as "the causal interactions between genes and their products, which bring the phenotype into being." Based on our current knowledge, we could define epigenetics as "inherited DNA activity that doesn’t depend on the underlying DNA sequence." This inheritance is easier to understand during mitosis (the transmission process through which a cell divides to produce daughter cells in the cell cycle) or, even, more provocatively, during germ cells meiosis; and, therefore, our epigenetic information is transmitted to our descendants.
Epigenetics refers to the dynamic chemical modifications that come about in the DNA and to their later association with regulatory proteins [Berdasco and Esteller, Cell Dev 19 (5):698-711, 2010].
DNA methylation plays a key role in gene activity and nuclear structure. In human beings, DNA is produced in the cytosine of the CpG dinucleotides. These CpG sites aren’t distributed at random in the human genome; areas rich in CpG, known as CpG islands, are often associated with the regulatory region of many genes and are generally not methylated in normal genes. This non-methylated state corresponds with the CpG island’s ability to transcribe its associated genes when the necessary transcription activators are present. Nevertheless, there is a subset of CpG islands that is highly methylated in normal tissue, and these are often associated with specific tissue genes, imprinted genes and genes subject to inactivation on the X chromosome in women. Furthermore, repetitive genome sequences are also highly methylated. Maintaining this state of methylation may play an important role in protecting DNA integrity by preventing chromosome instability. DNA methylation isn’t an isolated epigenetic marker. It is often associated with chemical modifications in the N-terminal tails of the proteins called histones. Years ago, histones were dismissed as packing material for DNA, while they are now seen as central storage points for epigenetic information through a complex group of posttranslational modifications, like lysine acetylation and methylation, arginine and serine phosphorylation, among others. It has been proposed that the different modification patterns seen in histone tails make up a histone code that determines gene activity.
Epigenetic alteration is an important characteristic of human cancer. Reduced DNA methylation in human tumors, compared to normal tissue, was one of the first epigenetic alterations described in tumors. This loss comes about mainly through DNA hypomethylation in repetitive sequences and demethylation of gene bodies (codifying regions and introns). Global DNA hypomethylation contributes to the origin of cancer cells through the generation of chromosomal instability, reactivation of transposable elements and loss of genetic imprinting. The most important thing, which is known as the DNA methylation paradox, is that there are localized areas of DNA that increase CpG methylation: CpG islands in the promoter area of many tumor suppressor genes, like hMLH1, BRCA1 and p16INK4a, lead to the inactivation of these anticancer proteins.
Even more recently, it has been shown that microRNA with tumor-suppressing functions and other types of non-codified RNA are also silenced in cancer cells through DNA hypermethylation [Esteller, Nat Rev Genet 12 (12):861-74]. In terms of histones, human tumors also show a distorted code, and for types of leukemia we know that pathognomonic translocations involve histone acetyltransferases and methyltransferasegenes.
If we analyze cancer in terms of cell evolution, epigenetics seems to play a key role. Human tumors undergo huge changes in their natural evolution. The cancer can not only metastasize in distant areas —creating new blood and lymph vessels to feed on and eliminate metabolites— but can also change if treated with chemotherapy, hormonotherapy or radiotherapy. The cancer cell’s ability to subject itself to rapid genetic changes in order to adapt to a hostile microenvironment is limited. Nevertheless, Darwinian selection comes about in cancer cells through the generation of adapted survivors as a result of rapid epigenetic changes. After 48 hours of external stimuli, DNA methylation and histone modification patterns in transformed cells may have completely changed. Take for example breast cancer. The cell adhesion gene, called E-cadherin, can become methylated and silenced in cancer, leading to metastasis in the ribs, but the cancer cells already located in the bone need to establish an interaction with their new environment and there is later a loss of DNA methylation in these cells’ drive to survive. Another interesting case is a glioma in which DNA methylation associated with the inactivation of the DNA repair enzyme (MGMT) predicts a positive response to a family of chemotherapy drugs. However, once the treatment has been initiated, the tumor evolves, with the aim of surviving, and selects those cells that are not methylated to MGMT: chemoresistance has been created for a purely epigenetic reason.
One of the fundamental differences between human cancer genetics and epigenetics is that DNA methylation and histone modification are reversible changes, under the right circumstances. Therefore, epigenetic alterations are the chink in the cancer cell’s armor because, with the right drug regimen, hypermethylated tumor-suppressing genes can be woken from their long sleep to carry out their normal growth inhibitor functions.
Two families of epigenetic drugs —DNA demethylating agents and histone deacetylase inhibitors— are emerging as the most promising compounds in this area, and five medications have been approved to treat specific leukemia and subtypes of lymphoma [Rodríguez-Parets and Esteller, Nat Med 17 (3):330-92011]. Now their positive track record with this type of tumors must be proven in solid epithelial tumors, and oncologists have been encouraged to do so. And for biotechnology? Well, epigenetics provides excellent biomarkers for the disease in order to better diagnose, and predict response to therapy, improving prognosis. And this is patentable, right?