Hybrid Stomach-Intestinal Chromatin States Underlie Human Barrett's Metaplasia

Epigenomic heterogeneity as a source of tumour evolution

In the past decade, remarkable progress in cancer medicine has been achieved by the development of treatments that target DNA sequence variants. However, a purely genetic approach to treatment selection is hampered by the fact that diverse cell states can emerge from the same genotype. In multicellular organisms, cell-state heterogeneity is driven by epigenetic processes that regulate DNA-based functions such as transcription; disruption of these processes is a hallmark of cancer that enables the emergence of defective cell states. Advances in single-cell technologies have unlocked our ability to quantify the epigenomic heterogeneity of tumours and understand its mechanisms, thereby transforming our appreciation of how epigenomic changes drive cancer evolution. This Review explores the idea that epigenomic heterogeneity and plasticity act as a reservoir of cell states and therefore as a source of tumour evolution. Best practices to quantify epigenomic heterogeneity and explore its various causes and consequences are discussed, including epigenomic reprogramming, stochastic changes and lasting memory. The design of new therapeutic approaches to restrict epigenomic heterogeneity, with the long-term objective of limiting cancer development and progression, is also addressed.

Cellular Origins and Lineage Plasticity in Cancer

All cancers arise from normal cells whose progeny acquire the cancer-initiating mutations and epigenetic modifications leading to frank tumorigenesis. The identity of those "cells-of-origin" has historically been a source of controversy across tumor types, as it has not been possible to witness the dynamic events giving rise to human tumors. Genetically engineered mouse models (GEMMs) of cancer provide an invaluable substitute, enabling researchers to interrogate the competence of various naive cellular compartments to initiate tumors in vivo. Researchers using these models have relied on lineage-specific promoters, knowledge of preneoplastic disease states in humans, and technical advances allowing more precise manipulations of the mouse germline. These approaches have given rise to the emerging view that multiple lineages within a given organ may generate tumors with similar histopathology. Here, we review some of the key studies leading to this conclusion in solid tumors and highlight the biological and clinical ramifications.

The pluripotency factor NANOG contributes to mesenchymal plasticity and is predictive for outcome in esophageal adenocarcinoma

BACKGROUND

Despite the advent of neoadjuvant chemoradiotherapy (CRT), overall survival rates of esophageal adenocarcinoma (EAC) remain low. A readily induced mesenchymal transition of EAC cells contributes to resistance to CRT.

METHODS

In this study, we aimed to chart the heterogeneity in cell state transition after CRT and to identify its underpinnings. A panel of 12 esophageal cultures were treated with CRT and ranked by their relative epithelial-mesenchymal plasticity. RNA-sequencing was performed on 100 pre-treatment biopsies. After RNA-sequencing, Ridge regression analysis was applied to correlate gene expression to ranked plasticity, and models were developed to predict mesenchymal transitions in patients. Plasticity score predictions of the three highest significant predictive models were projected on the pre-treatment biopsies and related to clinical outcome data. Motif enrichment analysis of the genes associated with all three models was performed.

RESULTS

This study reveals NANOG as the key associated transcription factor predicting mesenchymal plasticity in EAC. Expression of NANOG in pre-treatment biopsies is highly associated with poor response to neoadjuvant chemoradiation, the occurrence of recurrences, and median overall survival difference in EAC patients (>48 months). Perturbation of NANOG reduces plasticity and resensitizes cell lines, organoid cultures, and patient-derived in vivo grafts.

CONCLUSIONS

In conclusion, NANOG is a key transcription factor in mesenchymal plasticity in EAC and a promising predictive marker for outcome.

Genes Selectively Expressed in Rat Organs

Background

Understanding organic functions at a molecular level is important for scientists to unveil the disease mechanism and to develop diagnostic or therapeutic methods.

Aims

The present study tried to find genes selectively expressed in 11 rat organs, including the adrenal gland, brain, colon, duodenum, heart, ileum, kidney, liver, lung, spleen, and stomach.

Materials and Methods

Three normal male Sprague-Dawley (SD) rats were anesthetized, their organs mentioned above were harvested, and RNA in the fresh organs was extracted. Purified RNA was reversely transcribed and sequenced using the Solexa high-throughput sequencing technique. The abundance of a gene was measured by the expected value of fragments per kilobase of transcript sequence per million base pairs sequenced (FPKM). Genes in organs with the highest expression level were sought out and compared with their median value in organs. If a gene in the highest expressed organ was significantly different (p < 0.05) from that in the medianly expressed organ, accompanied by q value < 0.05, and accounted for more than 70% of the total abundance, the gene was assumed as the selective gene in the organ.

Results & Discussion

The Kyoto Encyclopedia of Genes and Genomes (KEGG), and Gene Ontology (GO) pathways were enriched by the highest expressed genes. Based on the criterion, 1,406 selective genes were screened out, 1,283 of which were described in the gene bank and 123 of which were waiting to be described. KEGG and GO pathways in the organs were partly confirmed by the known understandings and a good portion of the pathways needed further investigation.

Conclusion

The novel selective genes and organic functional pathways are useful for scientists to unveil the mechanisms of the organs at the molecular level, and the selective genes' products are candidate disease markers for organs.

Gastric intestinal metaplasia: progress and remaining challenges

Most gastric cancers arise in the setting of chronic inflammation which alters gland organization, such that acid-pumping parietal cells are lost, and remaining cells undergo metaplastic change in differentiation patterns. From a basic science perspective, recent progress has been made in understanding how atrophy and initial pyloric metaplasia occur. However, pathologists and cancer biologists have long been focused on the development of intestinal metaplasia patterns in this setting. Arguably, much less progress has been made in understanding the mechanisms that lead to the intestinalization seen in chronic atrophic gastritis and pyloric metaplasia. One plausible explanation for this disparity lies in the notable absence of reliable and reproducible small animal models within the field, which would facilitate the investigation of the mechanisms underlying the development of gastric intestinal metaplasia (GIM). This review offers an in-depth exploration of the current state of research in GIM, shedding light on its pivotal role in tumorigenesis. We delve into the histological subtypes of GIM and explore their respective associations with tumor formation. We present the current repertoire of biomarkers utilized to delineate the origins and progression of GIM and provide a comprehensive survey of the available, albeit limited, mouse lines employed for modeling GIM and engage in a discussion regarding potential cell lineages that serve as the origins of GIM. Finally, we expound upon the myriad signaling pathways recognized for their activity in GIM and posit on their potential overlap and interactions that contribute to the ultimate manifestation of the disease phenotype. Through our exhaustive review of the progression from gastric disease to GIM, we aim to establish the groundwork for future research endeavors dedicated to elucidating the etiology of GIM and developing strategies for its prevention and treatment, considering its potential precancerous nature.

The Multidisciplinary Approach and Surgical Management of GE Junction Adenocarcinoma

Gastroesophageal (GE) junction adenocarcinoma is an aggressive malignancy of growing incidence and is associated with public health issues such as obesity and GERD. Management has evolved over the last two decades to incorporate a multidisciplinary approach, including endoscopic intervention, neoadjuvant chemotherapy/chemoradiation, and minimally invasive or more limited surgical approaches. Surgical approaches include esophagectomy, total gastrectomy, and, more recently, proximal gastrectomy. This review analyzes the evidence for and applicability of these varied approaches in management, as well as areas of continued controversy and investigation.

Ectopic expression of HNF4α in Het1A cells induces an invasive phenotype

Barrett's oesophagus (BO) is a pathological condition in which the squamous epithelium of the distal oesophagus is replaced by an intestinal-like columnar epithelium originating from the gastric cardia. Several somatic mutations contribute to the intestinal-like metaplasia. Once these have occurred in a single cell, it will be unable to expand further unless the altered cell can colonise the surrounding squamous epithelium of the oesophagus. The mechanisms by which this happens are still unknown. Here we have established an in vitro system for examining the competitive behaviour of two epithelia. We find that when an oesophageal epithelium model (Het1A cells) is confronted by an intestinal epithelium model (Caco-2 cells), the intestinal cells expand into the oesophageal domain. In this case the boundary involves overgrowth by the Caco-2 cells and the formation of isolated colonies. Two key transcription factors, normally involved in intestinal development, HNF4α and CDX2, are both expressed in BO. We examined the competitive ability of Het1A cells stably expressing HNF4α or CDX2 and placed in confrontation with unmodified Het1A cells. The key result is that stable expression of HNF4α, but not CDX2, increased the ability of the cells to migrate and push into the unmodified Het1A domain. In this situation the boundary between the cell types is a sharp one, as is normally seen in BO. The experiments were conducted using a variety of extracellular substrates, which all tended to increase the cell migration compared to uncoated plastic. These data provide evidence that HNF4α expression could have a potential role in the competitive spread of BO into the oesophagus as HNF4α increases the ability of cells to invade into the adjacent stratified squamous epithelium, thus enabling a single mutant cell eventually to generate a macroscopic patch of metaplasia.

Strong Annexin A10 Expression Supports a Pancreatic Primary and Combined Annexin A10, Claudin 18, and SOX2 Expression Supports an Esophagogastric Origin in Carcinomas of Unknown Primary

Primary tumor site determination for gastrointestinal (GI) tract and pancreaticobiliary (PB) tree carcinomas that present as metastasis of unknown primary can be problematic. Annexin A10 (ANXA10), claudin 18 (CLDN18), and trefoil factor 1 (TFF1) have been identified through expression profiling as markers of gastric lineage commitment; sex-determining region Y (SRY)-box transcription factor 2 (SOX2) expression has been reported in several tumor types, including gastric adenocarcinomas. We evaluated the diagnostic utility of immunohistochemistry for ANXA10, CLDN18, SOX2, and TFF1 for determining the site of origin for GI/PB adenocarcinomas. Immunohistochemistry for all 4 markers was performed on tissue microarrays including 559 GI/PB tumors and 421 other tumors. H-scores were calculated as the product of the intensity (0 to 3) and extent (percentage, 0% to 100%) of staining. Positive staining was defined as >5% staining. ANXA10 expression was most frequent in pancreatic adenocarcinomas when compared with all other GI/PB tumors (96.4% vs. 43.5%, P <0.001). Strong staining for ANXA10 (H-score ≥200) distinguished pancreatic ductal adenocarcinoma from intrahepatic cholangiocarcinoma and adenocarcinomas of the gallbladder and colorectum (69.6% vs. 0%, P <0.001). Triple positivity for ANXA10, CLDN18, and SOX2 was more frequent in esophagogastric tumors than in other GI/PB tumors (22.6% vs. 4.1%; P <0.001). TFF1 expression was observed in nearly all tumor types. Staining for ANXA10, CLDN18, and SOX2 as part of a panel may aid in distinguishing esophagogastric adenocarcinomas from lower GI/PB tumors. ANXA10 staining may be particularly useful in distinguishing pancreatic adenocarcinomas from intrahepatic cholangiocarcinoma and adenocarcinomas of the gallbladder and colorectum.

Shared features of metaplasia and the development of adenocarcinoma in the stomach and esophagus

Introduction: Plasticity is an inherent property of the normal gastrointestinal tract allowing for appropriate response to injury and healing. However, the aberrancy of adaptable responses is also beginning to be recognized as a driver during cancer development and progression. Gastric and esophageal malignancies remain leading causes of cancer-related death globally as there are limited early disease diagnostic tools and paucity of new effective treatments. Gastric and esophageal adenocarcinomas share intestinal metaplasia as a key precancerous precursor lesion.

Methods: Here, we utilize an upper GI tract patient-derived tissue microarray that encompasses the sequential development of cancer from normal tissues to illustrate the expression of a set of metaplastic markers.

Results: We report that in contrast to gastric intestinal metaplasia, which has traits of both incomplete and complete intestinal metaplasia, Barrett's esophagus (i.e., esophageal intestinal metaplasia) demonstrates hallmarks of incomplete intestinal metaplasia. Specifically, this prevalent incomplete intestinal metaplasia seen in Barrett's esophagus manifests as concurrent development and expression of both gastric and intestinal traits. Additionally, many gastric and esophageal cancers display a loss of or a decrease in these characteristic differentiated cell properties, demonstrating the plasticity of molecular pathways associated with the development of these cancers.

Discussion: Further understanding of the commonalities and differences governing the development of upper GI tract intestinal metaplasias and their progression to cancer will lead to improved diagnostic and therapeutic avenues.

Targeting CLDN18.2 in cancers of the gastrointestinal tract: New drugs and new indications

Cancers of the gastrointestinal (GI) tract greatly contribute to the global cancer burden and cancer-related death. Claudin-18.2(CLDN18.2), a transmembrane protein, is a major component of tight junctions and plays an important role in the maintenance of barrier function. Its characteristic widespread expression in tumour tissues and its exposed extracellular loops make it an ideal target for researchers to develop targeted strategies and immunotherapies for cancers of the GI tract. In the present review, we focus on the expression pattern of CLDN18.2 and its clinical significance in GI cancer. We also discuss the tumour-promoting and/or tumour-inhibiting functions of CLDN18.2, the mechanisms regulating its expression, and the current progress regarding the development of drugs targeting CLDN18.2 in clinical research.

SATB2 loss in inflammatory bowel disease-associated small intestinal metaplasia of the distal colon

Epithelial metaplasia is a common adaptation to chronic inflammatory processes and can be associated with increased risk of dysplasia and cancer. The distal colon of patients with inflammatory bowel disease (IBD) commonly shows crypt architectural distortion and Paneth cell metaplasia (PCM), and IBD patients also carry increased risk of colitis-associated dysplasia and cancer (CAC). Loss of SATB2 expression (Special AT-rich binding 2 protein, a colon-restricted chromatin remodeler) has recently been shown to distinguish colitis-associated dysplasia and CAC from sporadic disease. Here we report non-diffuse heterogeneous patterns of SATB2 loss across non-dysplastic distal colon biopsies from IBD patients (n=20). This cohort was specifically curated to include biopsies with well-developed histologic features of villiform growth and PCM. Notably, CDX2 was strongly expressed and P53 showed a wild-type immunolabeling pattern across our non-dysplastic cohort, regardless of SATB2 immunolabeling pattern. Our findings fit with recent murine studies in which colon-specific Satb2 deletion resulted in histologic conversion of colonic mucosa to small intestinal-like mucosa, including emergence of villi and Paneth cells. Taken together, we show that SATB2 loss is associated with a pre-neoplastic metaplastic response to chronic injury in human IBD and chronic colitis, reframing PCM more broadly as small intestinal metaplasia. We propose that inflammation-associated SATB2 loss mediates a remodeled chromatin landscape permissive for dysplasia and CAC.

COOBoostR: An Extreme Gradient Boosting-Based Tool for Robust Tissue or Cell-of-Origin Prediction of Tumors

We present here COOBoostR, a computational method designed for the putative prediction of the tissue- or cell-of-origin of various cancer types. COOBoostR leverages regional somatic mutation density information and chromatin mark features to be applied to an extreme gradient boosting-based machine-learning algorithm. COOBoostR ranks chromatin marks from various tissue and cell types, which best explain the somatic mutation density landscape of any sample of interest. A specific tissue or cell type matching the chromatin mark feature with highest explanatory power is designated as a potential tissue- or cell-of-origin. Through integrating either ChIP-seq based chromatin data, along with regional somatic mutation density data derived from normal cells/tissue, precancerous lesions, and cancer types, we show that COOBoostR outperforms existing random forest-based methods in prediction speed, with comparable or better tissue or cell-of-origin prediction performance (prediction accuracy-normal cells/tissue: 76.99%, precancerous lesions: 95.65%, cancer cells: 89.39%). In addition, our results suggest a dynamic somatic mutation accumulation at the normal tissue or cell stage which could be intertwined with the changes in open chromatin marks and enhancer sites. These results further represent chromatin marks shaping the somatic mutation landscape at the early stage of mutation accumulation, possibly even before the initiation of precancerous lesions or neoplasia.

PredictiveNetwork: predictive gene network estimation with application to gastric cancer drug response-predictive network analysis

Background

Gene regulatory networks have garnered a large amount of attention to understand disease mechanisms caused by complex molecular network interactions. These networks have been applied to predict specific clinical characteristics, e.g., cancer, pathogenicity, and anti-cancer drug sensitivity. However, in most previous studies using network-based prediction, the gene networks were estimated first, and predicted clinical characteristics based on pre-estimated networks. Thus, the estimated networks cannot describe clinical characteristic-specific gene regulatory systems. Furthermore, existing computational methods were developed from algorithmic and mathematics viewpoints, without considering network biology.

Results

To effectively predict clinical characteristics and estimate gene networks that provide critical insights into understanding the biological mechanisms involved in a clinical characteristic, we propose a novel strategy for predictive gene network estimation. The proposed strategy simultaneously performs gene network estimation and prediction of the clinical characteristic. In this strategy, the gene network is estimated with minimal network estimation and prediction errors. We incorporate network biology by assuming that neighboring genes in a network have similar biological functions, while hub genes play key roles in biological processes. Thus, the proposed method provides interpretable prediction results and enables us to uncover biologically reliable marker identification. Monte Carlo simulations shows the effectiveness of our method for feature selection in gene estimation and prediction with excellent prediction accuracy. We applied the proposed strategy to construct gastric cancer drug-responsive networks.

Conclusion

We identified gastric drug response predictive markers and drug sensitivity/resistance-specific markers, AKR1B10, AKR1C3, ANXA10, and ZNF165, based on GDSC data analysis. Our results for identifying drug sensitive and resistant specific molecular interplay are strongly supported by previous studies. We expect that the proposed strategy will be a useful tool for uncovering crucial molecular interactions involved a specific biological mechanism, such as cancer progression or acquired drug resistance.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-022-04871-z.

Single cell cancer epigenetics

Bulk sequencing methodologies have allowed us to make great progress in cancer research. Unfortunately, these techniques lack the resolution to fully unravel the epigenetic mechanisms that govern tumor heterogeneity. Consequently, many novel single cell-sequencing methodologies have been developed over the past decade, allowing us to explore the epigenetic components that regulate different aspects of cancer heterogeneity, namely: clonal heterogeneity, tumor microenvironment (TME), spatial organization, intratumoral differentiation programs, metastasis, and resistance mechanisms. In this review, we explore the different sequencing techniques that enable researchers to study different aspects of epigenetics (DNA methylation, chromatin accessibility, histone modifications, DNA-protein interactions, and chromatin 3D architecture) at the single cell level, their potential applications in cancer, and their current technical limitations.

A human Barrett's esophagus organoid system reveals epithelial-mesenchymal plasticity induced by acid and bile salts

The pathogenesis of subsquamous intestinal metaplasia (SSIM), in which glands of Barrett's esophagus (BE) are buried under esophageal squamous epithelium, is unknown. In a rat model of reflux esophagitis, we found that columnar-lined esophagus developed via a wound-healing process involving epithelial-mesenchymal plasticity (EMP) that buried glands under ulcerated squamous epithelium. To explore a role for reflux-induced EMP in BE, we established and characterized human Barrett's organoids and sought evidence of EMP after treatment with acidic bile salts (AB). We optimized media to grow human BE organoids from immortalized human Barrett's cells and from BE biopsies from seven patients, and we characterized histological, morphological, and molecular features of organoid development. Features and markers of EMP were explored following organoid exposure to AB, with and without a collagen I (COL1) matrix to simulate a wound-healing environment. All media successfully initiated organoid growth, but advanced DMEM/F12 (aDMEM) was best at sustaining organoid viability. Using aDMEM, organoids comprising nongoblet and goblet columnar cells that expressed gastric and intestinal cell markers were generated from BE biopsies of all seven patients. After AB treatment, early-stage Barrett's organoids exhibited EMP with loss of membranous E-cadherin and increased protrusive cell migration, events significantly enhanced by COL1. Using human BE biopsies, we have established Barrett's organoids that recapitulate key histological and molecular features of BE to serve as high-fidelity BE models. Our findings suggest that reflux can induce EMP in human BE, potentially enabling Barrett's cells to migrate under adjacent squamous epithelium to form SSIM.NEW & NOTEWORTHY Using Barrett's esophagus (BE) biopsies, we established organoids recapitulating key BE features. During early stages of organoid development, a GERD-like wound environment-induced features of epithelial-mesenchymal plasticity (EMP) in Barrett's progenitor cells, suggesting that reflux-induced EMP can enable Barrett's cells to migrate underneath squamous epithelium to form subsquamous intestinal metaplasia, a condition that may underlie Barrett's cancers that escape detection by endoscopic surveillance, and recurrences of Barrett's metaplasia following endoscopic eradication therapy.

Impact of the Tumor Microenvironment for Esophageal Tumor Development—An Opportunity for Prevention?

Simple Summary

Researchers increasingly appreciate the tumor microenvironment (TME) for its role in the development and therapy resistance of cancers like esophageal adenocarcinoma. A better understanding of the TME fueling carcinogenesis is necessary for tailored prevention and therapies. Here, we highlight recent insights into tumor initiation, interactions with the immune system and possible novel preventative measures.

Abstract

Despite therapeutical advancements, and in contrast to other malignancies, esophageal adenocarcinoma (EAC) prognosis remains dismal while the incidence has markedly increased worldwide over the past decades. EAC is a malignancy of the distal esophageal squamous epithelium at the squamocolumnar junction with gastric cells expanding into the esophagus. Most EAC patients have a history of Barret’s esophagus (BE), a metaplastic adaption to chronic reflux, initially causing an inflammatory microenvironment. Thus, the immune system is highly involved early on in disease development and progression. Normally, anti-tumor immunity could prevent carcinogenesis but in rare cases BE still progresses over a dysplastic intermediate state to EAC. The inflammatory milieu during the initial esophagitis phase changes to a tolerogenic immune environment in BE, and back to pro-inflammatory conditions in dysplasia and finally to an immune-suppressive tumor microenvironment in EAC. Consequently, there is a huge interest in understanding the underpinnings that lead to the inflammation driven stepwise progression of the disease. Since knowledge about the constellations of the various involved cells and signaling molecules is currently fragmentary, a comprehensive description of these changes is needed, allowing better preventative measures, diagnosis, and novel therapeutic targets.

High-Resolution ATAC-Seq Analysis of Frozen Clinical Tissues

The ATAC-seq method enables the genome-wide analysis of accessible chromatin revealing transcriptionally active and poised regulatory elements. The ATAC-seq analysis of clinical specimens at a single-cell resolution reveals the cellular composition of the tissue contributing to the understanding of intra-tissue heterogeneity. Here we describe our method for nuclei isolation from frozen specimens with wide applicability across tissue types, producing nuclei suitable for a number of molecular profiling methods including ATAC-seq in bulk and at a single-cell resolution.

Transcription factor-mediated intestinal metaplasia and the role of a shadow enhancer

Here, Singh et al. show extensive but selective recruitment of intestinal enhancers by CDX2 in gastric cells and that HNF4A-mediated ectopic CDX2 expression in the stomach occurs through a conserved shadow cis-element. These findings identify mechanisms for TF-driven intestinal metaplasia and a likely pathogenic TF hierarchy.

Barrett's esophagus (BE) and gastric intestinal metaplasia are related premalignant conditions in which areas of human stomach epithelium express mixed gastric and intestinal features. Intestinal transcription factors (TFs) are expressed in both conditions, with unclear causal roles and cis-regulatory mechanisms. Ectopic CDX2 reprogrammed isogenic mouse stomach organoid lines to a hybrid stomach–intestinal state transcriptionally similar to clinical metaplasia; squamous esophageal organoids resisted this CDX2-mediated effect. Reprogramming was associated with induced activity at thousands of previously inaccessible intestine-restricted enhancers, where CDX2 occupied DNA directly. HNF4A, a TF recently implicated in BE pathogenesis, induced weaker intestinalization by binding a novel shadow Cdx2 enhancer and hence activating Cdx2 expression. CRISPR/Cas9-mediated germline deletion of that cis-element demonstrated its requirement in Cdx2 induction and in the resulting activation of intestinal genes in stomach cells. dCas9-conjugated KRAB repression mapped this activity to the shadow enhancer's HNF4A binding site. Altogether, we show extensive but selective recruitment of intestinal enhancers by CDX2 in gastric cells and that HNF4A-mediated ectopic CDX2 expression in the stomach occurs through a conserved shadow cis-element. These findings identify mechanisms for TF-driven intestinal metaplasia and a likely pathogenic TF hierarchy.

Stem cells and origins of cancer in the upper gastrointestinal tract

The esophagus and stomach, joined by a unique transitional zone, contain actively dividing epithelial stem cells required for organ homeostasis. Upon prolonged inflammation, epithelial cells in both organs can undergo a cell fate switch leading to intestinal metaplasia, predisposing to malignancy. Here we discuss the biology of gastroesophageal stem cells and their role as cells of origin in cancer. We summarize the interactions between the stromal niche and gastroesophageal stem cells in metaplasia and early expansion of mutated stem-cell-derived clones during carcinogenesis. Finally, we review new approaches under development to better study gastroesophageal stem cells and advance the field.