In addition to measuring various types of exposures, an emerging challenge is the need to integrate all the information, to be able to visualize cumulative effects over the lifetime. “A good analogy is with meteorologists, who rely on very precise tools to measure different parameters and use sophisticated techniques to integrate them,” says Dr. Einstein. While complex approaches to integrate multiple layers of scientific data could open the possibility to model future behavior and predict response to therapy, the integration and modeling represent, at this time, some of the main challenges in the emerging field of exposomics.
“If we look at the approach that is being used for predicting storms, several models are run iteratively, allowing comparisons across them, and scientists can subsequently go back to identify specific areas where the models can be improved, and this strategy could be useful in studying the exposome,” says Dr. Einstein.
“We study how cells lining the respiratory tract respond to environmental toxins that they are exposed to,” says Avrum E. Spira, M.D., professor of medicine, pathology, and bioinformatics at Boston University. Several options are available to examine the response to inhaled toxins, and these may rely on assessing the amount of cigarette smoke that an individual is exposed to, measuring the breakdown products of specific chemicals, or monitoring the response of individual genes in the airway epithelium to exposure.
Researchers in Dr. Spira’s lab showed that genome-wide changes in microRNA and mRNA levels that occur in respiratory epithelial cells in response to smoking can be integrated to assess the biological response to exposure. These modifications were used to establish biomarkers with potential diagnostic relevance, which can predict which subset of smokers have or are at higher risk for lung cancer. “We can measure expression of all 20,000 genes in these cells at one time, and look at how patterns differ and how these differences relate to lung disease,” says Dr. Spira.
“The exposome can provide the opportunity to look at all the exposures that a person had during their lifetime, and this includes not just what they perceive or is associated with environmental exposures but, in addition, endogenous exposures that originate within the organism and are often missed in traditional environmental health,” says Stephen M. Rappaport, Ph.D., professor of environmental sciences at the University of California, Berkeley.
While typically less than 10% of an individual’s risk is associated with the genetic makeup for most cancers, the contribution of nongenetic factors is assuming increasing medical and public health relevance, but also represents the most challenging component to explore. Historically, environmental exposures have been explored by asking individuals about their contact with specific chemicals.
“This works for some exposures, such as cigarette smoke,” explains Dr. Rappaport. However, for many other exposures, individuals might not remember or even be aware of having been exposed, making this approach unreliable and prompting the need for more sophisticated tools. “By looking at the -omics signals in the blood, we can obtain a much more reliable picture of what people have been exposed to than has been possible by traditional methods to assess exposure,” says Dr. Rappaport.
In analyzing small molecules from the blood, one of the challenges is the inability to predict the ones that are meaningful indicators of exposure. “I have proposed the idea of an exposome-wide study to identify from the many chemicals that a person is exposed to throughout life the molecules that one should focus on,” says Dr. Rappaport. This approach involves obtaining biospecimens from individuals with a specific disease and matched control subjects, comparatively investigating the profile of -omics signals, validating the results on additional cohorts, and studying the emergence of the respective modifications over time, prior to disease, as a result of exposure.
“Through a very ordered set of experiments, this would allow us to zero in on probably the very few relatively simple chemicals that can be predictive, and understand what we can do to reduce or mitigate exposures,” explains Dr. Rappaport. Conceptually similar to the strategy employed in genome-wide association studies, exposome-wide studies present a much stronger potential to identify factors that are important in shaping phenotypes.
In making the comparison with genome-wide association studies, it is also important to reminisce about the fact that gene variants associated with disease risk were often not the ones that investigators predicted ahead of time, based on initial candidate gene studies, to be the most important ones. “The best geneticists were not able to predict which genes are playing the most significant roles, and this is a valuable lesson because we do not have any reasons to think that it will be different in the case of the exposome,” says Dr. Rappaport.
While technological advances helped unveil the intricacies of the human genome and shed light on interindividual genetic variation, significant efforts are currently focusing on advancing the ability to survey epigenetic perturbations in a high-throughput fashion. Among epigenetic marks, visualizing 5-hydroxymethylcytosine, an epigenetic modification increasingly described in several tissues and still insufficiently understood functionally, has been challenging from a technological perspective.
“What is interesting is that, to date, bisulfite conversion could not distinguish between 5-methylcytosine and 5-hydroxymethylcytosine, but with some of our newest techniques we can selectively analyze the two modifications,” says Marc E. Van Eden, Ph.D., vp for business development and marketing at Zymo Research.
Investigators at Zymo Research found that, while 5-methylcytosine makes opening of the double-stranded DNA helix more difficult than cytosine, the effect is reversed in the case of 5-hydroxymethylated cytosine. “These modifications act almost like a molecular switch, a DNA code above the one that is made by the four nucleotides,” says Dr. Van Eden.
At this time, Zymo Research is the only provider for genome-wide 5-hydroxymethylcytosine analysis services at the single-nucleotide resolution. “We have three platforms available for genome-wide methylation analyses, which can cover 10%, 30%, or the entire methylome,” says Dr. Van Eden.
The availability of this approach is critical for investigators looking at putative epigenetic biomarkers, and this is an area of science that promises to become increasingly important, as it will become easier to integrate advances in genetics and genomics with those in epigenetics, to dissect the causes of complex diseases. “A few years from now, these approaches will become routine, and these tools will undoubtedly also be embraced by diagnostics and pharmaceutics markets,” says Dr. Van Eden.
As an emerging concept, exposomics is sculpturing a novel framework to integrate genetic factors with environmental contributions, incorporating all the external and internal exposures that an individual encounters throughout life, starting with the intrauterine period.
Advances defining this vibrant multidisciplinary field have become reality thanks to strides in genetics and genomics, the advent of additional -omics sciences, including metabolomics and metabonomics, and the epigenetics revolution that we are currently witnessing. Forecasting an exciting era for the biomedical world, the exposome promises to fill a long-existing gap in preventive medicine and public health, one that historically has been the most challenging to scrutinize and exploit.