Efforts Now Underway To Operationalize Exposomics

December 6, 2022

By Deborah Borfitz 

December 6, 2022 | Truly implementing precision medicine requires that environmental exposures be factored into the equation, according to Rick Woychik, Ph.D., director of the National Institute of Environmental Health Sciences (NIEHS) and the National Toxicology Program (NTP) of the U.S. Department of Health and Human Services. Woychik was a keynote speaker on the opening day of the Mayo Clinic’s recent Individualizing Medicine Conference focused on exploring the exposome, the sum of external factors a person is exposed to across a lifetime. 

The list of potential environmental exposures influencing human health is vast and the usual suspects—lead, asbestos, and air pollution—are not the only ones, he says. He foresees a future where geneticists and environmental health sciences work together to better predict clinical outcomes patient by patient. “It just can’t be sequence variations.” 

People don’t only look different, notes Woychik. They differ biologically and therefore have different disease susceptibilities and respond to therapeutics in their own way. “In the environmental health sciences community, we also discovered that we respond to environmental exposures differently, so what may be dangerous and unsafe for one individual may not apply to another individual.” 

Published studies by the NIEHS and its grantees have concentrated on the health effects of a variety of industrial chemicals, including “forever chemicals” such as PFAS (per-and polyfluoroalkyl substances) that comprise 12,000 molecular species, as well as agricultural chemicals (i.e., pesticides and herbicides), and “toxins that are in the air we breathe and in the water that we drink and... the soil that we use to grow our crops,” says Woychik. Additionally, the agency has been looking at the impacts of lifestyle factors ranging from nutrition and exercise to vaping and psychosocial influences such as structural racism. 

As the former president and CEO of The Jackson Laboratory in Bar Harbor, Maine, he pulled together a cross-section of the environmental health sciences community to learn how different inbred strains of mice respond to the toxic effects of acetaminophen, a common analgesic.  “With inbred strains you are essentially dealing with genetically identical twins... it is a very clean way to assess the impact of genetic background,” he says. 

The results clearly demonstrated that genetics play an important role in dictating an individual’s response to the environment, says Woychik. Some strains of mice had “exquisite sensitivity” to the toxicity effects of acetaminophen while other metabolically different strains were essentially immune to any of the toxicity effects at the same dose. 

The NTP replicated the experiment a few years later to look at clearance or toxicity of benzene, a widely used industrial chemical. “Again, it was vastly different in different inbred strains,” he reports. 

To complicate matters further, epigenetics also must be considered when looking at responses to environmental exposures, Woychik says. Biological outcomes can be influenced by changing the expression of genes by methylating the DNA or through post-translational modifications of various histones that bind to DNA. 

Epigenetics and the Microbiome 

More than 30 years ago, Woychik first studied the agouti gene responsible for the normal coat color of mice. Scientists in his lab cloned the gene to study different mutant alleles, including a fully penetrant version of one that gave the mice a totally yellow coat color but also caused a signaling problem in the brain making them massively obese with type 2 diabetes. Their genetically identical litter mates—all carrying the agouti gene predisposing them in its fully penetrant form to the all-yellow coat color and unhealthy phenotype—instead had varying degrees of mottling and were comparatively slender and healthy.  

The reason some of the mice with the agouti gene were phenotypically normal, despite having the mutation, was epigenetic suppression of the agouti gene, he explains. The phenotypic consequences of this epigenetic regulation of the mutant gene were readily apparent.   

Subsequently, the research team demonstrated that hundreds of other genes are also epigenetically modifiable. The phenomenon is again being studied at the NIEHS because the number of mice that are fully yellow and obese with type 2 diabetes relative to those that are phenotypically normal is dictated by what you feed mothers during pregnancy, Woychik says. 

The epigenetic signatures established during embryonic development that may have an impact on the clinical outcomes of the animals later in life is of growing scientific interest and exploration, he adds. And the epigenetic variables at work are influenced by environmental exposures, notably feeding mothers high concentrations of chemical entities that are “methyl donors.”  

As scientists now well know, the complex mixture of bacteria and other organisms populating the gut likewise play a critical role in dictating the health status of individuals, says Woychik. Next-generation sequencing tools can also be used to establish the presence of those organisms by looking at the presence of their DNA. 

“What we determined in the environmental health sciences community is you can influence the composition of the microbiome through chemical exposure... and the microbiome may be predisposing [people] to obesity or other important phenotypes, so if you want to do precision medicine or individualized medicine you have to factor that variable into the equation as well,” he says. The microbiome may be partly responsible for some of the differential effects to environmental toxins. 

The emerging concept of “precision environmental health,” which is about addressing individual variability associated with environmental exposures, is complementary to personalized medicine in that it factors in genetic variability and other variables such as epigenetic modifications in individuals exposed to different biochemicals, Woychik says. The objective is disease prevention based on individual-specific risks. 

The data management challenges are huge, says Woychik, who advocates for the development of “seamlessly integratable” data repositories to mitigate the need to hire armies of bioinformaticians. 

How To ‘Collect The Environment’ 

Ultimately, environmental health sciences researchers want to do studies in individual humans to identify the sequence variations within genes that are responsible for complex traits predisposing them to disease, says Woychik. But to have the statistical power to tease out the critical variations will take millions if not tens of millions of individuals, which will require collaboration between large-scale federal programs in the U.S. (i.e., All of Us Research Program of the National Institutes of Health) and other countries around the world. Standardized tools will also be needed for how data gets collected and genomes get sequenced, he adds, so the data can be integrated to the benefit of everyone. 

Given the 80,000 different chemicals in the environment, the question on the mind of many health scientists is how to structure an experiment that incorporates the environment into the way they collect data, he continues. Operational practices are needed on how to “collect the environment” and where and how to deposit that data. 

It will be important for members of the International Common Disease Alliance (ICDA) to collect data on such environmental exposures as lead, arsenic, PFAS (found in some Teflon, Scotchgard, and flame-retardant products), pesticides (potentially ingested with otherwise healthy fruits and vegetables), and air pollution (i.e., inhalable particles 2.5 micrometers and smaller)—all of which can have deleterious health effects, says Woychik. Collecting only phenotype data will leave a “big gap” in the ICDA’s ability to meet its goal of unlocking the biological basis of diseases. 

Work done by the NTP found that a condition known as obliterative bronchiolitis (aka “popcorn lung”) was being caused by breathing in diacetyl, which has a direct connection with the butter-flavored compound in microwave popcorn, he shares. But the only people being affected were people working in microwave popcorn factories who were exposed to high concentrations of the chemical entity. “If you are sequencing genomes from these individuals, you can sequence until you are blue in the face and you are not going to find the sequence variation that directly connects to popcorn lung” absent data on their high-level exposure to diacetyl.  

Similarly with small-particle air pollution (PM 2.5), “if it’s not part of how you assess exposures in your populations you are going to miss a whole bunch of different things,” continues Woychik, including respiratory, cardiovascular, and neurological effects. Many neurologists have been surprised to learn that exposure to air pollution has associations with the development of a whole variety of neurological diseases—enough to warrant asking their patients where they live. 

In terms of immune function, Woychik says he wonders if differential responses to SARS-CoV-2 relates partly to the fact that some people live closer to highways than greenways and are breathing in higher concentrations of PM 2.5. It may be that air pollution suppresses the immune system in such a way to cause a more aggressive clinical course for someone infected with COVID-19. 

Environmental exposures have also been linked to metabolic and neonatal effects, says Woychik. The developmental emergence of health and disease is now a big topic area for the environmental health sciences community, including exposures of both the mother and father prior to conception that may be influencing the epigenetic status of genomes that in turn influence how genes are expressed—in some cases, much later in the life of their offspring. 

Exposures to agricultural chemicals known as organophosphates have been linked to Parkinson’s disease, he adds, notably individuals with mutations in the PON1 gene important in metabolizing and neutralizing these chemicals. “This is a relatively simple case where you just find one gene with a sequence variation that may be contributing to gene-environment effects.” 

The NIEHS is already in discussions with the NIH about collecting environmental exposure data on participants in the All of Us Research Program, potentially using location mapping, wearable devices, and exposure biomarkers. The magnitude of the undertaking could be a holdup, as environmental health sciences research typically measures exposures on no more than a few hundred individuals, says Woychik.  

Following a workshop with the NIH over the summer, 17 proposals for ancillary projects were submitted to the NIEHS for funding that are now under review that would bring environmental exposures into the All of Us Research Program, he notes. Woychik says he wants to see practicing pediatricians and internal medicine physicians convinced that environmental health exposures are potentially harmful to human health, so they start asking their patients during their annual physical about where they live and work and the harms they have likely been exposed to. 

Building Momentum 

As was first proposed in 2005, “if we truly want to understand this impact of the environment, we have to get beyond one exposure at a time,” says Woychik. The “totality of exposures” could well include PFAS as well as metabolites from dichlorodiphenyltrichloroethane (DDT), an insecticide once routinely used in the U.S. that also persists for a long time in the environment.   

PFAS compounds are particularly vexing because of the strength of their carbon-flourine bonds that nothing on planet earth was designed to break. The chemicals were originally produced in West Virginia in the 1940s and 1950s for the first generation of Teflon and “they are still there... contaminating the water supplies,” Woychik says. 

The onus is on the environmental health sciences community to come up with an operational definition of how to do an exposome experiment to galvanize research on a global scale, he adds. People need to know what data to collect, and how much that is going to cost, and, in the future, high-throughput, high-resolution mass spectrometry will likely be needed to analyze all the samples.  

The newly created Advanced Research Projects Agency for Health (ARPA-H), intended to improve the government’s ability to speed biomedical and health research, is going to be “a very important factor in building some of those new tools,” predicts Woychik. These would include new technologies for whole-genome methylation analysis. 

In the interim, the exposome can be studied in pragmatic, doable ways, Woychik says. He points to the NIH-funded Human Health Exposure Analysis Resource (HHEAR), located at Mount Sinai in New York, which supports both targeted and untargeted analysis of various environmental chemicals present in human samples.  

Scientists across the biomedical spectrum are encouraged to join the effort to operationalize exposomics, which began with a series of five workshops this past summer. Over 40 colleagues participated in discussions around 64 distinct topics, says Woychik.