Epigenetics panel: (left to right) Jay Chung, NHLBI; Shiv Grewal, NCI; Mirit Aladjem, NCI; Munira Basrai, NCI; David Schrump, NCI

"...not everything that is inherited is genetic."

—Boris Ephrussi
geneticist and embryologist (1901–1979)

Munira Basrai, of the NCI Genetics Branch, opened the Research Festival symposium on "Epigenetics and Cell Cycle Control: From DNA Replication to Cancer Therapy" with the above quote from Boris Ephrussi, whose insights presaged the elucidation of epigenetics.

Epigenetics is the study of DNA modifications that persist during cell division but that do not change the DNA sequence itself. NCI’s Shiv Grewal, Mirit Aladjem, and David Schrump discussed ways in which epigenetic phenomena control gene expression, the timing of DNA replication, and the potential of cells to form tumors.

Grewal, of the Laboratory of Molecular Cell Biology, explained how gene expression is turned off, or silenced, in particular regions of the genome of fission yeast. Silencing occurs, he said, when specific proteins bind to the DNA, converting it into a form called heterochromatin.

How does the cell know which parts of the genome should be silenced? The answer, at least in some cases, Grewal said, lies in the RNAi pathway, a system that has lately come to the fore with its increasing use by researchers to knock out the function of genes of their choice ina variety of experimental organisms.

Some DNA regions destined to be silenced contain a short piece of repeated sequence. RNA made from the repeated sequence forms a double-stranded conformation, which is chopped up into small pieces by enzymes in the RNAi pathway.

The pieces, in turn, help recruit the enzymes and structural proteins needed to make heterochromatin to the homologous DNA sequence from which RNA was made.

Heterochromatin formation is epigenetic because it alters the DNA in a nonpermanent way that nevertheless is very stable and can be passed on to the next generation.

Replication Timing

Aladjem, of the Laboratory of Molecular Pharmacology, described new insights into how cells control when different parts of the genome are replicated.

In general, she said, regions with high levels of gene expression replicate early and silent regions replicate late.

To investigate why this is, Aladjem and her colleagues inserted the DNA sequences that code for the blood protein b-globin into a new home—a region of chromosome 15 that normally replicates late. In one orientation, the transplanted DNA was expressed and replicated early; in the other orientation, it was not expressed and was replicated late.

A particular chemical modification of the DNA known as methylation was associated with late replication. In fact, if the b-globin DNA was methylated before it was inserted, it replicated late in both orientations.

b-Globin DNA that was altered so that it could not be methylated, however, showed a normal pattern of replication timing, suggesting that methylation is not required for late replication. A portion of the b-globin sequence called the locus control region proved to be important for regulating replication timing.

Finally, Aladjem said, although late replication and gene silencing normally go hand in hand, the two processes are distinct, as newly inserted b-globin sequences replicate late even before they are fully silenced, a process that takes several weeks.

Victor Marquez More Epigenetics ON DECK: ZEBULARINE Epigenetics also took center stage at Hispanic Scientist Day on October 13. Keynoter Victor Marquez, chief of the Laboratory of Medicinal Chemistry, NCI-Frederick, discussed a promising anticancer drug called zebularine, which reduces DNA methylation much like DAC, the drug studied by Schrump’s group. Chemically speaking, zebularine closely resembles cytidine, one of the normal DNA bases. Incorporated into DNA at a low rate, it binds tightly to the enzyme responsible for DNA methylation, trapping it and blocking its further activity. The overall reduction in DNA methylation stimulates increased expression of important antitumor genes such as p16 and p53. This mechanism of action is very similar to that of DAC. Although DAC is the more potent of the two drugs, zebularine has a couple of important advantages. First, it is far less toxic than DAC because its activity is restricted mainly to tumor cells—the enzymes necessary to process zebularine so that it can get into DNA are expressed at much higher levels in tumor cells than in normal cells. Second, it can be administered orally—tumor growth was slowed in mice that drank water containing zebularine. Marquez, in collaboration with the NCI Cancer Therapy Evaluation Program, is currently preparing for a clinical trial of zebularine. So far, the NCI Developmental Therapeutics Program has produced large quantities of the drug and is in the process of completing toxicity studies in animals needed for inclusion in an IND. n —Karen Ross

Epigenetics at the Bedside

Two drugs that induce epigenetic modifications of DNA are the subjects of clinical trials involving lung cancer patients, reported Schrump, of the Thoracic Oncology Section of the Surgery Branch.

Tumor cells undergo a complex pattern of epigenetic changes in their DNA that affect gene expression, Schrump said.

Early in tumor formation there is widespread demethylation of DNA, which leads to a general increase in gene expression; however, a few critical tumor suppressor genes undergo the opposite modification—they are hypermethylated and silenced.

Schrump is investigating whether two drugs—5-aza-2'-deoxycytidine (DAC) and depsipeptide (DP)—can awaken these antitumor genes and halt the progression of lung cancer. DAC inhibits DNA methylation; DP is one of a class of drugs called histone deacetylase inhibitors that block the removal of acetyl groups from histones. (Histones are DNA-binding proteins that organize and compact the DNA. Acetylation of histones is associated with increased gene expression.)

In a phase I clinical trial with DAC, Schrump reported, patients’ tumors stabilized, although they did not shrink, and the expression of some tumor suppressor genes did increase. In addition, some tumors exhibited increased expression of tumor antigens that are known to be regulated by epigenetic mechanisms. A phase II trial with DP and a trial in which DAC and DP were given sequentially yielded similar results.

Schrump and his colleagues are now testing sequential doses of DP and flavopiridol (FLA), a drug that influences gene expression by an entirely different mechanism. Early evidence suggests that FLA may increase the effectiveness of DP and make cells more susceptible to programmed cell death.

The Multifaceted Per2

Per2, a protein that affects gene expression and cell cycle, turns out to have a surprising new function in metabolism. Jay Chung, Laboratory of Biochemical Genetics, NHLBI, discussed this protein of interest in another context: its role in regulating food intake and glucose levels in mice.

Per2 mutant mice develop tumors in their salivary glands, and their cells have an abnormal response to DNA damage. The mutants also cannot maintain a steady sleep-wake cycle if they are kept in constant darkness, which indicates that they have defects in their circadian clock.

Chung’s group discovered that Per2 mutant mice, particularly females, gain weight unusually quickly when fed a high-fat diet. Like people with type 2 diabetes, the mice are insulin resistant, unable to control their blood sugar despite producing adequate levels of insulin.

Per2, Chung noted, appears to be part of the preopimelanocortin/a-melanocyte stimulating hormone pathway, which is known to regulate appetite. n

Return to Table of Contents