Environmental Influences on Health, Starting Before Birth

Emerging field of "epigenetics" studies the signals that tell DNA to work
Share Share This
Feb 12, 2014

Cathrine Hoyo’s first job out of university was as a statistician for health surveillance in her native Zimbabwe. But after a few years of “counting numbers of sick people,” she decided to learn more so she could do more.

Today, she's an associate professor at Duke where she focuses on preventing rather than tabulating disease as an epidemiologist. An epidemiologist's first step in prevention is to identify the factors that cause disease and—just as important—to understand when those factors come into play.

Cathrine HoyoCathrine Hoyo is looking for the roots of chronic disease, starting in the womb. (Duke Photo - Jared Lazarus)

Now Hoyo and a growing team of Duke researchers are looking for answers in a group of children they’ve been following since even before they were born.
The work is contributing to a growing body of evidence that suggests many diseases and chronic conditions with an environmental component have their origins in the womb, when exposure to different chemical compounds can switch genes on or off.
This silencing and activation of genes can have subtle or dramatic effects on human development. This relatively new study of how gene switches operate is called epigenetics, the prefix epi meaning above or beyond genetic, and it adds an entirely new level of complexity to the study of the human genome. 
One way biology turns genes on or off is with molecules called methyl groups, which attach to DNA and change how it interacts with other molecules. Methyl groups come from nutrients that a mother eats, but how well they work can be affected by toxins, such as pesticides or compounds in cigarette smoke or even excess fat molecules circulating in her blood.
When Hoyo came to Duke in 2002, she was inspired and intrigued by the epigenetics research of Duke’s Randy Jirtle (now at the University of Bedfordshire and University of Wisconsin). He had recently published a now-famous study showing that supplementing the diet of pregnant mice with folic acid and other nutrients caused methylation changes in the offspring that seemed to make the difference between genetically identical mice being born with brown fur or with yellow fur. Just as startling, the yellow mice quickly became obese while their brown siblings did not. (See the press release) (And the paper)
“The idea you could take a nutrient and give it to a mother and it will change the epigenetic profile of the offspring and the obesity status was incredible to me,” Hoyo remembers.  She saw epigenetics as a potentially powerful strategy for improving public health.
Take obesity for example. “We’ve never been so obese,” she says. “We have come to a point it’s hard to believe it’s just eating poorly and not exercising that’s causing this obesity.” If obesity results in part from exposure to particular chemicals in the womb, reducing exposure to those chemicals among pregnant women might reduce obesity rates more effectively than trying to get people to eat less and exercise more.

It’s a simple idea, but incredibly difficult to prove. First you have to show that prenatal exposure to a particular chemical results in an epigenetic change, then you have to show that the epigenetic change results in obesity—or whatever condition you might be interested in. And while you can use animal studies to guide your work, ultimately you need to demonstrate the relationships in humans who are strikingly diverse and notoriously bad at remembering or knowing what they ate or were exposed to several months or years ago.

agouti miceThe classic photo from Randy Jirtle's work. These genetically identical mice experienced different maternal environments, one emerging obese and pale and the other lean and dark, solely because of mom's diet.

cloned miceThese genetic twins from the University of Sydney have a kinked tail mutation that shows up differently, depending on epigenetic imprinting. (image - Emma Whitelaw, University of Sydney, Australia)

But Hoyo couldn’t ignore the idea because it offered so much promise. “We know that children that are born to obese women tend to be obese too. How does epigenetics really influence this process? If we can demonstrate [the role of epigenetics], we can practice public health differently,” says Hoyo.
Hoyo wanted to recruit a cohort of pregnant women, and then follow their children for several years, collecting data on environmental exposures, epigenetics, and outcomes—particularly obesity and behavioral issues. As an epidemiologist, she knew how to design and maintain a cohort, but not how to do molecular analysis or assess behavior, so she teamed up with molecular biologist Susan Murphy and child psychologist Bernard Fuemmeler. As Fuemmeler quips, “It takes a village of scientists to understand the developmental origin of disease—a village of multidisciplinary scientists.”
Between 2004 and 2011, Hoyo recruited 2,500 pregnant women to be part of a cohort she calls the Newborn Epigenetic Study or NEST. In the early going, she recruited women in the hospital shortly before they gave birth. Later, she figured out how to reach women at their first prenatal appointment, where she collected a blood sample to provide a snapshot of the women’s current and recent chemical exposures. The study also took a cord blood sample from the baby at birth.
Hoyo  and her colleagues continue to collect data from the children every year or so, including height, weight, and detailed information about nutrition and behavior. They also collect swabs of cheek cells, from which they can extract DNA. She's now setting up a new lab and assuming a faculty position at NC State's Centennial campus in the new department of biological sciences, but will keep a partial appointment at Duke.
So far, data coming out of the cohort has shown that folic acid, tobacco smoke, antibiotics, maternal depression, and maternal and paternal obesity can all be associated with epigenetic differences among babies.
But there’s much more to come. “We’ve just barely begun looking,” Susan Murphy says. “There are still a lot of questions that can be addressed.” The blood samples are all in the freezer, available for further analysis. Researchers can continue to ask new questions by using all the information already collected and gathering other kinds of information from the families on future visits.
The project continues to receive funding and more Duke scientists have become involved, including Heather Stapleton and Avner Vengosh from the Nicholas School of the Environment; Theodore Slotkin and Rochelle Schwartz-Bloom from pharmacology and cancer biology; Edward Levin and Scott Kollins from psychiatry and behavioral sciences; Lisa Satterwhite from Biomedical Engineering; Barbara Engelhardt of biostatistics and bioinformatics, and others.
Studies underway or soon-to-be underway are looking at the effects of heavy metals, flame retardants, and pesticides. A $7.8 million, 5-year study headed by Murphy will use rodent models and the NEST cohort of humans to investigate epigenetic links between tobacco smoke exposure in pregnancy and childhood and ADHD. They're calling it NICHES, the Center for Study of Neurodevelopment and Improving Children’s Health following Environmental tobacco Smoke exposure.
For his part, Fuemmeler is fascinated with discovering the causes behind differences in self-regulation and impulsivity in children. “We’re wanting to understand if behavioral differences we see in children at ages three to five are related to variations in epigenetic marks on these genes that we know are susceptible to environmental exposure,” he says.
And everyone wants to know whether these epigenetic changes might be reversed.
In an early NEST project, the researchers showed that the epigenetic profiles of children born to mothers who had quit smoking when they found out they were pregnant were the same as the children born to nonsmokers, and they differed from the children of women who continued to smoke throughout pregnancy.
“The suggestion that there may be some way to correct that methylation profile by quitting smoking was really exciting," Murphy says. "I was very surprised by that. I thought once you smoked, even if you quit the damage wouldn’t be undone.”
Fuemmeler wonders whether pro-social interventions might actually change the epigenetic profile a child is born with. “What’s fascinating is the idea that epigenetics is malleable and that a positive environment might actually restore perturbations in the epigenome,” he says.
Since 2004, NEST has grown from a small cohort studied by a small team of scientists to a larger cohort studied by a growing team of scientists.  “There’s still a lot to do,” says Hoyo, who never hesitates to invite other scientists to join in the fun. “I’ve always felt like if you need expertise you go and get it, and I love to collaborate. It’s enriched the study a lot.”
Fuemmeler says, “It’s like going to the moon; you can’t get there with one set of scientists. You need to pull in people with different expertise. This is our moon shot. As we continue to work, it will mean a lot for our different disciplines and for public health.”
For Hoyo, a successful 'moon landing' would be identifying epigenetic signatures that link exposures to specific outcomes—whether obesity, cancer, ADHD, or others. If she and her colleagues could do that, they could tell pregnant women which healthy exposures to seek and which unhealthy exposures to avoid. For children and adults with known epigenetic signatures, their chances to thrive could be improved with early intervention or increased screenings. “That’s the ultimate prize for me,” Hoyo says. “It’s a big question to ask for one’s career, but I think there’s something to it.”
NEST and NICHES have been funded by grants from the National Institutes of Health, the US Environmental Protection Agency and the Fred and Alice Stanback Foundation.