9 ways to change your epigenome

Alan Horsager, Ph.D.
June 18, 2018

I sometimes describe epigenetics as molecular dust. It's a set of molecules that sit on top of your DNA and may, well, be mistaken for dust. If it wasn't so damn important. The implications for epigenetics are huge and that molecular dust I'm talking about could be more important than the DNA itself. Epigenetics, in my opinion, changes the way we view health and disease.

People think of DNA as relatively immutable. And, for the most part, it is. Think of DNA as the dictionary. Epigenetics turns those words into the story of you. Unlike DNA, the epigenome can change over time and is impacted by various lifestyle factors such as diet, exercise, and sleep pattern.

The "epi" in epigenetics means "above" or "upon". Epigenetics is the study of any molecular process that changes how genes are expressed into proteins, without changing the underlying DNA sequence. In other words, how genes are turned on and off.

Every cell in the human body has the same DNA sequence that includes about 20,500 genes. However, each of these genes isn't activated in each of the different cell types in your body. Furthermore, genes can be turned on and off at different points in time. Gene expression, and the epigenome that controls it, is the spatiotemporal symphony that makes up the human body.

The most well-known epigenetic "marks" are DNA methylation (methyl groups that sit on top of DNA and silence gene activity) and histones (the spools that DNA wrap around; how tightly spooled the DNA is around these histones can control the accessibility of various gene expression activators (i.e., transcription factors).

As scientists have become able to study epigenetics with advancements in molecular biology and DNA sequencing, hundreds of research groups have looked at how the epigenome changes with environmental factors. This evergreen post is intended to look at current research that supports the idea that the epigenome is dynamic and can change with various lifestyle influences.

Diet

There is growing evidence that a person's diet can serve as a source of epigenetic signaling. Researchers are now exploring how diet can actually modify a person's epigenome, and the potential impact on their offspring. Being able to understand the diet-epigenome relationship could allow us to identify nutritional adjustments that could prevent or ameliorate diseases such as diabetes, obesity, and heart disease.

One of the most famous studies looking at the diet-epigenome relationship involved victims of the Dutch Hunger Winter. In 1944, Nazis stopped trains from delivering food to the Netherlands, leading to a severe famine that killed an estimated 20,000 people. Pregnant women who gave birth during that winter were living on 400-800 calories a day. When studies were conducted on the adult offspring, they found increased rates of obesity, cardiovascular disease, diabetes, and schizophrenia. When they looked at the epigenomes of these individuals, they found certain genes were epigenetically silenced compared to healthy controls, suggesting that the severe hunger of the mothers had lead to changes in the unborn child at the level of the DNA.

It's not just the mother's diet that could have an impact. A number of studies have now shown that paternal diet around the time of conception can impact the epigenome of the offspring. In one such study, offspring of male mice fed a low-protein diet had increased expression of genes involved with obesity and high cholesterol.

Okay, so parental diet around the point of conception can impact the epigenome of the offspring. Is there any evidence that you can change your own epigenome through changes in diet?

See the research supporting the diet-epigenome connection

One study looked at whether or not eating a Mediterranean diet changed the epigenome, a diet that is thought to have positive effects on cardiovascular risk, blood pressure, and inflammation. The epigenome in 36 subjects with high cardiovascular disease risk was evaluated at baseline and again 5 years after maintaining a Mediterranean diet. They found significant changes in DNA methylation in 8 genes important for inflammation and immune response. 

In another study, a research team that is part of the Integrative Cardiac Health Program studied 63 people with heart disease who followed a very low-fat diet and compared them to a control group. The control group showed no improvements in health but the low-fat diet group lost weight and blood pressure improved by approximately 10%. After 12 weeks on the low-fat diet, gene expression had changed in 26 genes. After 12 months, 143 genes had changed, including genes important for inflammation and blood vessel damage. 

Take Away: 1) Diet of both the mother and father during peri-conception can impact the epigenome of the offspring. 2) Changes to your diet can lead to changes in your own epigenome.

Exercise

It's pretty easy for us to all agree that going for a run, a walk, or a bike ride improves our health and reduces the risk of certain diseases. The data supports this. But how does this happen? What is going on in our bodies that converts exercise into a healthier you?

Turns out that, at least in part, there are changes happening at the level of the DNA. Numerous studies now show that exercise changes the epigenetic profile of our genes, and these changes correlate with improved health and fitness.

There have been several recent studies that have established changes in DNA methylation are associated with exercise. However, it is difficult studying freely moving humans and, subsequently, it difficult knowing exactly what caused the epigenetic changes - was it exercise, diet, or some other change in behavior?

See the research showing how exercise changes the epigenome

To address this question, one study from the Karolinska Institute invited 23 men and women to their lab to exercise just one leg for 45 minutes, 4 times per week. The other leg they left unexercised. The idea was both legs would be subject to epigenetic changes from diet and other behaviors, but the effects of exercise would be isolated to the one leg.

After 3 months of training, the exercised leg was certainly stronger than the other leg. No shocker there. However, they also found more than 5000 sites on the genome were epigenetically different between the two legs from muscle biopsies. Many of the gene that saw epigenetic changes are involved in metabolism, insulin response, and inflammation. Pretty strong evidence for epigenetics affecting the health of our bodies.

Okay, fine. Exercise changes the epigenome of muscle cells. But what about other cells in the body?

In a 2015 paper, an Australian research group looked at epigenetic changes in sperm cells. 24 men were split into two groups - one group engaged in 3 months of sprint interval training, the other group did nothing. The investigators collected sperm samples at the start of the study and at the end and looked for epigenetic changes in the sperm.

They found that exercise training significantly changed the epigenome of the sperm. What was particularly striking was that they found increased methylation (suggestive of gene silencing) at regions associated with debilitating diseases such as schizophrenia, Parkinson’s disease, cervical cancer, leukemia, and autism. Many of these methylation changes occurred in paternally imprinted genes that escape DNA methylation reprogramming after fertilization, suggesting these exercise-induced changes may be heritable across generations!

Another interesting finding is that they found a global decrease in methylation after exercise training. In many different tissue types, aging is associated with increased DNA methylation. Perhaps exercise could serve as a method for counteracting the effects of aging? An intriguing idea.

The evidence for exercise-related changes in your epigenome is strong. Now you can feel confident that you're not only going to look and feel better, but your DNA will work better, too!

Sleep

Sleep is still mostly a mystery; we're still not entirely sure why we need it to survive. The longer we stay awake or the more active we are, the greater our sleep debt. Certainly, sleep disorders have been associated with an increased risk for heart disease, high blood pressure, stroke, and diabetes. How is it that poor sleep hygiene could increase the risk for other diseases? There is now compounding evidence that shows lack of sleep can create changes to your epigenome.

In humans, as well as other mammals, there appears to be a core set of genes that regulate our sleep and wakefulness cycles, as well as other associated physiological systems. These are called the "clock genes".

What science supports the sleep-epigenome connection?

In a group of 15 otherwise healthy men, epigenetic changes occurred after just one night of sleep deprivation. These changes occurred in CRY1, PER1, and BMAL1, all genes thought to be important for the maintenance of circadian rhythm. The study is certainly not well powered nor is it clear how long-lasting those effects are. However, the study does provide some good initial evidence that sleep disturbances can manifest into changes in gene expression.

We've all had the experience of a bad night of sleep. Your brain just doesn't work right. You're not alert. You're forgetful. In another study, a group of researchers from the University of Montreal establishes that acute sleep loss impacts brain gene expression broadly. In mice, they tested the hypothesis that epigenetics coordinate the gene expression response to sleep deprivation. Compared to their well-rested controls, mice suffering sleep deprivation DNA methylation differences in genes important for signaling, neurotransmission, and synapse formation (i.e., the connections between neurons). Like in the previous study, it's unclear how persistent these changes are.

The reality is that we experience some level of sleep disturbance from time to time. Maybe these changes underly some ability of our body to adapt to stress? Whatever the case may be, we know that chronic insomnia is never good and this includes changes in DNA function.

Yoga & Meditation

Stress is pernicious and, at times, relentless. We'll get into all the harmful effects of stress later in this post but, for now, let's focus on two healthy ways (yoga and meditation) to manage it and the evidence for their positive impact on the epigenome.

Yoga and meditation have definitively been shown to be effective at stress reduction. Now epigenetic research is finding that these stress reductions methods may be changing the structure of your DNA.

Research supporting the positive impact of yoga and meditation on your epigenome

Chronic inflammation is associated with a number of health-related issues including depression, cancer, and obesity. Research is now showing that inflammation can be controlled using mindfulness meditation.

In a study evaluating the stressful effects of caring for patients with dementia, 45 caregivers either participated in mindfulness meditation or listened to relaxing music for 12 minutes each day over a 8 week period. In the meditation group, 68 genes were found to be differentially expressed, including a number of genes important for regulating inflammation. This was after adjusting for potential confounds including sex, illness, and BMI. These data suggest that even a brief meditation session could help reverse the harmful effects of chronic stress.

A study by Dean Ornish at UCSF found that men with prostate cancer who engaged in 60 minutes of mindfulness meditation each day had decreased the expression of genes associated with cancer. Specific oncogenes from the RAS family were down-regulated as a function of this intervention. Although this was really a preliminary study, it does suggest that meditation as a intervention could result in silencing of genes related to the development of cancer.

Another study looked at the impact of yoga on reducing stress and increasing well-being in women that reported chronic stress. As mounting evidence implicates immune response and inflammation as the molecular hallmarks of stress response, his initial study specifically looked for epigenetic changes. This study evaluated 28 women before and after an 8-week yoga intervention against a 116 person control group, measuring  interleukin-6 (IL-6), tumour necrosis factor (TNF) and C-reactive protein (CRP) protein levels, and the DNA methylation of these genes and the global indicator, LINE-1. All data was age and BMI matched to control for possible confounds.The researchers showed reduced methylation in the TNF region of the yoga group, while no other genes showed a significant difference. Given that the intervention was only 8 weeks in duration, it's amazing to see any statically significant difference at all. That being said, the results are compelling and suggest more research needs to be conducted to measure epigenetic changes from yoga practice.

Although the data is relatively early, these studies establish that practices focused on calming the mind could lead to a reduction in stress and an associated change to you epigenome.

Weight Loss

There around approximately 500 million people that are obese throughout the world. It is known as a significant risk factor for many diseases such as heart disease, cancer, and diabetes. Obesity is well know to be heritable. Indeed, children of obese fathers are at higher risk of developing obesity. Environmental factors certainly impact the epigenetic profile of cells, but could these factors associate with either weight gain or loss?

There is lots of controversy regarding the epigenetic inheritance of various traits. However, this study showed very compelling evidence that this can happen for obesity and diabetes. Using an in vitro fertilization technique to ensure inheritance via epigenetics rather than behavioral or dietary influence, these researchers show that parents consuming a high-fat diet renders offspring significantly more at risk of developing obesity and diabetes. These data suggest that the epigenetic inheritance of metabolic disorders may contribute to our current pandemic of obesity and diabetes.

Research behind the weight loss-epigenome connection

To determine if epigenetic changes occur after rapid weight loss (and how heritable these changes may be), a group of researchers from Denmark and Sweden looked epigenetic difference in sperm DNA in two different cases: 1) obese vs. lean men, and 2) morbidly obese men before and after gastric bypass surgery. The results of this study were remarkable. The DNA methylation profile in sperm were significantly different in lean and obese men and, astoundingly, the profile of obese men changed after gastric bypass surgery!

Another study examined whether methylation patterns in blood cells differ in individuals who successfully lost enough weight  to go from obese to normal weight (n = 16) vs obese (n = 16) and already normal weight (n = 16) individuals. In 3 different obesity related genes (RYR1, MPZL3, and TUBA3C), the successful weight loss group more closely resembled the normal weight group than the obese group. This suggests that losing weight causes your DNA to look more like normal weight individuals.

Marijuana Smoking

The past few years has seen a remarkable acceptance of marijuana use throughout the United State, with nearly 10% of population smoking pot.  While there is tremendous support for it's use in ameliorating symptoms of chronic pain, cancer, and depression, marijuana use is not without its downsides. Marijuana use can increase the risk of developing schizophrenia in those with a genetic susceptibility for the disease, suppression of the immune system, and increasing risk of male infertility (although reversible if you stop).

Marijuana is hardly the national killer but don't kid yourself. It's far from harmless. Let's go in eyes wide open.

See the research on the marijuana-epigenome relationship

The epigenome provides a cellular fingerprint of environmental exposures, including drug use, which makes it a highly relevant target for evaluating the impact of marijuana. So, are their epigenetic changes associated with pot smoking?

For those interested in digging in, there is a useful review of the scientific literature in Biological Psychiatry that outlines what we know about epigenetic changes that are linked to marijuana.

One of the first epigenetic associations with marijuana use involved the COMT gene, which is implicated in both schizophrenia and substance abuse. The gene is thought to play a critical role in reward, motivation, and other behaviors thought to be linked to addiction. Here, it is thought that marijuana use uncovered a risk for schizophrenia that is hidden in the COMT gene. Given that the DNA methylation status in the COMT gene depended on the frequency of marijuana use, it remains unclear if cessation would revert the epigenetic status of the gene back to the normal state.

Although animal studies are certainly not as compelling as human studies, they provide the opportunity for a much more controlled experiment. In particular, it is much easier to learn about how certain substances may impact the cells of the brain. A number of early studies demonstrated that prenatal THC exposure resulted in changes in gene expression and the resulting behavior of the offspring. Interestingly, exposure to THC has caused changes in epigenetic profile of the Drd2 and Penk genes, both of which are important for substance abuse and dependence.

Very few studies have considered the lifelong or multi-generational epigenetic impact of marijuana use. The early evidence is interesting but still needs to be followed up with significant studies to establish the gene X marijuana relationship. My gut tells me it falls somewhere in the range of alcohol or tobacco use, which we'll discuss next.

Tobacco Smoking

Smoking has been known to cause lung cancer, heart disease, stroke, chronic obstructive pulmonary disorder, and even infertility. There is now a growing body of evidence that shows smoking tobacco can change your epigenome. Previous research has established smoking cigarettes with a distinct epigenetic profile. It is also found as the greatest risk factor for accelerated biological aging when measured with the epigenetic clock.

Research showing the impact of cigarette smoking on the epigenome

In a study conducted by researchers at Johns Hopkins University, researchers grew human lung cells in the lab and exposed them to liquid cigarette smoke (the equivalent of smoking 1-2 packs per day) for 15 months. They recorded genetic, epigenetic, and other molecular changes and compared them to normal bronchial cells that had not been exposed to the liquid cigarette smoke.

“Our study suggests that epigenetic changes to cells treated with cigarette smoke sensitize airway cells to genetic mutations known to cause lung cancers,” says Stephen Baylin, M.D., one of the primary authors of the study. 

The group found DNA damage occurred even after 10 days from the start of the experiment. They also found changes in DNA methylation. In cancer, DNA methylation on tumor suppressor genes can silence the expression of important proteins that suppress tumor growth. Additionally, when methylation is removed from oncogenes, tumors can begin to proliferate because the gene is no longer silenced. Cigarette smoking is a powerful agent to adjust the DNA methylation profile in a very negative way.

Cigarette smoking appears to cause changes in the epigenetic profile and the impact can persist for extended periods of time. One research groups was interested in understanding the molecular underpinnings of the latency between exposure and disease onset. Additionally, how reversible are epigenetic changes due to smoking after you quit? They evaluated the epigenomes of 1,454 individuals with chronic obstructive pulmonary disorder (COPD). Interestingly, they found 15 epigenetic sites significantly associated with current smoking, 2 sites associated with total smoke exposure, and, within the subset of former smokers, 3 sites associated with time since quitting cigarettes. This may help our understanding of extended risk of smoking-related diseases, even after quitting.

However, another study found that quitting does result in improvements in the epigenome. They conducted an epigenome wide association study in 464 individuals (22 current smokers and 263 ex-smokers), and validated their findings in an independent sample of 356 twins (41 current smokers and 104 ex-smokers). They found 15 different epigenetic regions where significantly different in smokers vs. nonsmokers. Interestingly, they found that all of those epigenetic regions would revert back to their normal epigenetic state, at least partially, even after just 3 months of quitting smoking.

So, epigenetically, it's absolutely worth giving up the habit.

Alcohol

There are approximately 2,500,000 deaths globally that are attributed to alcohol abuse, most of which occur in people that haven't been diagnosed with an alcohol use disorder. When something isn't measurable, it's very difficult to manage and treat. Measurement of epigenetic changes in folks that abuse alcohol offers an opportunity to diagnose the problem, measure the impact, and assess the potential risk of passing on the problem to our future offspring.

Research showing alcohol's impact on your epigenome

The epigenome actually serves as a very reliable biomarker for alcohol consumption, making it an excellent tool fo the diagnosis and treatment of alcohol-related diseases. One group established this through a study where they performed an epigenome-wide association of DNA methylation in relation to alcohol intake in 13,317 people (54% women). They identified 144 different epigenetic sites that allows for the identification of heavy alcohol drinkers. Interestingly, two of the epigenetic sites they found were on genes that form the GABA receptor, and alcohol is known to bind to GABA and causes the calming effect we get from drinking. Changes in GABA receptor expression may actually be a physical response to chronic alcohol intake, controlled by the epigenome.

Early studies established that children of alcoholics frequently have a decreased sensitivity to alcohol, which is a risk factor for developing an alcohol use disorder. The heritability of alcohol use disorders is around 50%, suggesting that a significant portion of the disease in passed on from one generation to the next. 

Focusing just on human studies, several groups have established that offspring of alcoholic fathers have an increased risk for psychiatric disorders, personality changes, and an increased risk of ADHD. While these findings are arguably confounded by social and environmental factors given that many of the offspring were raised by alcoholic fathers, the results also raise the possibility of acquired changes to the male gametes (i.e., epigenetic changes to sperm) are being inherited by the offspring. Given that genetic studies have not been able to uncover genetic mutations that can account for the heritability of alcohol use disorders, epigenetics is an interesting target.

While sperm have historically been seen as "just needing to show up" and provide their DNA, recent studies have demonstrated the importance of epigenetic modifications encoded in the sperm for offspring development. There are a number of studies that support sperm epigenetics' impact on the offspring phenotypes. And alcohol is well established as an epigenetic disrupter in sperm. In a mouse study, alcohol injections for two weeks decreased sperm counts, testosterone levels, and increased oxidative stress, which is associated with decreased methylation in human sperm. Alcohol likely has a very significant impact on DNA methylation as the sperm are being made, as alcohol disrupts the mechanisms of methylation. Alcohol given to rats 3 times per week for 9 weeks showed a a decrease in expression of a protein important for methylation in sperm DNA. Extending these studies into humans, it's been found that men with moderate alcohol consumption show a decrease in methylation at a set of genes call imprinted genes. These genes are critical for proper early development of the embryo post conception.

There is some good news here as well. Evidence suggests that abstinence can reverse many of the epigenetic modifications that occur with chronic alcohol abuse. So, if you're thinking of having children you may want to consider this as part of your pregnancy plan.

Combined, the data for epigenetic changes due to alcohol abuse are pretty compelling. Furthermore, there are good early data that show these changes can be inherited across generations. Finally, a planned and sustained abstinence can have a positive impact in reversing some of these effects.

Stress & Trauma

Epigenetics moves the concept of nature vs nurture from intellectually intriguing to scientifically measurable. As with the other factors we've discussed in this post, there is meaningful data that support stress and trauma are environmental exposures that can cause alterations in your epigenome. 

What we know is that the environment plays a crucial, albeit highly complex, role in shaping the functional aspects of our DNA, nicely illustrated by the tragic story of the Dutch famine. This historical event shows how changes in the epigenome are inherited from parent to offspring, and to the offspring's offspring, a process called transgenerational epigenetic inheritance. Although there is likely a broad impact of such events on the epigenome, and the cause-effect relationship is almost certainly heterogeneous, genes affected are associated with diseases such as diabetes, schizophrenia, and bipolar disorder.

For those interested in a deeper dive on the relationship between stress, trauma, and the epigenome, I highly recommend the Aeon article by Pam Weintraub.

See the data supporting the trauma-epigenome relationship.

Studies have shown how the environment and trauma can change the epigenomes of identical twins in diverging ways. Identical twins are, by definition, genetically identical. However, their epigenomes start out as identical as well but begin to diverge over time, showing how environmental factors, including stress or trauma, can alter the epigenetics of genes associated with depression, anxiety, and obesity.

One study evaluated the epigenome of African American subjects (age and sex-matched) with a diagnosis of post-traumatic stress disorder (PTSD) and a history of child abuse. They found global DNA methylation was increased in subjects with PTSD. Additionally, they found a number of gene-specific changes in DNA methylation that had been previously associated with inflammation (immune system dysregulation has been associated with trauma history in previous research). These results establish a relationship between psychosocial stressors and alterations in global and gene-specific DNA methylation, which may be associated with immune system dysregulation.

Another study shows that learned fears can be inherited across generations, suggesting that traumatic events in the parents can be inherited and passed through to the offspring. In this study, mice were trained to fear the smell of a specific chemical, acetophenone (smells like cherries). Even without the same training, this fear response was passed on to their offspring. Despite never even smelling acetophenone in their lives, the offspring showed a fear response when they were exposed to the chemical. The next generation of offspring also demonstrated this response, as did mice conceived through IVF (a way of controlling for any learned fear). The researchers also found structure differences in the brain associated with the inherited fear response. The research team suggest that DNA methylation explains the inherited effect. The gene responsible for sensing the chemical had decreased methylation in animals that showed the fear response to the chemical, which suggests a change in expression of oderant receptor.

I couldn't easily find any evidence that epigenetic aberrations associated with trauma could be reverted through some specific intervention. I'll keep a look out and, if I find anything, I'll add it to the post. Given what we know about the dynamic and flexible nature of the epigenome, I would be shocked if there wasn't some what to reverse the epigenetic impact of these traumatic events.

Epigenetics, a mechanism for learning and instinct

We are decidedly not born as a blank slate - fish immediately know how to swim, bees know how to buzz around, and newborns know how to suckle without any training. How are these behavioral instincts developed? Perhaps these instincts evolved from learned behaviors over time and are passed from generation to generation through an epigenetic mechanism.

Looking at "epigenetics reason for being" in this light, it starts to make sense. Maybe all of these changes to our epigenomes are simply a way of short-term evolution or adaptation to occur. Or, the development of a new instinct. Maybe an epigenetic mechanism underneath the heritable tolerance for alcohol is the way of our biology saying, "hey, you're gonna be exposed to a lot of alcohol. Let's set things up so you don't get too intoxicated too quickly".

Just an idea.

Intriguingly (to me at least), this sets up an opportunity for identifying and measuring biological mechanisms for learned behaviors that we need to adjust for. A cognitive therapy, of sorts. If I know my biology is set up process alcohol, metabolize foods, or respond to certain stimuli (think PTSD) differently than normal, perhaps I can make adjustments to my behavior to work around this. Or, better yet, if it's truly measurable, perhaps I can change my behavior and work it out of my system. Quite literally, I'm changing my destiny through science.

This is at least part of the motivation behind Relative Health. Give people access to their personal health data, the ability to interrogate it, and make comparisons to other populations. This way, people are empowered to control their biology rather than react unwittingly to it.

I will make the effort to continue to update this page as the field of epigenetics advances. If there are errors you see in what I've presented or you think there are ideas missing, please get in contact. Our goal is to make this as useful as possible.

© 2018 | Relative Health, Inc.