Primary Motility  Disorders of the  Esophagus
 The Esophageal
 Esophagogastric  Junction

  Browse by Author
  Browse by Movies
Volume: Barrett's Esophagus
Chapter: Etiology and origins of Barrett's epithelium

Is there immunocytochemical evidence to support origin of Barrett's epithelium from adjacent squamous epithelium?

J.A. Salo, E.I.T. Sihvo (Helsinki)

Barrett's esophagus is considered to be a consequence of severe, long-lasting gastroesophageal reflux disease. It is histologically defined as a intestinal epithelium in the tubular esophagus. It is unclear whether columnar Barrett's epithelium originates from real metaplastic change of the squamous esophageal epithelium or whether it is derived from a columnar cell colonization from the gastric cardia.

Intermediate filaments are cytoplasmic fibrils assumed to function as integrators of the cytoplasmic space and anchoring elements of the cell nucleus [1, 2]. The major cell types in mammalian tissues (mesenchyme, muscle, epithelial, neural and astrocytic) express a specific intermediate filament type that can be distinguished by immunocytochemical methods [1-3]. Cytokeratins (CK) are subunit proteins of epithelial cell intermediate filaments that can be identified by immunohistochemical staining techniques [1-3].

The tissue-specific expression of immunoreactive proteins as well as the epitheliumspecific expression of CK type of intermediate filaments has been widely applied in the elucidation of the origin and state of differentiation of tumors, especially different types of carcinomas [1-5]. In epithelial cells of human tissues 20 CK polypeptides have been identified so far [2, 4, 5]. Each epithelium has its typical cytokeratin profile determined genetically, reflecting the origin and state of differentiation of the epithelium [2, 4, 5]. For example, squamous epithelium has a characteristic CK profile different from simple epithelium.

Material and methods

This study included 35 consecutive patients with Barrett's esophagus and 10 normal controls. Antireflux operations had been performed earlier in 12 of the patients, including nine fundoplications and three partial gastrectomies with selective vagotomy and Roux-en-Y reconstruction. Long-lasting gastroesophageal reflux (GER) was the presenting symptom in 32 patients, whereas three patients had no reflux-related symptoms at all. At endoscopy, the length of the Barrett's mucosa extended from 3 to 14 cm (mean 7 cm). All 35 patients had continuous Barrett's esophagus, comprising histologically columnar epithelium including specialized intestinal type with dysplasia in three and adenocarcinoma in four cases.

Indirect immunofluorescence microscopy

For determination of the cytokeratin profile by indirect immunofluorescence microscopy, 10-20 biopsies were taken at endoscopy from normal esophagus, Barrett's mucosa and from the gastric cardia. For control purposes biopsies from normal colon and duodenum were studied. The biopsy specimens were frozen in liquid nitrogen and stored at -88 C until used. For indirect immunofluorescence 4 m, sections were cut, fixed in acetone, and precooled to -20 C, for 10 minutes. Then the specimens were exposed to primary monoclonal antibodies (Mab): the Mab 2A4 against CKs 8, 18 and 19 [6]; the Mab 4F5 against CKs 7, 17 and 19 [6]; the Mab M20 against CK 8 [7]; the Mab 4B11 against CK 18 [6]; the Mab K4.62 against CK 19 [8]; and the Mab KsIA3 against CK13 of squamous epithelia from Sigma Chem. Co. (St. Louis, MO). The Mabs KA11 reacting with some squamous epithelial cell CKs [9], Mab KA5 against CKs 1, 10, 11 present within keratinizing squamous epithelia [9] and KA1, reacting widely with stratified squamous epithelia [9] were kindly provided by Professor R. A. Nagle (University of Arizona). Then, the sections were reacted with fluorescein isoniocyanatecoupled rabbit antimouse IgG antiserum (Capped Laboratories, Orgamon Teknika, West-Chester, PA), cover-stripped and examined in a Leitz Aristoplan microscope. The study was approved by the Ethical Committee of the Department of Surgery at the Helsinki University Central Hospital.


Indirect immunofluorescence microscopy of specimens from normal esophageal epithelium of patients and controls revealed variable immunoreactivity with the Mabs 2A4, K4.62 and 4F5, either solely confined to the basal cell population or brightly to basal and less to supra-basal epithelium and suggesting the expression of CK 7 and 19 as the Mabs M20 and 4B11, CK 8 and 18, respectively, did not reveal any reactivity. On the other hand, the Mabs KS1A3 and KA11 reacted brightly with all epithelial cells and the Mab KA5 with supra-basal epithelial cells. Only the Mabs 4B11, M20, 2A4, K4.62 and 4F5 reacted with normal cardia epithelium, giving a bright reaction of both surface, and glandular epithelial cells, the Mabs KsIIA3 and KA5 giving a negative reaction. A similar negative reaction was obtained with specimens from colon and duodenum. In cases of Barrett's epithelium the reactivity with these Mabs was also consistent. However, the Mab KsIA3 and and KA11 now revealed in different types a bright but somewhat variable epithelial reactivity in both cryptlike and glandular cells. In immunohistochemistry both the esophageal submucosal glands as well as the cardiac glands were negative for CK 13.


Our preliminary results were published already in 1991 [10]. The present study suggests that human Barrett's esophagus has a different cytokeratin profile as compared to the epithelium of normal esophagus and gastric cardia. The principal finding was the abundant expression of CK 13 in normal esophagus and Barrett's epithelium, whereas CK 13 could not be verified in the gastric cardia. Cytokeratin 13 is characteristic of non-keratinized stratified squamous epithelium, such as that of esophagus, cervix and vagina and is not present in the simple glandular epithelium of the cardia [2]. Because of the strong organ specificity and genetic determination of CK, one can assume on the basis of the above findings that Barrett's esophagus originates from a metaplastic change of the squamous esophageal epithelium and is not formed by colonization of the columnar epithelium of gastric cardia into the distal esophagus. In immunohistochemistry both the esophageal submucosal glands as well as cardiac glands were negative for CK 13. The negativity of the esophageal glands for CK 13 may imply that they would not give origin to CK 13 immunoreactive epithelium. The discovery of CKs 8 and 18 in Barrett's mucosa as well as in gastric cardia can be explained by the re-expression of fetal epithelial characteristics [4]. However, Boch et al. [5] found a multilayered epithelium within Barrett's epithelium which expressed CK concurrently both squamous (CK 2, CK 11) and glandular (CK 6, CK 11). Their findings suggest a multipotential cell like the cell of origin of Barrett's epithelium. The metaplastic nature of Barrett's epithelium is also supported by the experiments of Gillen et al. [13] who found out that columnar re-epithelialization of the dog esophagus after mucosectomy originated from cells intrinsic to the esophagus, In contrast, Bremner et al. [14] suggested that, in the presence of GER, Barrett's esophagus arose in the ulcerated dog distal esophagus from proximal migration of columnar epithelial cells of cardia, while in control animals without reflux the denuded ductal epithelium was replaced by downward growth of the esophageal squamous epithelium.


1. Osborn M, Wilson K. Tumor diagnosis by intermediate filament typing: a novel tool for surgical pathology. Lab Invest 1983;48:372-391.

2. Franke WW, Schmid E, Schiller DL, et al. Differentiation related patterns of expression of intermediate size filaments in tissues and cultural cells. Cold Spring Harb Symp Quant Biol 1982;46:431-456.

3. Virtanen I, Heikinheimo K, Hormia M, et al. Expression of intermediate filaments (IF) in tissues and cultured cells. Int J Dev Biol 1989;33:55-61.

4. Moll R, Franke WW, Schiller DL, et al. The catalog of human cytokeratins:patterns of expression in normal epithelia, tumors and cultured cells. Cell 1982;31:11-18.

5. Lasie EB, Alexander CM. Use of keratin antibodies in tumor diagnosis. Semin Cancer Biol 1990;1:165-179.

6. Virtanen I, Miettinen M, Lehto VP, et al. Diagnostic application of monoclonal antibodies to intermediate filaments. Ann NY Acad Sci 1985;455:635-648.

7. Jahn L, Fougnet B, Franke WW. Cytokeratin expression defines subtypes and endothelial cells and smooth muscle cells in xenous laecis and man. Differentiation 1987;36:234-239.

8. Gigi-Leitner O, Geiger B. Antigenic inter relationship between 40 kilodalton cytokeratin polypeptides and desnoplakins. Cell Motil Cytoskeleton 1986;6:628-632.

9. Nagle RB, Moll R, Weidaner H, et al. Different patterns of cytokeratin expression in the normal epithelia of the upper respiratory tract. Differentiation 1985;30:130-135.

10. Salo J, Kivilaakso E, Virtanen I. Barrett's esophagus originates from the squamous esophageal epithelium as judged from its cytokeratin profile. Gastroenterology 1991;100:A153.

11. Boch JA, Shields HM, Antonioli DA, et al. Distribution of cytokeratin markers in Barrett's specialized columnar epithelium. Gastroenterology 1997;112:760-765.

12. Regauer S, Franke WW, Virtanen I. lntermediate filament cytoskeleton of amnion epithelium and cultured amnion epithelial cells: expression of epidermal cytokeratins in cells of a simple epithelium. J Cell Biol 1985;100:997-1009.

13. Gillen P, Keeling P, Byrne PJ, et al. Experimental columnar metaplasia in the canine esophagus. Br J Surg 1988;75:113115.

14. Bremner CG, Lynch VP, Elis FH. Barrett's esophagus: congenital or acquired? An experimental study of esophageal mucosal regeneration in the dog. Surgery 1970;68:209-216.

Publication date: August 2003 OESO©2015