Surface Hydrophobicity of Gastric Mucosa in Peptic Ulcer Disease

Transcription

1 GASTROENTEROLOGY 199;98: Surface Hydrophobicity of Gastric Mucosa in Peptic Ulcer Disease Relationship to Gastritis and Campylobacter pylori Infection R. T. SPYCHAL, P. M. GOGGIN, J. M. MARRERO, S. H. SAVERYMUTTU, C. W. YU, C. M. CORBISHLEY, J. D. MAXWELL, and T. C. NORTHFIELD Departments of Medicine and Histopathology, St. George's Hospital Medical School, Cranmer Terrace, Tooting, London, England. The hydrophobicity of biopsy specimens of gastric mucosa in 228 dyspeptic subjects undergoing diagnostic endoscopy was assessed by measuring the plateauadvancing contact angle of saline drops using a goniometer. Subjects with duodenal ulcers (n = 49) and gastric ulcers (n = 17) had significantly lower mean contact angles than controls (n = 124) without ulcer (57 in duodenal ulcer, 59 in gastric ulcer vs. 66 in controls; p <.1). There was no change in contact angle after healing with H 2 -receptor antagonists by comparison with pretreatment (59 vs. 56 for duodenal ulcer, n = 15; 57 vs. 59 for gastric ulcer, n = 5). Controls with gastritis had lower contact angles than those without (61, n = 5, vs. 7, n = 63; P <.1). The presence of Campylobacter pylori was associated with a significant decrease in contact angle in controls (59, n = 39, vs. 7, n = 75; P <.1). Peptic ulcer disease has traditionally been explained in terms of an imbalance between luminal aggression and mucosal protection (1). A proposed defense mechanism, based on animal studies, is the physical barrier provided by a hydrophobic surface that repels aqueous solutions, including acid (2,3). The inherent hydrophobicity of a tissue is measured by the plateau-advancing contact angle of a drop of saline (2-6). We have shown that hydrophobicity can be measured reliably on human endoscopic biopsy specimens, and that this property is greater in the stomach than in the duodenum and rectum (6). Our aims were therefore to study this property of gastric mucosa in peptic ulcer disease and to assess the relationship to gastritis and to Campylobacter pylori (CP) infection. Subjects We studied 228 subjects with abdominal pain or dyspepsia who underwent diagnostic upper gastrointestinal (GI) endoscopy. Forty-nine had active duodenal ulcer (DU) and 17 active gastric ulcer (GU). Fifteen DU and 5 GU patients were restudied after healing with H 2 -receptor antagonists (cimetidine, 8 mg, or ranitidine, 3 mg, daily for 4-8 wk). We also studied 38 subjects with peptic ulcer diagnosed on previous endoscopy but absent on this occasion (inactive ulcers; 29 with DU and 9 with GU). We studied 124 control subjects in whom no peptic ulcer was found on endoscopy and who had no history of ulcers. The age and sex distribution of each group is shown in Table 1. Patients with malignancy; those taking steroids, nonsteroidal antiinflammatory drugs, or anticoagulants; and those who had undergone previous gastric surgery were excluded. Local hospital ethical committee approval was given for this study, and written informed consent was obtained from all patients before endoscopy. Methods Upper GI endoscopy was carried out after an overnight fast by an investigator who did not perform the hydrophobicity readings (S.H.S.). After examination of the esophagus, stomach, and duodenum, biopsy specimens were taken from non ulcerated areas of the stomach, from the antrum (within 2 cm of the pylorus), and from the body (upper greater curve). Two unfixed biopsy specimens from each region were used for hydrophobicity measurements, and two biopsy specimens from each region were fixed in Abbreviations used in this paper: CP, Campylobacter pylori; DU, duodenal ulcer; GU, gastric ulcer. 199 by the American Gastroenterological Association /9/$3.

2 May 199 HYDROPHOBICITY IN PEPTIC ULCER 1251 Table 1. The Distribution of Patients: Pathology, Male-Female Ratio, and Age Age (yr) Group n Male:female mean ± SEM (range) Control : ± 1.8 (21-88) Active DU 49 34: ± 2.1 (21-87) Active GU 17 12: ± 3.2 (34-76) Inactive DU 29 2: ± 2.9 (24-73) Inactive GU 9 7:2 53. ± 6.1 (31-81) formalin for subsequent histological examination. In each of 79 subjects, one antral biopsy specimen was used to perform a urease test (CLO test) (7) to confirm the presence of CPo Hydrophobicity Hydrophobicity was measured as previously described and validated (6). Briefly, fresh biopsy specimens were rinsed in.15 M saline, placed on clean glass slides and oriented under a dissecting microscope (Bausch and Lomb StereoZoom), and then placed on the specimen stage of a goniometer (Rame-Hart 1/) for contact angle measurements of a drop of.15 M saline. The biopsy specimens were dried in air, without blotting, and the plateauadvancing contact angle was measured when the mucosa assumed a matt dullness, which as previously shown is not associated with mucosal disruption at the light microscopic level (6). The plateau-advancing contact angle is the most reproducible contact angle on biological tissues (5,6). The value for each measurement was the mean of at least 3 contact angle readings. The hydrophobicity measurements were performed without knowledge of the endoscopic or histological findings. Histology After fixation in neutral buffered formol saline, the biopsy specimens were routinely processed. Sections were cut at 4 j!m and stained with H&E. The presence and severity of inflammation was assessed according to standard criteria (8) and graded mild, moderate, or severe by two pathologists (C.W.Y. and C.M.C.). The presence of CP organisms was also recorded. Histological assessment was performed without prior knowledge of clinical details. Statistics Contact angles are expressed in degrees and as the mean ± SEM. One-way analysis of variance (ANOV A) was used to detect an overall difference between groups. When a significant overall difference was found, Fisher's protected least significant difference test for multiple comparisons was used. Two-way ANOV A was performed to assess the effect of 2 variables on contact angle. Student's t- and x 2 -tests were used as appropriate. A probability level of <.5 was regarded as significant. Two-tailed significance tests were used throughout. Results Peptic Ulceration The antral contact angles for active DU and GU subjects are compared with control subjects in Figure 1. Both ulcer groups had significantly lower contact angles than controls (DU 56.8 ± 1.1, GU 59.3 ± 2.3 vs. controls 66.2 ±.8; p <.1). There was no difference within the peptic ulcer groups with respect to site of ulcer. Figure 2 shows the contact angles of the 15 DU and 5 GU subjects who were studied before and after healing with Hz-receptor antagonists. There was no change in contact angle compared with pretreatment values (DU 59.1 ± 2. vs ± 1.5, GU 56.9 ± 3.1 vs ±.9). The antral contact angles in the subj ects with inactive DU (n = 29, 57.6 ± 1.2) and GU (n = 9, 57.9 ± 1.2) were not different from the active ulcer groups, but were significantly lower than control subj ects (p <.1). There was no difference in antral contact angles between male and female subjects in any group (controls 65.7 ± 1.4 vs ± 1.; ulcer patients 57.3 ± 1.1 vs ± 1.1). There was no difference between antral and body contact angles in any of the groups studied (Table 2). Table 2 also shows the contact angles of mucosa from the duodenal bulb in 45 subjects. The bulbar contact angles in the subjects with active DU (n = 13, 51.3 ± 1.8) and inactive DU (n = 9, 52.4 ± 2.8) were significantly lower than in control subjects (n = 17, 63.4 ± 2.8; p <.1). Gastritis Because there was no difference in contact angles between the DU and GU subjects with respect ~. : a :::.. : E =..... c u Ie Controls DU GU Figure 1. Contact angles of antral mucosa in the active DU and GU subjects and in the nonulcer controls.

3 1252 SPYCHAL ET AL. GASTROENTEROLOGY Vol. 9B, No.5 CI) -eo c:: Cd... u Cd... c:: U Before Healing After Figure 2. Antral contact angle before and after healing of peptic ulcers with H 2 -receptor antagonists. e, DU;, GU. Table 3. Comparison of Antrum and Body for the Prevalence and Severity of Gastritis in Paired Biopsies for 72 Subjects Body Nil Mild Moderate Antrum Nil 19 Mild 1 5 Moderate B Only two ulcer patients were found not to have any inflammation;.therefore, the ulcer group could not be analyzed for the presence of gastritis. Table 4 shows the data for antral contact angle in relation to the presence or absence of peptic ulcer disease and the severity of gastritis. The ulcer group had a higher prevalence and greater severity of gastritis than controls (p <.1). The contact angle was unrelated to severity of gastritis in both groups. The body similarly showed no relationship between contact angle and severity of gastritis (data not shown). to the activity or site of ulcer, they were grouped together for the purpose of further analysis. One hundred fourteen control and 87 ulcer subjects underwent histological assessment of antral mucosa, and 41 control and 35 ulcer subjects underwent histological assessment of body mucosa. Two subjects (1 control and 1 ulcer) had tissue that could not be classified adequately for severity of inflammation. There was a small, nonsignificant improvement in the grading of the gastritis after healing of peptic ulcers. Table 3 shows the presence and severity of gastritis in antral and body mucosa for all 72 subjects who had both assessed. The antrum had a higher prevalence and greater severity of gastritis (p <.1). In the control group, those with gastritis had significantly lower contact angles than those without gastritis (antrum 61.4 ± 1.1, n = 5, vs ± 1., n = 63, P <.1; body 63.7 ± 1.2, n = 2.3, vs ± 1.4, n = 21, P <.5). However, the control subj ects with gastritis still had significantly higher contact angles than the ulcer subjects (p <.2). Campylobacter pylori Infection Of the patients on whom CLO tests were performed, 3 of 4 (7.5/) without CP on histology were CLO-test positive, and 2 of 39 (5%) with CP on histology were CLO-test negative. Results were analyzed on the basis of histology. Table 5 shows the relationship between antral contact angle and both the diagnosis of peptic ulcer disease and the presence of CPo Nonulcer controls who were CP positive had lower body contact angles than those who were CP negative (63. ± 2., n = 1, vs ± 1.2, n = 31; p =.2). In the control group, those with CP-positive antral gastritis had significantly lower antral contact angles than those with CP-negative gastritis (58.9 ± 1.2, n = 39, vs ± 1.5, n = 14; P <.1). Peptic ulcer subjects with CP-positive antral gastritis also had lower antral contact angles than those with CPnegative gastritis, but this difference was not significant (57.7 ±.8, n = 77, vs. 61. ± 3., n = 9). Table 2. Comparison of Contact Angles for Antrum and Body of Stomach and for Duodenal Bulb Contact angle (), mean ± SEM (n) Group Antrum Body Bulb Control 66.2 ± (I..B (119) 66.3 ±.9 (51) 63.4 ± 2.B (17) ActiveDU 56.9 ± 1.1 (46) 54.9 ± 1.1 (2) 51.3 ± 1.B (13) ActiveGU 59.3 ± 2.3 (16) 59.1 ± 2.5 (5) 54.6 ± 2.4 (3) Inactive DU 57.6 ± 1.2 (29) 57.B ± 1.B (15) 52.4 ± 2.B (9) Inactive GU 57.9 ± 1.2 (9) 56.6 ± 1.9 (B) 56.B ± 2.4 (3)

4 May 199 HYDROPHOBICITY IN PEPTIC ULCER 1253 Table 4. Relationship Between Antral Contact Angle and Severity of Gastritis in Ulcer and Control Subjects Contact Angle (), mean ± SEM (n) Severity of antral gastritis Group Nil Mild Moderate Severe Control Ulcer 69.5 ± 1. (63) 54.7 ± 2.2 (2) 6.2 ± 1.3 (34) 59.2 ± 1.2 (18) 63.7 ± 1.9 (17) 58.2 ± 1. (59) 62.5 (1) 53.4 ± 2.4 (7) Discussion Our results show that the surface hydrophobicity of human gastric mucosa is reduced in subjects with peptic ulcer disease compared with controls without ulcers and that this reduction is not dependent on the site or activity of the ulcer. Like others, we used the plateau-advancing contact angle after air drying as our index of hydrophobicity (2-6). The drying process is not physiological, but it is essential to obtain reproducible results (5,6). The introduction of an air interface may result in denaturation, exposure of otherwise masked hydrophobic domains, or reorientation of molecules at the tissue surface (2), and the trauma involved when obtaining endoscopic biopsies could theoretically affect the contact angle readings by disrupting the surface. However, because the conditions of measurement are the same for all subjects, the differences we found could not be causeq. by these effects. Furthermore, the ability of this technique to predict the biophysical properties of a tissue has been validated by comparison with non-air-drying methods (1). The method used here enables many subjects to be studied and allows serial studies to be performed to assess the effects of treatment or the time course of disease. If operative gastric resection specimens were to be used, they would be subject to the trauma of handling and dissection. This approach would also restrict the numbers studied because resective gastric surgery for peptic ulcer is uncommon today, and control groups would be limited to patients with cancer. To check on any association between gastritis and hydrophobicity, we divided our control subjects with Table 5. Relationship of Contact Angle to Campylobacter pylori Infection Control Ulcer CP absent Contact angle (), mean ± SEM (n) CP present 69.8 ±.9 (75) 58.9 ± 1.2 (39) 6.9 ± 2.7 (1) (77) Both the diagnosis (p <.1) and CP status (p <.1) had significant effects in reducing contact angle. The ulcer group had a higher proportion of CP-positive subjects than the control group (p <.1). non ulcer dyspepsia into those with and those without histological evidence of gastritis and found that those with gastritis had lower contact angles; however, the patients with peptic ulcers still had significantly lower contact angles than nonulcer controls with gastritis, suggesting that gastritis is not the only factor and may be secondary to the changes in hydrophobicity. We found that the presence of CP infection was associated with a significant reduction in the contact angle. This could be caused by the organism itself within the mucus layer either exerting a direct topical effect that makes the surface appear more hydrophilic or causing an indirect toxic effect due to release of prot eases (11) or lipases (12) that can alter hydrophobic ~ t y On. the other hand, the presence of these organ Isms may be secondary to a primary defect in the gastric mucosa because adherence of bacteria to tissue s ~ r f has a cbeen e s shown to be dependent on hydrophobic and thermodynamic interactions (13,14). Repeat measurements after eradication of the organism will be necessary to determine which is the primary event. Our finding of a persistent reduction in the surface hydrophobicity of gastric mucosa in peptic ulcer after successful healing with Hz-receptor antagonists may be due to the fact that these agents have no effect on the accompanying CP infection (15). The alterations in mucus rheology and glycoprotein structure found in peptic ulcer disease (16) may account for our results. A defective gel that fails to anneal may produce a rougher and more heterogeneous surface or be more prone to acid/peptic diges!ion. The effect of surface roughness and heterogene Ity on contact angles is known as hysteresis (17,18). Roughness increases the contact angle (13,18), whereas the effect of heterogeneity depends on the relative proportions of hydrophobic and hydrophilic molecules present at the surface (18). The hydrophobic component of a heterogeneous surface is measured by the advancing contact angle (17,18); therefore, if this component is reduced one would expect a lower contact angle. Constituents of mucus other than glycopr?teins, such as effete cells (19), glycolipids (2), fatty acids (2), and phospholipids (21), could also affect the hydrophobicity. Animal studies have shown that aspirin decreases mucosal hydrophobicity (3,22) while prostaglandins, which increase basal hydrophobicity, limit this reduction (22). It has been suggested that a

5 1254 SPYCHAL ET AL. GASTROENTEROLOGY Vol. 98, No.5 layer of surface-active phospholipids is responsible for the protective hydrophobic properties of gastric mucosa (2,3,21). At present we cannot comment on whether our findings of reduced gastric mucosal hydrophobicity in peptic ulcer disease can be explained in these terms because contact angle does not provide specific information about the molecular configuration on a tissue surface (17,18). In conclusion, we have found that the hydrophobicity of the gastric mucosa is reduced in peptic ulcer disease, regardless of the activity or site of the ulcer, and that it is also reduced in the presence of gastritis or CP infection. References 1. Schwarz K. Ueber penetrierende Magen-und lejunalgeschwure. Beitr Klin Chir 191;67: Hills BA. Gastric mucosal barrier: stabilization of hydrophobic lining to the stomach by mucus. Am I Physiol 1985;249:G342- G Hills BA, Butler BD, Lichtenberger LM. Gastric mucosal barrier: hydrophobic lining to the lumen of the stomach. Am I Physiol 1983;244:G561-G Hills BA, Cotton DB. Premature rupture of membranes and surface energy: possible role of surfactant. Am I Obstet Gynecol 1984;149: Absolom DR, Zingg W, Neumann AW. Measurement of contact angles on biological and other highly hydrated surfaces. I Colloid Interface Sci 1986;112: Spychal RT, Marrero 1M, Saverymuttu SH, Northfield TC. Measurement of the surface hydrophobicity of human gastrointestinal mucosa. Gastroenterology 1989;97: Marshall BJ, Warren IR, Francis GJ, Langton SR, Goodwin CS, Blincow ED. Rapid urease test in the management of Campylobacter pyloridis associated gastritis. Am I Gastroenterol1987;82: Whitehead R, Truelove SC, Gear MWL. The histological diagnosis of chronic gastritis in fibreoptic gastroscope biopsy specimens. I Clin Pathol1972;25: Holly Fl. Refojo MF. Water wettability of hydrogels. In: Andrade ID, ed. Hydrogels for medical and related applications. Washington: American Chemical Society, 1976; Neumann AW, Absolom DR, Francis DW, Omenyi SN, Spelt IK. Policova Z, Thompson C, Zingg W, van Oss C/. Measurement of surface tensions of blood cells and proteins. Ann NY Acad Sci 1983;416: Slomiany BL, Bilsky I, Sarosiek J, Murty VLN, Dworkin B, Van Horn K. Zielenski I, Slomiany A. Campylobacter pyloridis degrades mucin and undermines gastric mucosal integrity. Biochern Biophys Res Commun 1987;144: Sarosiek J, Slomiany A, VanHorn K, Zalesna G, Slomiany BL. Lipolytic activity of Campylobacter pylori: effect of Sofalcone (abstr). Gastroenterology 1988;94:A Van Oss C/, Gillman CF, Neumann AW. Phagocytic engulfment and cell adhesiveness as cellular surface phenomena. New York: Dekker, Absolom DP, Lamberti FV, Policova Z, Zingg W, van Oss C/, Neumann AW. Surface thermodynamics of bacterial adhesion. Appl Environ Microbiol1983;46: Hui WM, Lam SK, Chau PY, Ho I, Lui I, Lai CL, Lok ASF, Ng MMT. Persistence of Campylobacter pyloridis despite healing of duodenal ulcer and improvement of accompanying duodenitis and gastritis. Dig Dis Sci 1987;32: Younan F, Pearson I, Allen A, Venables C. Changes in the structure of the mucus gel on the mucosal surface of the stomach in association with peptic ulcer disease. Gastroenterology 1982; 82: Adamson AW. Physical chemistry of surfaces. 4th ed. New York: Wiley, lohnson RE Ir, Dettre RH. Wettability and contact angles. Surface Colloid Sci. 1969;2: Lipkin M. Cell turnover in gastrointestinal mucosa. In: Harmon IW, ed. Basic mechanisms of gastrointestinal mucosal cell injury and protection. Baltimore: Williams & Wilkins, 1981; Sarosiek J, Slomiany A, Takagi A, Slomiany BL. Hydrogen ion diffusion in dog gastric mucus glycoprotein: effect of associated lipids and covalently bound fatty acids. Biochem Biophys Res Commun 1984;118: Butler BD, Lichtenberger LM, Hills BA. Distribution of surfactants in the canine gastrointestinal tract and their ability to lubricate. Am I Physiol1983;244:G645-G Lichtenberger LM, Richards IE, Hills BA. Effect of 16,16- dimethyl prostaglandin E2 on the surface hydrophobicity of aspirin treated canine gastric mucosa. Gastroenterology 1985;88: Received February 14, Accepted November 9, Address requests for reprints to: Professor T. C. Northfield. Department of Medicine, St. George's Hospital Medical School, Tooting, London, S.W.17.RE, England. RT.S. was supported by Syntex Pharmaceuticals, the Cadbury's Trust, and the Oakdale Trust. P.M.G. was supported by the British Digestive Foundation. This research was also supported by Gist Brocades.