Let’s expand on what is known about genes and cancer. Perhaps no disease relies more on genome-related risks than cancer. To explain why, we need to backtrack a moment. As mentioned earlier, while still a student at Harvard Medical School, Rudy was thrilled to participate in the first study to ever find the gene for a disorder of unknown cause (Huntington’s disease). Since those pioneering studies using genetic analyses in the early 1980s, the hope has been that all of the mysteries of inherited disease could be solved by comparing the genome of patients versus that of their healthy counterparts. In that total of 6 billion letters, combining A, G, C, and T, inherited from our parents, only about 200 million are used to make up the genes. The sparsely distributed genes are like words in the story of life told by the genome. The remaining 5.8 billion letters serve to arrange and punctuate those words, potentially creating many variations of the same story. For the most part, after the Huntington’s disease gene discovery, from 1990 to 2010, geneticists spent most of their time looking for disease mutations only in the DNA sequence of the genes, like typos in the words of the genome story. But epigenetics now tells us that much of the story is in that intergenic DNA, the regions of the genome that we used to call “junk DNA” lying in between the genes. These regions determine how the story is read and which chapters matter most.
In an editorial in Nature accompanying the first data to emerge from the comprehensive catalog known as the Roadmap Epigenome Project, it was stated: “In human diseases, the genome and epigenome operate together. Tackling disease using information on the genome alone has been like trying to work with one hand tied behind [one’s] back. The new trove of epigenomic data frees the other hand. It will not provide all the answers. But it could help researchers decide which questions to ask.” It turns out that most common diseases with a genetic basis are highly complex, and a large number of factors, ranging from genome mutations inherited from our parents to epigenetic modifications brought on by life experiences, conspire together to determine one’s risk for specific diseases.
In the decades-long war on cancer, definite progress has certainly been made. But according to the American Cancer Society, as of 2015, over 1.6 million Americans are still diagnosed with cancer each year and nearly 700,000 succumb to cancers of all types. More than any other disease, cancer has led to incredible progress in understanding the genetic mutations responsible for the disorder. And, the current belief is that the development of cancer is due to the accumulation of gene mutations causing the cells to become cancerous and form tumors of various types. However, we now know that the risk for cancer is also dependent on the way in which epigenetic modifications to the genome render certain regions more prone to newly occurring mutations. (To date, the greatest evidence for the role of epigenetics in disease comes from cancer studies, in fact.) These mutations can be triggered by exposure to certain environmental toxins—for example, dioxin, a lethal family of chemicals found in pesticide manufacture and waste incineration, for which there is no safe dosage. The Environmental Protection Agency estimates that the damage being caused by dioxins outstrips that caused by DDT in the sixties. An environmental toxin can have the ability to cause new epigenetic alterations. These can modify how the genomic DNA in that region is folded, which in turn can potentially affect where new mutations are allowed to form.
Thus tumor formation involves multiple steps including both genetic and epigenetic alterations in the genome. Unlike gene mutations, the epigenetic modifications can be considered impermanent and even reversible. Some forms of cancer are brought on by genes that are activated via a process called hypomethylation (hypo is a Greek prefix meaning “under”). In this case the methyl marks on genes that silence their activity have somehow become removed. Without a suppressor to hold them back, the harmful genes are activated. In other cases, the reverse happens. Turning off certain genes via methylation can lead to tumor formation or can involve the addition of acetyl chemical groups to the histone proteins that wrap around the DNA.
New drugs are now being developed that would offset these tumor-causing epigenetic alterations. For example, drugs known as DNA methyltransferase inhibitors (DNMTIs) act as demethylating agents that can remove methyl marks from genes. Such drugs are already used successfully to treat forms of leukemia. Other drugs, called histone acetylase (HDAC) inhibitors, are also being used for treatment of leukemia and lymphoma. Of course, these so-called epidrugs are not without problems, since they are terribly specific in their actions on the genome. And while they are being used with some success in treating blood cancers, they have not yet been very effective against solid tumors. While we hope for the best with this new class of epidrugs, we must also consider the need for studies of lifestyle changes—for example, healthy diet, stress management, exercise, weight control, and the like—that would achieve the same outcomes.
Randomness is more than a theoretical issue—in our own lives cancer causes a major portion of human suffering. Twenty years ago, in the 1990s, it was thought that cancer was essentially random, putting almost everyone at equal risk. Genetics reinforced the public image of cancer as ruthlessly impersonal, striking any victim it chose. There were countering arguments. Those who thought that cancer was caused by toxins pointed to tobacco and asbestos as prime examples. Others who argued for viruses pointed to cervical cancer, which is caused by the human papillomavirus (HPV). It turned out that everyone had a piece of the puzzle, or as one leading cancer expert called it, each camp was like one of the blind men holding on to a different part of the answer.
The current view brings us back to our familiar image, the cloud of causes. Environmental toxins, viruses, and random mutations all play a role, and as with the puzzle of why Dutch men suddenly became the tallest in the world, the cloud isn’t very satisfactory when trying to link cause and effect. The only real certainty is that all roads lead eventually to the genome. Cancer of any kind is now known to need a trigger inside the cell, in the form of a cancer gene (oncogene). There are many such genes, and in recent years they’ve been cataloged by a worldwide effort to formulate the Cancer Atlas, a complete genetic road map to the disease. Besides turning on an oncogene, cancer can begin by turning off its opposite, the tumor suppression gene.
Once one talks about switches being turned on and off, epigenetics enters the equation, and so do questions concerning randomness, because the event that triggers the switch may not be random at all. Smoking cigarettes isn’t a random event. If you smoke, your risk of contracting lung cancer enters the realm of high probability. But the epigenetic explanation for cancer offers as many problems as solutions. For one thing, the futile hope that cancer might involve a single gene, which perished three decades ago in the 1980s, has been repeated in epigenetics—it turns out that while one gene mutation may lead to a certain type of cancer, the disease seems to involve up to fifty or one hundred genes. Cancer genes can continue to mutate as the cancer spreads, making the malignancy a fast-moving, extremely elusive target. Gene-targeted drugs have garnered headlines by curing specific cancers like one form of childhood leukemia that involves only a single gene.
After two decades of searching for similar drugs to wipe out a variety of cancers, however, success has been very limited. To make matters worse, drugs that work brilliantly at wiping out all traces of malignancy often have a tragically temporary effect. The patient returns after a few months with his cancer returned. On the surface, it would seem that cancer’s secret weapon is how quickly and randomly it can mutate, upholding the evolutionary dogma that randomness rules.
But there are signs pointing in a new direction. Of all diseases, none more than cancer has been clearly linked to epigenetic aberrations.
The epigenomes of specific types of cancer cells carry the same epigenetic fingerprint that matches the cell that started the cancer. This serves to reveal the tissue in which the cancer originated, no matter where in the body it is found. Such information could be of immense use in the future for diagnosing and treating different forms of cancer, because once it has spread, a tumor has often been extremely difficult to trace back to where it started. Further complicating the problem is the cancer cell’s habit of continual mutation. Hopefully, by comparing the epigenomes of healthy and malignant cells, we can better understand how the risk for disease can be influenced by much more than the genomes provided to us by our parents.
It turns out that carefully examining the epigenetic marks (methylation and acetylation) can actually be predictive of which kind of cancer will develop. This revelation turns out to be the opening wedge against random mutations. As you live your life, and your environment and experiences chemically govern your gene activities—we’ve discussed this extensively already—specific new mutations can arise that are the same for every cell in a particular type of tumor. So epigenetic modifications lead to predictable new mutations. Something that’s predictable steps away from being purely random.
This level of predictability doesn’t solve the entire mystery, however. By analogy, think of the weather. On a summer day in August, thunderstorms are very likely to arise, and their timing can be predicted with a fair degree of accuracy—as the heat of the day builds up, a storm is more likely in the afternoon or evening than in the cool of the morning. But the exact movement of air currents, moisture, and clouds is much less predictable, and if you want to know the cause of a specific thunderstorm down to the last molecule of air, it’s impossible. In cancer, many mutations often occur simultaneously, and not all lead to bad results. Thousands of possibilities arise, with great unpredictability. (Because something is unpredictable doesn’t make it random. The next thought you are going to have isn’t random, but it is unpredictable. Cancer research has yet to figure out if cancer is like that or not.)
This realization created immense discouragement following hard on the triumphant findings about the genetic causes of cancer. Oncologists began to mutter about cancer as a devious enemy whose arsenal of defenses kept increasing every time a solution seemed to be at hand (a good example of our point in the previous chapter that cancer unfortunately can draw upon the cell’s complete intelligence). Now hope is rising again, because the Cancer Atlas has been sorting out which mutations are the dangerous ones, but just as important—and perhaps the single best clue to curing the disease—it appears that cancer develops along some set pathways that are fairly small in number, perhaps only a dozen for every kind of malignancy. In other words, there’s a pattern that goes even further to undercut the orthodox view about random mutations.
One promising finding is that certain tumors take many years, even decades, to develop after the initial trigger starts a cell on an abnormal course. The thought is that a specific sequence—the genetic pathway that an abnormal cell must follow—involves a series of steps that must unfold in order. Here’s an analogy: You’ve probably seen the little handheld games that involve steel BBs rolling around on a board with holes in it, the object of the game being to tilt the board around until you manage to get all the BBs to fall through a hole. The holes are tiny, so it’s not an easy challenge. Now imagine that a cancer mutation is faced with a similar challenge. It must thread its way through a small opening (a specific genetic modification out of myriad possibilities) in order to move on to the next stage. Once that’s accomplished, the next small opening presents itself in the form of a new mutation out of myriad choices, and so on.
If a cancer is typically slow growing, as types of colon and prostate cancer are, it may take thirty or forty years for a cancer cell to follow the whole sequence. The hope is that if detection can be made as early as possible—detecting the predictable fingerprint of epigenetic markings—cancer will be conquered long before the first symptoms appear. This glimpse of light at the end of the tunnel follows from the discovery that the exact gene mutations of many types of tumors can now be predicted from the epigenomic signature of the cell type from which that cancer most likely originated.
We must then at least wonder, is it possible that when epigenetic mutations arise in adults as a result of toxins, stress, trauma, diet, and the like, predictable new mutations will arise in certain cells? If the mutation occurs in sperm and egg cells, could they be passed on to the next generation? We don’t yet know. But even the possibility would have made Darwin’s head spin and is today leading to a major revision of his theory.
If epigenetic alterations do lead to specific mutations beyond those that cause tumors, then one’s life experiences and environment could, at least theoretically, lead to expanded predictability. There could be epigenetic signatures of other chronic illnesses that appear long before the first symptoms. It would be even more amazing if prevention extended to unborn generations that have been inheriting these marks in the womb. At the time this book was being written, such possibilities were only a very intriguing set of conjectures. Yet it’s fascinating to think about what future studies in this area will reveal.
So far we’ve been focusing on the genetic contributions to disease risk, but there’s an elephant in the room—the impact of environmental toxins on our genes and epigenome. The Centers for Disease Control and Prevention has found 148 different environmental chemicals in the blood and urine of the U.S. population. Increasing evidence gives support to the notion that environmental pollutants likely cause various diseases by inducing epigenetic changes in our genome, thus altering the activities of specific genes. For example, arsenic in contaminated water dramatically affects methylation of the genome, leading to bladder tumors. Exposure to high levels of other heavy metals (nickel, mercury, chromium, lead, and cadmium) in food and water supplies can also cause changes in gene methylation, leading to various types of cancer, including lung and liver cancers. The bottom line is that there are an estimated 13 million deaths or more worldwide due to environmental pollutants, many of which have been linked to epigenetic modifications of the genome.
We are not alarmists, but it’s important to follow where the science leads. Perhaps no one has advanced our knowledge of this issue as much as Dr. Michael Skinner, a developmental biologist at Washington State University. In one study Skinner exposed pregnant rats to a chemical known to interfere with embryonic development, a fungicide called vinclozolin used to keep mold off vineyard grapes, along with other blights and rots on fruits and vegetables. Vinclozolin had already been shown to decrease fertility in male mice. The disturbing thing that Skinner found was that the progeny of the chemically treated mice, all the way down to the fourth or fifth generation, were also affected with low sperm counts. This result was successfully replicated fifteen times.
The reason for the disruption of sperm production brought on by vinclozolin wasn’t mutations in the DNA, but epigenetic modifications in the exposed adult mice (via methyl marks), which were then passed on to the next generations. (This is different from what we normally hear about, when actual mutated genes for disorders get passed on from parents to children, as in sickle cell anemia.) Thus another clue was being added to the existence of “transgenerational genetics.”
Moreover, Skinner and his colleagues found that there was a specific pattern to where the methyl marks were attached in the genome after exposing mice to different types of chemical toxins. Each toxin, whether it was insecticide or jet fuel, left its own distinctive pattern. In some cases, the shifts being caused in gene activity could then be inherited and predispose the offspring to specific disorders. For example, the insecticide DDT, which has long been banned in the United States because of its disastrous effects in the food chain of animals and birds, also has a specific epigenetic effect. Exposing mice to DDT has been shown to create a predisposition to obesity in later generations, along with obesity-associated diseases such as diabetes and heart disease.
The range of detrimental epigenetic changes brought on by pesticides is wide. The pesticide methoxychlor, used to protect livestock from fleas, mosquitoes, and other insects, has been shown to cause testicular and ovarian dysfunction in mice. Another pesticide, dieldrin, has dramatic effects on epigenetic modifications (acetylation) to histones leading in mice to nerve cell death associated with Parkinson’s disease. Skinner also showed in mouse studies that the common pollutant and carcinogen dioxin, a waste product of many industrial processes, causes epigenetic inheritance of prostate disease, kidney disease, and polycystic ovary disease.
One of the most carefully studied environmental toxins that can cause abnormal epigenetic changes is bisphenol, or BPA. It has been widely used to make the plastics used in food and beverage containers, including baby bottles. BPA is well known to cause epigenetic changes. We’ll cite just a sample of relevant studies. Research at Tufts University showed that BPA can change gene activity in mammary glands of rats exposed to the chemical in the womb, rendering them more vulnerable to breast cancer later in life. Previously BPA was demonstrated to leave male rats at higher risk for prostate cancer. In another set of studies, BPA produced epigenetic changes associated with changing the yellow color of a particular breed of mouse as well as increasing the risk for cancer. (Note: One way to avoid BPA exposure among infants is to use glass bottles and containers or look for the label “BPA free.”)
Finally, diethylstilbestrol (DES), which was used from 1940 to 1960 to prevent miscarriages in pregnant women, has been shown to increase the risk for breast cancer. We now know that this risk is associated with epigenetic changes. One must wonder, then, whether these changes are passed down to the next generations, along with the increased risk.
Air pollution, especially from particulate matter in vehicular exhaust, also causes epigenetic changes that can lead to inflammation throughout the body. Benzene, which is found in gasoline and other oil-based fuels, leads to altered DNA methylation associated with leukemia. In our water supply, chlorination leads to by-products with names like trihalomethane, triethyltin, and chloroform, all of which can induce epigenetic changes in the genome. Many of these chemicals have been studied for detrimental effects on health. Rats with triethyltin in their drinking water suffered increased incidence of brain inflammation and swelling associated with increased methylation activities. Chloroform and the trihalomethane known as bromodichloromethane increased methylation in liver cells in a gene associated with liver disease.
Even benign substances we don’t associate with such risks can have a hidden story in their production. Alarmingly, many Indian spices sourced from India have been found to be contaminated with heavy metals. The cause is likely the proximity of spice farms to smelting and mining operations and the resulting use of contaminated irrigation water. In 2013 alone, the FDA denied import of more than 850 spice shipments from around the world. To minimize such risks, U.S.-grown organic spices can be used safely, while care should be taken with those derived from India and China. Buying from reputable sources with known brand names can help. But one must be especially careful with spices obtained over the Internet or in unbranded, anonymous containers found, for example, in small neighborhood shops. In many cases, some specialty stores can obtain spices that bypass FDA inspection. While only about 2 percent of imported spices are found by the FDA to be contaminated, you significantly increase your odds of obtaining them when you buy unbranded spices from anonymous overseas sources.
Taken together, there is little doubt that a wide range of environmental toxins and pollutants can alter our epigenome, resulting in increased susceptibility to a host of different cancers (breast, liver, ovary, lung) and other diseases, including schizophrenia, diabetes, and heart disease. Each person’s exposure is unique and different, which vastly complicates the problem. Some experts foresee the day when we will visit the doctor to get complete scans of our epigenetic alterations in order to determine our future risk for disease. Will we be increasingly using epigenetic-based drugs like HDAC inhibitors and RNA-based therapeutics to offset these risks and treat disease?
These scenarios are beginning to turn into reality. In this book we’ve offered an alternative you can pursue today, changing your lifestyle to mitigate risk, and perhaps in the future this approach, too, will be fine-tuned to specific epigenetic marks for disease. An even bigger question, based on studies like the ones cited here, is whether the epigenetic changes in adults living today will be inherited by the next generations tomorrow. Dr. Michael Skinner seems to have little doubt: “In essence, what your great-grandmother was exposed to could cause disease in you and your grandchildren.”
Along these lines, it will be critically important to continue to be aware of how epigenetic modifications arise in response to environmental toxins and pollutants. This is the only way we can move forward, for the good of our own health and the health of generations as yet unborn.