Impact of the Exposome on the Epigenome in Inflammatory Bowel Disease Patients and Animal Models

Inflammatory bowel diseases (IBD) are chronic inflammatory disorders of the gastrointestinal tract that encompass two main phenotypes, namely Crohn’s disease and ulcerative colitis. These conditions occur in genetically predisposed individuals in response to environmental factors. Epigenetics, acting by DNA methylation, post-translational histones modifications or by non-coding RNAs, could explain how the exposome (or all environmental influences over the life course, from conception to death) could influence the gene expression to contribute to intestinal inflammation. We performed a scoping search using Medline to identify all the elements of the exposome that may play a role in intestinal inflammation through epigenetic modifications, as well as the underlying mechanisms. The environmental factors epigenetically influencing the occurrence of intestinal inflammation are the maternal lifestyle (mainly diet, the occurrence of infection during pregnancy and smoking); breastfeeding; microbiota; diet (including a low-fiber diet, high-fat diet and deficiency in micronutrients); smoking habits, vitamin D and drugs (e.g., IBD treatments, antibiotics and probiotics). Influenced by both microbiota and diet, short-chain fatty acids are gut microbiota-derived metabolites resulting from the anaerobic fermentation of non-digestible dietary fibers, playing an epigenetically mediated role in the integrity of the epithelial barrier and in the defense against invading microorganisms. Although the impact of some environmental factors has been identified, the exposome-induced epimutations in IBD remain a largely underexplored field. How these environmental exposures induce epigenetic modifications (in terms of duration, frequency and the timing at which they occur) and how other environmental factors associated with IBD modulate epigenetics deserve to be further investigated.


Introduction
Inflammatory bowel diseases (IBD) are chronic relapsing-remitting inflammatory disorders of the gastrointestinal tract encompassing two main phenotypes: Crohn's disease (CD) and ulcerative colitis (UC). The pathogenesis of IBD is not fully understood to date, but the most commonly accepted hypothesis is an inappropriate gut mucosal immune response towards the constituents of the gut microbiota, which cross an impaired epithelial barrier, in genetically predisposed individuals and under the influence of environmental factors [1]. Epidemiological studies (such as those carried out on monozygotic twins [2] and immigrants [3]), as well as the increase over time of the CD and UC incidence and prevalence (while the human gene pool is the same as before) [4], are all arguments that emphasize the importance of environmental factors in the occurrence of these inflammatory diseases. Epigenetics is a branch of life science that studies mechanisms regulating DNA-dependent processes (e.g., transcription, replication, recombination, repair, etc.) without primarily involving the nucleotide sequence of the DNA but, rather, the structure of how DNA is packed in the cell nucleus (chromatin structure), which can be inherited by daughter cells after cell division. Epigenetic mechanisms, including DNA methylation, post-translational histones modifications and non-coding ribonucleic acids (ncRNAs) [5] , [6], regulating gene expression provide plausible explanations for the influence of the environment on gene expression profiles that favor intestinal inflammation [7]. Supporting this line of ideas, DNA methylation profiles observed in older monozygous twins with different environmental histories shows that epigenetic imprinting occurs mainly during crucial periods of development, whereas epigenomic changes can also occur day after day and accumulate over time in response to the exposome [8][9][10].
The term exposome has been proposed to encompass all environmental influences over the life course, from conception to death, that may influence disease emergence and clinical outcomes [11,12]. External environmental factors influencing the occurrence of IBD include the maternal lifestyle and in utero events [13], breastfeeding [14], diet [7], smoking habits [15,16], drugs [16][17][18], physical activity [16], stress [19], appendicectomy [16], vitamin D/UV exposure [16], infections [20] and hygiene [21]. While it is possible that these different factors directly induce epigenetic changes in the host, it is also possible that they influence the microbiome, an internal component of the exposome, and contribute to the occurrence of IBD through the exposome-microbiome-epigenome axis [22].
The impact of exposomes on the epigenetics in IBD has been poorly studied and is probably underestimated. This review aims to identify all the elements of the exposome that may play a role in intestinal inflammation through epigenetic modifications, as well as the underlying mechanisms that may contribute to IBD pathophysiology.

Epigenetics in IBD
Epigenetic mechanisms of gene expressions are involved in the intestinal epithelium homeostasis and in the development and differentiation of the immune cells, as well as in the modulation of responses generated by the immune system to defend against potential pathogens [23]. These epigenetic changes are reversible [24]. The genomic DNA in the eukaryotic cell nucleus is organized into chromatin. Chromatin consists of nucleic acids (genomic DNA and different types of RNAs); histone proteins (H2A, H2B, H3, H4 and H1) and non-histone chromatin-associated proteins [5,25,26]. Nucleosomes constitute the functional and structural units of chromatin. A nucleosome is built by around 146 bp of genomic DNA surrounding a histone octamer, which consists of two H2A-H2B dimers and one (H3-H4)2 tetramer [27,28]. In other words, chromatin is the physiological template for all DNA-dependent biological processes, including transcription. This fact increases the complexity of transcription regulation, since it implies that the chromatin structure has to be dynamic to grant or block access of transcription regulators to their respective binding elements on the DNA and to the transcription machinery to the genomic information in the nucleotide sequence. The epigenetic mechanisms of transcriptional regulation involve DNA methylation, histone modifications, nucleosome remodeling, interaction with the nuclear matrix and regulation via long non-coding RNAs (lncRNAs) and micro RNAs (miR) [26,[29][30][31]. These mechanisms of transcription regulation establish cell-specific, heritable patterns of differential gene expression and silencing from the same genome and allow the cells to change these gene expression signatures in response to stimuli, such as changing conditions due to their environment [32,33].
Another epigenetic mechanism of transcriptional regulation involves post-translational modifications of histone proteins (further referred to as histone modifications). Histone proteins (H1, H2A, H2B H3 and H4) are relatively small and basic proteins that are abundant in the cell nucleus and are an essential part of the nucleosome, as described above. Due to structural characteristics of the nucleosome, histone proteins can undergo post-translational modifications at their N-terminal tails, which include acetylation, methylation, phosphorylation, ubiquitination and sumoylation, among others [53][54][55][56]. While DNA methylation is relatively stable in somatic cells, histone modifications are more diverse and dynamic, changing rapidly during the course of the cell cycle [6,30,53,54]. Acetylation at specific amino acids of histones (e.g., histone 3 lysine 9 acetylation; H3K9Ac) is generally associated with active chromatin and is mediated by histone acetyltransferases (HAT) and removed by histone deacetylases (HDAC). Histone methylation also occurs at specific amino acids of histone proteins and can be associated with both the repression (e.g., H3 lysine 27 trimethylation; H3K27me3) and activation (e.g., H3 lysine 4 trimethylation; H3K4me3) of gene expressions. There is a variety of enzymes mediating histone methylation (histone methyltransferases; HMT) and histone demethylation [57,58]. Similarly, the reactions leading to other histone modifications are catalyzed by a broad spectrum of enzymes in a regulated manner. Several environmental agents induce changes in histone modifications, thereby leading to changes in gene expression signatures.
In addition to these mechanisms, epigenetic regulation can also involve ncRNA, which are RNAs not translated into proteins, including miRs and lncRNAs. If miRs have a length of 18-25 nucleotides, lncRNAs are over 200 bases long [59]. These nucleic acid molecules can regulate gene expressions by interfering with messenger RNA (mRNA) translations by degrading them or through interactions with protein complexes involved in the regulation of gene expression [59,60]. The ncRNAs are differentially expressed between the control and IBD subjects, and there is also a difference in expression between CD and UC patients [61][62][63]. In IBD, miRs are involved in the regulation of the intestinal mucosal barrier, T-cell differentiation, the Th17 signaling pathway and autophagy [63]. In UC patients, miR-21, miR-16 and let-7 expressions are significantly increased in inflamed mucosa, while miR-192, miR-375 and miR-422b expressions are significantly reduced [61]. In CD patients, miR-23b, miR-106 and miR-191 are significantly increased in the inflamed mucosa, while miR-19b and miR-629 expressions are significantly decreased [61].
All these epigenetic mechanisms contribute to the development, progression and maintenance of IBD. They are usually triggered by a range of environmental factors. Some authors have mentioned three critical periods during which the environment can favor the onset of the disease: (1) during the prenatal period (in response to the maternal lifestyle), (2) in the early postnatal period (during gut microbiota colonization) and (3) just before the disease onset [64]. This review aims to study the impact of the exposome on the epigenome in IBD.

Methods
To identify exposome elements that could impact the epigenetics of IBD, we performed a scoping search using Medline. We used the following Medical Subject Heading (MeSH) terms ('epigenetics' OR 'epigenomics' OR 'DNA methylation' OR 'histone(s)' OR 'short noncoding RNA' OR 'long noncoding RNAs' OR 'microRNA' OR 'miR' OR "miRNA") AND ("Inflammatory bowel disease" OR "IBD" OR "intestinal inflammation" OR "Crohn's disease" OR "ulcerative colitis" OR "colitis"). Secondary references of the retrieved articles were reviewed to identify publications not captured by the electronic search. We excluded articles not written in English and those related to colitis-associated cancer.

Parental Exposition
Accumulating evidence has pointed out that in utero environmental exposure can influence the epigenetic programming of the offspring and have an impact on its fate, conditioning its health status or, on the contrary, its lifelong risk of inflammatory conditions [65][66][67][68]. This is explained by the fact that the occurrence of an epimutation in a stem cell during embryonic development is transmitted to all their daughter cells and affects many more cells than those occurring in adult stem and/or somatic cells during postnatal development [69]. These epigenetic changes can not only be transmitted during successive division but also are passed on from generation to generation, some authors mentioning a real transgenerational epigenetic inheritance [70][71][72][73][74][75][76][77][78][79][80]. These prenatal environmental induced-epigenetic modifications could therefore contribute to the IBD epidemic not only by contributing to this condition but also by passing on modifications to subsequent generations, contributing to familial IBD predisposition, as illustrated by immigration studies [64,[81][82][83][84][85].
There are few data on prenatal epigenetic plasticity in response to the environment in intestinal inflammation [64]. Some data suggest that this epigenomic reprogramming occurs in response to maternal diet modifications, and an excess of prenatal micronutrients (i.e., methyl donors routinely incorporated into prenatal supplements, such as folate, methionine, betaine and vitamin B12) in the maternal diet could confer an increased risk of colitis in the offspring [73]. The occurrence of maternal infection during pregnancy could also lead to the production of IL-6, known to induce epigenetic changes in fetal intestinal epithelial stem cells, which could induce long-lasting impacts on intestinal immune homeostasis and a predisposition toward inflammatory disorders [86]. In addition to diet and infections, maternal smoking during pregnancy could also have an impact on the risk of developing IBD [87]. A study of the impact of prenatal maternal smoking on the offspring's DNA methylation has made it possible to highlight 69 differentially methylated CpGs in 36 genomic regions, among which four CpG sites were associated with an increased risk of IBD [87]. Maternal smoking induced persistent alterations in DNA methylation (rather, global hypomethylation [88][89][90][91][92]) but also miR dysregulation in the exposed offspring, changes that can be transmitted to the next generation [90,[93][94][95][96][97][98]. Taken together, these data suggest that these maternal influences during prenatal development can induce epigenetic changes in the offspring, sometimes considered by some authors as the first step towards IBD development (by introducing a permanent change in the disease-relevant cell types) [64,99,100].

Microbiota
Occurring in this predisposing environment, a microorganism's gut colonization during the first hours of life can be considered as the second step toward the occurrence of IBD [64,99,100]. Influenced by the mode of delivery, the presence or absence of breastfeed-ing and early environmental exposure, the early-life gut microbiota sets trajectories for health or IBD [101,102]. This newly formed microbiome will modulate until the age of 3 years to reach a globally largely similar taxonomic composition as in adults and will act as an epigenetic modulator, modifying the epimutations induced in the prenatal period. Breastfeeding and early bacterial colonization appear to play an important role in DNA methylation in intestinal epithelial stem cells and to condition the lifelong gut health [103].
The microbiome can induce epigenetic changes both in the intestinal epithelium and in immune cells (Table 1). Comparing the epigenomes of germ-free mice or antibiotictreated mice to conventional mice, it appears that this microbiome can influence the host epigenetics through changes in DNA methylation, histone modifications and, also, through ncRNAs [104][105][106][107][108]. Species belonging to Firmicutes (especially Faecalibacterium prausnitzii and Roseburia species [109]) and Bacteroides genera, known to be reduced in IBD [110], have an epigenetically mediated anti-inflammatory action (HDAC inhibition) via the production of short-chain fatty acids (SCFAs) (the role of these will be discussed in more detail below) [111]. The commensal flora can also affect the bioavailability of methyl groups through their production of folate and affects the host DNA methylation [112,113].
Some germs may also contribute to the occurrence of IBD through their epigenetic mechanisms. Adherent-invasive Escherichia coli (AIEC), commonly associated with CD [114], upregulates the levels of miR-30c and miR-130a in intestinal epithelial cells (IECs), which reduces the levels of ATG5 and ATG16L1 and inhibits autophagy, leading to increased numbers of intracellular AIEC and the inflammatory response [115]. In turn, AIEC-infected IECs secrete exosomes that can transfer these same miR to recipient IECs with the same consequences, promoting the invasion and proliferation of infected tissues [116]. In addition, AIEC triggers an excessive mucosal immune response against the gut microbiota via the let-7b/TLR4 miR signaling pathway [117]. Mycobacterium avium subspecies paratuberculosis (MAP), also known to be associated with IBD [118], induces miR-21 expression in infected macrophages and decreases their ability to eliminate the bacteria, thus contributing to intestinal inflammation [119].
Microbial components such as lipopolysaccharides (LPS) and flagellin may also induce host epigenetic changes. LPS (a major component of the Gram-negative bacteria outer membrane) contributes to the development of intestinal inflammation by promoting the activation of NF-κB (nuclear factor-kappa B) pathways and the cytokines released by the downregulation of miR-19b, miR-497 and miR-215 in IECs [120], monocyte/macrophage cells [121] and fibroblast cells [122], respectively. LPS can also increase the level of H19 lncRNA in IECs that bind to miR (miR-34a and let-7), inhibiting cell proliferation and, thus, impairing the intestinal epithelial barrier [20]. In contrast, the flagellins of some bacteria-in particular, Roseburia intestinalis (found in a reduced abundance in IBD patients)-have rather epigenetically mediated anti-inflammatory actions [123,124]. Flagellin inhibits the activation of the NLRP3 (NOD-like receptor family, pyrin domain containing 3) inflammasome and proptosis in macrophages via miR-223-3p [123] and induces a lncRNA (HIF1A-AS2) that inactivates the NF-κB/Jnk (c-Jun N-terminal kinase) pathway [124].  [121] Although the taxonomic composition of the microbiota is stable at year 3, its composition can be influenced by a range of other environmental factors (including dietary habits, smoking and drugs, as discussed below), which may be responsible for the third step towards the occurrence of IBD [64].

Gut Microbiota-Derived Metabolites
Influenced by both the microbiota and diet, SCFAs are gut microbiota-derived metabolites that result from the anaerobic fermentation of nondigestible dietary fibers (found in fruits and vegetables). Acetate, butyrate and propionate, the three principal SCFAs, exert an anti-inflammatory role and promote the integrity of the epithelial barrier functions partly via the epigenetic pathways (Table 2) [141,142]. Among the SCFAs, butyrate is the most studied one. By inhibiting, in a reversible way, HDACs [143,144], cells exposed to butyrate present higher acetylation at specific lysine residues in histones, resulting in increased transcription of genes in both intestinal epithelial and immune cells [145]. The inhibition of HDAC in cells contributes to the reduction of inflammation by (1) the induction of IκBα expression, with a subsequent inhibition of the NF-κB pathway, (2) the inhibition of the IFN-γ/STAT1 (signal transducer and activator of transcription) signaling pathway and (3) the activation of the anti-inflammatory function of PPARγ (peroxisome proliferator-activated receptor γ) [145]. Butyrate also has more specific epigenetic actions on certain cell types. At the epithelial level, butyrate plays a role in the integrity of the epithelial barrier (by restoring tight junction proteins [146]) and the defense against the invading microorganisms (via a nucleotide-binding oligomerization domain 2 (NOD2)dependent pathway or via autophagy [147]). Butyrate also has an effect on various immune cells, such as (1) monocytes/macrophages (in which it induces monocyte-to-macrophage differentiation, promotes their antimicrobial activity through inhibition of HDAC3 [148], reduces the production of their inflammatory mediators [149] and induces the polarization of M2 macrophages [150]); (2) T cells (promotes Treg [151] and inhibits Th17 cell development [151]); (3) neutrophils (in which HDAC inhibition leads to proinflammatory cytokine reduction [152]) and (4) dendritic cells (inhibit IL-12 [153]). The epigenetic role of propionate and acetate has been less studied. Propionate promotes epithelial cell migration and contributes to intestinal epithelial restitution, a complex process important for tissue regeneration in IBD [142]. Table 2. Impact of the gut microbiota-derived metabolites on the epigenome in intestinal inflammation. CD, Crohn's disease; CEBPB, CCAAT/enhancer binding protein; DNA, deoxyribonucleic acid; DSS, dextran sulfate sodium; HDAC, histone deacetylases; IBD, inflammatory bowel disease; IECs, intestinal epithelial cells; IFN, interferon; IL, interleukin; lncRNAs, long non-coding RNAs; LPS, lipopolysaccharide; MCP-1, Monocyte chemoattractant protein-1; miR, micro-RNA; NF-κB, nuclear factor-kappa B; NOD2, nucleotide-binding oligomerization domain 2; SCFAs, short-chain fatty acids; STAT, signal transducer and activator of transcription; TNF, tumor necrosis factor; UC, ulcerative colitis.

Diet
Next, compared to a low-fiber diet [163], impacting the level of these SCFAs [163], other diets have been shown to induce epigenetic changes related to IBD (Table 3). Regarding the literature, elements of the Western diet, characterized by a low-fiber, low-fruit, low-vegetable and deficiency in micronutrients, as well a high-fat diet, may be associated with epigenetic changes in IBD. The Western diet has been shown to lead to a decrease in miR-143/145a, miR-148a and miR-152 in colonocytes with a consequent increase in ADAM17 (a disintegrin and metalloprotease 17) expression protein and colitis aggravation [164]. A low or deficient methyl diet can also contribute to intestinal inflammation by reducing SIRT1 (sirtuin 1) expression (a histone deacetylase), contributing to endoplasmic reticulum stress [165] and demethylating HIF-1-responsive elements (HRE), which leads to the abnormal gut expression of CEACAM6 (CEA Cell Adhesion Molecule 6), favoring AIEC colonization and subsequent inflammation [166]. Finally, it was shown that a high-fat diet can change the miR profile of the visceral adipose exosomes (switching the exosomes from an anti-inflammatory to a proinflammatory phenotype with an increase of miR-155, for example), predisposing the intestine to inflammation via promoting macrophage M1 polarization [167].

Macrophages
Wei M, et al. (2020) [193] High fat diet rich in n-6 linoleic acid    BRBs exert their anti-inflammatory effects is through decreasing NF-κB p65 expression leading to decrease of DNMT3B expression (but also histone deacetylases 1 and 2 (HDAC1 and HDAC2) and methyl-binding domain 2 or MBD2), which in turn reverse aberrant DNA methylation of tumor suppressor genes, e.g., dkk2, dkk3, in the Wnt pathway, resulting in their enhanced mRNA expression locally in colon and systematically in spleen and bone marrow and thus in decreased translocation of β-catenin to the nucleus prohibiting the activation of the pathway

Smoking
Smoking habits are the single best-established environmental factor that influences the CD phenotype, behavior and response to therapy [203]. While nicotine is the most prominent component released during smoking (and therefore the best-studied), other chemical components could also induce epigenetic changes, including polycyclic aromatic hydrocarbons; heavy metals (nickel, cadmium, chromium and arsenic); carbon monoxide and reactive oxygen species [203]. Well-studied in lung diseases (but never in IBD, to our knowledge), smoking-induced epigenetic modifications seem to be strongly associated with smoking habits, the dose and the duration of smoke exposure [204][205][206][207][208][209]. The methylation of certain genetic loci, post-translational modifications of histones and the level of expressed miR may be reversible after smoking cessation (after 5 years, according to some studies) [93,[204][205][206][207][208][209][210]. In contrast to these reversible epigenetic changes, others remain unchanged even after 30 years of smoking cessation, explaining that epigenetic modifications induced by smoking exposition confer long-term risks of adverse health outcomes but could also be transmitted to the next generation [93,204,[207][208][209][210][211]. The mechanisms by which tobacco may contribute to inflammation are multiple and involve changes in the enzymes involved in DNA methylation, post-transcriptional histone modifications and ncRNAs [65,93,203,[212][213][214].
Regarding smoking-induced DNA methylation, a meta-analysis performed by Joehanes and colleagues highlighted various genome-wide association studies showing that smoking-induced genes differentially methylated are enriched for variants associated with smoking-related diseases, including IBD, CD and UC [210,215,216]. The findings suggest that changes in methylation of the BCL3, FKBP5, AHRR and GPR15 genes are involved in the mechanism by which smoking increases the risk of CD [217,218].
Smoking exposure also alters ncRNAs in a dose-and-time-dependent manner, high doses of and long-lasting exposure being necessary to induce irreversible ncRNA alterations, which may be involved in smoking-related diseases [237,238]. While there are no data on lncRNAs, the impact of smoking on miRs in IBD has been better studied. Interestingly, these IBD-induced epigenetic changes could partly explain why smoking is rather protective in UC, whereas it is an important risk factor in CD. Indeed, nicotine enhances the miR-124 expression, which targets and downregulates IL6R, resulting in a shifting Th1/Th2 balance toward Th1 (in peripheral blood lymphocytes and colon tissues), thereby protecting against Th2-type UC and worsening Th1-type CD [239]. This increase in miR-124 in epithelial cells, lymphocytes and macrophages in response to nicotine also results in the phosphorylation of STAT3, in a decreased production of IL-6 at the transcriptional level, and prevents the conversion of pro-TNF-α to TNF-α, which also explains the protective role of tobacco in the UC [240,241]. Tobacco also induces changes in several miRs that are functionally related to inflammation [65]. Among those highlighted in the IBDs are miR-21, miR-132, miR-195 and miR-223 [65]. MiR-21 (increased in the colon of IBD patients [242]) is known to increase the intestinal epithelial permeability (through an action on the tight junctions) [242][243][244] and plays a crucial role in T-cell differentiation, apoptosis and activation [242,[245][246][247] and promotes the production of inflammatory cytokines (including TNF-α, IFN-γ and IL-1β) by immune cells, contributing to tissue inflammation and IBD pathogenesis [248][249][250]. The overexpression of a miR-195 precursor lowered the cellular levels of the Smad7 protein, leading to a decrease in c-Jun and p65 expression, and might contribute to the protective effect of tobacco in UC [251]. Lastly, smoking also downregulates miR-200 [252], known to repress epithelial-to-mesenchymal transition (or EMT), a process involved in intestinal fibrosis [242,252]. Consequently, the decrease of miR-200 in response to smoking could partly explain why smoking IBD patients are more likely to develop intestinal fibrosis (and fibrostenosis) [253][254][255].

Vitamin D
Vitamin D is an environmental factor involved in IBD pathogenesis. Its deficiency, which can be both a cause and a consequence of IBD, is associated to an increased risk of disease activity, mucosal inflammation, clinical relapse and a lower quality of life [301]. A vitamin D-deficient diet contributes to IBD through the following epigenetic mechanisms: (1) increase in miR-142-3p expression in intestinal tissues leading to autophagy dysregulation [302]; (2) reduction of the interaction between VDR and HDAC11, an important complex for the maintenance of the epithelial barrier [303], and (3) the upregulation of miR-125b expression and reduction of M1 macrophage polarization to the M2 subtype [304].

Physical Activity
Despite all the known benefits of physical activity (PA) in IBD [16], the way in which it modifies epigenetics has never been studied to date to our knowledge. In other inflammatory diseases, the general view is that regular moderate-intensity physical activity could have an anti-inflammatory effect, while prolonged or high-intensity PA can trigger inflammation, both by leading to epigenetic changes that, in turn, regulate inflammatory responses in peripheral tissues [305][306][307]. These peripheral epigenetic changes appear to be largely induced by the muscle secretome, also known as "myokinome", which corresponds to all the cytokines or proteins produced by the myocyte in response to muscular contractions [308,309]. These sport-induced epigenetic modulations (including both DNA methylations, histone modifications and miR modulations [309]) seem to vary according to the type of performed exercise (and the frequency, the intensity and the duration) [309]; the individual (in terms of age, gender and body composition) [305,309] and can vary from one tissue to another [310].
Studies suggest that PA is associated with DNA hypermethylation (although these results are not unanimous), contributing to a decreased expression of inflammation-related genes (such as the hypermethylation of IL-17A and IFN-γ promoter regions or TNF gene and the hypomethylation of IL-10) [311]. Exercise can also induce histones acetylation/deacetylation in a body mass index-dependent manner [312]. Finally, physical activity also leads to the release of a range of miRs from the muscle, known to play roles in macrophage polarization, dendritic cell activation, dendritic cell-mediated T-cell activation and the Th1 and Th17 differentiation of T cells, which are all pathophysiological processes involved in IBD [309].

Limitations to the Analysis of the Exposome Impact on the Epigenome in IBD
The study of the impact of the exposome on the epigenome is difficult because of the limitations of both the exposome study and the epigenome study. Human epidemiological studies are necessary to assess exposome-related epimutation. The first way to study the impact of the exposome is using retrospective case-control epidemiological studies (which compare the life of IBD patients with control cases to identify environmental influences based on surveys), but these studies are subject to a recall bias of past exposure factors. The second way of studying the exposome is a multi-omics approach (via the quantification and detection of external influences in a group of patients compared to a control group by different technologies), but unfortunately, it does not allow the detection of the factors responsible for the crucial pathophysiological changes explaining the occurrence of the pathology, it is very difficult to obtain a cause-and-effect relationship when using this method. A third approach, which counters this, consists of directly studying the supposed factors based on the basis of the pathophysiological hypotheses, but this exposes a selection bias and is not always easy to carry out in humans. Easier to perform on in vivo animal models or in vitro models, it does not reflect what is happening in humans [100,313]. All patients exposed to an environmental factor do not develop epimutation [314,315]. The onset of this and related diseases may depend on the duration, frequency and intensity of exposure to the environmental factor and the period of life during which it occurs [81]. Epigenetic changes can be influenced by age, sex and race but, also, possibly by the underlying host genotype [316][317][318]. Furthermore, the epigenetic changes induced by the exposome may vary according to the cell type and analyzed tissue (peripheral blood mononuclear cells, epithelial cells and biopsies) [316]. The tissue isolation and manipulation may also induce nonspecific epigenetic changes and mask the exposome-related one [100]. Finally, whatever the methodology used, it is always subject to several environmental factors at the same time that may induce concomitant and potentially interactive epigenetic changes, and the impact of an individual factor is not always easy to identify [319].

Conclusions and Challenge for the Future
This review is the first, to our knowledge, to study the impact of the exposome on the epigenome in IBD. Different elements of the exposome such as the maternal lifestyle, microbiota, diet, smoking, infection and vitamin D, as well as different drugs, may induce epigenetic changes related to IBD (Figure 1). Next to these factors, the impact of other environmental factors known to be involved in the pathophysiology of IBD on host epigenetics has not yet been studied. The influence of physical activity [16], appendicectomy in UC [16], processed and fast food or dietary [320,321] or psychologic stress/anxiety/depression [19], NSAIDs [256], oral contraception [257] or infections [20], as well as other factors potentially involved, deserve to be investigated. Regarding the identified environmental factors, how environmental exposure (in terms of the duration, frequency and timing at which it occurs) induces an epimutation, and whether this involves an exposome-microbiome-epigenome axis and becomes a critical factor for IBD development is largely unknown and remains to be further investigated. Finally, the exposome could be a tool to predict relapses [322]. The development of electronic technologies to continuously record a patient's exposome could allow disease-modifying exposures to be detected and acted on early to prevent relapse or disease progression [322].  [323]). The environmental factors epigenetically influencing the occurrence of intestinal inflammation are breastfeeding, microbiota, diet, smoking habits, drugs, infections, vitamin D and physical activity. Although present at all times, it is mainly during the prenatal period, at birth and just before the onset of the disease that these factors play a key role in triggering the disease. These environmental factors, by inducing DNA methylation, histone modifications and ncRNAs in different cell types, trigger the pathways involved in IBD pathophysiology and contribute to disease initiation.  [323]). The environmental factors epigenetically influencing the occurrence of intestinal inflammation are breastfeeding, microbiota, diet, smoking habits, drugs, infections, vitamin D and physical activity. Although present at all times, it is mainly during the prenatal period, at birth and just before the onset of the disease that these factors play a key role in triggering the disease. These environmental factors, by inducing DNA methylation, histone modifications and ncRNAs in different cell types, trigger the pathways involved in IBD pathophysiology and contribute to disease initiation. Abbreviations 5hmC, 5-hydroxymethyl cytosine; ADAM17, a disintegrin and metalloprotease17; AIEC, Adherent-invasive Escherichia coli; CD, Crohn's disease; CEACAM6, CEA Cell Adhesion Molecule 6; CpG, cytosine-phosphate-guanine; DNA, deoxyribonucleic acid; DNMT, DNA methyltransferase; EMT, epithelial-to-mesenchymal transition; HAT, histone acetyltransferase; HDAC, histone deacetylases; HDM, histone demethylases; HMT, histone methyltransferase; HRE, HIF-1-responsive elements; IBD, inflammatory bowel disease; IEC, intestinal epithelial cell; IL, interleukin; Jnk, c-Jun N-terminal kinase; LPS, lipopolysaccharide; lncRNA, long non-coding RNA; MAP, Mycobacterium avium subspecies paratuberculosis; MeSH, Medical Subject Heading; miR, micro-RNA; mRNA, messenger RNAs; ncRNA, non-coding RNA; NF-κB, nuclear factor-kappa B; NLRP3, NOD-like receptor family pyrin domain containing 3; NOD2, nucleotide-binding oligomerization domain 2; NSAID, non-steroidal anti-inflammatory drug; PPAR-γ, peroxisome proliferator-activated receptor γ; RNA, ribonucleic acid; SAM, S-adenosyl-L-methionine; SCFA, short-chain fatty acid; SIRT, sirtuin; STAT, signal transducer and activator of transcription; TET, ten-eleven translocation enzymes; TNF, tumor necrosis factor; UC, ulcerative colitis.