EPPIE LEDERER AND PAULINE PHILLIPS were one of the most famous pairs of identical twins in the United States during the 20th century. Born 17 minutes apart, both women became wildly popular syndicated columnists—as Ann Landers and Abigail Van Buren, respectively—and dispensed tart-tongued advice about love and other matters. Photos from their younger days reveal that the two women were uncanny look-alikes, both graced with fashion-model cheekbones and vibrant eyes.

Over the years, ever-changing hairstyles made it easier to tell them apart. But it was their dramatically diverging health, finally, that truly distinguished one from the other. Eppie died of multiple myeloma at age 83, while Pauline lived to be 94 before succumbing to Alzheimer’s disease this year. That may seem surprising—after all, as identical twins they have perfectly matched DNA. But it turns out that twins have rates of “disease discordance”—that is, if one has a medical condition, the other twin typically won’t get it—that are well over 50% for most conditions.

Yet if DNA is not destiny, how is it that genes and environmental influences interact to bring about disease? Part of the answer may come from the burgeoning study of epigenetics. Developmental biologist Conrad Waddington is credited with coining the term in the 1940s, and even today there is disagreement among scientists about how to define it. There is, however, agreement that humans and animals have a chemical infrastructure—an epigenome—that switches genes on and off. (The prefix epi derives from the Greek, meaning “over” or “above.”) Evidence suggests that environment and lifestyle choices can trigger epigenetic modifications, and that could help explain why identical twins end up being not so identical.

Tinkering with gene switches can have a profound effect on how they behave, and labs around the world are now scanning the genome in hopes of identifying epigenetic modifications—sometimes called “marks” or “tags”—that could serve as biomarkers, helping to predict and identify a wide range of conditions. Drug developers, meanwhile, are exploring compounds that may be capable of thwarting epigenetic modifications that cause key genes to become under- or overactive, thereby promoting cancer and other diseases. Several medications that work by altering the epigenome are already in use.

And in all of this, identical twins are increasingly seen as desirable study subjects; because their genomes are the same, eliminating one essential variable, differences in their epigenomes might offer clues to the molecular origins of disease. “Epigenetics provides the mechanistic link between nature and nurture,” says molecular biologist Nessa Carey, author of The Epigenetics Revolution.

AS THE NAME SUGGESTS, the epigenome acts directly on genes, the basic units of heredity. Genes are made up of DNA and govern production of proteins that form every tissue in the body. This process, known as gene expression, begins with transcription, in which a molecule called messenger RNA transfers the information in DNA out of the cell’s nucleus and into the cytoplasm, where it is translated into amino acids that form proteins.

The body’s complete library of DNA, known as the genome, is found in every cell. Yet only a portion of the genes within a cell are switched on and producing proteins at any given time. A variety of influences determine the on/off status of genes. For instance, mutations can turn off a gene or alter the types of proteins it makes. Moreover, the epigenome creates a pattern of modifications that help determine which genes are turned on and off.

“Most of the epigenome is set during embryogenesis and early development,” says Jean-Pierre Issa, director of the Fels Institute for Cancer Research and Molecular Biology at Temple University in Philadelphia. By helping regulate genes’ on/off status, the epigenome ensures that a developing liver cell doesn’t try to become a hair cell or neuron. What’s more, the epigenome helps ensure that patterns of gene expression are preserved when cells divide.

The most-studied form of epigenetic modification is DNA methylation. Scientists have long known that DNA can become tagged, or “decorated,” with common organic compounds called methyl groups. Methylation can turn off a gene’s protein-making machinery. It matters, as far as gene expression is concerned, only if it occurs in a “promoter” region, a sequence of DNA needed to turn a gene on or off. When methylation blocks a promoter site, RNA can’t transcribe that gene and it can’t be expressed, or make proteins.

The other common form of epigenetic tag is called histone modification. To fit inside the tiny nucleus of a cell, strands of DNA wrap around histones, which are spool-shaped proteins. The resulting package is called chromatin, and scientists have identified several tags that alter chromatin’s structure and affect a gene’s capacity to produce proteins. Some chromatin alterations loosen DNA’s tight wrap, which has to happen for transcription to occur. In other cases, histone modifications may block RNA from transcribing DNA and silence a gene.

Other factors, too, may influence gene expression. For example, some forms of noncoding RNA (meaning it’s not involved in translation to proteins) have epigenetic activity. Next-generation gene sequencing is allowing researchers to identify regions where modifications occur.

What’s more, some scientists believe that studying identical twins is helping to push epigenetics forward. While twins may start out epigenetically matched, their epigenomes appear to “drift” over time. In 2005, a team of researchers at the Spanish National Cancer Research Centre led a study that analyzed DNA methylation patterns to see how they affected gene expression in identical twins ranging in age from 3 to 74. It turned out that each pair of toddlers had virtually the same methylation profiles. But that was decidedly not the case with the 50-year-olds, who had very different methylation profiles—and the more time a pair of twins had spent apart, the more epigenetically distinct they had become.

Issa and other skeptics have questioned the methods used in this oft-cited study and doubt whether the differences detected in the older twins were all that large. “That paper gives the impression that it’s all environment and that there are huge changes, but neither of those things is true,” says Issa. Esteban Ballestar of the Bellvitge Biomedical Research Institute in Barcelona, co-author of the study, defends the paper’s conclusions and notes that he and his colleagues have launched a larger study to validate their work.

Other recent research has found that identical twins who are discordant for various diseases—such as autism, type 1 diabetes, systemic lupus erythematosus and psoriasis—often have very different DNA methylation patterns. And studies of identical twins and non-twins have found patterns of possibly unhealthy DNA methylation in neuropsychiatric disorders, including Alzheimer’s disease, bipolar disorder and schizophrenia.


ABNORMAL PATTERNS OF DNA METHYLATION are a hallmark of virtually every form of cancer, says cancer biologist Stephen Baylin, deputy director of the Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins University School of Medicine and an epigenetics pioneer—and indeed cancer has been a primary focus of epigenetic research, perhaps in part because gene modifications that produce malignancies stand out so clearly. Massachusetts General Hospital pathologist Bradley Bernstein has studied how alterations to chromatin drive glioblastomas, the most common form of brain tumor. “The difference between a glioblastoma cell and a healthy neuron is massive,” says Bernstein. “The epigenomic state of a cancer cell is totally different—it’s been reprogrammed.”

Baylin and others have shown that the reprogramming often silences tumor suppressor genes, which act as brakes on abnormal cell growth. Genetic mutations that alter tumor suppressors such as the p53 gene are known to help tumors grow, and epigenetic modifications, too, can deactivate genes like this one.

A major milestone in epigenetic research occurred in 1980 when University of Southern California molecular biologist Peter Jones discovered that a compound called azacitidine—already known to stop leukemia cells from dividing—also blocks DNA methylation. After decades of additional research, azacitidine (Vidaza) was approved by the Food and Drug Administration in 2004 for treating myelodysplastic syndrome (MDS), a cancer of the blood and bone marrow. Another methylation inhibitor, decitabine (Dacogen), is also approved for MDS.

Meanwhile, two other drugs that alter unwanted epigenetic modifications treat cutaneous T-cell lymphoma. Romidepsin (Istodax) and vorinostat (Zolinza) inhibit enzymes called histone deacetylases (HDACs) that tighten the coil around DNA and block expression of tumor suppressor genes. In 2011, Baylin and colleagues published a study showing that some lung cancer patients who haven’t responded to other therapies go into long-term remission when treated with a combination of azacitidine and an experimental HDAC inhibitor called entinostat. A larger version of that trial is under way.

In 2013, a noncancer drug that works by altering epigenetic status gained FDA approval. Mipomersen (Kynamro) reduces low-density lipoprotein (LDL) cholesterol by silencing a gene that produces apolipoprotein B-100, a building block of LDL; the drug can treat familial hypercholesterolemia, an inherited disease that may cause heart attacks at an early age. With many more such drugs in the pipeline, interest “is going to start snowballing as we understand how drugs modulate these pathways,” says University of Illinois pharmacologist Julio Duarte, whose paper, “Epigenetics Primer: Why the Clinician Should Care About Epigenetics,” appeared in Pharmacotherapy this year.

OF COURSE, DRUG TREATMENTS WOULDN’T be necessary if methylation and histone alterations didn’t go awry in the first place. Some of the epigenetic variation that occurs between identical twins is likely because of random changes, says Nessa Carey. “It’s probably impossible to have a system that is responsive but at the same time completely protected from random change,” she says.

Yet other changes seem to be linked to environmental exposures that can have important effects. “The small percentage of the epigenome that is subject to change over a lifetime can have a great impact on the behavior and function of genes,” says Ballestar. Identifying which environmental influences alter the epigenome has been a challenge, however.

Baylin’s research suggests that chronic inflammation may have a profound impact on the epigenome. “Inflammation is a toxic environment for cells,” says Baylin. Inflammation forces cells to repair themselves constantly, and during that process, some genes may become methylated and shut down. If such seemingly transient changes persist, they may alter the expression of genes needed to ward off cancer and other diseases.

The influence of tobacco on the epigenome is clear-cut. German researchers reported in 2013 that smokers have strikingly different methylation patterns from those found in nonsmokers and in people who quit using tobacco. Moreover, a 2012 National Institute of Environmental Health Sciences study, looking at more than 1,000 babies of mothers who smoked during pregnancy, identified 10 genes with abnormal methylation.

A large-scale study of identical twins may help elucidate some ways that environment and lifestyle choices alter epigenetics. Launched in 2010, EpiTwin was founded by epigenetic epidemiologist Tim Spector of King’s College in London and includes about 5,000 identical twins. Every three years, each twin’s blood is analyzed for possible changes in methylation patterns at 20 million sites in the DNA. “People called us mad when we started collecting DNA more than once, because the prevailing wisdom was that it couldn’t change,” says Spector, author of Identically Different: Why You Can Change Your Genes. Yet he and his colleagues have indeed found alterations in participants’ methylation patterns.

Still, not all epigenetics researchers are convinced that studying identical twins who are discordant for a disease will yield meaningful discoveries. “To assume that any difference between twins is being mediated by histone modification or DNA methylation is too much of a jump,” says MGH’s Bernstein.

But Spector argues that EpiTwin has already made valuable contributions—for example, through a study of 15 twin pairs discordant for breast cancer that pointed toward hypermethylation of the DOK7 gene, which he believes could one day be used as a biomarker to help identify women at risk for breast cancer years in advance. Other work, with findings about pain and type 2 diabetes, will soon be published.

Classic twin studies of the past focused primarily on the similarities in identical siblings, especially in the relatively rare pairs who were raised apart. Spector believes the greater value in studying identical twins is in understanding how they differ. “As twins who start the same end up different, we’re going to see how they epigenetically vary from 10 years before,” says Spector. “That’s going to be very exciting.”