Keywords

Introduction

Esophageal adenocarcinoma (EAC) is thought to develop from a pre-cancerous condition of the esophagus called Barrett’s esophagus (BE). BE is a metaplastic condition where specialized intestinal columnar epithelium replaces the normal squamous epithelium in the esophagus [76]. BE is suspected when a salmon pink-colored mucosal lining is visualized in the tubular esophagus during an endoscopic examination and is confirmed by histopathological evaluation demonstrating intestinal-type epithelium in biopsies obtained proximal to the gastro-esophageal junction (GEJ) [78]. Clinically, BE is important because it is recognized as the precursor to esophageal adenocarcinoma (EAC) ; although the absolute risk of BE progression is low (approximately 0.2 %/year), individuals with BE have an approximately 25× increased risk of developing EAC compared to the general population [76].

The incidence of EAC has been rapidly increasing in the US for reasons that are not entirely clear [6], but might be related to an increase in the prevalence of known risk factors for BE and/or EAC. Risk factors for EAC include smoking, overweight and obesity, central adiposity, and chronic gastroesophageal reflux disease, which is thought to trigger BE [73].

EAC appears to arise via a metaplasia-dysplasia-carcinoma sequence where Barrett’s metaplasia progresses through low- and high-grade dysplasia to invasive carcinoma [76]. Although the molecular and genetic events involved in initiation and progression of BE/EAC are still being investigated, certain histologic and associated genetic alterations have been described for EAC [21, 28, 58, 70, 71, 79, 88, 90]. Epigenetic alterations, mainly aberrant DNA methylation , have also been demonstrated to frequently occur in BE and EAC [11, 43, 50, 81]. Some epigenetically altered genes are known tumor suppressors , and in some cases, aberrant methylation is predicted to play a causative role in the pathogenesis of these EACs. Additionally, some of these aberrantly methylated genes (e.g., p16, RUNX3, HPP1, NELL1, TAC1, SST, AKAP12, and CDH13) might be useful prognostic markers to predict the progression of BE to EAC [11, 43].

Although elevated body mass index (BMI) has been associated with altered DNA methylation in several prior studies [17, 33, 37, 60], there is currently little information on the relationship between obesity, epigenetic alterations, and BE/EAC. In this chapter, we will: (1) briefly review epigenetic alterations in BE and EAC, (2) describe the relationship between obesity, gastro-esophageal reflux, and esophageal inflammation, and (3) further explore the relationship between obesity and epigenetic alterations in these conditions.

Epigenetic Alterations in DNA methylation in BE and EAC

Previous studies have evaluated global patterns of DNA methylation in BE and EAC and found that epigenetic alterations occur in BE tissue before the development of dysplasia or cancer [93]. Xu et al. noted that both BE (N = 77) and EAC (N = 117) samples were highly methylated compared to normal squamous esophagus (N = 94), providing evidence that DNA hypermethylation occurs early in the BE to EAC progression sequence. Another study used microarrays to compare DNA methylation between BE and EAC samples and found that methylation alterations in certain genes could distinguish between BE, BE with dysplasia, or EAC [3]. The authors found a panel of genes (SLC22A18, PIGR, GJA12, and RIN2) could accurately discriminate BE from dysplasia/EAC and could stratify patients into low-, intermediate-, or high-risk groups based on DNA methylation patterns .

Distinct genome-wide methylation patterns can be found when comparing normal squamous esophagus (SQ), BE, dysplastic BE, and EAC samples [50]. Additionally, there appear to be subsets of both BE and EAC that demonstrate relatively high methylation levels compared to other BE or EAC cases. Previously, Kaz et al. examined BE and EAC tissue samples using methylation microarrays and found there were subgroups with distinct methylation signatures (high and low methylation epigenotypes), suggesting that there may be a CpG Island Methylator Phenotype (CIMP) molecular class of BE and EAC [50]. This observation needs to be verified with additional studies.

Altered methylation of promoter CpG islands, which is associated with gene silencing in some cases, has been shown to occur frequently in BE, dysplastic BE, and EAC. Studies evaluating the methylation status of several dozen candidate genes that are epigenetically altered in other cancers have been evaluated in BE and EAC. One of the first tumor suppressor genes shown to be aberrantly methylated in BE was CDKN2A/p16, which normally inhibits CDK-mediated phosphorylation of the Rb protein and inhibits cell cycle progression. CDKN2A/p16 promoter hypermethylation combined with 9p21 chromosomal loss leads to inactivation of this gene in some cases of EAC or BE with dysplasia [52, 91]. CpG island hypermethylation of the CDKN2A/p16 promoter ranging from 3 to 77 % of BE cases has been reported in several publications, suggesting that CDKN2A/p16 methylation is in early event in BE pathogenesis [5, 23, 85, 92].

Other candidate tumor suppressor genes , such as APC, ESR1, and CDH1, also show aberrant promoter CpG island methylation in esophageal samples. In a study evaluating 107 distinct spatial locations in six esophagectomy specimens, which contained both BE and EAC, spatial methylation maps were created to define methylation patterns in the BE and adjacent EAC [22]. Eads et al. found a high incidence of methylated ESR1, APC, and CDKN2A in different sites of individual cases of BE, BE with dysplasia, and EAC in a pattern suggesting simultaneous methylation in large contiguous fields or clonal expansion of cells that acquired methylation early in the BE → EAC progression sequence [22]. Similar patterns of widely distributed genetic alterations consistent with clonal expansion in BE have been reported in studies that focused on LOH events or mutations of APC, TP53, and CDKN2A/p16 [4, 67, 92].

Other groups have focused on altered APC and CDH1 methylation in BE and EAC as well [7, 49]. In a study of 52 patients with BE and EAC, Kawakami et al. found hypermethylated APC in 39.5 % of cases of BE and 92 % of EAC cases, but not in matched normal esophagus. When they looked at plasma samples from these patients, they could detect methylated APC in 25 % of EAC patients; this was associated with reduced survival [49]. Meanwhile, Smith et al. found high levels of methylated APC in >95 % of BE and EAC tissues studied, supporting the concept that aberrant methylation of putative tumor suppressor genes occurs early in the BE → EAC sequence [74].

Numerous other genes implicated in carcinogenesis have been found to be aberrantly methylated in their CpG island promoter regions in BE/EAC, including the STAT-induced STAT inhibitors (SSIs), suppressors of cytokine signaling (SOCS-1 and -3) and Reprimo (RPRM), and members of the glutathione S-transferase (GST) and glutathione peroxidase (GPX) family [34, 65, 81]. Jin et al. identified aberrant methylation of somatostatin (SST), tachykinin-1 (TAC1), NELL1, and CDH13 and noted that the incidence of methylation of these genes was higher in BE, BE with dysplasia, and EAC vs. normal esophageal samples [44–48]. These investigators further showed that treatment of esophageal cell cultures with the demethylating agent 5-aza-deoxycytidine caused increased mRNA expression levels of these hypermethylated genes, supporting the association between methylation and transcriptional repression. It is clear from these and other studies that, in general, aberrant methylation of genes is detectable in cases of BE without dysplasia, suggesting that many of the epigenetic alterations that occur in EAC are already present in BE.

The Role of Obesity in Chronic Gastroesophageal Reflux Disease (GERD) and Esophageal Inflammation

Gastro-esophageal reflux is a strong risk factor for both esophageal inflammation and BE [66, 87]. Reflux of gastric acid and enzymes as well as bile from the duodenum are likely responsible for inducing an inflammatory response in the esophagus that is associated with Barrett’s metaplasia [29, 41]. GERD and reflux esophagitis have been associated with the presence of pro-inflammatory and pro-tumorigenic cytokines , including IL-8 and IL-1 beta, which are discussed in more detail below [27].

Obesity , in particular central adiposity, is also a well-documented risk factor for developing both BE and EAC [13, 14, 53, 87]. The current guidelines for selecting individuals to screen for BE reflect this risk; those with multiple risk factors for EAC, including elevated BMI and intra-abdominal distribution of body fat, are advised to undergo screening for BE with upper endoscopy [77]. The mechanism(s) by which obesity and/or central adiposity lead to the development of BE and EAC are not well-understood. One common hypothesis is that elevated BMI increases the intragastric pressure, which promotes gastroesophageal reflux, although the relative importance of this mechanism is debated [24, 25, 55]. Other related mechanisms of BE/EAC development associated with obesity and GERD include the presence of an increased gastroesophageal pressure gradient and anatomic disruption of the normal gastroesophageal junction [63].

Yet the association between obesity and BE/EAC has been shown to persist even in the absence of GERD. A recent meta-analysis that found a positive association between obesity and BE/EAC noted that obesity remained a risk factor for BE even after controlling for GERD (adjusted odds ratio (OR) = 2.04, 95 % confidence interval (CI) = 1.44-2.90) [73].

It is worth noting that BE and EAC are much more common among males: although GERD occurs with similar frequency among women and men, BE and EAC are roughly twice and seven times more common in men than in women, suggesting gender-related factors play a role in the formation of both BE and EAC [80]. A biologic explanation to explain this paradox is that the adipocytes themselves, which are metabolically active, might promote the development of BE and EAC in men preferentially [24]. Adipocytes located within the visceral compartment (mesentery and omentum) are more metabolically active than subcutaneous fat cells. This might account for the finding that central obesity, more than overall body weight, is a major risk factor for BE and EAC [12]. Thus, the fact that BE is more common in men than women might also be explained in part by the finding that male-pattern obesity is associated with excess abdominal adipose, i.e., an increased waist-to-hip ratio (WHR) [16]. For example, a case-control study by Edelstein et al. compared measures of central adiposity in a group of patients with newly diagnosed BE compared to matched controls and found high WHR was associated with a BE risk (OR = 2.4, 95 % CI = 1.4–3.9) [24].

Metabolically active intra-abdominal fat may lead to alterations in the expression of hormones, cytokines, and adipokines, which in turn may lead to the development of clinical metabolic disorders including insulin resistance and type 2 diabetes [35, 95]. Altered levels of these metabolically active substances have also been associated with the promotion of cancer, possibly related to their pro-inflammatory effects and their effects on angiogenesis, insulin signaling, apoptosis, and metastasis [31]. Leptin and proinflammatory cytokines , which are produced by visceral adipocytes, have been linked to inflammation seen in BE samples [61], and these cytokines have been shown to inhibit apoptosis and increase proliferation in BE and EAC cell lines [62].

Other adipokines, including free fatty acids, produced by adipocytes can lead to the development of insulin resistance which is associated with high blood insulin levels and the promotion of tumor formation [8, 86, 94]. In another case-control study of 284 newly diagnosed BE patients compared to 294 GERD control subjects and 285 population control subjects, adiponectin levels were positively associated with the risk of BE among patients with GERD [2]. However, another recent study evaluated 135 BE patients, 133 refractory GERD patients, and 1157 control subjects using multivariate logistic regression models for waist-to-hip ratio and found an inverse relationship between serum adiponectin levels and BE (tertile 1 vs. tertile 3 OR = 0.42, 95 % CI = 0.22–0.80) [30]. A meta-analysis , which included both of these studies, found that total serum adiponectin was not associated with the risk for BE compared with GERD control subjects (OR = 1.20, 95 % CI = 0.69–2.10) or population control subjects (OR = 0.79, 95 % CI = 0.46–1.34) [10]. However, this meta-analysis did find a relationship between BE and serum leptin levels (OR = 2.23, 95 % CI = 1.31–3.78) and serum insulin levels (OR = 1.74, 95 % CI = 1.14–2.65) when men and women were considered together or separately [51]. Thus, there is compelling but inconsistent evidence for adiponectin, insulin, and leptin affecting the formation and progression of BE/EAC; additional investigation will be important to produce a clearer understanding of their role(s).

Inflammation and Epigenetic Alterations in BE and EAC

Chronic inflammation is a predisposing factor for malignant transformation, and it has been estimated that roughly 25 % of all cancers are associated with chronic infection and/or inflammation [38]. Chronic inflammation in the esophagus might promote Barrett’s metaplasia via induction of transcription factors such as CDX1 and CDX2, which play critical roles in intestinal development [32]. Increased expression of these genes has been shown in BE and EAC tissues, but they are not expressed in the normal squamous esophagus nor in the gastric epithelium [32]. Exposure of esophageal cells to bile acid and hydrochloric acid, which is present in gastroesophageal refluxate, has been shown to activate the CDX promoters in esophageal cell lines, and CDX expression can be found in the chronically inflamed squamous esophagus as well as in metaplastic BE [75]. Therefore, it appears that chronic acid/bile reflux into the esophagus stimulates CDX expression which might mediate the development of BE.

Inflammation is known to promote direct DNA damage and genetic alterations , but has also been shown to affect DNA expression via epigenetic mechanisms such as DNA methylation [54]. For example, hypermethylation of CDH1, the gene for E-cadherin, and CDKN2A/p16, which has been demonstrated in metaplastic gastric mucosa of individuals infected with Helicobacter pylori, is thought to be a driver of gastric carcinogenesis [9, 57]. The mechanism mediating aberrant DNA methylation at sites of chronic inflammation is thought to be related to the release of HOCl and HOBr by neutrophils and eosinophils, which in turn leads to the production of 5-methylcytosine and 5-bromocystosine [82]. Because neither DNA methyl transferase-1 (DNMT-1) nor methyl-binding proteins are easily able to distinguish these 5-halocystosines from 5-methylcytosine, inappropriate de novo methylation may occur during DNA replication in the setting of inflammation [54, 82].

The inflammatory milieu seen in the esophagus of individuals with BE and EAC likely promotes both genetic and epigenetic alterations. As noted previously, numerous genes have been shown to be aberrantly methylated in BE and/or EAC, and altered DNA methylation appears to be any early event in the BE to dysplastic BE to EAC sequence [50]. A putative mechanism connected with inflammation-associated processes in the formation of BE involves the glutathione S-transferase (GST) and peroxidase (GPX) family of genes, which normally function to protect cells from the damage caused by reactive oxygen species (ROS), which have been shown to accumulate in the setting of chronic gastric reflux [56]. Peng et al. examined the promoter regions of 23 GST and GPX genes in normal esophagus, BE, dysplastic BE, and EAC cases in conjunction with gene sequencing and gene expression assays and found aberrant DNA methylation of GPX3 (62 %), GPX7 (67 %), GSTM2 (69 %), and GSTM3 (15 %) in EAC cases. DNA methylation and mRNA expression were inversely correlated for GPX3, GPX7, and GSTM2, and immunohistochemical (IHC) analysis using a tissue microarray confirmed weak/absent staining in EAC for these genes and moderate/strong staining in normal samples [56]. Thus, it is possible that persistent esophageal inflammation related to GERD leads to epigenetic inactivation of genes involved in antioxidant pathways, which could be an important mechanism in the development of EAC.

Another study examined levels of MLH1 promoter methylation in esophageal tissues, including cancer, BE, reflux esophagitis, as well as normal tissue [84]. This group found significant hypermethylation of MLH1 in cancer (63.5 %) and pre-cancer (53.8 %), with the highest level of MLH1 methylation seen in patients with GERD (88.8 %), suggesting a relationship between inflammation triggered by reflux and aberrant DNA methylation. Of note, altered DNA methylation has been described in other cancer-related inflammatory conditions as well, including ulcerative colitis, cholangiocarcinoma, and chronic pancreatitis [40, 64, 89].

Obesity, Epigenetics, and BE/EAC

Although the risk of developing BE and/or EAC is associated with obesity [72, 77], and altered DNA methylation is commonly seen in BE and EAC, studies on the effects of demographic factors, such as obesity, on the epigenome in the esophagus are very limited at the time this chapter was written. There is evidence that certain environmental, behavioral, and demographic factors can influence the epigenetic state, which suggests that the behavioral factors associated with BE and EAC may act by inducing alterations in the methylation status of DNA [1]. For instance, alterations in the promoter CpG island methylation status of genes associated with obesity, appetite control, and metabolism have been shown to occur in DNA isolated from blood and breast tissue of obese compared to lean individuals [18, 20, 33, 83]. Hoyo et al. examined IGF2 methylation in differentially methylated regions (DMR; regions of DNA where multiple adjacent CpG dinucleotides show concordant changes in methylation, which are discussed in further detail below) in umbilical cord blood of newborns and correlated it to newborn birth weight, given previous findings of altered IGF2 methylation and obesity [37]. This group found that reduced IGF2 methylation was associated with elevated plasma IGF2 protein levels, with the strongest association seen in infants born to obese women (pre-pregnancy BMI > 30 kg/m2, p < 0.0001). Elevated levels of IGF2 were associated with higher infant birth weight even after adjusting for several factors including pre-pregnancy BMI, gestational diabetes, and infant gender. This group concluded that circulating levels of IGF2, a risk factor for esophageal and other cancers, might be affected by altered IGF2 methylation, which in turn might be affected by pre-pregnancy obesity.

Another group recently evaluated 44 patients with squamous cell cancer of the esophagus (ESCC) in order to determine whether the tumor suppressor gene CDKN2A/p16 was aberrantly methylated in ESCC tumors and matched normal tissues, and whether this epigenetic alteration was associated with obesity or other risk factors [60]. The authors found aberrant CDKN2A/p16 methylation in 12/44 (27 %) of ESCC samples and no normal samples. Additionally, they noted that obesity status was positively correlated with CDKN2A/p16 methylation (p = 0.001), with logistic regression analysis demonstrating the risk of methylation for BMI ≥ 25 was 12 times higher than for individuals with BMI < 25 (OR = 12, p = 0.004). Although this is a small study, the authors suggest it provides evidence that obesity increases the risk of developing ESCC, possibly by promoting CDKN2A/p16 methylation.

While it is likely that both somatic genetic and epigenetic alterations play a role in the pathogenesis of BE and EAC, there is currently very little information about the relationship between Barrett’s esophagus and esophageal adenocarcinoma, obesity, and aberrant DNA methylation. In an attempt to provide insight into this important question, our research group has used methylation microarrays to examine epigenome-wide methylation patterns in a sizable collection of esophageal samples for which demographic information, including BMI, was available. We analyzed methylation patterns of 46 DNA samples isolated from individuals with BE (N = 15), BE with low-grade dysplasia (BE + LGD; N = 14), BE with high-grade dysplasia (BE + HGD; N = 9), and EAC (N = 8) cases using HumanMethylation 450 BeadChips (HM450, Illumina). We stratified these samples into those obtained from individuals with either high BMI (BMI > 30) or low BMI (BMI ≤ 30). We used the data from the HM450 array to compare the methylation levels of more than 485,000 individuals CpG dinucleotides between the high and low BMI cases. We considered a locus to be “differentially methylated” if the p value was <0.0001 and the methylation level (also known as the beta value) differed by at least 10 % between the low and high BMI cases. Using these criteria, we found a total of 974 differentially methylated loci (DML) in BE, dysplastic BE, and EAC samples when comparing the high and low BMI groups. In general, the high BMI cases showed increased methylation at the DML in the esophageal tissues, with 872 out of 974 DML (89.5 %) demonstrating increased methylation in high vs. low BMI cases.

The DML were found in various functional regions of the genome: 226 were located in gene promoters, 471 in gene bodies (intragenic), and 277 in between genes (intergenic). We also evaluated the location of DML with respect to CpG islands , including CpG dinucleotides located in promoter CpG islands, non-promoter CpG islands, and CpGs outside of islands. Analysis of the regions outside of promoter-related CpG islands is notable because an understanding of methylation alterations in areas with relatively low CpG density is becoming increasingly recognized to be important in diseases such as cancer. It has been shown that CpG- rich regions (i.e. CpG islands) demonstrate more stable DNA methylation across tissues and cell populations, whereas methylation is more dynamic in CpG shores (within 2 kb of a CpG islands) and CpG shelves (within 4 kb of a CpG island). Furthermore, the methylation status of CpG shores and shelves appears to regulate gene expression, which would provide a mechanism through which epigenetic alterations in these regions could affect BE and EAC formation [39, 97]. We found 182 DML were located in CpG islands and 376 were located in CpG island shores (within 2 kb of a transcription start site).

As part of our studies, we also assessed whether differentially methylated loci in the high vs. low BMI groups might be associated with esophageal adenocarcinoma (EAC). We defined “cancer associated loci ” as loci whose methylation status differed between a group of 12 normal squamous esophagus and 24 EAC cases which we assayed using HM450 methylation arrays. Using this definition, we found 352 DML (36.1 % of the total 974 DML) that were cancer-associated. This is more than 2× as many cancer-related DML than we would expect by chance alone since just 16 % of the total probes on the array are “cancer related” by our criteria. The top 20 most significant annotated DML associated with BMI are shown in Table 8.1.

Table 8.1 Top 20 differentially methylated loci : high vs. low BMI cases (BE, LGD, HGD/EAC combined)
Full size table

There are also differences in methylation patterns when comparing high to low BMI cases when examining the various esophageal tissue types separately. We compared methylation in the high BMI (N = 4) vs. low BMI (N = 11) BE cases, the high BMI (N = 7) vs. low BMI (N = 7) LGD cases, and the high BMI (N = 9) vs. low BMI (N = 8) HGD/EAC cases. Table 8.2 summarizes the DMLs found when comparing these groups. The methylation status of the high compared to low BMI BE cases with respect to genomic regions and CpG island location is shown in Fig. 8.1. In general, in the BE and EAC cases, methylation levels were increased in gene promoters, bodies, and intergenic regions of high BMI patients compared to the low BMI patients.

Table 8.2 Differentially methylated loci: high-BMI vs. low-BMI cases separated by tissue type
Full size table
Fig. 8.1
figure 1

Genomic location , relationship to CpG islands, and methylation status of DML when comparing high vs. low BMI esophageal samples. In each panel, “Hypo” refers to percentage of DML that are hypomethylated in high BMI vs. low BMI samples; “Hyper” refers to percentage of DML that are hypermethylated in high BMI vs. low BMI samples. On the Y axis, DMLs (%) refers to the percentage of the total DML that are associated with a particular genomic location (Panels A and D) or CGI relationship (Panels B and E). Percentages may be up to more than 100 % because some probes have been classified with more than one designation. Beta values are equivalent to percent methylation. Panel A: DML when comparing high BMI to low BMI BE cases by genomic region. Panel B: Location of DML when comparing high BMI to low BMI BE cases with respect to CpG island location. Panel C: Box and whiskers plot showing distribution of DML that are hypomethylated in the high vs. low BMI BE cases (left) and hypermethylated in the high vs. low BMI BE cases (right). Panel D: DML when comparing high BMI to low BMI HGD/EAC cases by genomic region. Panel E: Location of DML when comparing high BMI to low BMI HGD/EAC cases with respect to CpG island location. Panel F: Box and whiskers plot showing distribution of DML that are hypomethylated in the high vs. low BMI HGD/EAC cases (left) and hypermethylated in the high vs. low BMI HGD/EAC cases (right)

Full size image

Ultimately, alterations in methylation levels of particular genes in obese patients might promote metaplasia or neoplasia in the esophagus by affecting certain signaling pathways based on the expectation that a subset of the gene loci that show abnormal DNA methylation will have alterations in their expression levels. In order to investigate this, our group assessed the methylation status of CpGs located in genes associated with signaling pathways and biological mediators implicated in obesity-associated cancers [35, 36, 96] in the esophageal tissues from the subjects with low vs. high BMI. We were interested in determining whether alterations in methylation of obesity-related pathways might correlate with BMI status , which would provide a plausible mechanism for obesity-related gene expression changes. As described above, the insulin/IGF-1 pathways are frequently perturbed in obese individuals, and these alterations can be associated with an elevated cancer risk. With regard to these pathways, we observed increased methylation of IGFBP1 (average beta = 0.11 in low BMI cases and 0.27 in high BMI cases) and IRS2 (average beta = 0.11 in low BMI cases and 0.36 in high BMI cases) in the high BMI compared to low BMI BE cases. We also examined molecular pathways associated with adipose inflammation, which has been shown to mediate obesity-related cancer [36], and found the proinflammatory gene IL-1β (IL1B) to be hypermethylated in high vs. low BMI cases when we assessed the combined esophageal tissue sets. We also found hypermethylation of IL1B in the HGD/EAC cases from high BMI subjects. For the combined cases, the average beta was 0.25 in low BMI cases and 0.35 in high BMI cases, and for the HGD/EAC cases, average beta was 0.20 in low BMI cases and 0.38 in high BMI cases. We also evaluated the adiponectin and leptin pathways, which as discussed above have been implicated in obesity-associated cancer [42, 59]. In our studies, we did not observe any differences in the DNA methylation status of genes involved in leptin or adiponectin pathways in any of the esophageal tissue sets in the high vs. low BMI subjects.

Our group also utilized several databases to search for cancer-related or other molecular pathways that might be altered by differential methylation patterns in individuals with high vs. low BMI. We used the NCI Pathway Interaction Database (NCI-PID), Kyoto Encyclopedia of Genes and Genomes (KEGG) database, and the list of Gene Ontology (GO) terms to identify biological processes or pathways that were over- or underrepresented based on genes containing DML between the esophageal tissue sets in the subjects with either high- or low-BMI status. We restricted our NCI-PID analysis to only cancer-associated DML (DML when comparing EAC cases to squamous cases on the microarray) in order to improve the likelihood that altered molecular pathways would be biologically relevant.

Among the BE cases, we found one NCI-PID pathway, “direct p53 effectors,” which includes the differentially methylated gene RDX from our dataset, associated with methylation differences between high and low BMI groups. There were 13 KEGG pathways (including “cell adhesion molecules”) and 77 GO terms (including “response to growth hormone” and “biological adhesion”) that were represented in the differentially methylated genes in the BE samples from the high vs. low BMI subjects. With respect to the EAC cases, there were no NCI-PID pathways that were significantly associated with methylation differences between high and low BMI status after restricting our analysis to only cancer-related genes. There was one KEGG pathway (“Wnt signaling”) and 87 GO terms (such as “tissue morphogenesis” and “response to TGF-beta”) differentially methylated between HGD/EAC cases from subjects with high BMI vs. low BMI (p value <0.05).

TP53 , the gene for p53, is a well-known tumor suppressor gene that is frequently lost early in BE through mutation or loss of heterozygosity (LOH) [69]. TP53 LOH has been shown to identify a subset of BE patients who are at risk for progression to EAC [19, 68]. The finding of differential methylation involving the p53 pathway in BE from subjects with high vs. low BMI suggests a relationship between obesity and DNA methylation of cancer-related genes in the esophagus. Similar results have been found in other studies comparing methylation in obese to lean individuals. In a recent study of 345 breast cancer cases, the majority (87 %) of CpG sites analyzed showed elevated methylation in obese patients, particularly in estrogen receptor-positive tumors. Obesity was associated with the aberrant methylation of cancer-related genes involved with the immune response, cell growth, and DNA repair [33]. Several prior studies have compared DNA methylation in whole blood or peripheral blood leukocytes among obese and non-obese individuals [15, 17, 26]. In two of these studies, the gene HIF3A was found to be hypermethylated in the blood cells and adipose tissue of obese adults, suggesting perturbation of the hypoxia-inducible transcription factor pathway in those with elevated BMI.

Conclusions

Barrett’s esophagus, a metaplastic condition involving the esophagus which develops in the setting of chronic gastro-esophageal reflux and esophageal inflammation, is the precursor lesion for esophageal adenocarcinoma. Thus, GERD is a well-known risk factor for the development of BE and EAC. Obesity, in particular central adiposity, is another important risk factor for the development of these conditions. Traditionally, it has been assumed that obesity augmented the risk of BE/EAC by inducing mechanical or physical changes such as increasing intra-abdominal pressure and/or altering the integrity of the gastro-esophageal junction, leading to increased GERD and reflux esophagitis. However, it has become increasingly clear that the adipose tissue itself, in particular the metabolically active visceral fat more commonly seen in males, may directly promote inflammation and cancer development.

Epigenetic alterations, which are commonly seen in BE and EAC, have been associated with chronic inflammation in the esophagus and also with obesity in other tissues. Data regarding the relationship between aberrant methylation in the esophagus and obesity are limited, although there is preliminary evidence that elevated BMI is associated with altered methylation in BE, dysplastic BE, and EAC tissues. There is also some evidence that genes involved in cancer-related pathways (p53) and pathways implicated in obesity-related cancers and adipose inflammation (insulin, IGF-1) demonstrate altered methylation in individuals with elevated BMI compared to those with low BMI. Future studies, ideally combining gene methylation, gene expression, and demographic data, will be useful to clarify the mechanism by which obesity elicits alterations in DNA methylation that associate with the risk of BE or EAC.