When cancer manipulates the genome, it leaves behind epigenetic marks, a telltale pattern of turned on and turned off genes. The pattern amounts to a kind of fingerprint. Not only does this pattern differ from the pattern seen in healthy cells, it is also common to various kinds of cancerous cells, even those in different developmental stages.
This observation resulted from detective work on the part of researchers at Johns Hopkins. These researchers, led by Andrew Feinberg, M.D., and Rafael Irizarry, Ph.D., decided to pursue epigenetic clues that emerged from studies of colorectal cancer, which appears to be characterized by the presence of large hypomethylated blocks.
Methylation, in epigenetics, refers to the pattern of methyl groups attached to DNA. These groups help determine which genes are turned “on” or “off,” and ultimately how cells behave. Although methylation abnormalities have been associated with specific cancers, scientists have only recently acquired the tools to learn just how widespread these abnormalities are.
The Johns Hopkins researchers, in an article published August 26 in Genome Medicine, described how they performed a comprehensive genome-scale analysis of 10 breast, 28 colon, nine lung, 38 thyroid, and 18 pancreas cancers, as well as five pancreas neuroendocrine tumors. The researchers also scrutinized matched normal tissue from most of these cases, as well as 51 premalignant lesions. To analyze the methylation patterns on their DNA samples, the researchers used a statistical approach that helped them identify large hypomethylated blocks.
“All of the tumors had big blocks of DNA where the methylation was randomized in cancer, leading to loss of methylation over big chunks and gain of methylation in smaller regions,” said Winston Timp, Ph.D., one of the researchers. “The changes arise early in cancer development, suggesting that they could conspire with genetic mutations to aid cancer development.”
Additional details appeared in the Genome Medicine article, which was entitled, “Large hypomethylated blocks as a universal defining epigenetic alteration in human solid tumors.” In this article, the authors explained how they were able to analyze DNA methylation on a genomic scale to identify common features of the cancer epigenome in solid tumors and assess the timing of those changes.
“We find that hypomethylated blocks are a universal feature of common solid human cancer, and that they occur at the earliest stage of premalignant tumors and progress through clinical stages of thyroid and colon cancer development,” wrote the authors. “We also find that the disrupted CpG islands … including hypermethylated island bodies and hypomethylated shores, are enriched in hypomethylated blocks, with flattening of the methylation signal within and flanking the islands.”
The overall effect, Dr. Feinberg noted, appears to be that cancers can easily turn genes on or off as needed. For example, they often switch off genes that cause dangerous cells to self-destruct while switching on genes that are normally only used very early in development and that enable cancers to spread and invade healthy tissue. “They have a toolbox that their healthy neighbors lack, and that gives them a competitive advantage,” Dr. Feinberg added.
“These insights into the cancer epigenome could provide a foundation for development of early screening or preventive treatment for cancer,” Dr. Timp explained, suggesting that the distinctive methylation “fingerprint” could be used to tell early-stage cancers apart from other, harmless growths. It would be even better, he continued, to find a way to prevent the transition to a cancerous fingerprint from happening at all.
The study’s authors emphasize that methylation deregulation in cancer appears to be a large-scale phenomenon, rather than a matter of methylation at specific sites: “These results motivate a relatively new view of cancer epigenetics in which large-scale heterochromatin structures are disrupted generally, at least in solid tumors, leading to loss of both epigenetic and gene expression regulation, resulting in hyper-variability of gene expression. These changes could even have interaction with large-scale genetic domains important in cancer.”
This story has been corrected from an earlier version, which mistakenly listed the name of one of the researchers who led the study as Rafael Orozarru, not Rafael Irizarry. GEN regrets the error.