Introduction
Detrimental environmental exposures contribute to many acute and chronic diseases worldwide including cardiovascular disease, cancer, developmental, and neurological disorders, and many others. These exposures contribute significantly to worldwide mortality and have been an active area of study in biomedical fields. While traditional environmental health research was focused on the impacts of exposures to single agents or a few environmental factors, it is well-recognized that individuals are exposed to numerous environmental factors throughout their lifetimes and thus, research should shift towards studying the combined effects of all exposures. This led to the concept of the exposome, in 2005, which refers to the sum of exposures an individual experiences from conception onwards.1 Individuals are exposed to different profiles of environmental compounds, which can lead to differences or similarities in health outcomes depending on the degree of overlap in chemical structure and biological activity of these compounds and the genetic makeup of the individual. Health outcomes related to exposures can also vary depending on the timing of the exposure during the life course. Accordingly, this concept also highlights that multiple early-life exposures explain health consequences that emerge many years later.2
The exposome concept was further defined in 2012 by categorizing exposures into three major domains: internal, specific external, and general external.3 Internal exposures are processes occurring within the body such as metabolism, endogenous circulating hormones, body morphology, microflora composition/activity, inflammation, lipid peroxidation, oxidative stress, and aging. Specific external exposures refer to factors that have historically been major focuses of epidemiological studies linking exposure to health risks such as radiation, infectious agents, chemical contaminants/environmental pollutants, diet (eg, Mediterranean, nutrients, supplements, natural products), lifestyle factors (eg, tobacco, alcohol, drugs of abuse, physical activity), as well as occupational and medical interventions. Lastly, general exposures include broader social, economic, or behavioral/psychological factors such as social capital, education, financial status, psychological and mental stress, urban–rural environment, and climate. These domains form a picture of the entirety of exposures an individual may face. This socio-ecological framework that is built upon the exposure paradigm is referred to as the public health exposome.4 It should also be noted that exposures in different domains may be intercorrelated, and exposure to factors in one domain may modulate the effect of an exposure in a different domain.
Specific external exposures, especially environmental chemicals, are one of the most well-studied aspects of the exposome, as many compounds have defined biochemical mechanisms by which they interact with organisms to exert toxicological effects. Indeed, environmental toxins/toxicants participate in covalent and/or noncovalent interactions with host macromolecular structures (eg, receptors, enzymes, nucleic acids) to exert their mechanism of action, often leading to detrimental health effects.5 Moreover, advancements in precision medicine/therapeutics indicate that individuals respond differently when exposed to the same chemical agents. The toxicodynamic and the toxicokinetic (absorption, metabolism, distribution, and excretion [ADME]) profile of a toxin/toxicant depends on various factors including genetic polymorphisms, gene expression, epigenetic profiles, comorbidities, co-exposures, biological/chronological age, microbiome composition, inflammatory status, gender/sex, and behavioral factors (stress, exercise, diet).6-11 Notably, beneficial exposures (eg, prescribed drugs) also exert their biological effects through many of these mechanisms.12 Two major barriers to understanding these dynamics are the large, continually growing number of environmental compounds for which little is known and the fact that what we do know comes from studying exposures in isolation, which does not reflect the reality of exposures in populations. This highlights a need to 1) conduct large-scale exposome studies to identify human population exposure profiles, 2) use high-throughput techniques (omics technologies) to understand how these exposome profiles alter human biology, and 3) use this information to guide the creation of potential intervention strategies to modify the effects of exposures while recognizing that reducing or removing harmful exposures should first be attempted if possible. Large-scale efforts supported by NIH such as the Children’s Health Exposure Analysis Resource (CHEAR) and the Human Health Exposure Analysis Resource (HHEAR) aim to enhance our understanding of how exposures affect populations by providing researchers access to comprehensive exposure analysis tools, including mass spectrometry-based profiling of all small molecules (untargeted exposomics analysis), to add or expand environmental exposures to their health study.13,14 Our question is: how can this information be used to design intervention methods to modify the effects of exposures?
An answer to this question may lie in nutrition. As evidenced by HHEAR and CHEAR studies, exposure to environmental chemicals has significant effects on the metabolome,15-19 suggesting biochemical mechanisms are involved in the toxicological effects of these compounds. This is also the case for beneficial exposures (drugs) where the metabolome is increasingly being recognized as a readout for drug activity and efficacy.12 Therefore, modifying these metabolic perturbations may provide a strategy to mitigate/prevent adverse health outcomes or enhance beneficial health outcomes related to exposures (Figure 1). Nutrients directly incorporate into metabolic pathways (eg, substrates, products, co-factors, and regulatory elements) and participate in key biological processes (eg, gene transcription, DNA methylation, DNA synthesis/repair, protein synthesis/transport, hormonal signaling),20 thus dietary interventions may provide an effective means to counteract harmful metabolic imbalances caused by exposures to chemicals. Moreover, this provides a logistical advantage as exposure to environmental chemicals can be episodic and/or chronic. Therefore, an intervention strategy is needed that can be implemented consistently to at-risk individuals for disease (such as those with a known genetic variant, geographical location, or socioeconomic status that is known to associate with a disease outcome). It should be noted that the primary intervention for protecting against detrimental exposures is exposure reduction, however this is not always feasible for all individuals. In these instances, developing interventions to combat harmful exposures (eg, through nutrition) would be valuable alternative solutions if proven effective. Dietary modifications or nutrient supplementation are relatively safe to implement and can be tailored to an individual’s preferences to make the intervention more palatable and increase adherence. In this concept manuscript, we provide evidence in support of a “Nutritional Pharmacology/Toxicology” approach which uses metabolomics—alone or in combination with other high information data streams—to identify metabolic perturbations associated with exposures and uses this information to develop nutritional interventions to enhance positive exposure-associated effects (eg, prescription drugs) or mitigate negative exposure-associated effects (eg, environmental contaminants). Notably, the term “Nutritional pharmacology” was used in 1980 in a letter to the editor in The American Journal of Clinical Nutrition which stated that nutrients could be used in high concentrations to provide therapeutic effects beyond supplying essential nutrients.21 Herein, we expand upon this concept by using metabolomics technologies for the rational development of nutrient-exposure combinations. Additionally, we also offer perspectives of the importance of exposome research in health disparities/precision research through identifying nutritional needs for subpopulations and the role of nutrients in gene-environment interactions.
Schematic representation of the concept of nutritional pharmacology/toxicology in exposome research. Chemical agents that contribute to the exposome are composed of a diverse set of compounds including, but not limited to, medication/substance use, air pollution, drinking water contaminants, compounds from plastics, pesticides, mycotoxins/microbial toxins, and diet/diet constituents. Many of these compounds have been shown to affect a wide range of host biological processes which manifest as metabolic perturbations which can be detected with metabolomic technologies. These metabolic perturbations facilitate the transition from a healthy to a disease (eg, toxins/toxicants) phenotype, or a disease to a healthy (eg, medications/nutrients) phenotype. Nutritional intervention may provide a strategy to promote a healthy phenotype by either counteracting metabolic perturbations of detrimental substances or by enhancing metabolic perturbations of beneficial substances.
In certain contexts, nutritional and/or dietary factors have been demonstrated to protect against detrimental exposures or enhance beneficial exposures at the population level
To support this concept, one important question to address is: have nutrients and/or dietary patterns been shown to protect against environmental exposures in humans? Herein, we provide examples in the literature of nutritional/dietary interventions in human subjects protecting against the toxic effects of environmental exposures in certain contexts. Our focus is on studies that support nutrients targeting metabolic perturbations caused by exposures to exert protective effects. Notably, nutrients/dietary components protect against exposures by other mechanisms including altering ADME properties by acting as binding agents to prevent intestinal absorption of toxins/toxicants or to chemically modify toxins/toxicants prior to exposure to neutralize their toxic effects.21,22 Additionally, dietary interventions reduce the amount of exposure to toxins/toxicants, particularly when oral ingestion of the chemical agent in food is a major route of exposure.23 While these are important aspects to consider in the interplay between nutrition and the exposome, they are not the focus of this narrative. Lastly, to most efficiently demonstrate the translatability of the nutritional pharmacology/toxicology concept in exposome research, we focus solely on studies dealing with human subjects. Numerous studies have evaluated the modulatory effects of nutrients/dietary components toxins/drugs using in vitro and in vivo models and have proposed potential mechanisms for these effects. Readers are encouraged to view the numerous review articles summarizing these studies which are published elsewhere.24-27
While the number of human dietary intervention studies to protect against environmental exposures is sparse, there are instances in the literature that have demonstrated this principle, particularly using folate, polyunsaturated fatty acids (PUFAs), and antioxidants including polyphenols and vitamins C and E.26
Folate (vitamin B9) is an essential nutrient and co-enzyme that transfers one-carbon units. It regulates metabolic processes such as purine and pyrimidine synthesis and methylation reactions, and it is known to play a role in preventing neural tube defects.28 Gaskins et al. investigated the effects of folate supplementation on counteracting the developmental and reproductive toxicity of air pollution (nitrogen dioxide [NO2], ozone, fine particulate, and black carbon concentrations) in women using assisted reproductive technology. NO2 exposure was associated with lower rates of live birth for women consuming <800 μg day−1 of supplemental folate, but women consuming >800 μg day−1 of supplemental folate did not have this negative association.29,30 These findings suggest that supplemental folate intake may protect against the reproductive toxicity of air pollution. Folate also protects against endocrine disrupting compounds. Ormond et al. observed that occupational exposures to phthalates and hair spray during pregnancy led to increased risk of hypospadias, which was reduced with folate supplementation.31 Notably, there have been randomized controlled trials showing increased arsenic excretion with folic acid supplementation, providing strong evidence for exposure protection with this nutrient.32,33 These studies demonstrate that folate supplementation may have potential uses for public health in protecting against exposures related to developmental/reproductive toxicity.
PUFAs are a class of unsaturated fatty acids that contain more than one double bond in their carbon backbone and have an especially high prevalence in Mediterranean diets and in foods such as walnuts, fish, and sunflower seeds. Included in this class are omega-3 and omega-6 fatty acids which are known for their roles as signaling molecules and regulators of inflammation, with omega-3 PUFAs generally regarded as beneficial due to anti-inflammatory effects and omega-6 PUFAs generally regarded as detrimental due to pro-inflammatory effects.34 Importantly, these diets also contain high levels of monounsaturated fatty acids (MUFAs) and polyphenolics which contribute to health benefits associated with this dietary pattern.35 Multiple studies have shown that PUFAs, or diets high in PUFAs, protect against the toxicological effects of air pollutants. Brigham et al. investigated whether dietary omega-3 or omega-6 intake modulated asthma outcomes in children following exposure to indoor particulate matter (PM). Results showed that higher omega-6 intake amplified the effects of PM ≤2.5 μm while higher omega-3 intake reduced the effect of PMs.36 Chen et al. observed the association of omega-3 intake on cardiovascular outcomes following PM and ozone exposure and found that higher omega-3 consumption was associated with reduced adverse cardiovascular outcomes (total cholesterol, von Willebrand factor, tissue plasminogen activator, fibrin degradation products, very-low frequency heart rate variability (HRV), high-density lipoprotein, serum amyloid A, soluble intracellular adhesion molecule) following both ozone and PM ≤2.5 μm exposure.37 Lastly, Lim et al. observed that individuals exposed to higher levels of PM ≤2.5 μm had increased mortality risk for cardiovascular disease (CVD) and cerebrovascular disease (CER), and those exposed to higher levels of NO2 had increased mortality risk for CVD. This study also showed that individuals with a higher alternative Mediterranean Diet Index (aMED) score—a 9-point scale that measures an adherence to a Mediterranean diet—had lower rates of these air pollution-related mortalities.38 These findings support that high PUFA intake, particularly omega-3, may have a protective effect against air pollution-related health outcomes, although the contributions of other bioactive components of the Mediterranean Diet should not be ignored.
Dietary antioxidants (eg, polyphenolic compounds, carotenoids, and vitamin C/E) reduce cellular damage related to oxidative stress. They reduce reactive oxygen species (ROS) by participating in oxidation-reduction reactions or enhancing cellular antioxidant systems.39 High serum carotenoid concentrations reduced the risk of polychlorinated biphenyl (PCB)-related development of type 2 diabetes in an analysis of the National Health and Nutrition Examination Survey (NHANES) 2003-2004 data.40 Additionally, Nobile et al. showed that polyphenol-enriched supplements improved markers of pollution-induced oxidative stress and skin aging in a double-blind study.41 Vitamins C and E also improve exposure-related health outcomes. Péter et al. reviewed numerous studies that investigated the modulatory effect of vitamins C and E, either alone or in combination, on the pulmonary/inflammatory effects of air pollution. The seven studies included in this review all reported some degree of a positive effect of vitamins E and/or C supplementation in protecting against air pollution toxicity.42 Lastly, Heitzer et al. showed that vitamin C normalizes endothelial dysfunction in chronic smokers suggesting a potential role for antioxidant vitamins modifying the effects this exposure.43 However, it should be noted that some studies have observed an increase in lung cancer incidence in smokers with supplementation of antioxidants such as beta-carotenes.44 Therefore, more research is needed in this area to better understand how antioxidants modify the effects of smoking.
While nutrient-drug interactions have historically been viewed as negative, there’s also evidence that has shown that nutrients can enhance the effects of beneficial exposures (eg, drugs) in human subjects.12,45 For example, Liu et al. showed that supplementation with a low dose (60 mg) of caffeine increased the effectiveness of escitalopram (an antidepressant drug) in reducing depressive scores by nonpharmacokinetic means.46 Multiple studies have shown that folate can increase the effectiveness of antidepressant medications. Godfrey et al. found that individuals who were folate deficient had an improved response to antidepression treatment if supplemented with 15 mg of folate daily. Interestingly, this effect became greater over time and was also shown to improve psychotropic treatment for schizophrenia.47 In another study, Alpert et al. performed a study where 22 patients who only had a partial response or nonresponse to selective serotonin reuptake inhibitor (SSRI) therapy were augmented with 8 weeks of 15-30 mg day−1 of folinic acid (leucovorin). These individuals were observed to have a further improvement in depressive symptoms following this supplementation, with 19% of subjects experiencing remission.48 Another study observed that folic acid supplementation enhances lithium prophylaxis of bipolar and unipolar disorders. The study found that individuals with low folate levels experienced up to a 40% reduction in their affective morbidity following folate supplementation and subsequent increase in plasma folate levels.49
Other examples of beneficial nutrient-drug interactions include calcitriol and androgen-independent prostate cancer (AIPC). Beer et al. observed that five AIPC patients who completed an 8-week cycle of oral calcitriol administration (0.5 µg kg−1) had to a 50% reduction in prostate-specific antigen (PSA) in APIC after taking intravenous docetaxel.50 This was supported by a follow-up study which used a similar dosing regimen in 37 patients where 81% showed a positive response in PSA levels with 59% of patients showing a >75% reduction in PSA levels.50 Another well-studied, positive drug-nutrient interaction in oncology treatment is the combination of leucovorin (folinic acid) and 5-fluorouracil. This combination is commonly used to increase 5-FU efficacy, in addition to oxaliplatin, in the treatment of colorectal cancer, and is commonly referred to as the FOLFOX regimen.51,52 Vitamin D has been shown to increase the effectiveness of therapies in infectious diseases (hepatitis C, tuberculosis), osteoporosis, epilepsy, atopic dermatitis, multiple sclerosis, and chronic kidney disease.53 In a study by Liso et al., 47 patients with inflammatory bowel disease, Chron’s disease, and ulcerative colitis taking infliximab were enrolled in a study where 20 individuals received a purple corn supplement with a high anthocyanin content, and 27 individuals did not. Results showed that individuals taking the antioxidant-enriched purple corn supplement showed an improvement in inflammatory markers compared with individuals who did not take the supplement.54 Taken together, these studies demonstrate that beneficial drug-nutrient interactions have been seen in clinical settings. It should be noted that more research is needed to better understand the mechanisms that underlie these beneficial interactions, and more drug-nutrient interactions need to be studied.
Metabolomics workflows can be used to identify perturbations associated with exposures
The above studies support the concept that nutritional interventions may provide a promising approach to modulate the health effects of exposures; however more in-depth research is needed. For example, assessing the exposome of individuals as there are likely multiple co-exposures occurring in study participants that may confound or complicate results. Additionally, high information workflows are needed to understand how exposures modulate metabolism so that interventions can be designed to modify these effects. The example studies provided above were based on known associations with pre-formulated hypotheses. The Nutritional Pharmacology/Toxicology workflow does not depend on known associations, but rather uses metabolomics technologies to identify metabolic perturbations and their associations with exposures to derive nutritional interventions. Metabolomics has emerged as a high-throughput, high-dimensional tool to discover how low molecular weight (<1000 Da) metabolites of the host system associate with the onset, severity, or progression of disease, or the response to therapeutic intervention.55 In the field of nutrition, early efforts applied metabolomics to investigate the metabolic response of the host system to specific nutrients, dietary intake, and weight status.56 In the field of pharmacology and toxicology, studies evaluated the response of host metabolism to drugs and chemicals to gain insights into mechanisms and perturbations in target organs.57,58 We posit that the field is ready to combine these efforts, making connections between nutrients and exposures (toxins, drugs, etc.) through their effects on metabolic processes.
Recent technological advancements have led to the simultaneous detection of tens of thousands of features in commonly accessible biospecimens (urine, blood, stool, etc.), many of which are not readily identifiable/annotatable to known compounds giving rise to the concept of metabolomics “dark matter”.57,59 Over the past decade, knowledge bases have rapidly expanded and contain a high amount of spectral data and data mining tools to improve signal annotation and expand the definition of an individual’s metabotype.60 Although many of the signals remain unknown, high resolution mass spectrometry metabolomics affords the simultaneous detection of endogenous metabolites as well as exogenous metabolites produced by microbial systems, ingestion of foods and beverages, use of drugs, medications, caffeine, and tobacco, exposures to household or industrial chemicals, as well as automobile and industrial emissions.61 Because of this, metabolomics technologies are well-positioned to make connections between endogenous metabolites and exposures.
Figure 2 presents a general workflow whereby differential metabolic profiling of exposed and nonexposed individuals is performed to determine metabolic perturbations caused by toxins/toxicants. Because metabolomics workflows can detect exposome compounds, prior knowledge of exposure history is not needed. Rather, individuals can be classified as exposed or nonexposed post-analysis based on the detection of exposome compounds in the metabolomics method. This is particularly useful in large cohorts where the number of exposed individuals can reach large enough sample sizes for downstream analyses. Examination of metabolic perturbations can point to nutritional interventions to target metabolites and/or metabolic pathways to potentially prevent or mitigate toxic effects associated with the exposure. This paradigm can also be applied to beneficial exposures (eg, prescribed drugs and health-promoting phytochemicals in the diet) to compose nutritional interventions to enhance the health effects of a therapeutic substance. Importantly, because exposures target multiple metabolic pathways, these nutritional interventions should be designed to include multiple nutrients with multiple metabolic targets. Generating these “nutrient cocktails” has recently been proposed in some of our recent studies. In a study of osteoarthritis (OA), the metabolic profiles of obese individuals diagnosed with OA were compared with profiles of obese individuals diagnosed as not having OA. This study demonstrated a decrease in the short chain fatty acids propionate that is produced by gut microbes (eg, clostridia) and related to the amount of indigestible fiber in the diet, a decrease in glucosamine derived metabolites, and perturbations in leukotriene metabolism that can be related to the precursor PUFAs. A nutrient cocktail or diet controlling the ratio of omega-3/omega-6, and increased glucosamine, and short chain fatty acids was suggested as a potential intervention strategy to mitigate inflammation associated with OA, although this would need to be tested in future studies.62 In a study of opium use disorder (OUD), metabolic profiles of opium users (some diagnosed as OUD positive) were compared with the profiles of individuals reporting never using opium. This study demonstrated general metabolic disruptions in neurotransmitter metabolism, Krebs cycle, sugar metabolism, fatty acid metabolism, and one carbon metabolism, and a higher level of phthalates and volatile organic compounds in urine of the opium users. This study led to the conclusion that a nutrient cocktail composed of vitamins, vitamin-like compounds, fatty acids, and sugars may mitigate the onset or severity of addiction.19 This concept of deriving nutritional cocktails based on metabolomics analysis that we have presented is still in the early stages and much more research is needed to better understand the potential of this approach.
Metabolomics technologies allow for the design of nutritional interventions to modulate the effects of exposures. Differential metabolomic profiling on biospecimens from exposed and nonexposed individuals identifies metabolites and metabolic pathways affected by exposures, both detrimental and beneficial. Nutritional interventions can then be designed to target these metabolites/metabolic pathways and can be evaluated in controlled experimental models to assess for protection against the toxicological effects or enhancement of the pharmacological effects of the exposure.
Health disparities and gene-environment interactions: the potential to generate precision nutrition interventions for different subpopulations
Another important factor to consider when investigating nutrients to combat exposures are differences within subpopulations. There are significant health disparities across subpopulations regarding environmental exposures—race, sex, and other social determinants of health (eg, healthcare access and quality, food access and quality, education access and quality, social and community context, economic stability, and neighborhood and built environment) that play significant roles in the types and amounts of exposures an individual may experience.63-65 Even when exposed to the same chemical agent at the same level, many of these factors may cause an individual to have a more severe response to the exposure than another. As such, exposome studies are needed that assess subpopulation differences in exposure profiles as well as differences between subpopulations in metabolic disruptions following the same exposure. With this information, exposome measurements have the potential to be used for precision therapy approaches by which an individual's demographic information and exposure profile can be used to determine tailored nutritional interventions to maximize health benefits.
The importance of personalizing nutritional interventions to combat adverse effects of exposures is further demonstrated by the concept of gene-environment interactions. Gene-environment (GxE) interactions refer to the interplay between the genetic makeup of an individual and the physical and social environments resulting in a specific phenotype66 (Figure 3A). One example of this is the association with N-acetyltransferase 2 (NAT2) polymorphisms with increased risk of bladder cancer in individuals who were current smokers or who had occupational exposure to arylamines (including jobs in rubber/chemical industries which have high exposure to these compounds). NAT2 is a xenobiotic biotransformation enzyme which catalyzes acetylation reactions and it is important for clearing carcinogens such as arylamines and other tobacco constituents. A significant association was found between NAT2 variants leading to slower acetylation rates and bladder cancer in current smokers and those with occupational exposure to arylamines.67 Another example is with variants in the Glutathione S-transferase P1 (GSTP1) and microsomal epoxide hydrolase (EPHX1) genes and exposure to air pollution. EPHX1 and GSTP1 are involved in the metabolism of reactive polyaromatic hydrocarbons (PAHs) which are found in air pollution generated by combustion of fossil fuels and also in barbecued meat and cigarette smoke. Salam et al. showed that children with variants in GSTP1 and EPHX1 had an increased risk for developing asthma when exposed to traffic-related air pollution.68 Another example comes from the relationship between paraoxonase 1 (PON1) polymorphisms and Parkinson’s disease following pesticide exposure. PON1 encodes for an esterase/lactonase and it is able to hydrolyze organophosphorus compounds. Individuals with a polymorphism in PON1 had a faster progression of Parkinson’s symptoms following exposure to organophosphate pesticides.69 Finally, Merino et al. showed an added benefit of diet and physical activity on cardiovascular disease risk factors depending on an individual’s generic risk profile for coronary artery disease.70 Taken together, these examples demonstrate that gene-environment interactions play a large role in dictating one’s response to exposures, and that individuals exposed to the same environmental insult may have drastically different health outcomes depending on their genotype. Currently, nutritional intervention approaches in these situations is lacking as is information regarding the optimal window for intervention, and metabolomic approaches may be able to help in filling these knowledge gaps to uncover potential strategies for further testing. Taken together, genetic variant information could one day be used to identify individuals with the highest need for intervention (eg, nutritional) to protect against the adverse health effects of exposures.
Nutritional interventions may provide a precision therapeutic approach to augment gene-environment interactions. Gene-environment interactions fall into two major categories: (A) Two individuals exposed to the same exposure can have different health outcomes due to the presence of low or high risk genetic variants, and (B) two individuals with the same genotype can have different exposure profiles which leads to differences in epigenetic changes resulting in different health outcomes later in life despite the same underlying genotype. Nutritional interventions can be used as a precision therapy to aid those with high-risk genetic variants or to mitigate/reverse adverse epigenetic changes caused by exposures.
Another aspect of gene-environment interactions is that exposures can alter gene expression profiles through epigenetic changes. This is especially important when considering early-life exposures. Exposure to a toxin/toxicant at an early age (or during fetal development) can lead to persistent epigenetic changes which alter gene expression patterns, leading to a higher risk of developing adverse health outcomes later in life (Figure 3B). Notably, epigenetic patterns can be passed down from mother and to offspring, indicating that the effects of an individual’s exposure in early life can be trans-generationally passed down.71,72 In this way, environmental insults have lasting health impacts on multiple generations. For example, prenatal exposure to tobacco smoke is associated with increased risk of diseases later in life such as cancer and cardiovascular disease.73,74 Accordingly, prenatal exposure to tobacco smoke alters global and gene-specific DNA methylation patterns which persist later in life, providing a mechanism for this increased disease risk.75,76 Another example lies with early life exposure to heavy metals, particularly arsenic. Prenatal exposure to arsenic increases risk for developing cancer, cardiovascular disease, nonalcoholic fatty liver disease, and diabetes.77 Alterations in DNA methylation in cord blood have been observed following arsenic exposure during pregnancy, which may explain the increased risk for chronic diseases.78 This paradigm of early life exposures leading to chronic disease later in life is challenging to investigate—toxins/toxicants and their metabolites are often cleared from the body at the time of diagnosis, making it difficult to attribute the disease to exposures unless an individual is followed throughout their lifetime and currently, few intervention strategies (including nutritional) are available beyond exposure reduction strategies. More research is needed to identify epigenetic signatures associated with different exposures to better identify an individual’s exposure history and how that relates to current/future health outcomes.
In both scenarios, nutritional intervention may play a role in mitigating adverse health effects due to gene-environment interactions (Figure 3). Individuals who have a genetic variant that increases disease susceptibility can be given stronger dietary recommendations to give increased protection against the metabolic perturbations caused by the exposure. Nutritional interventions may also aid in preventing epigenetic modifications if administered during the time of exposure. Furthermore, nutrition has been shown to alter the epigenome and may provide a mechanism to reduce or reverse existing epigenetic perturbations caused by exposures to mitigate disease risk.79
On the pharmacological side, our workflow could be used to investigate individuals who are responsive to beneficial treatments versus those who are nonresponsive or partially responsive (Figure 4). Understanding variations in treatment response has been an area of great scientific interest in many pharmacological settings.80-85 Using metabolomics, responders and nonresponders for a pharmacological exposure can be analyzed to determine metabolic processes/networks that differentiate responder status. These results can be used to generate hypothesis regarding metabolic processes that can be targeted through nutritional interventions or other means to enhance responsiveness. Indeed, metabolomics approaches have been used in several pharmacological settings to identify differences in responders and nonresponders, indicating that responder status is often characterized by metabolic differences before, during, or after treatment.86-90 This indicates that metabolomics can successfully identify biological differences linked to responder status which may potentially play a causal role in therapeutic response. We propose that this is an exciting future direction of pharmaco-metabolomics research and more efforts are needed to explore this area.
Metabolomics technologies allow for the design of nutritional interventions to overcome resistance to beneficial exposures. Differential metabolomic profiling on biospecimens from responsive/sensitive individuals and nonresponsive/resistant individuals receiving the same pharmacologic exposure (eg, drugs, natural products) can identify candidate metabolites and metabolic pathways to target to increase efficacy. This approach allows for the generation of hypothesis to target these metabolites and metabolic pathways to potentially enhance treatment outcomes.
An additional internal interface that contributes to the exposome is the diet-gut microbiome axis. The microbiota that inhabit the intestine appear to confer beneficial or adverse effects to host physiology given the dietary context of which they are exposed. In the context of a high carbohydrate diet, they produce short chain fatty acids, by fermentation of indigestible fiber components, which promote colon health.91 In other contexts, such as when faced with diets high in phospholipids, they have been found to produce metabolites such as trimethylamines that are processed by the host liver to generate metabolites, such as trimethylamine oxides, harmful to the cardiovascular system.92 Similarly, other microbiota derived metabolites have been observed to be associated with many disease states.93,94 A diverse or heterogenous population of the gut microbiota is generally seen as beneficial to the host, and dietary exposure appears to be one of the principal driving factors determining an individual’s gut microbial composition.95 Hence, the diet-gut microbiome axis presents as a potentially intervenable target to tailor favorable outcomes. It should be noted that this may also provide additional complexities or challenges for designing nutritional interventions. For example, Ke et al. demonstrated that dietary thymidine and serine enhance 5-fluoro 2’deoxyuridine (FudR) toxicity through different microbial mechanisms, suggesting that the microbiome can alter the response of the host to a drug without altering the drug or host.96 The interaction of the microbiome and the exposome is complex and while it is not the topic of this article, this has been thoroughly reviewed elsewhere.10,97 As more research is performed to study the interaction of nutrition and exposures, care should be taken to consider the effects of the microbiome in these studies.
In silico experimentation and hypothesis generation for exploring nutrient-exposure relationships
Above, we have presented a workflow to use metabolomics methods to better understand the biological perturbations of the exposome and to potentially derive hypothesis for nutritional intervention strategies. However, there are many additional factors to consider in exposome studies which can potentially be available including genetic information, partial/complete mechanistic disease information, dietary patterns, demographic information, etc. These other factors may provide additional information beyond metabolomics/exposomics data to derive optimal intervention strategies. Therefore, a more intensive framework may be needed to integrate these other data sources with metabolomics/exposomics data for this purpose, and would ideally be able to incorporate data across multiple exposomics studies to more clearly identify precision intervention strategies.
One existing challenge with integrating these types of data is a lack of common standards and terminologies across exposome studies. A few research programs have leveraged ontologies (computational terminology standards) to standardize environmental health or nutrition data to alleviate these challenges. Applying ontologies, the CHEAR and HHEAR programs integrate exposure data and associated health outcome data across multiple studies and make them available to the research community.13,14 The FOODON ontology describes parts of animals, plants, and fungi bearing a food role.98 However, these programs are rare examples of computable exposure and nutrition data and have not yet been integrated with other resources. The absence of machine-readable nutrition data is a major roadblock to discovering connections between diet and the environment, and limits the field’s ability to derive nutritional intervention strategies to modulate the effects of exposures. Therefore, new approaches are needed to organize machine-readable data in a way that can be continually built upon as new data emerge.
A solution that we propose is the use of Knowledge Graphs (KGs). KGs and their underlying ontologies link information from different sources, similar to how people are linked in online social networking. Individuals who do not know each other are linked together via ‘friends of friends’ or ‘common hobbies’, etc. Similarly, a collection of partially overlapping data sets are combined into one large, interconnected data set (the KG) to answer complex questions that could not be answered before, thereby making data interpretation and hypothesis generation more comprehensive, easier, and faster. Using KGs for hypothesis generation (eg, drug repurposing) is an active area of research, with studies using a combination of manual exploration and automated link prediction.99-101 Using the standards in KG-Hub,102 improvements have been made in rare disease diagnosis within the Monarch Initiative,103 which focuses on genotype-phenotype data. However, environmental factors, including diet, are not yet available within the Monarch KG. Recently, landscape assessment of nutrition resources was performed to enhance ontologies and KGs for health interpretation of dietary risk factors as well as mechanistic discovery.104 Incorporation of knowledge from the Comparative Toxicogenomics Database (CTD)105—a manually curated database about chemical–gene/protein interactions, chemical–disease, and gene–disease relationships, the OSU/PNNL Superfund Research Program Analytics portal (SRP),106 the Micronutrient Information Center (MIC)107—a freely available, evidence-based resource on the role of nutrients and dietary factors in health and disease, among other sources, would greatly enhance computable knowledge of dietary and environmental health. Combining these data into a single, query-able graph would allow for the discovery of deep connections in the molecular processes that are affected by environmental exposures and nutrients (Figure 5). Metabolomics data, particularly from large-scale cohorts like those analyzed in HHEAR and CHEAR, can also be integrated into these KGs to better understand the link between metabolic perturbations, exposures, and other individual/population data—including dietary information—to better understand the relationship between nutrients and these other factors. Advantages of using KGs over other strategies includes: 1) semantic integration—KGs utilize ontologies and semantic relationships to integrate diverse datasets, drawing connections between different elements (eg, exposures, nutrients, health status, etc.) in a comprehensive analysis of data with enhanced visualization capabilities; 2) flexibility and scalability—KGs are able to easily accommodate new data sources, expand analysis scopes, and incorporate emerging knowledge to ensure analyses are kept up-to-date and comprehensive; 3) data exploration capabilities—KGs allow researchers to explore known and unknown deep connections for hypothesis generation, accelerating the discovery process including potential intervention strategies for testing; and 4) interoperability and collaboration—KGs promote data analyses across different research domains, bridging gaps between fields of study and enabling multidisciplinary research which would aid in a better understanding of the exposome and its impact on human health. A disadvantage of this approach is the necessity for unified ontology systems to facilitate data merging but as mentioned above, the exposomics field is moving in this direction which will help in overcoming this barrier. The use of KGs for hypothesis generation has great potential to accelerate the scientific process, but only if we develop repeatable, reliable methods for building KGs and using evidence to iteratively improve them. This has rarely been done in the past, and never by using dietary or environmental data. This approach would greatly increase the value of metabolomics data in exposome studies, allowing for deeper connections to be made between metabolic perturbations of exposures and more comprehensive hypothesis generation for intervention strategies.
ETL pipeline to transform source data into a single, query-able Knowledge Graph. Existing Monarch Initiative Infrastructure (Koza and KGX) will be leveraged to create a pipeline to map knowledge sources to ontologies and Biolink to enable integration of heterogeneous data across scale and modality (PEH-KG) in a structure that allows for reasoning (eg, link prediction) and researcher questions through a Neo4J graph database query endpoint. AGR Alliance of Genome Resources,111 BioThings,112 FDC Food Data Central,113 Reactome,114 Rhea,115 ICEES Integrated Clinical and Environmental Exposures Service,116 PubChem,117 Tox21.118
Conclusion
In conclusion, the exposome describes the totality of exposures individuals face across their lifetime and has complex effects on human health. Much of the effects of the exposome remain unknown as the majority of studies have focused on one or a small number of exposures. Regardless, the link between environmental chemical exposure and human health is clear, and interventions are needed to reduce or prevent exposure-related adverse health outcomes. High information workflows using metabolomics technologies can capture the metabolic perturbations caused by exposures and reveal novel mechanisms by which these compounds modify health outcomes. Revealing how these imbalances are related to our essential nutrients can point to nutritional targets to inform the development of intervention strategies. Herein, a concept of nutritional pharmacology/toxicology has been proposed to target these metabolic perturbations as a method to combat the adverse effects or enhance the beneficial effects of these compounds. This method not only has potential to be applied generally across populations, but it may also be used as a precision public health approach to target subpopulations based on race, sex, social determinants of health, responder status, genetic variants that predispose one to the exposure, or towards epigenetic perturbations that have resulted from prior exposures. Similar to drug development paradigms that use combination cocktails of drugs to hit multiple therapeutic targets,108-110 it is likely the field of nutrition will evolve to evaluate an individual’s exposure profile, metabotype, and genotype to derive combination cocktails of nutrients to return an individuals’ metabolic imbalance to their normal state. To fully study these factors, better experimental models are needed for exposome research and related interventions to make translatable discoveries. Also, much of the current work regarding nutrient-toxin and nutrient-drug interactions (including those presented above) focuses on generic targets such as oxidative stress, inflammation, or endothelial function as targets of nutrients rather than specific metabolites or metabolic processes. Therefore, more research is needed to better understand how nutrients target specific metabolic processes to modify an individual’s response to toxins or pharmacological agents. Additionally, more research is needed where exposures (both toxins and drugs) are studied as mixtures rather than single agents to reflect exposure profiles seen in populations, ideally in the presence or absence of select nutrients to better understand nutrient-environment interactions. Lastly, a better understanding of the clinical effectiveness of altering metabolism in these approaches is needed to better understand the utility of these approaches. Some barriers—including variability in metabolomics assays, lack of standardized data analytical approaches for mixtures, inadequate sample sizes, potential unintended consequences of food/supplement interventions—stand in the way of the widespread adoption of this concept. However, ongoing efforts in the metabolomics community for standardization practices and large metabolomics/exposomics consortium efforts such as HHEAR and CHEAR will aid in alleviating these issues. Advancements in these areas will aid in the development of nutritional interventions for public health and precision interventions to modify the effects of exposures.
Acknowledgments
The authors would like to thank Julie McMurry for the design of Figure 4, Melissa Haendel and Emily Ho for insightful conversations, Yuxia Cui and Claudia Figueroa-Romero for proofreading, and Pothur Srinivas for conceptual discussions and input concerning the interaction between the microbiome and the exposome. This work was supported by the NIEHS via a contract through ICF International HHSN273201600015U (AET), City University of New York grant number PSC-CUNY TRADA-54-101 (GAS), American Cancer Society (ACS) grant number DICRIDG-21-071-01 (A.R.), and US Department of Energy grant number DE-EM0005266 (A.R.).This study was also supported by the NIEHS through the Human Health Exposure Analysis Resource (HHEAR) program through grant 5U2CES030857 (S.S., B.R.) and by NCI through the Metabolomics and Clinical Assay Center of the Nutrition for Precision Health (NPH) program through grant 5U24CA268153 (S.S., B.R.).The authors would also like to acknowledge the HHEAR Webinar Series where this concept, among others, was discussed to form ideas and writing groups across the consortium.
Author contributions
Blake Rushing (Conceptualization [equal], Visualization [lead], Writing—original draft [lead], Writing—review and editing [lead]), Anne Thessen (Visualization [equal], Writing—review and editing [equal]) Ghada Soliman (Writing—review and editing [equal]), Aramandla Ramesh (Writing—review and editing [equal]), and Susan Sumner (Conceptualization [lead], Funding acquisition [equal], Supervision [equal], Writing—original draft [equal], Writing—review and editing [equal])
Data availability
No original data was produced for this manuscript.
Conflict of interest statement
None declared.
References
1 WildCP. Complementing the genome with an “exposome”: the outstanding challenge of environmental exposure measurement in molecular epidemiology. Cancer Epidemiol Biomarkers Prev. 2005; 14(8):1847-1850. doi: http://doi.org/10.1158/1055-9965.EPI-05-0456
2 VrijheidM. The exposome: a new paradigm to study the impact of environment on health. Thorax. 2014; 69(9):876-878. doi: http://doi.org/10.1136/thoraxjnl-2013-204949
3 WildCP. The exposome: from concept to utility. Int J Epidemiol. 2012; 41(1):24-32. doi: http://doi.org/10.1093/ije/dyr236
4 JuarezPD, Matthews-JuarezP, HoodDB, et al The public health exposome: a population-based, exposure science approach to health disparities research. Int J Environ Res Public Health. 2014; 11(12):12866-12895. doi: http://doi.org/10.3390/ijerph111212866
5 BoelsterliUA. Mechanistic Toxicology: The Molecular Basis of How Chemicals Disrupt Biological Targets. CRC Press; 2002.
6 TainioM, Jovanovic AndersenZ, NieuwenhuijsenMJ, et al Air pollution, physical activity and health: a mapping review of the evidence. Environ Int. 2021; 147:105954. doi: http://doi.org/10.1016/j.envint.2020.105954
7 HeindelJJ. The developmental basis of disease: update on environmental exposures and animal models. Basic Clin Pharmacol Toxicol. 2019; 125(S3):5-13. doi: http://doi.org/10.1111/bcpt.13118
8 DesalegnA, BoppS, AsturiolD, LamonL, WorthA, PainiA. Role of physiologically based kinetic modelling in addressing environmental chemical mixtures—a review. Comput Toxicol. 2019; 10:158-168. doi: http://doi.org/10.1016/j.comtox.2018.09.001
9 ChenL, LuoK, EtzelR, ZhangX, TianY, ZhangJ. Co-exposure to environmental endocrine disruptors in the US population. Environ Sci Pollut Res Int. 2019; 26(8):7665-7676.
10 TuP, ChiL, BodnarW, et al Gut microbiome toxicity: connecting the environment and gut microbiome-associated diseases. Toxics 2020; 8(1):19. doi: http://doi.org/10.3390/toxics8010019
11 BoyceWT, LevittP, MartinezFD, McEwenBS, ShonkoffJP. Genes, environments, and time: the biology of adversity and resilience. Pediatrics 2021; 147(2):1-12. doi: http://doi.org/10.1542/peds.2020-1651
12 ÖtlesS, SenturkA. Food and drug interactions: a general review. Acta Sci Pol Technol Aliment. 2014; 13(1):89-102.
13 BalshawD, CollmanG, GrayK, ThompsonC. The children’s health exposure analysis resource (CHEAR): enabling research into the environmental influences on children’s health outcomes. Curr Opin Pediatr. 2017; 29(3):385-389. doi: http://doi.org/10.1097/MOP.0000000000000491.
14 VietSM, FalmanJC, MerrillLS, et al Human health exposure analysis resource (HHEAR): a model for incorporating the exposome into health studies. Int J Hyg Environ Health. 2021; 235:113768. doi: http://doi.org/10.1016/j.ijheh.2021.113768
15 KingsleySL, WalkerDI, CalafatAM, et al Metabolomics of childhood exposure to perfluoroalkyl substances: a cross-sectional study. Metabolomics. 2019; 15(7):95. doi: http://doi.org/10.1007/s11306-019-1560-z.Metabolomics
16 AldereteTL, JinR, WalkerDI, et al Perfluoroalkyl substances, metabolomic profiling, and alterations in glucose homeostasis among overweight and obese Hispanic children: a proof-of-concept analysis. Environ Int. 2019; 126:445-453. doi: http://doi.org/10.1016/j.envint.2019.02.047
17 ChenZ, YangT, WalkerDI, et al Dysregulated lipid and fatty acid metabolism link perfluoroalkyl substances exposure and impaired glucose metabolism in young adults. Environ Int. 2020; 145:106091. doi: http://doi.org/10.1016/j.envint.2020.106091
18 BankerM, PuttabyatappaM, O'DayP, et al Association of maternal-neonatal steroids with early pregnancy endocrine disrupting chemicals and pregnancy outcomes. J Clin Endocrinol Metab. 2021; 106(3):665-687. doi: http://doi.org/10.1210/clinem/dgaa909
19 LiY-Y, GhanbariR, PathmasiriW, et al Untargeted metabolomics: biochemical perturbations in golestan cohort study opium users inform intervention strategies. Front Nutr. 2020; 7:584585. doi: http://doi.org/10.3389/fnut.2020.584585
20 KohlmeierM. Nutrient Metabolism: Structures, Functions, and Genes; Academic Press, 2015.
21 SpillerG. Nutritional pharmacology. Am J Clin Nutr. 1980; 33(7):1716.
22 RushingBR, SelimMI. Aflatoxin B1: a review on metabolism, toxicity, occurrence in food, occupational exposure, and detoxification methods. Food Chem Toxicol. 2019; 124:81-100. doi: http://doi.org/10.1016/j.fct.2018.11.047
23 HennigB, EttingerAS, JandacekRJ, et al Using nutrition for intervention and prevention against environmental chemical toxicity and associated diseases. Environ Health Perspect. 2007; 115(4):493-495. doi: http://doi.org/10.1289/ehp.9549
24 LewandowskaU, GorlachS, OwczarekK, HrabecE, SzewczykK. Synergistic interactions between anticancer chemotherapeutics and phenolic compounds and anticancer synergy between polyphenols. Postepy Hig Med Dosw. 2014; 68(4468):528-540. doi: http://doi.org/10.5604/17322693.1102278
25 WonCS, OberliesNH, PaineMF. Mechanisms underlying food-drug interactions: inhibition of intestinal metabolism and transport. Pharmacol Ther. 2012; 136(2):186-201. doi: http://doi.org/10.1016/j.pharmthera.2012.08.001
26 HoffmanJ, HennigB. Protective influence of healthful nutrition on mechanisms of environmental pollutant toxicity and disease risks. Ann N Y Acad Sci. 2017; 1398(1):99-107. doi: http://doi.org/10.1111/nyas.13365.Protective
27 HennigB, OrmsbeeL, McClainCJ, et al Nutrition can modulate the toxicity of environmental pollutants: implications in risk assessment and human health. Environ Health Perspect. 2012; 120(6):771-774. doi: http://doi.org/10.1289/ehp.1104712
28 GreenbergJ, BellS, GuanY, YuY. Folic acid supplementation and pregnancy—more than just neural tube defect prevention. Reprod Endocrinol. 2017; 4(34):57-63. doi: http://doi.org/10.18370/2309-4117.2017.34.57-63
29 GaskinsAJ, Mínguez-AlarcónL, FongKC, et al Supplemental folate and the relationship between traffic-related air pollution and livebirth among women undergoing assisted reproduction. Am J Epidemiol. 2019; 188(9):1595-1604. doi: http://doi.org/10.1093/aje/kwz151
30 WesselinkAK, HatchEE, WiseLA. Invited commentary: interaction between diet and chemical exposures. Am J Epidemiol. 2019; 188(9):1605-1607. doi: http://doi.org/10.1093/aje/kwz152
31 OrmondG, NieuwenhuijsenMJ, NelsonP, et al Endocrine disruptors in the workplace, hair spray, folate supplementation, and risk of hypospadias: case-control study. Environ Health Perspect. 2009; 117(2):303-307. doi: http://doi.org/10.1289/ehp.11933
32 AbuawadA, BozackA, Navas-AcienA, et al The folic acid and creatine trial: treatment effects on arsenic metabolite concentrations in blood. Env Heal Perspect. 2023; 131(3). doi: http://doi.org/10.1289/isee.2022.p-0722
33 PetersBA, HallMN, LiuX, et al Folic acid and creatine as therapeutic approaches to lower blood arsenic: a randomized controlled trial. Environ Health Perspect. 2015; 123(12):1294-1301. doi: http://doi.org/10.1289/ehp.1409396
34 Sokoła-WysoczańskaE, WysoczańskiT, WagnerJ, et al Polyunsaturated fatty acids and their potential therapeutic role in cardiovascular system disorders—a review. Nutrients 2018; 10(10):1-21. doi: http://doi.org/10.3390/nu10101561
35 SchwingshacklL, MorzeJ, HoffmannG. Mediterranean diet and health status: active ingredients and pharmacological mechanisms. Br J Pharmacol. 2020; 177(6):1241-1257. doi: http://doi.org/10.1111/bph.14778
36 BrighamEP, WooH, McCormackM, et al Omega-3 and omega-6 intake modifies asthma severity and response to indoor air pollution in children. Am J Respir Crit Care Med. 2019; 199(12):1478-1486. doi: http://doi.org/10.1164/rccm.201808-1474OC
37 ChenH, ZhangS, ShenW, et al Omega-3 fatty acids attenuate cardiovascular effects of short-term exposure to ambient air pollution. Part Fibre Toxicol. 2022; 19(1):12. doi: http://doi.org/10.1186/s12989-022-00451-4
38 LimCC, HayesRB, AhnJ, et al Mediterranean diet and the association between air pollution and cardiovascular disease mortality risk. Circulation. 2019; 139(15):1766-1775. doi: http://doi.org/10.1161/CIRCULATIONAHA.118.035742.Mediterranean
39 SeifriedHE, AndersonDE, FisherEI, MilnerJA. A review of the interaction among dietary antioxidants and reactive oxygen species. J Nutr Biochem. 2007; 18(9):567-579. doi: http://doi.org/10.1016/j.jnutbio.2006.10.007
40 HofeCR, FengL, ZephyrD, StrombergAJ, HennigB, GaetkeLM. Fruit and vegetable intake, as reflected by serum carotenoid concentrations, predicts reduced probability of polychlorinated biphenyl-associated risk for type 2 diabetes: national health and nutrition examination survey 2003-2004. Nutr Res. 2014; 34(4):285-293. doi: http://doi.org/10.1016/j.nutres.2014.02.001
41 NobileV, SchianoI, PeralA, GiardinaS, SpartàE, CaturlaN. Antioxidant and reduced skin-ageing effects of a polyphenol-enriched dietary supplement in response to air pollution: a randomized, double-blind, placebo-controlled study. Food Nutr Res 2021; 65:1-19. doi: http://doi.org/10.29219/fnr.v65.5619
42 PéterS, HolguinF, WoodLG, et al Nutritional solutions to reduce risks of negative health impacts of air pollution. Nutrients 2015; 7(12):10398-10416. doi: http://doi.org/10.3390/nu7125539
43 HeitzerT, JustH, MunzelT. Antioxidant Vitamin C improves endothelial dysfunction in chronic smokers. Circulation 1996; 94(1):6-9.
44 MiddhaP, WeinsteinSJ, MännistöS, AlbanesD, MondulAM. β-Carotene supplementation and lung cancer incidence in the alpha-tocopherol, beta-carotene cancer prevention study: the role of tar and nicotine. Nicotine Tob Res. 2019; 21(8):1045-1050. doi: http://doi.org/10.1093/ntr/nty115
45 BoullataJI, ArmentiVT. Handbook of Drug-Nutrient Interactions. Vol 26; Humana Press, 2004. doi: http://doi.org/10.1177/0115426504019006648
46 LiuQS, DengR, FanY, et al Low dose of caffeine enhances the efficacy of antidepressants in major depressive disorder and the underlying neural substrates. Mol Nutr Food Res 2017; 61(8):1-11. doi: http://doi.org/10.1002/mnfr.201600910
47 GodfreyPS, TooneBK, CarneyMW, et al Enhancement of recovery from psychiatric illness by methylfolate. Lancet. 1990; 336(8712):392-395.
48 AlpertJE, MischoulonD, RubensteinGEF, BottonariK, NierenbergAA, FavaM. Folinic acid (leucovorin) as an adjunctive treatment for SSRI-refractory depression. Ann Clin Psychiatry. 2002; 14(1) http://dx.doi.org/10.1016/j.jaci.2012.05.050http://dx.doi.org/10.1016/j.jaci.2012.05.050
49 CoppenA, ChaudhryS, SwadeC. Folic acid enhances lithium prophylaxis. J Affect Disord. 1986; 10(1):9-13. doi: http://doi.org/10.1016/0165-0327(86)90043-1
50 BeerTM, EilersKM, GarzottoM, EgorinMJ, LoweBA, HennerWD. Weekly high-dose calcitriol and docetaxel in advanced prostate cancer. J Clin Oncol. 2003; 21(1):123-128.
51 GuoY, XiongBH, ZhangT, ChengY, MaL. XELOX vs. FOLFOX in metastatic colorectal cancer: an updated meta-analysis. Cancer Invest. 2016; 34(2):94-104. doi: http://doi.org/10.3109/07357907.2015.1104689
52 MachoverD. A comprehensive review of 5-fluorouracil and leucovorin in patients with metastatic colorectal carcinoma. Cancer 1997; 80(7):1179-1187. doi: http://doi.org/10.1002/(SICI)1097-0142(19971001)80:7<1179::AID-CNCR1>3.0.CO;2-G
53 KarczmarewiczE, Czekuc-KryskiewiczE, PludowskiP. Effect of vitamin d status on pharmacological treatment efficiency impact on cost-effective management in medicine. Dermatoendocrinol. 2013; 5(2):299-304. doi: http://doi.org/10.4161/derm.25279
54 LisoM, SilaA, VernaG, et al Nutritional regimes enriched with antioxidants as an efficient adjuvant for IBD patients under infliximab administration, a pilot study. Antioxidants. 2022; 11(1):138. doi: http://doi.org/10.3390/antiox11010138
55 JohnsonC, IvanisevicJ, SiuzdakG. Metabolomics: beyond biomarkers and towards mechanisms. Nat Rev Mol Cell Biol. 2016; 17(7):451-459.
56 BrennanL. Metabolomics in nutrition research: current status and perspectives. Biochem Soc Trans. 2013; 41(2):670-673. doi: http://doi.org/10.1042/BST20120350
57 MussapM, LoddoC, FanniC, FanosV. Metabolomics in pharmacology—a delve into the novel field of pharmacometabolomics. Expert Rev Clin Pharmacol. 2020; 13(2):115-134.
58 RamirezT, DaneshianM, KampH, et al Metabolomics in toxicology and preclinical research. ALTEX. 2013; 30(2):209-225. doi: http://doi.org/10.14573/altex.2013.2.209
59 Da SilvaRR, DorresteinPC, QuinnRA. Illuminating the dark matter in metabolomics. Proc Natl Acad Sci USA. 2015; 112(41):12549-12550. doi: http://doi.org/10.1073/pnas.1516878112
60 MisraBB. New software tools, databases, and resources in metabolomics: updates from 2020. Metabolomics. 2021; 17(5):49. doi: http://doi.org/10.1007/s11306-021-01796-1
61 DavidA, ChakerJ, PriceEJ, et al Towards a comprehensive characterisation of the human internal chemical exposome: challenges and perspectives. Environ Int. 2021; 156:106630. doi: http://doi.org/10.1016/j.envint.2021.106630
62 RushingBR, McRitchieS, ArbeevaL, et al Fecal metabolomics reveals products of dysregulated proteolysis and altered microbial metabolism in obesity-related osteoarthritis. Osteoarthritis Cartilage. 2022; 30(1):81-91. doi: http://doi.org/10.1016/j.joca.2021.10.006
63 RauhVA, LandriganPJ, ClaudioL. Housing and health: intersection of poverty and environmental exposures. Ann N Y Acad Sci. 2008; 1136:276-288. doi: http://doi.org/10.1196/annals.1425.032
64 Payne-SturgesD, GeeGC. National environmental health measures for minority and low-income populations: tracking social disparities in environmental health. Environ Res. 2006; 102(2):154-171. doi: http://doi.org/10.1016/j.envres.2006.05.014
65 GeeGC, Payne-SturgesDC. Environmental health disparities: a framework integrating psychosocial and environmental concepts. Environ Health Perspect. 2004; 112(17):1645-1653. doi: http://doi.org/10.1289/ehp.7074
66 HunterDJ. Gene-environment interactions in human diseases. Nat Rev Genet. 2005; 6(4):287-298. doi: http://doi.org/10.1038/nrg1578
67 VineisP, MarinelliD, AutrupH, et al Current smoking, occupation, N-acetyltransferase-2 and bladder cancer: a pooled analysis of genotype-based studies. Cancer Epidemiol Biomarkers Prev. 2001; 10(12):1249-1252.
68 SalamMT, LinPC, AvolEL, GaudermanWJ, GillilandFD. Microsomal epoxide hydrolase, glutathione S-transferase P1, traffic and childhood asthma. Thorax. 2007; 62(12):1050-1057. doi: http://doi.org/10.1136/thx.2007.080127
69 PaulKC, SinsheimerJS, CockburnM, BronsteinJM, BordelonY, RitzB. Organophosphate pesticides and PON1 L55M in Parkinson’s disease progression. Environ Int. 2017; 107:75-81. doi: http://doi.org/10.1016/j.envint.2017.06.018.Organophosphate
70 MerinoJ, JablonskiKA, MercaderJM, et al Interaction between type 2 diabetes prevention strategies and genetic determinants of coronary artery disease on cardiometabolic risk factors. Diabetes 2020; 69(1):112-120. doi: http://doi.org/10.2337/db19-0097
71 LacalI, VenturaR. Epigenetic inheritance: concepts, mechanisms and perspectives. Front Mol Neurosci. 2018; 11:292. doi: http://doi.org/10.3389/fnmol.2018.00292
72 HoSM, JohnsonA, TaraporeP, JanakiramV, ZhangX, LeungYK. Environmental epigenetics and its implication on disease risk and health outcomes. ILAR J. 2012; 53(3-4):289-305. doi: http://doi.org/10.1093/ilar.53.3-4.289
73 BakkerH, JaddoeVWV. Cardiovascular and metabolic influences of fetal smoke exposure. Eur J Epidemiol. 2011; 26(10):763-770. doi: http://doi.org/10.1007/s10654-011-9621-2
74 RumrichIK, VilukselaM, VähäkangasK, GisslerM, SurcelH-M, HänninenO. Maternal smoking and the risk of cancer in early life—a meta-analysis. PLoS One. 2016; 11(11):e0165040. doi: http://doi.org/10.1371/journal.pone.0165040
75 RogersJM. Smoking and pregnancy: epigenetics and developmental origins of the metabolic syndrome. Birth Defects Res. 2019; 111(17):1259-1269. doi: http://doi.org/10.1002/bdr2.1550
76 BretonCV, ByunHM, WentenM, PanF, YangA, GillilandFD. Prenatal tobacco smoke exposure affects global and gene-specific DNA methylation. Am J Respir Crit Care Med. 2009; 180(5):462-467. doi: http://doi.org/10.1164/rccm.200901-0135OC
77 YoungJ, CaiL, StatesJC. Impact of prenatal arsenic exposure on chronic adult diseases. Syst Biol Reprod Med. 2018; 64(6):469-483. doi: http://doi.org/10.1080/19396368.2018.1480076.Impact
78 VaisermanA. Epidemiologic evidence for association between adverse environmental exposures in early life and epigenetic variation: a potential link to disease susceptibility? Clin Epigenetics. 2015; 7(1):96. doi: http://doi.org/10.1186/s13148-015-0130-0
79 TiffonC. The impact of nutrition and environmental epigenetics on human health and disease. Int J Mol Sci. 2018; 19(11):3425. doi: http://doi.org/10.3390/ijms19113425
80 FokJS, KolkhirP, ChurchMK, MaurerM. Predictors of treatment response in chronic spontaneous urticaria. Allergy Eur J Allergy Clin Immunol. 2021; 76(10):2965-2981. doi: http://doi.org/10.1111/all.14757
81 DellonES, GuptaSK. A conceptual approach to understanding treatment response in eosinophilic esophagitis. Clin Gastroenterol Hepatol. 2019; 17(11):2149-2160. doi: http://doi.org/10.1016/j.cgh.2019.01.030
82 MacherJP, CrocqMA. Treatment goals: response and nonresponse. Dialogues Clin Neurosci. 2004; 6(1):83-91. doi: http://doi.org/10.31887/dcns.2004.6.1/jpmacher2
83 VarkerT, KartalD, WatsonL, et al Defining response and nonresponse to posttraumatic stress disorder treatments: a systematic review. Clin Psychol Sci Pract. 2020; 27(4):1-36. doi: http://doi.org/10.1111/cpsp.12355
84 PennyHA, BaggusEMR, RejA, SnowdenJA, SandersDS. Non-responsive coeliac disease: a comprehensive review from the NHS England national centre for refractory coeliac disease. Nutrients. 2020; 12(1):216. doi: http://doi.org/10.3390/nu12010216
85 PrihantonoFM. Breast cancer resistance to chemotherapy: When should we suspect it and how can we prevent it? Ann Med Surg (Lond). 2021; 70(May):102793. doi: http://doi.org/10.1016/j.amsu.2021.102793
86 GhiniV, LaeraL, FantechiB, et al Metabolomics to assess response to immune checkpoint inhibitors in patients with non-small-cell lung cancer. Cancers (Basel). 2020; 12(12):1-16. doi: http://doi.org/10.3390/cancers12123574
87 WeiS, LiuL, ZhangJ, et al Metabolomics approach for predicting response to neoadjuvant chemotherapy for breast cancer. Mol Oncol. 2013; 7(3):297-307. doi: http://doi.org/10.1016/j.molonc.2012.10.003
88 ParkJE, JeongGH, LeeIK, et al A pharmacometabolomic approach to predict response to metformin in early-phase type 2 diabetes mellitus patients. Molecules. 2018; 23(7):1579. doi: http://doi.org/10.3390/molecules23071579
89 Kaddurah-DaoukR, BoyleSH, MatsonW, et al Pretreatment metabotype as a predictor of response to sertraline or placebo in depressed outpatients: a proof of concept. Transl Psychiatry. 2011; 1(7):e26–7. doi: http://doi.org/10.1038/tp.2011.22
90 HintonDJ, VázquezMS, GeskeJR, et al Metabolomics biomarkers to predict acamprosate treatment response in alcohol-dependent subjects. Sci Rep. 2017; 7(1):2496-2498. doi: http://doi.org/10.1038/s41598-017-02442-4
91 MorrisonDJ, PrestonT. Formation of short chain fatty acids by the gut microbiota and their impact on human metabolism. Gut Microbes. 2016; 7(3):189-200. doi: http://doi.org/10.1080/19490976.2015.1134082
92 SitaramanR. Phospholipid catabolism by gut microbiota and the risk of cardiovascular disease. J Med Microbiol. 2013; 62(Pt 6):948-950. doi: http://doi.org/10.1099/jmm.0.053587-0
93 PostlerTS, GhoshS. Understanding the holobiont: how microbial metabolites affect human health and shape the immune system. Cell Metab. 2017; 26(1):110-130. doi: http://doi.org/10.1016/j.cmet.2017.05.008
94 MartinezKB, LeoneV, ChangEB. Microbial metabolites in health and disease: navigating the unknown in search of function. J Biol Chem. 2017; 292(21):8553-8559. doi: http://doi.org/10.1074/jbc.R116.752899
95 LeemingER, JohnsonAJ, SpectorTD, RoyCIL. Effect of diet on the gut microbiota: Rethinking intervention duration. Nutrients. 2019; 11(12):1-28. doi: http://doi.org/10.3390/nu11122862
96 KeW, SabaJA, YaoCH, et al Dietary serine-microbiota interaction enhances chemotherapeutic toxicity without altering drug conversion. Nat Commun. 2020; 11(1):2587. doi: http://doi.org/10.1038/s41467-020-16220-w
97 WeersmaRK, ZhernakovaA, FuJ. Interaction between drugs and the gut microbiome. Gut 2020; 69(8):1510-1519. doi: http://doi.org/10.1136/gutjnl-2019-320204
98 DooleyDM, GriffithsEJ, GosalGS, et al Food on: A harmonized food ontology to increase global food traceability, quality control and data integration. NPJ Sci Food. 2018; 2(1):23. doi: http://doi.org/10.1038/s41538-018-0032-6
99 MacLeanF. Knowledge graphs and their applications in drug discovery. Expert Opin Drug Discov. 2021; 16(9):1057-1069.
100 TiddiI, BallietD, TeijeAT. Fostering scientific meta-analyses with knowledge graphs: a case-study. In: The Semantic Web. Springer International Publishing; 2020:287–303. doi: http://doi.org/10.1007/978-3-030-49461-2
101 OttS, Barbosa-SilvaA, SamwaldM. LinkExplorer: predicting, explaining and exploring links in large biomedical knowledge graphs. Bioinformatics. 2022; 38(8):2371-2373. doi: http://doi.org/10.1093/bioinformatics/btac068
102 ReeseJT, UnniD, CallahanTJ, et al KG-COVID-19: a framework to produce customized knowledge graphs for COVID-19 response. Patterns. 2021; 2(1):100155.doi: http://doi.org/10.1016/j.patter.2020.100155
103 ShefchekKA, HarrisNL, GarganoM, et al The monarch initiative in 2019: an integrative data and analytic platform connecting phenotypes to genotypes across species. Nucleic Acids Res. 2020; 48(D1):D704-D715. doi: http://doi.org/10.1093/nar/gkz997
104 ChanL, VasilevskyN, ThessenA, McMurryJ, HaendelM. The landscape of nutri-informatics: a review of current resources and challenges for integrative nutrition research. Database. 2021; 2021:1-20. doi: http://doi.org/10.1093/database/baab003
105 DavisAP, GrondinCJ, JohnsonRJ, et al Comparative toxicogenomics database (CTD): update 2021. Nucleic Acids Res. 2021; 49(D1):D1138-D1143. doi: http://doi.org/10.1093/nar/gkaa891
106 OSU/PNNL Superfund Research Program Analytics Portal. Accessed December 20, 2022. srp.pnnl.gov/portal
107 Micronutrient Information Center. Accessed December 20, 2022. lpi.oregonstate.edu/mic
108 GiladY, GellermanG, LonardDM, O'MalleyBW. Drug combination in cancer treatment—from cocktails to conjugated combinations. Cancers (Basel). 2021; 13(4):1-26. doi: http://doi.org/10.3390/cancers13040669
109 WhiteJM, SchifferJT, Bender IgnacioRA, et al Drug combinations as a first line of defense against coronaviruses and other emerging viruses. Mbio. 2021; 12(6):e0334721. doi: http://doi.org/10.1128/mbio.03347-21
110 YusufS, AmirA, JackieB, et al Combination pharmacotherapy to prevent cardiovascular disease: present status and challenges. Eur Heart J. 2014; 35(6):353-364. doi: http://doi.org/10.1093/eurheartj/eht407
111 AgapiteJ, AlbouLP, AleksanderS, et al Alliance of genome resources portal: unified model organism research platform. Nucleic Acids Res. 2020; 48(D1):D650-D658. doi: http://doi.org/10.1093/nar/gkz813
112 XinJ, AfrasiabiC, LelongS, et al Cross-linking BioThings APIs through JSON-LD to facilitate knowledge exploration. BMC Bioinformatics. 2018; 19(1):30-37. doi: http://doi.org/10.1186/s12859-018-2041-5
113 LarrickB, KretserA, McKillopK. Update on “a partnership for public health: USDA global branded food products database”. J Food Compos Anal. 2022; 105:104250. doi: http://doi.org/10.1016/j.jfca.2021.104250
114 JassalB, MatthewsL, ViteriG, et al The reactome pathway knowledgebase. Nucleic Acids Res. 2020; 48(D1):D498-D503. doi: http://doi.org/10.1093/nar/gkz1031
115 BansalP, MorgatA, AxelsenKB, et al Rhea, the reaction knowledgebase in 2022. Nucleic Acids Res. 2022; 50(D1):D693-D700. doi: http://doi.org/10.1093/nar/gkab1016
116 FechoK, PfaffE, XuH, et al A novel approach for exposing and sharing clinical data: the translator integrated clinical and environmental exposures service. J Am Med Inform Assoc. 2019; 26(10):1064-1073. doi: http://doi.org/10.1093/jamia/ocz042
117 KimS, ChenJ, ChengT, et al PubChem 2019 update: improved access to chemical data. Nucleic Acids Res. 2019; 47(D1):D1102-D1109. doi: http://doi.org/10.1093/nar/gky1033
118 ThomasR, PaulesR, SimeonovA, et al The federal Tox21 program: a strategic and operational plan for continued leadership. ALTEX. 2018; 35(2):163-168.