A recent toast to James Watson highlights a tolerance for bigotry many want excised from the scientific community.
Low doses of environmental chemicals can make animals gain weight. Whether they do the same to humans is a thorny issue.
November 1, 2015|
© GRACE RUSSELL
In 2005, Mike Skinner’s group at Washington State University published a disturbing observation: pregnant rats exposed to high levels of a commonly used fungicide had sons with low sperm counts as adults. When the males did succeed in impregnating a female, they bore sons who also had fewer sperm, and the gametes were less viable. The problem perpetuated through multiple generations, as Skinner’s lab observed the rats over several years.1
“We sat on [the results] for four years because it was a major observation, so we wanted to get as much on the mechanism as possible,” Skinner says. He and his colleagues found that altered DNA methylation patterns in the germ line were to blame.
To see if other environmental chemicals could have the same effect, they screened a host of potentially toxic chemicals: jet fuel, plastics ingredients, and more pesticides. Again, exposed animals had offspring with reproductive problems, which were passed down for generations. The researchers also saw another phenotype pop up again and again: obesity. Skinner first saw fat rats in his experiments after he’d injected females with a mixture of bisphenol A (BPA) and phthalates, substances used to make plastic products and, like the fungicide the researchers originally tested, known to be endocrine disruptors. The rats’ pups and their pups’ pups—animals that had direct exposure to the chemicals—showed other abnormalities, but were of normal weight. However, roughly 10 percent of third-generation (F3) rats descended from exposed females became obese.2
The results were interesting, but not particularly striking to Skinner—until his team tested DDT, a pesticide used widely in the U.S. before it was banned in the 1970s because of its impact on bird populations and concerns that it could harm human health. Again, rats whose mothers or grandmothers had been exposed to the chemical had normal body size. “But by F3, 50 percent of the population, both male and female, had obesity,” Skinner recalls. “We said, ‘Wow, this is sort of a major deal.’”3
Skinner’s thoughts turned to the dramatic rise in obesity rates among US adults over the past few decades; currently, more than a third of American adults are obese. “My guess is there’s probably not a woman who was pregnant in the 1950s who wasn’t exposed to DDT,” he says. “When we starting seeing the obese animals, it clicked. . . . Maybe these 1950s exposures had something to do with today’s human situation.”
Increasingly, we’re finding that environmental exposures to chemicals may be an under-recognized third factor in the epidemic.—Leonardo Trasande,
New York University School of Medicine
His idea is entirely speculative, and Skinner is quick to point out there is no direct evidence that ancestral pesticide exposures cause weight gain in future generations of humans. But the idea that chemicals in the environment conspire to make us susceptible to obesity is gaining traction. In the past decade, researchers have identified dozens of chemicals that can cause obesity in animals or metabolic disruption at the cellular level. And observational studies in humans have suggested a link between environmental chemical exposures and greater body mass index (BMI).
“This is not to minimize diet and physical activity; they are still the leading causes of the obesity epidemic,” says Leonardo Trasande of New York University (NYU) School of Medicine. “It’s just that, increasingly, we’re finding that environmental exposures to chemicals may be an under-recognized third factor in the epidemic.”
In the early 2000s, Bruce Blumberg of the University of California, Irvine, was at a meeting in Japan when he heard a talk about tributyltin (TBT), a chemical used in marine paints to prevent organisms from growing on the hulls of ships. Blumberg studies endocrine disruptors, and his group was looking at whether certain chemicals, including TBT, could activate a nuclear hormone receptor called the steroid and xenobiotic receptor; among other things, it is important for drug metabolism. The presentation described how TBT could cause sex reversal in fish, and Blumberg wondered what exactly TBT was up to.
Blumberg asked his team back home in California to test TBT on its entire collection of nuclear hormone receptors in vitro. The group found that the compound activated a fatty acid receptor called PPARγ.4 “There’s only one way you can go with that data,” says Blumberg. “This receptor is the master regulator of fat-cell development.” The researchers went on to show that TBT can spur adipocyte precursors to differentiate into fat cells in vitro,4 that live frogs exposed to it develop fat deposits around their gonads, and that mice exposed to TBT in utero have greater fat stores as adults. Generations of the exposed animals’ progeny are also prone to increased adiposity.
In a 2006 review, Blumberg and UC Irvine colleague Felix Grün coined a new term for such environmental chemicals linked with fat gain: obesogens.5 Although Blumberg’s work was not the first to implicate such substances in obesity, the term obesogen defined an emerging line of inquiry that questioned the strict calories-in-calories-out dogma of weight regulation. “It just takes somebody to say something that catches people’s attention,” says Jerry Heindel, a scientific program administrator at the National Institute of Environmental Health Sciences (NIEHS) who had been pushing for the agency to prioritize obesogen research. “When [Blumberg and Grün] said there are these chemicals we’re calling obesogens, that caught people’s attention.”
“Until these new environmental factors to obesity came around, most people believed obesity came from an energy imbalance: eating too much or expending too little,” agrees De-Kun Li, an epidemiologist at Kaiser Permanente in Oakland, California. Although no one denies that food and exercise are crucial to body weight, they aren’t everything, he says. A small but growing number of scientists are now convinced that chemical exposures—in particular, those that mess with hormonal pathways—make the human body susceptible to obesity in the face of modern, waistline-challenging lifestyles.
LUCY CONKLINNot long after Blumberg and his colleagues published their initial work on TBT’s activation of PPARγ, endocrine-disorder specialist Rob Sargis began a fellowship at the University of Chicago to study metabolic disease. Sargis wondered if obesogens could also act through other hormonal pathways. In particular, physicians have long known that too much cortisol—a glucocorticoid hormone responsive to stress—can cause a condition called Cushing’s syndrome, which can involve diabetes, weight gain, and even obesity. Could environmental endocrine disruptors cause obesity and metabolic problems through glucocorticoid signaling, too?
Sargis and his collaborators decided to screen for compounds that could disrupt glucocorticoid signaling in adipose cell cultures. Four stood out: BPA; dicyclohexyl phthalate (a plasticizer); and two pesticides, endrin and tolylfluanid. Each compound activated the glucocorticoid receptor and promoted fat-cell differentiation and lipid accumulation.6 “We knew we were onto something,” says Sargis. “The question was, what was the molecular mechanism?”
They started with insulin signaling, given Sargis’s interest in metabolism and glucocorticoids’ known ability to interfere with this glucose-regulating pathway. Further experiments revealed that tolylfluanid jammed normal insulin signaling by downregulating a member of the insulin signaling cascade. This caused the cells to become resistant to the hormone.7 “We discovered this specific defect, which told us it wasn’t overt toxicity but a specific disruption in cell signaling,” says Sargis. And earlier this year, Sargis’s team showed that mice fed tolylfluanid became insulin resistant and gained weight and fat mass.8
© LUCY CONKLINMeanwhile, Blumberg and others have continued to hash out how obesogen disruption of endocrine signaling contributes to fat storage, the production of fat cells, and overall metabolic dysfunction. For instance, Blumberg and his colleagues have shown that, in mice exposed to endocrine-disrupting TBT in utero, bone marrow– and adipose tissue–derived mesenchymal stem cells become fat cells (as opposed to bone, cartilage, or muscle) in far greater numbers than do the corresponding cells in untreated mice. Most recently, Blumberg found that when exposed animals are fed a high-fat diet, they get fatter faster (the results are not yet published). “They handle calories differently, and that’s what we always thought.”
Researchers have shown in mice and in vitro that exposure to a flame retardant (2,2′,4,4′-tetrabrominated diphenyl ether, or BDE-47) or a fungicide (triflumizole) also bulks up adipose tissue, and the chemicals do so via PPARγ activation, just like TBT. And other teams have produced evidence that phthalates used in plastics spur fat-cell production in cell culture and in animals by activating PPARγ as well. Endocrine disruptors that act as estrogen mimics can also predispose animals to obesity. Take BPA, for instance, which binds to estrogen receptors. Just as with exposure to other obesogens, BPA given to mice during pregnancy can lead to fatter offspring. A metabolite of BPA, called BPA-G, causes lipid accumulation and expression of fat-cell differentiation markers in cultures of both mouse and human adipocyte precursors.
In addition to disrupting cell signaling, some obesogens appear to leave specific, long-lasting epigenetic marks on cells’ DNA. Skinner has found, for example, that the methylation profile of a rat exposed to DDT is different from that of one exposed to plastics compounds. Although Skinner is still working out the functional consequences of such epimutations, they may serve as signposts to identify pathways disrupted by the chemicals. And more research could lead to reliable biomarkers of exposure based on methylation signatures, Skinner adds. “The application of what we’re finding in rodent models would be a significant advance for health care.”
As data accumulate on the effects of obesogens in vitro and in animal models, the question of how these compounds affect humans remains largely unanswered, meaning regulators have little clinical data to go on when determining acceptable exposure levels for the chemicals. “There is always going to be some uncertainty,” says Leonardo Trasande of New York University School of Medicine. “And we deal with uncertainty in all aspects of human life and policy. The question is: What is the threshold to act?”
For pesticides, there is a fairly standard process by which the US Environmental Protection Agency (EPA) measures animal toxicity, estimates expected human exposures to determine whether the chemicals should be restricted or banned, and then acts on those assessments. But for tens of thousands of other compounds, the path is less clear-cut and, often, leads nowhere.
Next year, the Toxic Substances Control Act (TSCA)—the main law regulating chemicals found in products other than pesticides, food, and cosmetics—will reach the ripe old age of 40. “The law is old and out of date,” says Richard Denison, the lead senior scientist at the Environmental Defense Fund. “It hasn’t kept up with science.”
TSCA places most chemical regulation in the hands of the EPA, though food additives and packaging, including plastics in bottles, fall to the US Food and Drug Administration. When the law was passed in 1976, some 60,000 chemicals were on the market; nowadays, there are more than 85,000. But the EPA hasn’t tried to ban a chemical under the TSCA since the 1980s. The agency spent the better part of a decade working to ban asbestos, and seemingly succeeded in 1989. But in 1991, a court of appeals overturned the ruling, declaring that EPA hadn’t sufficiently shown that the benefits of banning asbestos outweighed the costs.
Because the EPA could not get a ban approved for asbestos—a known carcinogen—it essentially threw up its hands. “As a result, in the more than three and a half decades since the passage of TSCA, the EPA has only been able to require testing on just a little more than 200 of the original 60,000 chemicals listed on the TSCA inventory, and has regulated or banned only five of these chemicals under TSCA’s Section 6 authority,” James Jones of the EPA’s Office of Chemical Safety and Pollution Prevention testified before Congress last April. And the lack of oversight is cause for concern, Denison says. “We’re looking at decades of an agency not even trying to restrict chemicals we know pose significant risks.” The United States is currently one of only a few industrialized nations where asbestos is not entirely banned, despite being classified as a known human carcinogen by the EPA, the US Department of Health and Human Services, and the International Agency for Research on Cancer.
But Denison and others are hoping TSCA will soon be getting an extreme makeover. Two bills currently in Congress—one that has passed the House and another that is awaiting a full vote in the Senate as this article goes to press—aim to enable the EPA to prioritize chemicals for testing, order appropriate risk assessments, decide whether the results warrant restrictions, and implement regulations. But the bills do not dictate “how EPA is supposed to do its risk assessment,” says Mark Duvall, a principal at environmental law firm Beveridge & Diamond and outside counsel for the American Chemistry Council. “The legislative language at this point leaves EPA a lot of discretion to determine, with the best available science, the weight of scientific evidence.”
The EPA has a distinct program for testing endocrine disruption, first outlined nearly two decades ago but only implemented a few years ago. Of an initial screen of 52 chemicals published on the EPA’s website this summer, 32 were shown to have endocrine activity in vitro or in animals. The agency had previously determined that 14 of these are safe, but the rest will be subject to further scrutiny in animal model assays, and many more await screening. The EPA is also adopting a high-throughput program for evaluating chemical safety, called ToxCast, that involves an initial in vitro screening followed by more-extensive experiments if the results point to potential harm. Should the TSCA reform make it into law, the EPA might finally have the muscle it needs to act upon these results.
Although most obesogen research points to the chemicals’ roles in obesity or metabolic disruption, the effects are not always consistent, in part because assays, exposure doses, and model systems vary. BPA in particular has incited considerable debate, not least over whether studies on adipocyte precursors in culture or in rodents can reliably predict what happens in people. Although new chemicals are tested for safety, there are no randomized, controlled clinical trials to examine the effects of environmental substances on obesity or any other condition. (See “Exposure Control” at right.)
There are, however, observational studies that support the idea that obesogens may have effects in humans similar to those they have in mice. Several years ago, NYU’s Trasande and his colleagues collected data from a large national survey of children on levels of BPA in urine and BMI. Among the white children in the study, they determined that higher BPA levels were linked with a greater likelihood of being obese.9 “In the least exposed group, one in 10 were obese, whereas in every other sample, one in five were obese. So the effect was sizable,” says Trasande. (Black and Hispanic children did not show such a relationship.)
Li has also found a relationship between BPA and obesity. Among children in China, Li found that preteen girls (not boys) with higher BPA levels in their urine were more likely to be in the heaviest category.10 “Generally the results are consistent,” he says. “There is a correlation.” But BPA is metabolized quickly, so a one-time pee sample doesn’t reveal a person’s lifetime exposure, Li and Trasande note. It’s also impossible from these studies to determine which came first, the exposure or the condition.
Human data on other obesogens are even more scant. A systematic review of studies on phthalates’ potential link to obesity came up short when the researchers couldn’t find enough methodological consistency across studies.11 Indeed, scientists don’t have good assays to measure some of these substances in relevant tissues. For example, in the 1990s, R. Thomas Zoeller, who studies thyroid hormone at the University of Massachusetts Amherst, found that polychlorinated biphenyl (PCB) exposure in fetal rats affected thyroid action in the brain. Given the identical structures of thyroid receptors in humans and rats, “it’s extremely naive to suggest those results are not useful to humans,” he says. But he had no way to prove it; going into the brains of people and measuring thyroid hormone activity is impossible. And measuring hormone activity in the blood—as is typically done—doesn’t capture the neuro-disruption.
The lack of well-understood and relevant modes of action means some chemicals may fall through the cracks of safety testing, says Zoeller, but new assays that rely on newly found biomarkers may solve the problem. “Scientifically, I think we can approach these things and find answers for them,” he says.
But even as better biomarkers become available, getting good, long-term data on the relationships between chemicals and obesity in humans is expensive and time-consuming. As a result, the obesogen field remains on the periphery of clinical practice and environmental policy. “I doubt the clinical obesity medicine community has much, if any, knowledge on this,” Scott Kahan, a weight-management physician at the National Center for Weight and Wellness, says in an email. Heindel of NIEHS agrees: in animal models, “we can show chemicals increase weight, we can show they increase fat, [and] we can show some mechanisms of that,” he says, but without stronger evidence in humans, “people aren’t really accepting this as an important part of the obesity epidemic.”
Even if physicians did have obesogens on their radars, there’s very little they could do to limit or treat exposure in their patients. Certain chemicals may be difficult to avoid, either because of their ubiquity or their persistence in the environment or because the damage was done generations ago. Sargis says that’s all the more reason to study them.
“We can despair and say, ‘Things are not going to go away, we can’t remediate these things,’” says Sargis. Or, scientists can face the challenge of environmental toxins as an opportunity to understand the health damage they can wreak and put a stop to it—and maybe even learn some new biology along the way. “You have to go in with that attitude; otherwise it gets really depressing.”
A sampling of potential obesogens†
|NAME||USE||EVIDENCE OF HARM||MECHANISM|
|Tributyltin (TBT)||Fungicide and disinfectant; added to marine paints to discourage growth of barnacles and other organisms; also found as a nonintentionally added substance in some plastics||Lipid accumulation in preadipocytes in culture; mice exposed in utero develop larger fat deposits, and effects perpetuate for multiple generations||Activates PPARγ/RXR transcription factors, among other effects|
|Organobromines||Flame retardants and other uses||Male rats gain weight and fat mass; exposed human infants have low birth weight||Not yet detailed; human cord blood and rodents show low thyroid hormone levels|
|Organochlorines (e.g., DDT, PCBs, tolyfluanid)||Pesticides; electronics manufacture||Weight gain, increased fat mass, and metabolic dysfunction in rodents; associated with higher BMI in humans||Glucocorticoid receptor and PPARγ activation; antiandrogenic activity|
|Organophosphates||Insecticides||Weight gain and metabolic dysfunction in exposed rats||Unknown|
|Bisphenol A (BPA)||Plastics production||Lipid accumulation in preadipocytes in culture; rodents exposed in utero or postnatally have greater fat mass and weight as adults; linked to obesity and type 2 diabetes in humans||Activates estrogen and glucocorticoid receptors and PPARs, among other actions|
|Phthalates (e.g., diethylhexylphthalate)||Plastics production||Lipid accumulation in preadipocytes in culture; progeny of exposed mice have increased fat mass and higher body weight; linked to type 2 diabetes and increased fat mass in women||Activate PPARs and glucocorticoid receptors, among other actions|
|Heavy metals (e.g., cadmium, arsenic, lead)||Mining, fertilizer, plastics production, wood preservatives||Associated with an increased risk of developing type 2 diabetes in humans; female mice exposed to arsenic in utero become obese||Mimic estrogen; disrupt glucose metabolism|
|Perfluorooctanoic acid (PFOA)||Nonstick coatings and other uses||Increased body weight among exposed female mice; linked to higher BMI in humans||Unknown|
†Adapted from A.S. Janesick et al., “Environmental Chemicals and Obesity,” in Handbook of Obesity, Vol. 1, 3rd ed., G.A. Bray, C. Bouchard, eds. (Boca Raton, FL: CRC Press, 2014), 471-88.
Correction (November 3): In the first paragraph, we mistakenly referred to some of the rodents in a study as mice; they were all rats. The Scientist regrets the error.
November 3, 2015
Just to make life complicated most of these chemicals enter through the digestive system which has its own biome which may be influenced to make other other disrupting chemicals thus complicating the pathway or may modify the chemicals themselves
November 6, 2015
If the obesogen hypothesis were true, why don't we see widespread obesity in wildlife?
December 14, 2015
Nicolas asks "why don't we see widespread obesity in wildlife?" The answer: We do. Read "Canaries in the coal mine: a cross-species analysis of the plurality of obesity epidemics" by Klimentidis et al. 2010, published in the Proceedings of the Royal Society B http://bit.ly/1P2exWB They examined over 20,000 animals from 24 populations representing eight species living with and around people in industrialized socieites. In ALL populations, the trend in weight increased. "The probability of all trends being in the same direction by chance is 1.2 x 10^-7."