Short Communkation Lewis, Secretor, and ABO Phenotypes, and Sulfomucin Expression in Gastric Intestinal Metaplasia1

The closely interrelated Lewis, secretor, and ABO phenotypes have long been linked to the risk of peptic ulcers and gastric cancer and may modulate the interaction between Helicobacter pylon and the gastric surface epithelium. We explored the association between the expression of sulfomucins in gastric intestinal metaplasia, a known marker of preneoplastic progression, and the expression of Lewis, secretor, and ABO phenotypes, in 523 subjects from Nari#{241}o,Colombia, and 856 subjects from northern Spain. In both study populations, Lewis (a+/b-) and nonsecretor phenotypes showed a significant positive association with the expression of sulfomucins (odds ratios, 2.4 and 2.6, respectively). Introduction Lewis and ABO, major human alloantigens, are structurally interrelated (1) and jointly referred to as “histo-blood group” antigens, given their expression in erythrocytes and multiple tissue locations (2), and as “onco-developmental” antigens for their role in differentiation and oncogenesis (3). They are more abundantly expressed in gastrointestinal epithelia and secretions than in most other tissue locations (1). The biological significance of this observation has not been clarified. The secretor status is defined by the expression (secretor) or lack of expression (nonsecretor) of ABO antigens in body fluids and specific tissue locations such as the gastric surface epithelium (4), where the secretor status also controls the expression of the Lewis phenotype: nonsecretors express the Lewis (a+Ib-) phenotype, and secretors express the Lewis (a-/b+) phenotype. Lewis (a-/b-) individuals do not express Lewis antigens, but still display ABO antigens with either a Received 8/5/96; revised 12/13/96; accepted 1/3/97. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. I This work was supported by grants from “Fondo de Investigacidn Sanitaria”, “Consejerla de Salud del Gobierno Vasco”, “Sociedad Espafiola de Enfermedades del Aparato Digestivo” and by grant P0l-CA-28842 from the National Cancer Institute, National Institutes of Health. 2 To whom requests for reprints should be addressed, at Department of Pathology, Louisiana State University Medical Center, Box P5-1, New Orleans, Louisiana, 701 12. Phone: (504) 568-6035; Fax: (504) 599-1278. secretor or a nonsecretor pattern according to their secretor status (1). In the present study, we explored the association of Lewis, secretor, and ABO phenotypes with the expression of sulfomucins in gastric intestinal metaplasia, in two populations with different frequencies ofblood group phenotypes and at different risk of gastric cancer. This inquiry was prompted by the following preliminary observations: (a) sulfomucins are proven markers of preneoplastic progression in the stomach (5); (b) the expression of sulfomucins in columnar-type cells is a hallmark of intestinal metaplasia type III, incomplete, or colonic, and identifies an advanced stage in the gastric premalignant evolution (5); (c) the simultaneous expression of sulfomucins and aberrant Lewisa antigen is associated with a greater risk than the expression of either marker alone (6); and (d) Lewis, secretor, and ABO phenotypes are related to the risk of peptic ulcers and gastric cancer (7), and they may also affect the adherence of Helicobacter pylon to the gastric epithelium (8). Materials and Methods Subjects of this study were individuals from the general population of Nariflo, Colombia, and consecutive patients referred for endoscopy in various communities of northern Spain. Colombian subjects were enrolled in a long-term study of gastric carcinogenesis. Spanish subjects were enrolled in a study of intestinal metaplasia involving regional hospitals of the Guipuzcoa, Oviedo, Soria. and Navarra provinces. These studies were approved by the Louisiana State University Medical School Institutional Review Board and parallel committees in Colombia and Spain. All participants gave informed, written consent. Four gastric biopsies (one corporal, three antral) were obtained routinely, fixed in 10% buffered formalin, and embedded individually in paraffin. Only subjects with intestinal metaplasia documented by microscopic examination of H&Estained sections were included in this study. The H. pylon status was evaluated with the modified Steiner stain (9). Sulfomucins were identified in areas of metaplasia with high-iron diamine Alcian blue stain (10), and evaluated as absent or present in goblet and nongoblet (columnar) cells (6). Lewis, secretor, and ABO phenotypes were determined by immunohistochemistry in nonmetaplastic gastric mucosa, as reported previously (6). Anti-Lewis5 (7 LE, BioGenex, working dilution 1:250) and anti-Lewist’ (2.25 LE, BioGenex, 1:250) were used to determine the Lewis phenotype. Anti-A (A58l, DAKO Corp., 1 :40), anti-B (A582, DAKO Corp., 1 :40) and anti-H type 2 (A583, DAKO Corp., 1:40) were used to determine the ABO phenotype. Detection of ABO antigens in the surface epithehum defined the secretor phenotype; a negative reaction identified the nonsecretor phenotype. Samples from Colombia were processed at Louisiana State University Medical Center in New Orleans. All samples from Spain were processed at Ar#{225}nzazu Hospital in San Sebastian. Identical histochemistry and immunohistochemistry protocols were used in both study locations. Positive and negative conon August 15, 2017. © 1997 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from 288 Short Communication: Gastric Intestinal Metaplasia 3 The abbreviations used are: OR, odds ratio; CI, confidence interval. Table I Percentage frequency of Lewis, secretor, and ABO phenotypes among individuals with intestinal metaplasia in Colombia and Spain Colombia Spain (n 523) (n 856) Lewis (a+/b-) 3.8 19.2 Lewis (a-/b+) 74.0 70.4 Lewis (a-/b-) 22.2 10.4 Nonsecretor 6.3 20.7 Secretor 93.7 79.3 0 68.6 43.1 A 20.5 47.5 B 9.6 6.2 AB 1.3 3.2 Table 2 Percentage p revalence of sulfomucins by blood grou Colombia and Spain p phenotype in Total Goblet cells Columnar n positive only cells Colombia Lewis (a+/b-) 20 90.0 60.0 30.0 Lewis (a-/b+) 387 57.4 38.0 19.4 Lewis (a-/b-) 1 16 56.9 37.9 19.0 Nonsecretor 33 84.9 57.6 27.3 Secretor 490 56.7 37.5 19.2 0 359 60.2 40.7 19.5 A 107 55.1 37.4 17.7