Epigenetic enigma

Research conducted in the United States with Brazilian participation shows how epigenetic information propagates during cellular division and how, in cancer cells, this process is subverted

By Fábio de Castro

Agência FAPESP – Research conducted in the United States with Brazilian participation has revealed how epigenetic information—genomic information that is not part of the DNA sequence—propagates during cell division, and how this process is subverted in cancer.

The study was published in PLoS Genetics magazine. One of the authors, Brazil’s Daniel Diniz de Carvalho, is doing his post-doctorate studies at the University of Southern California in Urology, Biochemistry and Molecular Biology.

A veterinary doctor educated at Universidade de Brasília (UnB), Carvalho finished his doctorate in Immunology at Universidade de São Paulo’s Biomedical Sciences Institute (ICB-USP) in 2009 with a FAPESP fellowship.

Under the orientation of ICB-USP’s Gustavo Amarante-Mendes, Carvalho’s doctoral thesis produced important results recently published in the Nature group’s Oncogene magazine. The study’s initial objective was to understand how the methylation patterns of DNA remain unchanged in normal cells. Methylation is the main epigenetic mechanism: a methyl group is transferred to some DNA cytosine bases. The process is fundamental for switching off genes that provoke alterations in the genetic code.

“The new mechanism described in the study explains how normal somatic cells manage to maintain a faithful DNA methylation pattern throughout many cellular divisions. This suggests that during a tumor’s genesis process, this mechanism is deregulated, allowing methylation to occur in regions where it should not—which could explain how tumor suppressor genes are inhibited,” Carvalho told Agência FAPESP.

All the cells in an organism have the same DNA, but each one is specialized for a specific function. This happens because the epigenetic alterations guarantee this differentiation. “In cancer, this process is subverted. Methylation occurs in the wrong places, switching off genes that should suppress the tumor, allowing it to manifest,” he explains.

But with the models used to date, it is not clear how methylation is maintained in healthy cells, in exact locations on the DNA during cellular division. “Our main finding refers exactly to this: how aberrant methylation does not occur in the normal cell,” he said.

Two enzymes are responsible for controlling the methylation mechanism: DNMT3A and DNMT3B. Both can methylate DNA in any location. “These enzymes are dangerous, as they can set off methylation in places where it should not happen,” he affirms.

The two enzymes only manage to be stable when they are anchored in the nucleosome containing methylated DNA. This allows for a homeostatic mechanism that means the enzymes only occur in that region. “We discovered that, in the tumor, one of these enzymes breaks out of the nucleosome and becomes free in the nucleus. Then it manages to methylate in locations on the DNA that shouldn’t be methylated,” he remarks.

Control mechanisms

The researchers also discovered that enzyme detachment occurs because of a protein deletion: a piece of the enzyme is lost and it can then establish itself outside the nucleosome. “But the most important was to describe how the mechanism for maintenance of normal DNA methylation patterns occurs in a healthy situation. In cellular division, this pattern must be transferred exactly to the daughter cell. We managed to describe how this mechanism is maintained with such fidelity,” says Carvalho.

The enzymes, according to the scientist, need to remain anchored to the nucleosome. This regulatory mechanism is what allows for faithful propagation and keeps aberrant methylation from occurring.
“We will try to come to a better understanding of how tumor cells manage to overcome this mechanism. We have arrived at the first clues: deletion of the enzyme’s delta isoforms, which allows it to detach from the nucleosome. Now, we want to understand how the tumor manages to get around these control mechanisms that we described,” he affirms.

The article entitled Nucleosomes Containing Methylated DNA Stabilize DNA Methyltransferases 3A/3B and Ensure Faithful Epigenetic Inheritance (doi:10.1371/journal.pgen.1001286), by Daniel Diniz de Carvalho and others can be read at: www.plosgenetics.org.