Helicobacter pylori anti-CagA antibodies : Prevalence in symptomatic and asymptomatic subjects in Turkey

1Fatih University, Faculty of Science and Literature, Department of Biology/Microbiology Unit; 2Samatya Hospital, Gastroenterology Department/Endoscopy Unit, Istanbul, Turkey Correspondence: Dr Barik Salih, Fatih University, Faculty of Science and Literature, Department of Biology/Microbiology Unit, Buyuk Cekmece, Istanbul, Turkey. Telephone +90-212-8890810, fax +90-212-8890832, e-mail basalih@fatih.edu.tr Received for publication February 22, 2002. Accepted July 4, 2002 MF Abasiyanik, E Sander, BA Salih. Helicobacter pylori antiCagA antibodies: Prevalence in symptomatic and asymptomatic subjects in Turkey. Can J Gastroenterol 2002;16(8):527-532.

Several tests have been used to diagnose H pylori infection; these are divided into biopsy-based tests (CLO, culture, histology, polymerase chain reaction) and nonbiopsybased tests (serology, urea breath test) (20).Serological tests are noninvasive and are used to detect immunoglobulin (Ig) G, IgA and IgM antibodies in the sera of H pyloriinfected subjects.The ELISA test is the most widely used test for the detection of such antibodies in epidemiological studies (4,21,22).Immunoblot assay enables the detection of antibodies against specified H pylori antigens, such as CagA and VacA antigens (21,23).The aim of the present study was to detect H pylori antibodies in symptomatic and asymptomatic subjects by the ELISA test, and to determine the prevalence of anti-CagA antibodies and other H pylorispecific antigens in both groups by immunoblotting.

Subjects
Symptomatic subjects: Sixty-six patients attending the endoscopy unit at the Samatya Hospital, Istanbul, Turkey, from March to July 2000, with dyspepsia and abdominal pain as their main complaint, were selected randomly.The study group comprised 38 men and 28 women, 16 to 74 years of age (average age 45 years).All patients underwent endoscopy, and three biopsies were taken from the antral part of the stomach for CLO, culture and histology.A serum sample was obtained from each patient and stored at -20°C.Asymptomatic subjects: One hundred nineteen subjects (blood donors, school teachers and health workers), 20 to 65 years of age (average age 37 years), with no history of abdominal pain were selected randomly.A serum sample was obtained from each subject and stored at -20°C.A questionnaire was filled out for each subject.

CLO test
One biopsy was placed into the CLOtest (Clia Waived, USA) and kept at room temperature.The results were recorded as specified by the manufacturer.

Culture
Biopsies for culture were dissected into small pieces using a sterile blade and then inoculated onto Colombia agar medium (Oxoid, United Kingdom), with 5% defibrinated sheep blood and selective supplements (H pylori-selective supplement, Oxoid).The cultures were incubated in anaerobic jars, in humidified, microaerophilic conditions, using CompyGen Gas pack (Oxoid) at 37°C for five days.The organisms were identified as H pylori by Gram staining, colony morphology, and urease, oxidase and catalase-positive reactions.

Histology
Antral biopsies placed in 10% formalin were sent to the histopathology laboratory.Sections were stained with hematoxylin and eosin stain, and modified Giemsa stain, and examined by an experienced histopathologist.

Enzyme immunoassay
The ELISA-kit SIA-H pylori (Sigma, USA) was used to detect IgG antibodies to H pylori in human sera.The test was conducted according to the manufacturer's instructions.Samples were washed with the EL×50 Auto Strip Washer, and the results were read with the EL×800 Universal Microplate Reader (Bio-Tek Inc, USA).All samples were tested at the same time.

Immunoblot assay
The Helico-Blot 2.1 (Genelabs Diagnostics, Singapore) was used to test the ELISA-positive serum samples from both groups.The test is based on Western blot assay of H pylori whole-cell antigens that is made from bacterial lysate of a strain of H pylori that is known to cause ulcers.Strips were marked at the 116 (CagA), 89 (VacA), 37, 35, 30 and 19.5 kDa antigen lines and at an additional antigen line, designated the current infection marker (CIM).The recommended criteria for determining H pylori seropositivity as specified by the manufacturer is any one of: • the detection of 116 kDa (CagA) band with one or more bands of 89, 37, 35, 30 (UreA) and 19.5 kDa together, or with CIM; • any one band of the 89, 37 or 35 kDa, with or without the CIM; or • detection of both the 30 and 19.5 kDa bands, with or without the CIM.
Imaging and analysis of immunoblots were done using Gel Doc 2000 and Quantity One quantification software (Bio-Rad, USA).

Statistical analysis
Data were analyzed using SPSS 10.1 (SPSS Inc, USA).P<0.05 was considered to be statistically significant.

Symptomatic subjects
Biopsies were taken, and the presence of H pylori was determined through CLO, culture and histology.Patients were considered to be infected if one test was positive.Fiftyseven of 66 patients were found, by either one of the biopsybased tests, to be infected with H pylori. Results of the ELISA test were compared with those of the biopsy-based tests.ELISA was shown to have 89% sensitivity (95% CI 81.5 to 95.5%), 100% specificity, 100% positive predictive value and 67% negative predictive value (95% CI 55.5 to 78.3%).Among the 66 symptomatic subjects included in the study, 32 of 38 (84%) men and 22 of 28 (79%) women were found to be H pylori-positive by ELISA -a total of 54 of 66 (82%) subjects.
Immunoblot assay was performed on 30 randomly selected ELISA-positive sera from symptomatic subjects.Table 1 shows the correlation between the prevalence of CagA antibodies and age.Twelve of 14 (86%) patients older than 45 years of age and 12 of 16 (75%) patients younger than 45 years of age were positive for CagA antibodies.Statistical analysis showed no significant difference between these age groups.CagA antibodies were detected in patients with the following histological findings: 11 of 14 (79%) with chronic gastritis, 12 of 13 (92%) with duodenal ulcer and three of three (100%) with gastric cancer (one patient also had gastric ulcer) (  were detected in eight of 14 (57%) subjects with chronic gastritis, 11 of 13 (85%) with duodenal ulcer and two of three (67%) with gastric cancer.Antibodies to the other four antigens varied also in their frequency pattern (Table 2).The CIM appeared in 25 of 30 (83%) positive strips.

Asymptomatic subjects
Among the 119 asymptomatic subjects, 76 (64%) were found to be H pylori positive by ELISA.The sera from 40 of these subjects (randomly selected) were analyzed by immunoblotting.

DISCUSSION
In this study, the prevalence of H pylori in gastric biopsies of symptomatic subjects was determined by CLO, culture and histology.If one test was positive, patients were considered to be infected.Others have reported using a similar approach (18).In comparison with the biopsy-based tests, serology proved to be useful for the detection of H pylori infection (24).A review of the overall performance of the commercially available serology kits that measure IgG antibodies showed that serology is an accurate method of diagnosing H pylori in patients (25).Comparison between serology and the combination of CLO, histology and culture revealed 94% sensitivity and 88% specificity (18).Compared with the biopsy-based tests, ELISA was 89% sensitive and 100% specific in detecting H pylori infection in symptomatic subjects.These results were in agreement with those of an earlier study (25).
H pylori is associated with several gastroduodenal diseases such as gastritis, gastric ulcer, duodenal ulcer, gastric adenocarcinoma and mucosa-associated lymphoid tissue (MALT) lymphoma (2,26,27).CagA has been associated with the development of peptic ulcer disease and gastric cancer.It is expressed in approximately 60% of the isolates and the protein is highly immunogenic (4), while vacA gene, although present in nearly all strains, is expressed in only 50% of the isolates (18).
Immunoblotting is a more sensitive serological test than ELISA for the diagnosis of H pylori infection and the detection of specified antigens (6,15,19).With immunoblot testing, no significant age difference in the response to CagA antigen between patients older than 45 years and those younger than 45 years of age was found, indicating that the prevalence of CagA-positive strains is not restricted to older age and that subjects in different age groups mount a good antibody response to the CagA antigen.An increase in the prevalence of anti-CagA antibodies has been reported in patients with peptic ulcer disease and those with gastric cancer (2,4,20,28).There appears to be major geographic differences in the prevalence of CagApositive strains.In Australia, the prevalence of antibodies to CagA in patients with duodenal ulcer was higher than that in those with nonulcer dyspepsia and in asymptomatic people, while in China, no significant difference in the prevalence of CagA antibodies between asymptomatic and gastric cancer groups was found (4).In another study, host and environmental factors were found to be more important in the development of the disease (29).In the present study, the detection of such antibodies in the sera of symptomatic and asymptomatic subjects indicates that CagApositive H pylori strains prevailed in both groups.These results appeared to be in agreement with those of previously published reports (4,(15)(16)(17)27,30).Rocha et al (23) indicated that immunoblotting using the Helico-Blot kit is useful for the diagnosis of H pylori infection in children and that the frequency of immunoreactive bands to CagA antigen is higher in children with duodenal ulcer than in children without the disease.However, Mitchell et al (31) reported in an earlier study that antibody to CagA is not a marker for specific disease in children.More recently, Figueiredo et al ( 24) compared the Helico-Blot kit with two other noncommercially developed enzyme immunoassays and concluded that the detection of anti-CagA antibodies is strongly dependent on the test used.The limitations reported by Figueiredo et al (24) in their interpretation of the results were due to several factors that included observer visual scoring, low intensity of some signals and variation between different strip batches that prevent accurate alignment.These limitations were overcome in our study by the use of Gel Doc and the Quantity One software program for imaging and analysis.
Differences in the frequencies of the immunoreactive bands to the six antigens between the symptomatic and asymptomatic subjects in the present study were attributed to individual and strain variability.In addition, the human gastric mucosa colonized by H pylori shows a variety of responses that may vary greatly in the intensity and distribution of the histological responses and correlation with the clinical outcome (11).The detection of CIM did not seem to add any further information to the interpretation of the results.

CONCLUSIONS
Antibodies to CagA antigen are prevalent in both symptomatic and asymptomatic subjects, regardless of the gastroduodenal disease.This gene product might be insufficient to act as a marker for the development of the disease in some populations.