Unraveling the Exposome

Humans are exposed to a milieu of environmental stressors of a chemical, physical, and social nature that may change over time. Interaction of these stressors with various intrinsic factors such as genetics, sex, life stage, and health status determines susceptibility to related diseases. Cumulative risk assessment seeks to determine the combined risks to health from exposures to multiple agents or stressors. This can be achieved by expanding beyond a G × E approach—where “G” represents genetic susceptibility and “E” (environment) represents a limited range of exposures—to an I × E approach—where “I” (intrinsic) represents the many inter-related biological factors that contribute to disease susceptibility and “E” (extrinsic) represents all nongenetic factors including the exposome. Exposomics is poised to advance this concept and make significant advances in environmental health science and our understanding of the causes of chronic diseases. The internal exposome can be assessed using targeted and untargeted exposomics tools to measure individual chemicals, groups of chemicals, or the totality of chemicals acting on a particular receptor or biological pathway in a functional assay. Comprehensive data on the internal, external, and public health components of the exposome together could inform risk assessment and ultimately guide risk management. These approaches could be applied in vulnerable populations such as migrants or those burdened with multiple types of stressor simultaneously as identified through map- or indicator-based approaches. Development and refinement of additional exposomics tools that can be applied in prospective human epidemiology studies should be a focus of future studies.

The exposome has been described by Wild as the measure of all of the exposures an individual has in a lifetime and how those exposures relate to health. The exposome represents the totality of exogenous (external) and endogenous (internal) exposures from conception onwards, simultaneously distinguishing, characterizing, and quantifying etiologic, mediating, moderating, and co-occurring risk and protective factors and their relationship to disease.

The exposome provides a systems science approach to bringing together and organizing data needed to model the relationships, mechanisms, and pathways among and between external exposures, endogenous exposures, health outcomes, and population-level health disparities. It holds promise for identifying completed exposure pathways from source of exposure in the natural, built, social, and policy environments to route of entry into the body, biomarkers of exposure, biomarkers of disease, disease phenotype, clinical outcomes, and population level disparities, across the lifespan, and between generations. This chapter proposes a new taxonomy for sequencing the public health exposome first described by Wild as the eco-exposome. The aim of this chapter is to identify a common taxonomy for conceptualizing and operationalizing environmental exposures as an important step towards articulating a science of health disparities.

The US military has great potential to study and utilize the exposome, as well as great need. By their very nature, military Service members serve in diverse environments, under a variety of stressors, with pressure to perform and execute tasks under any conditions. The US military is unique in the range of environments they deploy to, as well as the kinds of industrial chemicals and materials they are exposed to, making them one of the most dynamic occupational health populations in the world. Unique environmental exposures include documented incidents such as burn-pit exposure, sulfur fires at the Mishraq sulfur mine, and water quality at Camp Lejeune, versus subtle exposures such as lead from munitions training, diesel exhaust, or JP-8 jet fuel exposure. Balancing the environmental health concerns of individuals with the operational needs of a unit can be challenging and dynamic. This has led the military to prioritize efforts for exposure surveillance, mitigation strategies, and leading-edge research initiatives. In this chapter, we will discuss the unique operational environments and exposures warfighters encounter, as well as the biomonitoring, military records for exposures, and how this relates to individual exposomes. The military has unique assets for exposome monitoring, such as uniform electronic health records (EHR), individual longitudinal exposure records (ILER), serum collection pre- and post-deployment for biobanking and surveillance, and a more normalized population relative to nutrition and demographics. This makes the military exposome unique, and provides important avenues for study and application.

The time periods that influence fetal and early life development are identified in this chapter as key windows of susceptibility to exposures and critical developmental stages of preconception, and the prenatal, perinatal, and postnatal periods. We highlight in this chapter these key developmental windows that characterize the fetal and early life exposome, and present a review of studies that have identified fetal and early life external and internal domains of the exposome. We also present a discussion of issues in exposome study design, including choice of biological samples and statistical complexities, specific to the key developmental times of fetal and early life. While notable studies and consortia have been established to investigate the exposome during the times of fetal development and early life, we argue that future exposome research must expand to incorporate the preconception period, build upon the existing and large body of knowledge of reproductive and peri/pre-natal epidemiological methods and study design, and utilize methods of causal inference. Collectively, this will aid in strengthening both the internal and external validity of our studies, and in the identification of potential causal mechanisms underlying many preventable diseases. Such advancements will lead to better risk assessments and potential policy and medical interventions.

Epigenetic regulation is hereditable but can be influenced by environmental stimuli, in utero circumstances, and aging. Different layers of epigenetic remodeling including DNA methylation, modifications of histone tails, and noncoding RNAs control the spatial and temporal transcriptomic activity. In addition, the epigenome is involved in sustaining chromosome stability. Genomic DNA isolated from blood cells or other pertinent tissues is being expansively exploited for the discovery of biomarkers of effect and exposure. Technology to measure epigenetic marks on a genomic scale complemented with novel tools for data-analysis have recently been developed and continue to be enhanced. Here, we describe common techniques that are applied for untargeted approaches; and to measure regional modifications and gene-specific aberrations. Alterations in epigenetic marks have been associated with various exposures such as tobacco smoke, air pollution, and metal exposures in population-based studies. On the other hand, deviant DNA methylation is a major epigenetic mechanism of epigenetic silencing in a wide range of human diseases including cancers. Epigenetic modifications might play a prominent role in explaining biological mediation of exposures and their effect on health. This is of particular significance in early life exposures where epigenetic alterations can explain how diseases linked to in utero or childhood conditions occur later in life. We discuss relevant examples of how epigenetic remodeling by environmental stimuli affects several health outcomes in adults and in early life.

The exposome concept places substantial weight on the internal chemical milieu of individuals, as this is the primary integrator of the human genome and the wider external environment. Small molecule metabolites of both endogenous and exogenous origin are involved in a plethora of cellular and systemic functions, and collectively contribute to the mechanistic linkage of exposures, responses, and associated adverse outcomes. Temporal and spatial responses of metabolic phenotypes to various environmental stimuli provide a direct report on multiple interacting and conditional processes that are modulated by numerous factors including diet, lifestyle, pharmaceutical use, microbial activity, age, sex, and many others. Measuring and integrating information about the human metabolome represents a critical part of the path toward understanding the environmental determinants of chronic disease.

The size and diversity of the chemical space that metabolites occupy means that measuring the human metabolome via serum, urine, or other biofluids or tissues represents a huge analytical challenge, addressed by the application of high-resolution platforms, typically incorporating liquid- and/or gas-chromatography for separation, and nuclear magnetic resonance spectroscopy and/or mass spectrometry for detection. Advances in the performance of these platforms now permit the measurement of many hundreds or thousands of metabolites, in either targeted or untargeted assays. The focus of this chapter is on the utility of the different analytical platforms, their complementarity, and application to large-scale sample set analysis. Considerations for data analysis and integration with other omics, exposure, and outcome data are discussed, alongside approaches for interpreting findings in the context of the human exposome.

The advent of omics technologies has enhanced significantly our capacity to interpret mechanistically the association between environmental exposure and disease. Although understanding these interactions requires capturing perturbations at different levels of biological organization, transcriptomics holds a key role. Modulation of gene expression represents the initial biological perturbations due to environmental exposure. This is of particular importance when assessing real-life exposure that involves multiple stressors in highly variable time regimes. This chapter aims at (a) demonstrating the place of transcriptomics in modern risk assessment and environmental health associations, highlighting the respective bioinformatics tools that are necessary for the interpretation and (b) demonstrating the feasibility of transcriptomics of understanding environmental risk associated to real-life ubiquitous mixtures. Although environmental exposures occur to mixtures of chemicals rather than to individual agents, most of the toxic effects of air pollutants are ascribed to single chemicals. There is a growing feeling in both the scientific and regulatory communities, however, that there is a need for more comprehensive approaches toward managing the potential impact of complex environmental chemical mixtures on human health. In this perspective, it is expected that toxicogenomics would be the appropriate screening method for assessing biological effects of complex chemical mixtures, allowing us to review the whole spectrum of potential biological response rather than focusing on a predefined number of endpoints as in classical toxicological analysis. In this chapter, beyond the overview of the analytical and computational aspects necessary for implementing toxicogenomics in the context of the exposome, a concrete example of such an application on a typical indoor air mixture as defined in the EU-wide review study INDEX and on a mixture of polyaromatic hydrocarbons (PAHs) isolated from urban air in the city of Milan is given with the aim to identify specific sets of biomarkers for each of the two types of exposure (indoor or outdoor). A human cell line derived from a bronco-pulmonary system (A549) was used as the appropriate in vitro model to support the investigation of the molecular basis for adverse outcomes that are attributed to indoor and/or outdoor air pollution based on epidemiological evidence. Applying a Total Gene Expression assay by Applied Biosystems Microarrays, large sets of genes modulated by single mixtures exposure were profiled. This process led us to identify common biochemical pathways and specific molecular responses. Indoor air mixtures induced a higher level of gene modulation than ambient air PAHs. A closer look at the differences in biological response confirmed major discrepancies in the mode of action of the two mixtures. Indoor air induced primarily modulation of genes associated to protein targeting and localization including in particular cytoskeletal organization; PAHs modulated mostly the expression of genes related to cell motility and gene networks regulating cell–cell signaling, as well as cell proliferation and differentiation. These results provide biological information useful for articulating mechanistic hypotheses linking exposure to xenobiotic mixtures and physiological responses. The evidence on the latter is supported by a large amount of epidemiological evidence, associating exposure to urban air pollution with respiratory allergies, chronic obstructive pulmonary disease, cardiovascular disease, and cancer. Lately, such evidence has been extended to include associations of exposure to polluted ambient and indoor air with kidney disease and even neurodegenerative disorders, and in particular dementia.

Estimating the food exposome, that is, the totality of dietary exposures, represents a substantial challenge because of the wide nature of foods consumed and the variability of the amount and frequency of intake according to food preference, season, and other individual characteristics. Classical approaches to estimating dietary intake have used dietary assessment instruments such as questionnaires and food composition tables. However, these are subject to a number of biases and errors, including misreporting and recall bias. The use of dietary biomarkers provides a more objective approach to assess exposures to dietary compounds and their food sources. Approximately 150 dietary biomarkers have been measured so far in cohort or biomonitoring studies and many new candidate biomarkers are now regularly proposed following the rapid development of mass spectrometry techniques and metabolomics. Several 100 food-derived compounds can be simultaneously measured as part of the human exposome in very small human biospecimens collected in large populations. These approaches and their implementation in dietary-wide association studies are reviewed. This could significantly advance nutritional epidemiology and its utility in linking dietary exposures with disease outcomes.

Contact with indoor dust has been shown to be an important source of contaminant exposure in humans. The composition of dust and its associated contaminants varies by a number of factors including location, nearby objects, and activities. Indoor locations that contain furniture, appliances, and other electronic goods produce dust enriched with chemical additives used in the manufacturing process. Classes of chemicals commonly found in dust include plasticizers, perfluorocarbons, flame retardants, lubricants, and metals used in electronics, batteries, power supplies, and imaging devices. Other types of hazardous chemicals associated with dust are contaminants that are formed after burning or combustion. These include polycyclic aromatic hydrocarbons (PAHs), chlorinated and brominated PAHs, halogenated dioxins, and dibenzofurans. These contaminants are established components of dust and are frequently the focus of environmental monitoring efforts. However, nontargeted chemical analysis of some workplace dust suggests the presence of other organic and inorganic compounds, which remain to be characterized. In addition, many of these contaminants are environmentally persistent and bioaccumulate in seafood and other types of food, complicating exposure assessments and the importance of dust as an exposure source. However, recent research has begun to identify differences in the contaminant profiles between dust and food that can allow better source appointment for some classes such as PBDEs.

The exposome comprises exposures arising from both endogenous processes and the external environment, which include not only chemical compounds but also nutrients, drugs, infectious agents, the microbiome, physical stress, and psychosocial stress. But where do those exposures come from, and what can we do about them? This chapter will focus on two distinct but interrelated questions: (1) Why understanding the environment outside of the body is critical for unraveling the exposome; and (2) How to measure the exposome from the outside the body. The external environment is critical to unraveling the exposome for it provides a depth of context that analysis of biological samples cannot, it also informs data interpretation by providing linkage between external sources of exposure and the internal dose, and in some cases, external assessment is the only way to objectively evaluate an exposure such as social and built environment as well as behavioral factors. The grand challenge lies in how to provide a comprehensive and rigorous characterization of the external environment on the exposome scale. This chapter will discuss

Over the past decade, high-resolution molecular profiles using OMICS technologies have accumulated and have given rise to an unprecedented source of information to explore the effective biological effects of external stressors and to detect drivers of subsequent disease risk. Although the volume, dimensionality, and complexity of OMICs data are constantly increasing, several methods enabling their analysis are now available. The exploration of these data relies on statistical approaches including univariate models coupled with multiple testing correction, dimensionality reduction techniques, and variable selection approaches. While these methods are established, their application in an exposome context is raising specific methodological challenges. In addition, the isolated exploration of an OMIC profile offers the possibility to capture stressor-induced biological/biochemical alterations, potentially impacting individual risk profiles, but this may only yield a fractional picture of the complex molecular events involved, therefore limiting our understanding of the effective mechanisms mediating the effect of the exposome. Despite efficient developments over systems biological approaches, such integrations remain at best data-specific, usually disease-specific, and more systematically restricted to the exploration of (few) predefined hypotheses. The challenging task of exploring the ‘mechanome’ as defined by the ensemble of stressor-induced molecular mechanisms occurring throughout the life course and determining the individual’s risk of developing adverse conditions can be decomposed in three interdependent streams focusing on (1) OMICs profiling, (2) OMICs data integration, and (3) the exploration of molecular mechanisms involved in the exposure effect mediation towards (chronic) disease development.

The promise of a unified way to measure the human exposome is the discovery of novel environmental factors associated with and potentially causative of disease. The human exposome has been tentatively defined as the totality of environmental exposures such as dietary nutrients, pharmaceutical drugs, infectious agents, and pollutants encountered from birth to death. Much as human genetics has benefited from high-throughput profiling in the form of genome-wide association studies (GWAS), a data-driven paradigm for the exposome is needed to systematically and reproducibly discover the environmental determinants of disease. In this chapter, we describe methods for associating the exposome with phenotypic state such as disease. Specifically, this chapter will describe hands-on analytic examples and data to search the exposome for correlates with phenotype, called the “environment/exposome-wide association study” (EWAS). First, we will describe the philosophy behind such a study, including transparency and mitigation of the chances for selection biases. Second, we will describe how to mitigate chances of type 1 error and investigate the possibility of true signals in a sea of possible false positives. We will describe open-source tools for visualization and display of correlated data to enable investigators to efficiently ascertain patterns in phenotypic associations. We end by describing a few success stories of the approach.

The HERCULES Exposome Research Center was first funded under the Environmental Health Sciences Core Center P30 Program by the National Institute of Environmental Health Sciences in 2013. The Core Center Program is designed to support infrastructure for research in environmental health. Emory took a unique approach and focused their proposal on the singular topic of the exposome. The idea was to build intellectual and physical infrastructure that would facilitate exposome research. Cores were developed to expand analytical capabilities in targeted and untargeted mass spectrometry and to provide support for data analysis. A major goal of HERCULES was to promote the exposome concept, which has been accomplished through a series of workshops, seminars, and courses. HERCULES has supported the development of new research centers that use the exposome approach and anticipates continued expansion of exposome research at the home institutions.

Systems biology has been driven by technology (the development of omics) and by statistical modelling and bioinformatics. We aim to bring biological thinking back. We suggest that three traditions of thought need to be considered: (a) causality in epidemiology, for example the “sufficient-component-cause framework”, and causality in other sciences, for example the Salmon and Dowe approach; (b) new acquisitions about disease pathogenesis, for example the “branched evolution model” in cancer, and the role of biomarkers in this process; (c) the burgeoning of omic research, with a large number of “signals” that need to be interpreted. To address the new challenges of epidemiology, the concept of the “exposome” has been proposed. We show examples from recent projects in the field, namely, new omic approaches applied to epidemiological studies; and in particular, the identification of hallmarks of cancer as intermediate steps between exposure to carcinogens and the cancer phenotype, according to the “meet-in-the-middle” concept. We use examples derived from the study of mutational spectra in tumours and benzo(a)pyrene and bisphenol A as model carcinogens. We suggest conceptualising the detection and tracing of signals in terms of information transmission.

Exposome appears as a very promising tool for better understanding the complexity of interactions between genome and environment, especially when investigating large population studies. The HEALS project aims at identifying the complex links among genes, environment, and human disease on allergies and asthma, neurodevelopmental/neurodegenerative and metabolic disorders based on individual exposome characterization and how should this be implemented in large cohorts. HEALS relies on the re-analysis of existing cohort studies and the deployment of a Pilot European Exposure and Health Examination Survey. Although the analysis will start from the collection of biomonitoring data, a wide array of omics technologies (completed by confirmatory in vitro testing) will be employed. Lifetime exposure assessment will involve novel technologies such as sensors and agent- based modelling. Mapping the different omics responses onto regulatory networks and disease pathways will allow understanding the intermediate stages from exposure to disease at individual as well as population level. HEALS is expected to provide additional insights into the way to synthesize different data and methodological tools for assessing the internal and external exposome overall aiming to a better understanding of both the potential mechanisms and the origin of disease. This includes (1) how different environmental factors contribute cumulatively to disease and (2) the common nodes of exposure and molecular events resulting in phenomenally different health outcomes. HEALS is a comprehensive methodological advance aiming to provide the way of linking interdisciplinary research towards the understanding of genome and lifetime environmental interaction at individual and population level.

Since it was first defined by Christopher Wild, to today, the concept of the exposome has evolved into a valuable tool to evaluate human exposure and health. In the chapters of this book the definition of the exposome paradigm and concept is discussed. The most recent techniques based on OMICs, targeted and untargeted analysis for its characterization are described. The challenges arising from the amount of data that needs to be processed to understand the complex mechanism behind exposure and health outcome, as well as the latest statistical and data analysis methods have been discussed. Finally, multiple projects across the globe have or are currently tackling the application of the exposome concept to real studies, and these studies have been presented in this book. In this section, we will provide a chapter summary, and describe how the exposome paradigm has advanced since its definition. We will also explore question such as: what have we learned so far? Does the exposome paradigm provide valuable information? And what needs to be improved? Finally, we will discuss the possibility of future applications of the exposome in disease causality and personalized medicine.