Primary Motility  Disorders of the  Esophagus
 The Esophageal
 Esophagogastric  Junction

  Browse by Author
  Browse by Movies
Volume: Barrett's Esophagus
Chapter: Dysplasia

Extrapolating from an ex vivo model of Barrett's esophagus, is it possible to define a cell population with potential high risk for dysplasia?

R.C. Fitzgerald, R. Lascar, B.S. Kaur, G. Triadafilopoulos (London, Detroit, Palo Alto, Stanford)

Adenocarcinoma in Barrett's esophagus does not arise de novo but rather follows a multistep process in which the metaplastic epithelium sequentially develops low-grade and highgrade dysplasia, early adenocarcinoma and eventually invasive cancer. It is presently unclear why only a fraction of patients with Barrett's esophagus develop dysplasia and/or adenocarcinoma and what genetic, molecular, or environmental factors are involved [1]. Identifying a cell population with high risk for dysplasia would - if clinically applicable - reduce the frequency of surveillance in patients with Barrett's esophagus.

Recent studies using endoscopic mucosal biopsy specimens of patients with Barrett's esophagus have allowed the recognition of several molecular changes in this epithelium that are associated with dysplasia and adenocarcinoma. Furthermore, experiments with ex vivo organ cultures of Barrett's samples have identified that several components of the gastroduodenal refluxate, either alone or in combination, may further enhance these molecular changes. Using this ex vivo model, we have identified the following factors that may predispose a group of Barrett's cells to dysplasia.

Cell populations at risk for dysplasia

Increased cell proliferation

Expanded proliferative compartment and deregulated proliferating cell nuclear antigen (PCNA) expression is noted in various premalignant and malignant states. Uncontrolled cell proliferation in Barrett's esophagus patients is accompanied by genomic alterations. In contrast, hypoproliferative states are protective against tumor formation. PCNA immunolocalization is a widely available and simple technique and a valid indicator of cellular proliferation in Barrett's esophagus. In organ culture experiments using mucosal biopsy samples of Barrett's esophagus, PCNA expression correlates with 3H-thymidine labeling [2]. In addition, there is a strong positive (r = 0.76) correlation between PCNA (a marker of proliferation) and degree of dysplasia in Barrett's esophagus [3]. Therefore, Barrett's esophagus cell populations exhibiting high proliferative activity are predisposing to dysplasia and cancer formation. In turn, efforts to reduce proliferation in Barrett's esophagus should be expected to reduce dysplasia risk.

Decreased differentiation

The Barrett's esophagus epithelium is heterogeneous, since it may contain cellular characteristics of the intestinal mucosa, including goblet cells that contain sialomucins, and characteristics of the gastric mucosa, such as precursor mucous vesicles. Notably, the degree of cell differentiation varies considerably across regions of Barrett's esophagus as evidenced by studies using Villin, an actin-binding protein of the microvillus, and Ep-cam, an epithelial glycoprotein, as differentiation markers [4]. Villin is barely detectable in patients with esophageal adenocarcinoma, which suggests that cell differentiation may be inversely related to disease progression in Barrett's esophagus.

The impact of acid exposure on cell proliferation and differentiation was evaluated ex vivo using biopsy specimens from Barrett's esophagus and normal esophagus [2]. In order to simulate physiologic conditions, the explants were exposed in culture to acidified medium (pH 3.5) either continuously for 24 hours or as a 1-hour pulse and compared with samples cultured in a pH 7.4 media. Cell differentiation was quantified by Villin expression, which was present in 25% of Barrett's esophagus specimens before organ culture. After 6 or 24 hours of continuous exposure to acid, Villin expression was detected in 50% and 83% of specimens, respectively, whereas it did not increase when incubated in pH 7.4 media. The changes in Villin expression correlated with ultra-structural maturation of the brush border, which contained long, fairly uniform micro-villi on the apical surface. In contrast, Villin expression was not increased by pulse exposure to acid nor was it detected in any specimens from normal esophagus.

In the same study, cell proliferation (see [1], above) was determined by tritiated thymidine incorporation and PCNA expression. Cell proliferation in Barrett's esophagus specimens was higher than in normal esophageal specimens under all pH conditions tested. Pulse acid exposure increased cell proliferation in Barrett's esophagus specimens, while continuous acid exposure reduced cell proliferation compared with incubation in neutral pH. After 1-hour pulse of acid, thymidine incorporation was 6-fold higher in Barrett's esophagus specimens than in control specimens. Staining of PCNA in Barrett's esophagus specimens also increased significantly after pulse acid from 10% of gland cells and 1% of surface cells after incubation at neutral pH to 35% of gland cells and 12% of surface cells after pulse acid exposure.

Therefore, acid affects cell proliferation and differentiation in Barrett's esophagus in a reciprocal fashion, with the effect being dependent on the pattern of acid exposure. Continuous acid exposure promotes a differentiated phenotype, whereas pulse acid exposure favors an undifferentiated phenotype. If these data accurately reflect what happens in Barrett's esophagus, then variable patterns of esophageal acid exposure typical with reflux may be expected to contribute to the heterogeneous epithelium seen in Barrett's esophagus and the variable risk for progression to dysplasia and adenocarcinoma. Cells exposed to pulses of acid would be expected to proliferate preferentially and in turn have a higher risk for developing dysplasia. In comparison, cells exposed to continuous acid would be expected to undergo differentiation and have a lower risk for progressing to dysplasia. These data have implications for therapy. Since continuous acid exposure rarely if ever occurs in Barrett's esophagus and patients have pulses of acid reflux at baseline, acid suppression will need to be very profound and continuous to completely abolish esophageal acid exposure and favor a more differentiated state in Barrett's esophagus epithelium.

Increased cyclooxygenase-2 expression

Cyclooxygenase-2 (COX-2), or prostaglandin synthase-2, is a membrane-bound glycoprotein that functions as a rate-limiting enzyme in the generation of prostanoids from arachidonic acid. COX-2 expression is restricted in certain tissues but it may be induced in response to various stimuli. Recent evidence has implicated COX-2 in colorectal, gastric and esophageal carcinogenesis. For example, increased COX-2 expression was demonstrated in colorectal adenomas and carcinomas, gastric carcinomas, and Barrett's esophagus and adenocarcinomas. Also, COX-2 inhibition has resulted in suppression of neoplastic polyps in the APC knockout mouse, a model of familial adenomatous polyposis.

We have examined the pattern of COX-2 expression in endoscopic esophageal mucosal biopsy specimens of patients with Barrett's esophagus, Barrett's dysplasia and esophageal adenocarcinoma by immunoblotting and immunohistochemistry [5]. Immunoblots revealed minimal constitutive expression of COX-2 in normal esophagus and duodenum. COX-2 protein expression was significantly higher in patients with Barrett's metaplasia, dysplasia and adenocarcinoma, when compared to normal squamous esophageal or columnar duodenal epithelia and was heterogeneous in different regions of the Barrett's esophagus surface. Immunohistochemistry revealed prominent staining in the glands of Barrett's esophagus, dysplasia and adenocarcinoma and faint staining in the basal layers of squamous esophagus and the surface of the duodenum. In response to pulses of acid or bile salts in an ex vivo organ culture system, COX-2 expression increased significantly in Barrett's esophagus tissues and this effect was attenuated by the selective COX-2 inhibitor NS-398. These changes in COX-2 expression were also paralleled by changes in prostaglandin E2 release in the organ culture media. These results suggest that COX-2 expression and its associated PGE2 release play a role in Barrett's esophagus progression to dysplasia and esophageal adenocarcinoma. Such COX-2 regulation of expression and activity by exposure to acid or bile salts may predispose certain populations of Barrett's esophagus cells to dysplasia and neoplasia.

Increased PKCe expression and activity

Protein kinase C (PKC) is comprised of a family of more than ten closely related isoforms that regulate various cellular processes by phosphorylation of proteins on serine and threonine residues. PKC is a key enzyme in signal transduction and mediates a number of functions in eukaryotic cells, including the control of growth and differentiation. PKC isoforms exhibit partial homology, however each isoform also has a unique conserved sequence that may determine a distinct role for each enzyme in signal transduction. PKC isoenzymes are unique, not only with respect to primary structure, but also on the basis of expression patterns and their subcellular localization.

To further understand the molecular mechanisms of bile salt-mediated cell proliferation, we have examined the possible role of PKC in our ex vivo model of Barrett's esophagus. Our experiments demonstrated a bile salt-mediated modulation of PKC activation and they identified increased expression of a specific isoform, PKCe, in response to bile salt pulse that correlates well with increased proliferating cell nuclear antigen (PCNA) expression as a marker of proliferation [6].

Exposure to pulses of acid or bile salts

In the organ culture model of Barrett's esophagus, cell proliferation markers (PCNA and 3H-thymidine incorporation) increase with 1-hour pulses of acid (pH 3.5) or bile salts (1mM). In contrast, cell proliferation markers (PCNA and 3H-thymidine incorporation) decrease with pulses of combined acid (pH 3.5) with bile salts [6]. These results raise the possibility that variation in acid or bile exposure may contribute to the proliferative alterations and the molecular and structural heterogeneity of Barrett's esophagus and the risk for dysplasia (Figure 1).

Figure 1. Hypothetical model for the variable effect of bile salts with or without acid on Barrett's esophagus in vivo. The effect of multiple pulses of bile salts without acid and multiple pulses of bile salts with acid on cell proliferation is compared with the baseline condition of no acid or bile salt exposure. This dynamic effect of bile salts with or without acid results in a heterogeneous cell population with potential implications on dysplasia risk (From [6]).

Increased sodium-hydrogen exchanger activity

The sodium-hydrogen exchanger (NHE) controls intracellular pH (pHi) and cell volume in many cell systems. NHE is an acid extruder in response to pulses of extra-cellular acid exposure. At the end of an acid reflux event, enhanced NHE activity leads to an intracellular alkaline overshoot effect that accelerates cells into S phase of cell cycle [7]. Hence, populations of Barrett's esophagus cells with enhanced NHE activity may have a survival advantage and possibly progress to dysplasia.

Increased transforming growth factor beta expression

Transforming growth factor beta-one (TGF-beta-1) is a multifunctional cytokine that may mediate many key events in normal growth and development as well as tissue repair after injury. Although TGF-beta was originally recognized by its ability to transform normal rat fibroblasts in vitro, it has been subsequently shown that it is chemo-attractant for fibroblasts, stimulates fibroblast proliferation and enhances fibroblast collagen and fibronectin synthesis, while inhibiting collagenase gene expression in vitro.

Immunoblot analysis of endoscopic mucosal biopsy homogenates has revealed the expression of TGF-beta-1 to be significantly higher in patients with erosive esophagitis when compared to normal squamous esophageal epithelium [8]. When studied repeatedly over time in a select group of patients with esophagitis and Barrett's esophagus, TGF-beta1 expression remained increased ( two to sixfold over control). As compared to both squamous esophageal epithelium and columnar duodenal epithelium, TGF-beta-1 expression was even more pronounced in Barrett's esophagus and in esophageal adenocarcinoma. Immunoperoxidase staining complemented the western analysis results. Immunoblotting for TGF-beta-receptor II revealed an increase in receptor expression only in esophagitis and not in Barrett's esophagus or adenocarcinoma. In an ex vivo organ culture system, TGF-beta-1 expression progressively decreased in an acid-independent fashion.

It appears therefore that although TGF-beta-1 is constitutively expressed in normal esophageal epithelium, its expression is enhanced in the inflamed epithelium of gastroesophageal reflux disease, the metaplastic Barrett's esophagus epithelium, and in esophageal adenocarcinoma in a progressively increasing fashion. The consistent and striking over-expression of this cytokine in the premalignant and malignant Barrett's epithelia, together with the strong historical correlate of elevated TGF-beta-1 in inflamed and malignant epithelia, suggest that expression of this cytokine may put such epithelia at risk for dysplasia and cancer.

Decreased E-cadherin expression

Cadherins are cell adhesion molecules that play a vital role in cell-cell adhesion; loss or down-regulation of their expression in various cell populations has been implicated in neoplasia. Expression of E-cadherin has been found to be significantly lower in Barrett's esophagus samples, compared with normal squamous epithelium. Furthermore, an even lower expression was noted in adenocarcinoma samples as compared to normal controls or Barrett's esophagus samples, suggestive a progressive down-regulation of expression as the metaplasia progresses to dysplasia and neoplasia [9].

Src activation in Barrett's esophagus

The cellular oncogene c-src encodes for Src tyrosine kinase and is elevated in various preneoplastic epithelia, such as colon polyps or the inflamed epithelia of ulcerative colitis. Src specific activity is 3 to 4-fold higher in Barrett's esophagus and 6-fold higher in Barrett's adenocarcinoma as compared to normal esophagus or duodenum. Different regions of Barrett's esophagus from the same patient have shown heterogeneity in Src activity compared to the uniform Src activity observed in different regions of normal esophagus or duodenum. Furthermore, Src activation is an early event, before the development of dysplasia. Hence, the activation and micro-heterogeneous distribution of Src in premalignant and malignant Barrett's esophagus epithelia, together with the strong historical correlate Src activity in cell transformation, suggest that this kinase may play a part in the Barrett's esophagus progression to malignancy [10].


From the aforementioned data, one could theoretically define a patient population with Barrett's esophagus who would be at high risk to develop dysplasia and adenocarcinoma. Such group of patients would exhibit increased PCNA expression, decreased Villin expression, increased COX-2 expression and PGE2 release, increased PKCe expression and activity, increased TGF-beta expression, decreased e-cadherin expression and increased Src specific activity. However, the mechanisms and regulation of these molecular changes and their interaction with environmental factors, such as acid and/or bile are still unknown and will require further study.


1. Fitzgerald RC, Triadafilopoulos G. Recent developments in the molecular characterization of Barrett's esophagus. Dig Dis 1998;16:63-80.

2. Fitzgerald RC, Omary MB, Triadafilopoulos G. Dynamic effects of acid on Barrett's esophagus: an ex vivo proliferation and differentiation model. J Clin Invest 1996;98:2120-2128.

3. Ouatu-Lascar R, Fitzgerald RC, Triadafilopoulos G. Differentiation and proliferation in Barrett's esophagus and the effects of acid suppression. Gastroenterology 1999;117:327-335.

4. Kumble S, Fajardo L, Omary MB, Triadafilopoulos G. Multifocal heterogeneity in Villin and Ep-CAM expression in Barrett's esophagus. Int J Cancer 1996;66:48-54.

5. Shirvani VN, Ouatu-Lascar R, Kaur BS, Omary MB, Triadafilopoulos G. Cyclooxygenase-2 expression in Barrett's esophagus and esophageal adenocarcinoma: ex vivo induction by bile salts and acid exposure. Gastroenterology 2000;118:487-496.

6. Kaur BS, Ouatu-Lascar R, Fitzgerald RC, Omary MB, Triadafilopoulos G. Bile salts induce or blunt cell proliferation in Barrett's esophagus in an acid-dependent fashion. Am J Physiol (Gastrointest Liver Physiol) 2000;278:G1000-1009.

7. Fitzgerald RC, Omary MB, Triadafilopoulos G. Altered sodium-hydrogen exchange activity is a mechanism for acidinduced hyperproliferation in Barrett's esophagus. Am J Physiol 1998;275:G47-55.

8. Triadafilopoulos G. What conclusions can be drawn from the enhanced expression of growth factors, cell adhesive molecules and oncogenes in the process of neoplasia? In: Giuli R, ed. The esophagogastric junction. Paris:John Libbey Eurotext, 1998:978-986.

9. Swami S, Kumble S, Triadafilopoulos G. E-cadherin expression in gastroesophageal reflux disease, Barrett's esophagus, and esophageal adenocarcinoma: an immunohistological and immunoblot study. Am J Gastroenterol 1995;90:18081813.

10. Kumble S, Omary MB, Cartwright C, Triadafilopoulos G. Src activation in malignant and premalignant lesions of Barrett's esophagus. Gastroenterology 1997;112:348-356.

Publication date: August 2003 OESO©2015