Current Environmental Health Reports

, Volume 2, Issue 2, pp 204–213 | Cite as

The Pregnancy Exposome

  • Oliver RobinsonEmail author
  • Martine Vrijheid
Early Life Environmental Health (J Sunyer, Section Editor)
Part of the following topical collections:
  1. Topical Collection on Early Life Environmental Health


The exposome concept takes a holistic approach facilitated by new and emerging technologies to describe ‘the totality of human environmental (i.e. non-genetic) exposures from conception onwards, complementing the genome’. It provides a framework to advance the environmental epidemiology field that has until now focused almost exclusively on single-exposure health effects. The exposome includes an external domain, measured by methods including geo-spatial modelling, questionnaire and biomonitoring of external exposures while the internal domain is commonly assessed through molecular omics platforms. The internal domain, in part, reflects the biological response to the external domain. New statistical frameworks are required to integrate and assess exposome-health effects. The pregnancy period is a key starting point to describe the dynamic exposome, due to its heightened sensitivity and potential lifetime impact. A handful of studies have started to move towards an exposome approach in assessing the effects of the multiple exposures during pregnancy on child development. New research projects are underway to test the exposome approach on a large scale.


Exposome Child health Pregnancy Multi-pollutant Omics Exposure 


The idea of an ‘exposome’ was first proposed by Wild [1] as a call to arms for health research to complement the impressive advances made in measuring the human genome with similar strides in measuring the environmental component of disease aetiology. Like genomics and other rapidly proliferating ‘omics’ fields, the key focus of the exposome concept is to take a holistic approach facilitated by new and emerging technologies to describe ‘the totality of human environmental (i.e. non-genetic) exposures from conception onwards, complementing the genome’. The hope is that through the use of data-driven approaches pioneered in the genomics fields, advances can be made in the environmental epidemiology field that has until now focused almost exclusively on single-exposure health effects. The exposome has been delineated to include three overlapping and complimentary domains [2]: a general external domain including macro-level factors such as climate, social and economic context and psychosocial factors; a specific external domain including agents such as environmental pollutants, diet and drugs; and a specific internal domain including inflammation, metabolism and the gut microflora (Fig. 1). How to incorporate the hierarchical structure of the interlinked domains (if at all) into a putative exposome analysis is just one complexity still to be addressed.
Fig. 1

The three domains of the exposome as defined by Wild [2], with some examples of each domain

In this review, we first discuss initial attempts to put the exposome concept into practice and some of the challenges that are being faced, before discussing the importance of the pregnancy exposome. We then review the handful of studies that have so far addressed the relationship between the external and internal domains of the pregnancy exposome and child health and development. We conclude with a discussion on future perspectives.

Measuring the Exposome

The exposome has also been divided by how each part can be feasibly measured [3••]: The external exposome includes both those factors encountered outside that are principally characterized by geo-spatial modelling and location (e.g. air pollution, noise and the built environment) and those external exposures assessed on an individual basis through questionnaire or biomonitoring (e.g. diet, smoking, persistent organic pollutants (POPs) and pesticides). The internal exposome is then assessed through high-throughput molecular omics methodologies including methylomics, transcriptomics, proteomics and metabolomics and are expected to reflect parts of the measured external exposome [4]. Furthermore, some investigators [5, 6] limit the exposome to what can be internally measured in biological tissues or fluids arguing that, for a given exposure to have a health consequence, there must be a measurable component or response and that this ‘top-down’ approach is most likely to yield novel risk factors. However, many technical challenges still need to be overcome, and for some important exposures, no biomarkers are available [7].

There are still few practical applications of the exposome. Patel et al. [8] applied the genome-wide association study (GWAS) analytical approach of multiple testing corrected for false discovery to U.S. National Health and Nutritional Examination Survey biomonitoring data of 266 environmental factors, which provided a relatively wide coverage of the external exposome. They identified environmental risk factors for type 2 diabetes with similar effect size to genes commonly indentified through GWAS. This is an important proof of concept, although its cross-sectional nature limits the conclusions that can be drawn. Nieuwenhuijsen et al. [9] developed and tested a personal monitoring kit, worn by volunteers, that continuously measures multiple exposures encountered outside. While promising for future exposome studies, its practicality and cost in its current form limit its use in larger studies. Chadeau-Hyam [10] tested the ‘meet-in-middle’ concept for linking both external exposures and disease to the same internal exposome markers. In a small pilot study, they identified urinary metabolic markers associated with dietary fibre intake in pre-clinical samples from the European Prospective Investigation into Cancer and Nutrition study that were also associated with later development of colorectal cancer, demonstrating the utility of using internal exposome markers to demonstrate mechanistic links between the external exposome and disease. In a large prospective study, Wang et al. [11] took an internal exposome approach and applied agnostic metabolome-wide measurement of serum and identified novel metabolites, choline, trimethylamine N-oxide and betaine, that predicted cardiovascular disease and could be traced, through careful experiment, back to their dietary source, phosphatidylcholine. This key study demonstrated that it is possible to trace internal exposome markers back to their external exposure origins and identified a novel aetiological pathway of cardiovascular disease risk. Overall, available studies show the complexities of comprehensively measuring and analysing the exposome; although none has achieved ‘complete’ exposome coverage (or is ever likely to), they share a similar holistic, rather than single exposure, approach that defines them as exposome studies.

The Dynamic Exposome

A further important feature of the exposome is its longitudinal and dynamic nature, in contrast to the static and fixed genome. Over the course of their lifetime, an individual is exposed to levels of different environmental factors that vary on an hourly to yearly basis, with certain exposures being greatest during certain periods of life such as occupational exposures during the working portion of a lifetime [12]. Since continuous exposome monitoring would be impossible, it is widely envisaged that the longitudinal or ‘life course exposome’ would be built up from cross-sectional snapshots at key life periods such as in utero, early childhood, adolescence, adulthood and old age. The choice of time point at which the exposome will be surveyed is crucial since the health effects of exposures vary over time for two main reasons. Firstly, sensitivity to the exposome will vary according to age and developmental stage, with key development windows in utero particularly influenced by the exposome which may then have lifetime impacts [13]. Secondly, the effects of certain agents may be altered when co-exposed to other time-varying agents through synergistic or interactive effects. For instance, exposure to the widely distributed pollutant, dichlorodiphenyltrichloroethane (DDT) alters expression of hepatic metabolic enzymes, thereby enhancing susceptibility to other environmental toxicants [14]. The improved understanding of the combined health effects of exposure mixtures generally encountered in reality is one of the promises of the exposome.

Linking the Exposome to Health

The exposome concept is defined in terms of its relationship to health, and as such a crucial part of putting the exposome into practice is developing the necessary statistical frameworks to link these data to health outcomes. Recent reviews have outlined a number of modelling methodologies [15•, 16] that address the statistical challenges, such as the multiple testing burden and dense correlations structure, inherent in exposome analysis. Related methodological issues of exposome analysis such as varying confounding sets, dose-response relationships and differential measurement error will also need to be addressed. Some of these issues are discussed in recent multiple air pollution analyses [17, 18], and it is increasingly recognized within air pollution research that multi-pollutant analyses are needed to better reflect reality and improve public health interventions [19]. Billionnet et al. [20] reviewed the air pollution literature to identify studies that have moved towards the multi-pollutant approach and the statistical methodologies used. They indentified 18 studies with a range of methodologies including hierarchical Bayesian approaches, data reduction and clustering methods and more traditional multivariate regression models. The multivariate regression approach is applicable when co-linearity is not too large and individual effects are strong. Otherwise, methods that describe the ‘typology’ of air pollution are more suitable, and the authors suggested better understanding of the biological pathways through which toxicity is exerted may allow more appropriate grouping of pollutants. Within the literature examining suspected endocrine disrupting contaminants, Lenters et al. [21] recently applied a novel exposomic approach to 18 exposures in four xenobiotic classes (non-essential metals, organochlorines, perfluoroalkyl acids and phthalates) and its relation to male reproductive function. Using sparse partial least squares, a combined data reduction and variable selection method that is able to incorporate correlated exposures, they were able to extract statistically robust associations between certain organochlorines and phthalate metabolites and reproductive parameters. It is apparent that different statistical approaches will be needed for the different objectives of exposome analyses which can include identifying the most important individual exposures (crucial for regulatory policy), describing exposure mixture or exposome patterns (as most normally encountered in reality) and indentifying synergistic or interactive effects, which mainly due to issues of power, have been relatively unexplored in the epidemiological literature. Lampa et al. [22] performed a simulation study using real environmental exposure data and found that the boosted regression trees method is suitable for identifying a range of interaction scenarios. Although the method is susceptible to identifying spurious interactions, it appears a relatively powerful approach for epidemiological study.

The Pregnancy Exposome

The developing fetus is especially vulnerable to effects of environmental exposures [13] since its organs are rapidly growing and developing, its metabolism is immature and the received dose may be greater relative to its body weight [23]. In utero, exposure to environmental stressors during critical windows can disrupt developmental processes and permanently alter body structure, metabolism and physiology, leading to chronic pathologies in later life. There is now good evidence for the effect of prenatal exposure to environmental contaminants including air pollution, metals and industrial agents on fetal growth [24, 25], neurological development [24, 25] and the respiratory system [25, 26]. Evidence is also growing for effects on metabolic signalling [27]. Under The Developmental Origin of Health and Disease hypothesis, these effects may be expected to have a lifetime impact [28, 29]. Therefore, the ‘pregnancy exposome’ is a key both as a starting point to develop a lifetime exposome and due to the heightened impact of the exposome during the in utero period. Figure 2 illustrates a typical external exposome profile of environmental contaminants during pregnancy and continuing into early life indicating the dynamic nature of the exposome.
Fig. 2

The dynamic pregnancy and early-life external exposome [12]. Representative of the temporal variation of environmental measurements of a child who is breastfed, is exposed to maternal smoking during pregnancy and childhood, and does not change address. 1. Red line: persistent organic pollutants, including organochlorines, brominated and perfluorinated compounds (breastfeeding). 2. Orange line: mercury and lead (cumulative during pregnancy). 3. Blue line: arsenic (not cumulative and reduced exposure during breastfeeding). 4. Grey line: environmental tobacco smoke. 5. Black line: air pollution, noise, and green spaces. 6. Purple line: ultraviolet radiation, organophosphates pesticides, benzophenone-3 (seasonal exposure). 7. Green line: non-persistent pollutants, including phthalates, phenols, and triclosan, and water disinfection by products

External Pregnancy Exposome Studies

A Pubmed search ((pregnancy or in utero) and (exposome or multiple pollutant)) identifies only a small number of studies of health outcomes that take an external exposome or multiple pollutant approach. These early examples of ‘pregnancy exposome’ studies are described in the following section and in Tables 1 and 2. Warren et al. [30] jointly examined the effects of two air pollutants, PM2.5 and ozone, continuously measured throughout pregnancy on the incidence of preterm birth (PTB) in TX, USA. Using a hierarchical Bayesian framework, they identified critical windows of effect that differed between the pollutants. PM2.5 had the stronger effect overall with exposure in weeks 4 to 22 most relevant while ozone was found to have the greatest impact on PTB in weeks 1 to 5, illustrating the importance of understanding the temporal nature of the pregnancy exposome. Le et al. [31] examined the effects of air pollutants CO, SO2, NO2 and PM10 on PTB and term small for gestational age (SGA) births in Detroit, USA, taking into account interactive effects with social context. Due to the diverse pollutant sources in the area, correlations between exposures were modest allowing the use of multi-pollutant regression models to separate the effects of each pollutant individually. Significant effects at various critical windows were found for SGA births for CO (first and second trimester), SO2 (all trimesters), NO2 (first month and all trimesters) and PM10 (first month and first trimester). For PTB births, odds ratios were elevated during only first month SO2 and NO2 exposures. Effect sizes were greater among mothers with lower educational level and black ethnicity, while there was heterogeneity in effects by pollutant between smokers and non-smokers. Positive associations with SGA were observed for among smokers for NO2, but associations with CO and PM10 were only observed among non-smokers. Dadvand et al. [32•] extended their analysis of term low birth weight (LBW) in Barcelona, Spain, to include additional components of the community level external exposome including the built environment (proximity to major roads), temperature, noise, air pollutants and the buffering effects of roadside tree planting. Living within 200 m of a major road while pregnant was associated with a 46 % increased risk of LBW, and measured air pollutants and heat together explained around one third of this association. There was some evidence for attenuation of the effect size with increasing tertiles of tree cover. Vafeidi et al. [33] measured serum levels of multiple POPs, including pesticides, polychlorinated diphenyls and polybrominated diphenyl ether (PBDE)-47, among pregnant women during the first trimester in Crete, Greece. Combining these correlated exposures into single model using an overall exposure score, they observed a 42-g decrease in birth weight with unit increase in exposure score, which was robust to adjustment for social and maternal factors including gestational weight gain.
Table 1

External pregnancy exposome studies


Study design



Main findings

Warren et al. [30]

American prospective birth registry

PM2.5 and ozone

Time varying hierarchical Bayesian

Different critical windows were found for the effect of each pollutant on preterm birth

Le et al. [31]

American prospective birth registry

CO, SO2, NO2 and PM10

Multivariate regression

Associations found with small for gestational age for all pollutant during various critical windows; associations with preterm birth found for first month SO2 and NO2 only

Dadvand et al. [32•]

Spanish prospective birth cohort

PM2.5, PM2.5–10, PM10, PM2.5 absorbance, NO2, NOx, noise, temperature, distance to major roads, tree cover

Multivariate regression, mediation analysis

Proximity to major road associated with low birth weight; air pollution and temperature explained a third of the association; some evidence for buffering effect by tree cover

Vafeidi et al. [33]

Greek prospective birth cohort

10 Persistent organic pollutants

Regression using exposure score

Exposure score inversely associated with birth weight

Braun et al. [34]

American prospective birth cohort

52 Endocrine-disrupting chemicals

Semi-hierarchical Bayesian

Positive associations found for two chemicals with autistic behaviours in childhood

Yorifugi et al. [35]

Faroese prospective birth cohort

Lead and methylmercury

Multivariate regression with interaction terms

Associations between lead and childhood cognitive function observed in lowest methylmercury exposure group

Golding et al. [36]

British prospective birth cohort

1755 Parental, prenatal and neonatal environment and lifestyle factors

Two-stage multiple univariate testing and final regression model

Association between unhappy childhood of the mother and motor skills in the child

References selected from PubMed search (pregnancy or in utero) and (exposome or multiple pollutant) to represent state of the science

Table 2

Internal pregnancy exposome studies


Study design



Main findings

Koestler et al. [37]

American prospective birth cohort


450 K DNA methylation array in cord blood

Dose-response linear relationships with methylation of genes encoding oestrogen receptor alpha and a nuclear receptor co-activator

Fry et al. [38]

Thai prospective birth cohort


Microarray of mRNA in cord blood

11 Transcripts highly predictive of arsenic exposure; activation of inflammation, stress and cell death pathways

Rager et al. [39•]

Mexican prospective birth cohort


Microarray of mRNA and miRNA in cord blood

Activation of pathways involved in cytokine activity, immune response, inflammatory response and stress

Ahmed et al. [40]

Bangladeshi birth cohort


Cytokine panel in cord blood and T cell count in placental tissue

Reduction in T cells in placenta and association with cytokine activity

Suter et al. [41]

American prospective birth cohort

Maternal smoking

27 K DNA methylation array and microarray of mRNAs in placenta

Correlation between altered methylation and expression in 428 genes primarily involved in oxidative stress pathways

Joubert et al. [42•]

Norwegian birth cohort

Maternal smoking

450 K DNA methylation array in cord blood

Altered methylation within 10 genes, including AHRR and CYP1A1 involved in detoxification of xenobiotics

Hochstenbach et al. [43]

Norwegian birth cohort

Dioxin-like compounds, estrogenic and androgenic agents, and adducts of acrylamide and glycidamide

Microarray of mRNA in cord blood

Significant differences between sexes in exposure response with activation of TNF-alpha-NF-kB signalling and Wnt-pathway in boys

Maitre et al. [44•]

Greek prospective birth cohort

Metabolomic analysis

1H NMR analysis of first trimester urine samples

Metabolites acetate, formate, trimethylamine and tyrosine as associated with fetal growth restriction. N-acetyl glycoprotein resonances associated with induced preterm birth

References selected from PubMed search (see Table 1) and recent reviews [4, 45, 46] (including citing and cited references) to represent state of the science

The effect of multiple external pregnancy exposures on child neurodevelopment has also been investigated by a few studies. Braun et al. [34] measured levels of 52 endocrine-disrupting chemicals among 175 pregnant women in a prospective birth cohort in Cincinnati, USA, and examined the relationship with autistic behaviours in children at 4 and 5 years of age using a semi-hierarchical Bayesian analysis to account for co-linearity and multiple sampling. Increasing concentrations of the pesticide trans-nonachlor and PBDE-28 were associated with more autistic behaviours; however, the small study size prevented the dismissal of chemicals for which null associations were found. Yorifugi et al. [35] examined the association between cord blood lead concentration and cognitive function at 7 and 14 years of age in a Faroese birth cohort, while accounting for exposure to methylmercury. Clear associations between lead and cognitive function were only observed in the lowest methylmercury exposure group and evidence suggested a less-than additive or antagonistic interaction. This was explained by the toxicants competing to induce the observed effect through similar pathways and demonstrates the importance of assessing co-pollutants when investigating health effects. Golding et al. [36] applied the exposome approach, in the sense of hypothesis-free multiple testing, to examine environmental or lifestyle factors associated with gross motor skills at 8 years old in a British birth cohort. They conducted multiple univariate analyses on 1755 variables covering factors spanning the lives of both parents and then incorporated the most significant variables into a final regression model. They identified the novel association between an unhappy childhood of the mother and motor skills in the child. However, with only minimal adjustment for repeated testing to account for the generation of false positives, it is difficult to draw conclusions from this study.

Internal Pregnancy Exposome Studies

Analyses investigating the internal exposome and its response to external environmental exposures have also been developing in recent years, using the maturing omics platforms. Studies were identified through the Pubmed search described above and from recent reviews [4, 45, 46] (including citing and cited references). Most studies have looked at single exposures that can be well measured during pregnancy to identify associated internal exposome profiles and improve mechanistic understanding (Table 2). Arsenic is a known human carcinogen and an important public health hazard since millions of people worldwide are chronically exposed to arsenic in their drinking water, while dietary sources, including rice, other grains and fruit juices are also important. A few studies have now looked at the effects of in utero exposure on internal exposome profiles of newborns at various levels of biological organization. Koestler et al. [37] examined the relationship between maternal arsenic exposure and DNA methylation, involved in epigenomic regulation of gene expression, in cord blood of newborns. Although they did not detect global changes in methylation across the genome, a number of differentially methylated CpG dinucleotide sites were identified, predominately with greater methylation in the high-exposed group. Several of these loci exhibited dose-response linear relationships between methylation and arsenic exposure including genes encoding oestrogen receptor alpha and a nuclear receptor co-activator, which given the importance of hormonal signalling in fetal development may have health consequences. A study in Thailand [38] compared RNA transcripts in the cord blood of newborns born to mothers differentially exposed to arsenic using microarray analysis. They found that prenatal arsenic exposure had a strong impact on the transcriptome of the newborn, with a subset of 11 transcripts highly predictive of arsenic exposure. Pathway analysis of gene differentially expressed across the genome identified activation of pathways including inflammation, stress and cell death. Rager et al. [39•] performed a similar analysis in a Mexican birth cohort; however, they also measured levels of miRNAs, small non-coding RNAs involved in regulation of messenger RNA (mRNA) translation. They identified 12 miRNAs associated with arsenic exposure, many of which were known to have roles in carcinogenesis, and 334 mRNA transcripts including 28 identified in the study of Fry et al. [38]. They found that a third of miRNA-mRNA pairings, predicted computationally based on target sequences, were correlated in their analysis. At the biological pathway level, analysed using both miRNAs and mRNAs, there was a similar enrichment of genes involved in cytokine activity, immune response, inflammatory response and stress. Interestingly, since arsenic exposure in utero is associated with health outcomes related to a repressed immune system such as increased morbidity during infancy from infectious diseases, they identified genes involved in innate and adaptive immunity. Ahmed et al. [40] also identified inflammatory markers at the proteomic and cellular level in a Bangladeshi cohort associated with high maternal arsenic exposure including reduction in levels of placental T cells and perturbations to cord blood concentrations of cytokines.

Maternal smoking is another exposure that has been relatively well investigated and provides an important model for the use of omics in identifying environmental exposure pathways. Suter et al. [41] used parallel arrays to measure DNA methylation and gene expression in placenta sampled from smoking and non-smoking mothers. They found altered expression of 623 genes and methylation of 1024 CpG sites among smokers, with significant correlation between transcription and methylation in 428 genes primarily in oxidative stress pathways. Altered methylation of six of these genes was significantly associated with birth weight reduction. The study demonstrates that smoking results in site-specific methylation changes resulting in transcriptional disregulation. Joubert et al. [42•] performed an epigenome-wide analysis of methylation changes in cord blood related to maternal plasma cotinine levels, a validated smoking biomarker, in a Norwegian birth cohort. The identified altered methylation in 26 CpG sites mapping to 10 genes that passed conservative correction for multiple testing and were replicated in a smaller American birth cohort. Intriguingly, two of the genes AHRR and CYP1A1 encode proteins involved in detoxification of xenobiotics, while AHRR was also identified independently in an epigenome-wide study of lymphoblasts and pulmonary alveolar macrophages in adult smokers [47], demonstrating that in utero, exposure induces epigenetic changes that are also observed in exposed adults.

Hochstenbach et al. [43] took a more global approach to their external exposure assessment in their study of global gene expression changes in cord blood in newborn boys and girls. They measured exposure to dietary and environmental carcinogens using gene reporter assays to determine exposure to dioxin-like compounds, estrogenic and androgenic activity and haemoglobin adducts of acrylamide and glycidamide. They noted significant differences between the sexes in gene expression response with, in particular, male-specific activation of TNF-alpha-NF-kB signalling and the Wnt-pathway, which they suggest may present a mechanistic explanation for the higher rates of childhood leukaemia observed in boys.

Metabolomics, the study of small metabolites present in biological samples, is considered a promising exposomic platform since the metabolome is both environmentally and biologically determined, with a twin study estimating that 60 % of variability observed depends on the environment [48] and is the final expression of all biological processes thereby integrating the environment with all of the biochemical interactions (including with the gut microflora) taking place in the body. A few studies have investigated the pregnancy metabolome with the largest so far conducted by Maitre et al. [44•] on first trimester urine samples collected from a Greek birth cohort. Using 1H NMR spectroscopy that is less sensitive but more robust than other metabolomic platforms, they identified metabolites acetate, formate, trimethylamine and tyrosine as associated with fetal growth restriction. These metabolites were inversely correlated with blood insulin and are associated with activity of the gut microflora and may represent a specific microbial distribution or dietary pattern conducive to such a distribution. N-acetyl glycoprotein resonances, thought to reflect inflammation, were predictive of induced preterm birth and correlated with the body mass index of the mother.

Conclusions and Future Perspectives

The exposome concept is a broad idea and has generated much discussion of how it should be interpreted. It is part of a wider trend in related fields such as exposure assessment, toxicology, systems biology and public health policy to take a more holistic approach that is more reflective of the exposure mixtures we encounter day to day. Although implementing exposome research remains challenging and expensive, characterization of the ‘pregnancy exposome’ is a logical and feasible area to focus efforts due to its lifetime importance and defined exposure window.

While exposome research is still in its infancy, as we have seen, a number of studies have started to implement the concept during pregnancy through a variety of approaches. As with other omics platforms, such in GWAS studies where imputation is required to provide more complete coverage or in metabolomics where coverage will depend on the matrix or analytical platform used, only a partial evaluation of the exposome has so far been achieved. Efforts to move towards wider coverage of the external exposome have been made through measurement of multiple pollutant biomarkers or incorporating geo-spatially derived exposure estimates. Emerging technologies and methods are allowing numerous environmental variables to be measured simultaneously [49]; however, exposure misclassification remains a key concern of exposure assessment, and care must be taken that the exposome analyses do not multiply these problems. Future pregnancy exposome studies must address this issue with, for example, multiple sampling to fully capture exposure to short-lived non-persistent pollutants (a class within which new varieties are continuously produced by the chemical industries) or more detailed information on how mothers move through their external environment [23]. Numerous studies have also characterized the internal pregnancy exposome at various biological levels. To move towards a unified exposome approach, efforts are needed to piece these parts together. Current focus is on linking internal exposome profiles to single or a few external exposures, and it may turn out some of the internal exposome responses emerging are more generalisable to whole classes of external exposures, which may eventually minimize the need for comprehensive external exposure assessment. Alternatively, the internal exposome should be considered as important exposures in their own right rather than external exposome mediators, and ‘exposome-wide’ health analyses should analyse these domains side by side, with emphasis on interaction between and within the domains. Metabolomics may prove a platform capable of integrating the exposome domains in a single analysis; however, the sensitivity of current machines is still a long way off detecting markers of environmental pollutant exposures alongside currently detectable diet, drug and metabolic profiles [5].

Around the world, research projects are now underway to integrate these parts of the pregnancy exposome within a large number of subjects. In Europe, several projects are ongoing; the HELIX project will measure both the external and internal exposome domains among mother-child pairs and link these to important childhood developmental areas such as pre- and postnatal growth, obesity, behaviour and neurodevelopment, and asthma and allergies [3••]. Among up to 32,000 mother-child pairs, the health effects of the outdoor exposome measured at the community level (air pollution, temperature, noise, UV radiation, green spaces and the built environment) will be analysed, and in a subset of 1200 mother-child pairs, the external exposome to be supplemented with multiple pollutant biomarkers and questionnaire derived variables to build a comprehensive picture of the external exposome. Multiple sampling, personal monitoring and smartphone-derived activity data will also improve external exposome estimates. In parallel, the internal exposome will be measured in the same individuals through multiple omics platforms, which should address some of the questions of how the domains are linked. Statistical methodologies are also being developed to best integrate and analyse the pregnancy and early-life exposome. The Exposomics project [50] uses multiple cohorts covering various life periods and focuses on improved assessment of air and water pollutants and linking these both to acute and chronic internal exposome response. The expectation is that exposome signals from pregnancy exposure measured in archived samples collected in birth cohorts will be replicated in samples from cohorts at different life stages building a life-course exposome picture.

In conclusion, the exposome concept provides an important new framework to integrate and improve knowledge on the environmental component of disease aetiology, with the pregnancy period crucial to full understanding of the human exposome. A more holistic approach, incorporating information on how the various components of the pregnancy exposome co-exist and interact will improve our knowledge of the causes of the complex and multifactoral developmental disorders of childhood, some of which are becoming increasingly common, and ultimately lead to better prevention strategies.



We thank Maribel Casas for drawing Fig. 2.

This research received funding from the European Community’s Seventh Framework Programme (FP7/2007-2013) under grant agreement 308333–the HELIX project.

Compliance with Ethics Guidelines

Conflict of Interest

Oliver Robinson and Martine Vrijheid declare that they have no conflict of interest.

Human and Animal Rights and Informed Consent

This article does not contain any studies with human or animal subjects performed by any of the authors.


Papers of particular interest, published recently, have been highlighted as: • Of importance •• Of major importance

  1. 1.
    Wild CP. Complementing the genome with an “exposome”: the outstanding challenge of environmental exposure measurement in molecular epidemiology. Cancer Epidemiol Biomarkers Prev. 2005;14:1847–50.CrossRefPubMedGoogle Scholar
  2. 2.
    Wild CP. The exposome: from concept to utility. Int J Epidemiol. 2012;41:24–32.CrossRefPubMedGoogle Scholar
  3. 3.••
    Vrijheid M, Slama R, Robinson O, Chatzi L, Coen M, van den Hazel P, et al. The human early-life exposome (HELIX): project rationale and design. Environ Health Perspect. 2014;122:535–44. This paper describes the design and rationale of the HELIX project, one of the first european-wide Exposome projects, and should ultimately show the utility of the Exposome approach. PubMedCentralPubMedGoogle Scholar
  4. 4.
    Wild CP, Scalbert A, Herceg Z. Measuring the exposome: a powerful basis for evaluating environmental exposures and cancer risk. Environ Mol Mutagen. 2013;54:480–99.CrossRefPubMedGoogle Scholar
  5. 5.
    Rappaport SM, Barupal DK, Wishart D, Vineis P, Scalbert A. The blood exposome and its role in discovering causes of disease. Environ Health Perspect. 2014;122:769–74.PubMedCentralPubMedGoogle Scholar
  6. 6.
    Miller GW, Jones DP. The nature of nurture: refining the definition of the exposome. Toxicol Sci. 2014;137:1–2.CrossRefPubMedCentralPubMedGoogle Scholar
  7. 7.
    Peters A, Hoek G, Katsouyanni K. Understanding the link between environmental exposures and health: does the exposome promise too much? J Epidemiol Community Health. 2012;66:103–5.CrossRefPubMedGoogle Scholar
  8. 8.
    Patel CJ, Bhattacharya J, Butte AJ. An Environment-Wide Association Study (EWAS) on type 2 diabetes mellitus. PLoS One. 2010;5:e10746.CrossRefPubMedCentralPubMedGoogle Scholar
  9. 9.
    Nieuwenhuijsen M, Donaire-Gonzalez D, Foraster M, Martinez D, Cisneros A. Using personal sensors to assess the exposome and acute health effects. Int J Environ Res Public Health. 2014;11:7805–19.CrossRefPubMedCentralPubMedGoogle Scholar
  10. 10.
    Chadeau-Hyam M, Athersuch TJ, Keun HC, De Iorio M, Ebbels TMD, Jenab M, et al. Meeting-in-the-middle using metabolic profiling—a strategy for the identification of intermediate biomarkers in cohort studies. Biomarkers. 2011;16:83–8.CrossRefPubMedGoogle Scholar
  11. 11.
    Wang Z, Klipfell E, Bennett BJ, Koeth R, Levison BS, Dugar B, et al. Gut flora metabolism of phosphatidylcholine promotes cardiovascular disease. Nature. 2011;472:57–63.CrossRefPubMedCentralPubMedGoogle Scholar
  12. 12.
    Athersuch TJ. The role of metabolomics in characterizing the human exposome. Bioanalysis. 2012;4:2207–12.CrossRefPubMedGoogle Scholar
  13. 13.
    Buck Louis GM, Yeung E, Sundaram R, Laughon SK, Zhang C. The exposome–exciting opportunities for discoveries in reproductive and perinatal epidemiology. Paediatr Perinat Epidemiol. 2013;27:229–36.CrossRefPubMedCentralPubMedGoogle Scholar
  14. 14.
    Robinson O, Want E, Coen M, Kennedy R, van den Bosch C, Gebrehawaria Y, et al. Hirmi Valley liver disease: a disease associated with exposure to pyrrolizidine alkaloids and DDT. J Hepatol. 2014;60:96–102.CrossRefPubMedGoogle Scholar
  15. 15.•
    Sun Z, Tao Y, Li S, Ferguson KK, Meeker JD, Park SK, et al. Statistical strategies for constructing health risk models with multiple pollutants and their interactions: possible choices and comparisons. Environ Health. 2013;12:85. Provides an important first simulation study on potential Exposome statistical approaches. CrossRefPubMedCentralPubMedGoogle Scholar
  16. 16.
    Chadeau-Hyam M, Campanella G, Jombart T, Bottolo L, Portengen L, Vineis P, et al. Deciphering the complex: methodological overview of statistical models to derive OMICS-based biomarkers. Environ Mol Mutagen. 2013;54:542–57.CrossRefPubMedGoogle Scholar
  17. 17.
    Mostofsky E, Schwartz J, Coull BA, Koutrakis P, Wellenius GA, Suh HH, et al. Modeling the association between particle constituents of air pollution and health outcomes. Am J Epidemiol. 2012;176:317–26.CrossRefPubMedCentralPubMedGoogle Scholar
  18. 18.
    Winquist A, Kirrane E, Klein M, Strickland M, Darrow LA, Sarnat SE, et al. Joint effects of ambient air pollutants on pediatric asthma emergency department visits in Atlanta, 1998–2004. Epidemiology. 2014;25:666–73.CrossRefPubMedGoogle Scholar
  19. 19.
    Dominici F, Peng RD, Barr CD, Bell ML. Protecting human health from air pollution: shifting from a single-pollutant to a multipollutant approach. Epidemiology. 2010;21:187–94.CrossRefPubMedCentralPubMedGoogle Scholar
  20. 20.
    Billionnet C, Sherrill D, Annesi-Maesano I. Estimating the health effects of exposure to multi-pollutant mixture. Ann Epidemiol. 2012;22:126–41.CrossRefPubMedGoogle Scholar
  21. 21.
    Lenters V, Portengen L, Smit LA, Jonsson BA, Giwercman A, Rylander L, et al. Phthalates, perfluoroalkyl acids, metals and organochlorines and reproductive function: a multipollutant assessment in Greenlandic, Polish and Ukrainian men. Occup Environ Med. 2014. doi: 10.1136/oemed-2014-102264.
  22. 22.
    Lampa E, Lind L, Lind PM, Bornefalk-Hermansson A. The identification of complex interactions in epidemiology and toxicology: a simulation study of boosted regression trees. Environ Health. 2014;13:57.CrossRefPubMedCentralPubMedGoogle Scholar
  23. 23.
    Vrijheid M. Child health and the environment: where next with birth cohort research? Occup Environ Med. 2014;71:663–4.CrossRefPubMedGoogle Scholar
  24. 24.
    Pedersen M, Giorgis-Allemand L, Bernard C, Aguilera I, Andersen AM, Ballester F, et al. Ambient air pollution and low birthweight: a European cohort study (ESCAPE). Lancet Respir Med. 2013;1:695–704.CrossRefPubMedGoogle Scholar
  25. 25.
    Wigle DT, Arbuckle TE, Turner MC, Berube A, Yang Q, Liu S, et al. Epidemiologic evidence of relationships between reproductive and child health outcomes and environmental chemical contaminants. J Toxicol Environ Health B Crit Rev. 2008;11:373–517.CrossRefPubMedGoogle Scholar
  26. 26.
    Gascon M, Morales E, Sunyer J, Vrijheid M. Effects of persistent organic pollutants on the developing respiratory and immune systems: a systematic review. Environ Int. 2013;52C:51–65.CrossRefGoogle Scholar
  27. 27.
    Thayer KA, Heindel JJ, Bucher JR, Gallo MA. Role of environmental chemicals in diabetes and obesity: a National Toxicology Program workshop review. Environ Health Perspect. 2012;120:779–89.CrossRefPubMedCentralPubMedGoogle Scholar
  28. 28.
    Godfrey KM, Gluckman PD, Hanson MA. Developmental origins of metabolic disease: life course and intergenerational perspectives. Trends Endocrinol Metab. 2010;21:199–205.CrossRefPubMedGoogle Scholar
  29. 29.
    Van den Bergh BR. Developmental programming of early brain and behaviour development and mental health: a conceptual framework. Dev Med Child Neurol. 2011;53 Suppl 4:19–23.CrossRefPubMedGoogle Scholar
  30. 30.
    Warren J, Fuentes M, Herring A, Langlois P. Spatial-temporal modeling of the association between air pollution exposure and preterm birth: identifying critical windows of exposure. Biometrics. 2012;68:1157–67.CrossRefPubMedCentralPubMedGoogle Scholar
  31. 31.
    Le HQ, Batterman SA, Wirth JJ, Wahl RL, Hoggatt KJ, Sadeghnejad A, et al. Air pollutant exposure and preterm and term small-for-gestational-age births in Detroit, Michigan: long-term trends and associations. Environ Int. 2012;44:7–17.CrossRefPubMedCentralPubMedGoogle Scholar
  32. 32.•
    Dadvand P, Ostro B, Figueras F, Foraster M, Basagana X, Valentin A, et al. Residential proximity to major roads and term low birth weight: the roles of air pollution, heat, noise, and road-adjacent trees. Epidemiology. 2014;25:518–25. This is one of the first studies to incorporate many parts of the community level external exposome during pregnancy into a single analysis. CrossRefPubMedGoogle Scholar
  33. 33.
    Vafeiadi M, Vrijheid M, Fthenou E, Chalkiadaki G, Rantakokko P, Kiviranta H, et al. Persistent organic pollutants exposure during pregnancy, maternal gestational weight gain, and birth outcomes in the mother-child cohort in Crete, Greece (RHEA study). Environ Int. 2014;64:116–23.CrossRefPubMedGoogle Scholar
  34. 34.
    Braun JM, Kalkbrenner AE, Just AC, Yolton K, Calafat AM, Sjodin A, et al. Gestational exposure to endocrine-disrupting chemicals and reciprocal social, repetitive, and stereotypic behaviors in 4- and 5-year-old children: the HOME study. Environ Health Perspect. 2014;122:513–20.PubMedCentralPubMedGoogle Scholar
  35. 35.
    Yorifuji T, Debes F, Weihe P, Grandjean P. Prenatal exposure to lead and cognitive deficit in 7- and 14-year-old children in the presence of concomitant exposure to similar molar concentration of methylmercury. Neurotoxicol Teratol. 2011;33:205–11.CrossRefPubMedCentralPubMedGoogle Scholar
  36. 36.
    Golding J, Gregory S, Iles-Caven Y, Lingam R, Davis JM, Emmett P, et al. Parental, prenatal, and neonatal associations with ball skills at age 8 using an exposome approach. J Child Neurol. 2014;29(10):1390-8.Google Scholar
  37. 37.
    Koestler DC, Avissar-Whiting M, Houseman EA, Karagas MR, Marsit CJ. Differential DNA methylation in umbilical cord blood of infants exposed to low levels of arsenic in utero. Environ Health Perspect. 2013;121:971–7.CrossRefPubMedCentralPubMedGoogle Scholar
  38. 38.
    Fry RC, Navasumrit P, Valiathan C, Svensson JP, Hogan BJ, Luo M, et al. Activation of inflammation/NF-kappaB signaling in infants born to arsenic-exposed mothers. PLoS Genet. 2007;3:e207.CrossRefPubMedCentralPubMedGoogle Scholar
  39. 39.•
    Rager JE, Bailey KA, Smeester L, Miller SK, Parker JS, Laine JE, et al. Prenatal arsenic exposure and the epigenome: altered microRNAs associated with innate and adaptive immune signaling in newborn cord blood. Environ Mol Mutagen. 2014;55:196–208. This study integrates epigenomic regulation with the transcriptome during pregnancy and identifies signatures of arsenic exposure in plausible disease related pathways. CrossRefPubMedCentralPubMedGoogle Scholar
  40. 40.
    Ahmed S, Mahabbat-e Khoda S, Rekha RS, Gardner RM, Ameer SS, Moore S, et al. Arsenic-associated oxidative stress, inflammation, and immune disruption in human placenta and cord blood. Environ Health Perspect. 2011;119:258–64.CrossRefPubMedCentralPubMedGoogle Scholar
  41. 41.
    Suter M, Ma J, Harris A, Patterson L, Brown KA, Shope C, et al. Maternal tobacco use modestly alters correlated epigenome-wide placental DNA methylation and gene expression. Epigenetics. 2011;6:1284–94.CrossRefPubMedCentralPubMedGoogle Scholar
  42. 42.•
    Joubert BR, Haberg SE, Nilsen RM, Wang X, Vollset SE, Murphy SK, et al. 450K epigenome-wide scan identifies differential DNA methylation in newborns related to maternal smoking during pregnancy. Environ Health Perspect. 2012;120:1425–31. Using smoking, which may be considered a first model of environmental exposure response, they identify specific methylation sites including gens involved in detoxification. CrossRefPubMedCentralPubMedGoogle Scholar
  43. 43.
    Hochstenbach K, van Leeuwen DM, Gmuender H, Gottschalk RW, Løvik M, Granum B, et al. Global gene expression analysis in cord blood reveals gender-specific differences in response to carcinogenic exposure in utero. Cancer Epidemiol Biomark Prev. 2012;21:1756–67.CrossRefGoogle Scholar
  44. 44.•
    Maitre L, Fthenou E, Athersuch T, Coen M, Toledano M, Holmes E, et al. Urinary metabolic profiles in early pregnancy are associated with preterm birth and fetal growth restriction in the Rhea mother-child cohort study. BMC Med. 2014;12:110. One of the first and largest pregnancy metabolomic studies, which may prove a useful exposome platform. CrossRefPubMedCentralPubMedGoogle Scholar
  45. 45.
    Hou L, Wang D, Baccarelli A. Environmental chemicals and microRNAs. Mutat Res. 2011;714:105–12.CrossRefPubMedCentralPubMedGoogle Scholar
  46. 46.
    Hou L, Zhang X, Wang D, Baccarelli A. Environmental chemical exposures and human epigenetics. Int J Epidemiol. 2012;41:79–105.CrossRefPubMedCentralPubMedGoogle Scholar
  47. 47.
    Monick MM, Beach SR, Plume J, Sears R, Gerrard M, Brody GH, et al. Coordinated changes in AHRR methylation in lymphoblasts and pulmonary macrophages from smokers. Am J Med Genet B Neuropsychiatr Genet. 2012;159B:141–51.CrossRefPubMedCentralPubMedGoogle Scholar
  48. 48.
    Nicholson G, Rantalainen M, Maher AD, Li JV, Malmodin D, Ahmadi KR, et al. Human metabolic profiles are stably controlled by genetic and environmental variation. Mol Syst Biol. 2011;7:525.CrossRefPubMedCentralPubMedGoogle Scholar
  49. 49.
    Patel CJ, Ioannidis JP. Placing epidemiological results in the context of multiplicity and typical correlations of exposures. J Epidemiol Commun Health. 2014;68(11):1096-100.Google Scholar
  50. 50.
    Exposomics. 2014 [cited 29/10/2014]; Available from:

Copyright information

© Springer International Publishing AG 2015

Authors and Affiliations

  1. 1.Centre for Research in Environmental Epidemiology (CREAL)BarcelonaSpain
  2. 2.Universitat Pompeu Fabra (UPF)BarcelonaSpain
  3. 3.CIBER Epidemiología y Salud Pública (CIBERESP)BarcelonaSpain

Personalised recommendations