Diagnostic Value of Detection of IgM Antibodies to Helicobacter pylori

Experimental and Molecular Pathology 72, 77–83 (2002) doi:10.1006/exmp.2001.2408, available online at http://www.idealibrary.com on

Diagnostic Value of Detection of IgM Antibodies to Helicobacter pylori

Mehdi Alem,*,1 Neda Alem,* Hartley Cohen,†,2 Tom England,‡ Nina Hamedi,* Mitra Moussazadeh,* Jo Ann Roth,§ and Guo Qiu Shen‡ *Micro Detect, Inc., Tustin, California 92780; †GI Division, University of Southern California School of Medicine, Los Angeles, California 90033,‡ Specialty Laboratories, Santa Monica, California 90404; and §UniLab, Tarzana, California 91356

Received July 20, 2001

This report describes the diagnostic value of anti-Helicobacter pylori IgM detection. Serum samples from 9043 symptomatic and asymptomatic individuals were evaluated with ELISA for the presence of antiH. pylori IgG, IgM, and IgA. The specificity of detected IgM was confirmed by inhibition and cross-reactivity assays. Treatment of IgMpositive specimens with 1% 2-mercaptoethanol resulted in approximately 90% inhibition. Our data suggest a low level of cross-reactivity (5%) between H. pylori and four different enteropathogenic bacteria tested. The specificity of anti-H. pylori IgM was also demonstrated by Western blot and linearity studies. Data show that the detected IgM is highly specific. Western blot analysis revealed a variable IgM response to H. pylori antigens among patients, with the most reactive antigenic fractions being in the range of 55- to 100-kDa. Overall, the data confirm the diagnostic value of anti-H. pylori IgM detection. The prevalence of IgM antibodies to H. pylori in tested sera was significantly higher in symptomatic patients (10.4%) than in asymptomatic individuals (1.1%). Likewise, the percentage of sera positive for IgM alone was higher in symptomatic than in asymptomatic groups (3.8 vs 0.22%). About 5% of sera were positive only for IgA. We concluded that ELISA can be used for the detection of specific IgM to H. pylori and that the presence or absence of IgM antibodies to H. pylori may reflect whether or not an acute infection exists. 䉷 2002 Elsevier Science Key Words: H. pylori; IgM; diagnostic; ELISA.

INTRODUCTION

Detection of specific IgG and IgA antibodies to Helicobacter pylori has been shown to be very accurate for the diagnosis of H. pylori infections. Even after eradication of H. pylori, the titers of IgG and IgA antibodies to H. pylori may remain detectable indefinitely. Hence, detection of IgG and IgA antibodies to H. pylori does not differentiate between current and past infections. Testing sera for IgM to H. pylori may provide data confirming current or acute infection. However, the diagnostic value of detecting H. pylori IgM antibodies has been disputed in some reports (1–5); Jones et al. (6) have reported that IgM antibodies to H. pylori are in fact present in approximately 10% of H. pyloriinfected patients irrespective of their particular gastrointestinal conditions. Idiopathic gastritis is a chronic condition (7); therefore, failure to show high IgM values in all infected patients may be related to obtaining the specimens after acute IgM seroconversion (1). Morris et al. (8) reported that the presence of IgM antibodies to H. pylori might reflect acute rather than chronic infection. Newell and Rathbone (9) reported that circulating anti-H. pylori antibodies are mainly IgG and rarely IgM, presumably because the disease generally manifests itself as a chronic rather than as an acute infection. They concluded that high levels of IgM in those

1

To whom correspondence should be addressed at: Micro Detect, Inc., 2852 Walnut Avenue, Suite H-1, Tustin, CA 92780. Fax: (714) 832-8231. E-mail: [email protected]. World Wide Web page: www. microdetect.com. 2 Present address: 1245 Wilshire Boulevard, Suite 917, Los Angeles, CA 90017.

0014-4800/02 $35.00 䉷 2002 Elsevier Science All rights reserved.

77

78 patients might provide further information regarding an acute state of infection. In a case report, Easthan et al. (10) evaluated two children with antral nodular hyperplasia and active gastritis associated with H. pylori. They reported the presence of high titers of IgG in both patients, together with serum IgM in one of them. In their study, they suggested that the presence of specific IgM to H. pylori is associated with relatively recent infection and possible mucosal invasion. Kosunen et al. (11) reported decreases in IgG, IgA, and IgM antibody titers after eradication of H. pylori. In their study on a selected group of 144 infected patients, 5.5% were found to be H. pylori IgM-positive. They concluded that the detection of IgM to H. pylori might be useful in the diagnosis of acute infections. Rosenstock et al. (12) studied the relation between H. pylori infection and GI symptoms and reported that individuals with epigastric pain and increased IgM antibodies represent the acute stage of H. pylori infection. Dobbs et al. (13) reported an association between increasing age and H. pylori IgM antibodies. They also concluded that increasing age and H. pylori seropositivity are each associated with a downshift in circulating IgM. Rosenstock et al. (14) studied the seroconversion in H. pylori infection and found that an increase in IgM antibody levels significantly increased the likelihood of an IgG seroconversion. Since a high titer of IgM antibodies to H. pylori may indicate recent or early stages of infection, the primary purpose of this study was to determine the utility of detecting IgM antibodies to H. pylori in evaluation of symptomatic and asymptomatic individuals.

MATERIALS AND METHODS

Patient population. A total of 9043 serum specimens obtained from four different sites were evaluated in this study. One hundred fifty-nine serum specimens were obtained from patients attending the Gastroentrology Endoscopy Unit at Los Angeles County and University of Southern California Medical Center for endoscopic evaluation of upper gastrointestinal symptoms. During endoscopy, paired biopsy specimens were obtained routinely from the gastric antrum and gastric corpus of each patient. One of the biopsy specimens of each pair was used for culture, and the second was used for routine histological evaluation. Additional biopsy specimens from antrum were also obtained for CLO test (Delta West L., Bentley, Australia). Methods of evaluation of

ALEM ET AL.

biopsy specimens were the same as previously described (15). Four hundred fifty specimens were from individuals attending a clinical lab in Southern California for evaluation of various medical disorders. A significant number of these specimens (random specimens) was sent to the laboratory mainly for clinical chemistry tests. The third group, consisting of 1621 specimens, was sent to a second clinical laboratory in Southern California (Specialty Laboratories, Santa Monica, CA) for evaluation of H. pylori infection. The fourth group, consisting of 6808 specimens, was sent to a third clinical laboratory in Southern California (UniLab, Tarzana, CA) for evaluation of H. pylori infection. Specimens were obtained from both genders, different races, and age groups ranging from 1–99 years (Table 1). ELISA method. The statuses of serum IgG, IgA, and IgM to H. pylori for specimens in groups one and two were determined at Micro Detect, Inc., (Tustin, CA). Evaluations of specimens in groups three and four were performed at sites three and four, respectively. Briefly, in all sites, frozen or fresh specimens were diluted 1:100 and anti H. pylori antibodies were measured using commercially available test kits (PYLORI DETECT IgG, IgM, and IgA, manufactured by Micro Detect). For detection of IgM, each patient specimen was first diluted 1:10 in 1 mg/mL anti-human IgG before adjustment of the final dilution to 1:100. The procedures for detection of antibodies to H. pylori were those described in the package insert of each product. Briefly, the status of

TABLE 1 Number, Race, Gender, and Age of Patients Site (group)

Number

% Race

% Gender

Age (years)

1

159

46 male 54 female

20–74

2 3

455 1621

77.4 H 6.9 B 9.4 C 6.3 A Unknowna Unknown

10–90 1–99

4

6808

Unknown

Unknown 41 male 59 female 46 male 54 female

2–98

Note. Sites (groups): 1, Los Angles County and University of Southern California Medical Center, Los Angeles, CA; 2, random specimens were obtained from a clinical laboratory in Southern California, Los Angeles, CA; 3, Specialty Laboratories, Santa Monica, CA; and 4, UniLab, Tarzana, CA. H, Hispanic; B, Black; C, Caucasian; A, Asian. a Unknown, but mostly Asian and Hispanic.

79

DETECTION OF IgM ANTIBODIES TO H. pylori

serum antibodies to H. pylori was determined by using an antigen mixture of two strains of H. pylori (ATCC 43504 and ATCC 43629). Culture bacteria were harvested in saline, and specific antigens of H. pylori consisting of combined antigenic fractions with molecular sizes of 120, 84, 55–66, 30, and 14 kDa were purified and used for coating ELISA plates in PYLORI DETECT IgG, IgM, and IgA test kits and in a Western blot study. SDS–PAGE and Western blot analysis. The protein components of H. pylori antigen used for evaluating the specificity of IgM antibodies to H. pylori were separated by 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE), and the resulting fractions were electrophoretically transferred to a nitrocellulose membrane (Bio-Rad, Richmond, CA). The membrane was blocked with 3% bovine serum albumin, washed, and incubated with patient sera. To show the specificity of H. pylori IgM and reduce the possible false-positive results due to rheumatoid factor, both positive and negative specimens were first treated with anti-human IgG to remove IgG from the sera. The treated sera were then added to the membrane and incubated for 6 h at room temperature. The membrane was then washed, incubated with alkaline phosphatase-conjugated anti-human IgM, and subsequently stained with naphthylphosphate FAST-red substrate (Sigma Chemical Co., St. Louis, MO). Linearity. Linearity of the assay used for detection of specific IgM antibodies to H. pylori was obtained by measuring the absorbance values of two fold serial dilutions of five H. pylori IgM-positive specimens. The absorbance values at 450 nm were compared to the log2 of dilutions by standard linear regression. To obtain absorbance values in a detectable range, the serial dilutions of sera were started from 1:100. Inhibition and cross-reactivity assays. The specificity of the assay for detection of IgM antibodies to H. pylori was demonstrated by an inhibition assay. Seven sera with different levels of IgM antibodies to H. pylori were first treated with anti-human IgG and were then tested in ELISA. The absorbance values were compared side-by-side with the absorbance values of the same specimens, which were treated with 1% 2-mercaptoethanol prior to ELISA. Tests for possible cross-reactivity of H. pylori antibodies with antigens of enteropathogenic bacterial species were also conducted by inhibition assay. Briefly, sera absorbed with anti-human IgG were further neutralized with 10 ␮g/mL of partially purified antigens of Campylobacter jejuni, Campylobacter coli, Campylobacter fetus, and Escherichia coli and then used in H. pylori IgM assay. The level of cross-reactivity was calculated by comparing the absorbance of untreated specimens with that of neutralized specimens.

RESULTS Detection of IgM antibody in serum. Of 159 biopsyconfirmed specimens, 125 were positive and 34 were negative by endoscopic methods. Only one of the endoscopynegative specimens in this group tested positive for IgM antibodies to H. pylori. This specimen was also positive for IgA and IgG antibodies to H. pylori by ELISA. Of 125 confirmed positives, only 13 (10.4%) tested positive for IgM to H. pylori. Of 13 IgM-positive specimens, only 2 specimens were positive for IgM alone; 4 were positive for both IgM and IgG. The remaining 7 were positive for all three IgA, IgG, and IgM antibodies. Of 455 random specimens, 10 specimens (2.2%) tested positive for IgM to H. pylori. The numbers of IgA and IgG positives were 17 and 39%, respectively. Of 10 IgMpositives, 5 were positive for all three IgG, IgM, and IgA antibodies. One was only IgM-positive, and the remaining 4 were positive for IgG and IgM. Of 445 random specimens in this group, 3.8% were only IgA-positive. Of 8429 specimens in group three and group four, 33.7% were positive for a combination of IgG, IgA, and IgM and 66.3% were negative for all three antibodies. In these groups, 8.6% were IgM-positive, and 3.8% of the IgM-positive were only IgM-positive. Of all 8429 specimens in group three and group four, 4.9% were only IgA positive (Table 2). Western blot. In order to demonstrate the presence and complexity of H. pylori-specific IgM, 19 H. pylori IgMpositive and 4 H. pylori-negative specimens were analyzed by Western blot. All specimens positive for IgM antibodies by ELISA showed reactivity in the Western blot, but the response to the different H. pylori antigen fractions was not consistent. We observed different patterns of reactivity, with the most reactive antigenic fractions being in the 55- to

TABLE 2 Percentage Distribution of IgM and IgA Antibodies in the Total Population under Investigation Group(s) 1 2 3 and 4

IgM⫹ alone IgM⫹a IgA⫹ alone

N 159 455 8429

b

1.3% 0.22% 3.8%

10.4% 1.1% 8.6%

1.9% 3.8% 4.9%

Ab⫹

Ab⫺

NA NA 42.1% 57.9% 33.7% 66.3%

Note. N, number. Ab⫹, positive for combination of any three immunoglobulins; Ab⫺, negative for all three immunoglobulins; NA, not applicable. a Total IgM-positive. b Selective group.

80 100-kDa range (Fig. 1). Specimens obtained from patients without evidence of H. pylori infection showed no reactivity with any of the H. pylori antigenic fractions. Assay linearity. Absorbance values for reciprocal dilutions of five sera positive for IgM antibodies to H. pylori were determined at 450 nm. The resulting data indicated a linear relationship between measured absorbance (A)450 and log2 of dilution (r 2 ⫽ 0.9–0.93). Inhibition and cross-reactivity assays. The results of inhibition assays are presented in Table 3. The data show that destruction of IgM in the test samples with 1% 2mercaptoethanol yields negative or nonreactive specimens. In these assays, treatment of specimens with 1% 2-mercaptoethanol resulted in about a 90% reduction in the level of IgM in positive specimens. This confirms the specificity of anti-H. pylori IgM antibodies. The specificity of the antigen used in the H. pylori IgM ELISA was also demonstrated in a series of cross-reactivity tests. In these tests, H. pylori antibodies were neutralized with antigens from four different bacterial species and the IgM titers were then determined in ELISA. Our data suggest that there is a low level of crossreactivity between H. pylori and other enteropathogenic bacteria; however, this low level of cross-reactivity did not affect the performance of the IgM assay. C. jejuni is the closest of all to H. pylori and causes acute enteritis in man. Of five H. pylori IgM-positive sera treated with antigens of C. jejuni, C. coli, C. fetus, and E. coli, all produced positive results (Table 3).

DISCUSSION H. pylori has been isolated from essentially all patients with chronic active type B gastritis and peptic ulcer and is rarely detected in individuals with normal gastric mucosa. In response to H. pylori infection, the host produces specific local and systemic immune responses. This has been shown by experimental infections in humans and animals. Elimination of the organism leads to a distinct improvement of gastritis; however, for a relatively long period of time the titers of IgG and IgA antibodies remain high. This means that high levels of IgG antibodies to H. pylori do not demonstrate the acute, chronic, or previously treated infections. The use of a more specific IgM test would allow for direct screening of specimens and serve as a diagnostic tool for establishing active or recent infection. The exact mode of initial H. pylori transmission to humans and the process and possibility of reinfection by H. pylori have not been deciphered.

ALEM ET AL.

FIG. 1. Western immunoblot analysis of reactivity of H. pylori IgM antibodies with H. pylori antigen fractions. (A) Lanes 1–11, sera positive for IgM to H. pylori; lanes 12 and 13, sera negative for IgM to H. pylori. (B) Lanes 1–8, sera positive for IgM to H. pylori; lanes 9 and 10, sera negative for IgM to H. pylori. All H. pylori IgM-positive sera reacted with protein fractions at 60–66 kDa, which are subunits of urease. Urease is considered to be one of the predominant antigens of H. pylori. One of the low-molecular-weight antigenic fractions of H. pylori is in the range of 14 kDa. Nearly 50% of positive sera evaluated by Western blot in this study showed reactivity with this fraction. In contrast, very weak reactivity was observed with fraction 30 in a small number of tested sera. Most antigen preparations of H. pylori contain a small amount of the 55-kDa protein (flagellin), which cross-reacts with other entropathogenic bacteria; a weak reactivity was observed between this fraction and a number of positive sera. The species-specific antigens of H. pylori are located in the protein range of 80–120 kDa. Of these proteins, two of them, one 87 kDa (VacA) and one 120 kDa (CagA), are considered highly specific to H. pylori. Studies reported that antibodies to each one of these proteins might be markers of H. pylori strains associated with the greatest clinical impact. Not all H. pylori strains are positive for both CagA and VacA, and even in CagA- and VacA-positive strains, the total amounts of these proteins are relatively low. Most positive sera tested in this study showed low reactivity to proteins in the range of 75–100 kDa. Specimens with no evidence of H. pylori infection showed no reactivity with any of the H. pylori antigenic fractions.

81

DETECTION OF IgM ANTIBODIES TO H. pylori TABLE 3 Inhibition Assay of Helicobacter pylori IgM ELISA Specimen nos.a 1 2 3 4 5 6 7 Mean inhibition

Untreated OD

MEb OD

Cj OD

Cc OD

Cf OD

Ec OD

0.286 0.179 1.391 1.342 1.258 1.284 1.237 0

0.081 (72%↓) 0.061 (66%↓) 0.114 (92%↓) 0.177 (92%↓) 0.106 (91%↓) 0.133 (90%↓) 0.131 (90%↓) 85%↓

0.281 (2%↓) 0.165 (8%↓) 1.257 (10%↓) 1.281 (5%↓) 1.155 (8%↓) 1.223 (5%↓) 1.131 (9%↓) 7%↓

0.275 (1%↓) 0.170 (5%↓) 1.365 (2%↓) 1.264 (6%↓) 1.173 (7%↓) 1.253 (2%↓) 1.197 (3%↓) 4%↓

0.292 (0%↓) 0.171 (5%↓) 1.344 (3%↓) 1.263 (6%↓) 1.198 (5%↓) 1.206 (6%↓) 1.180 (5%↓) 4%↓

0.269 (6%↓) 0.171 (5%↓) 1.318 (5%↓) 1.284 (4%↓) 1.184 (6%↓) 1.220 (5%↓) 1.189 (4%↓) 5%↓

Note. Seven sera with different titers of IgM antibodies to H. pylori were inhibited by 1% 2-mercaptoethanol or antigens from four different entropathogenic bacterial species. OD, optical density; ↓, decrease; Cj, Campylobacter jejuni; Cc, Campylobacter coli; Cf, Campylobacter fetus; Ec, Escherichia coli. a Specimens 1 and 2 were confirmed negative and specimens 3–7 were confirmed positive for IgM to H. pylori. b ME, 1% 2-mercaptoethanol.

The specificity of IgG to H. pylori may be used as a tool for examining H. pylori infection; however, detection of this antibody by itself cannot provide sufficient information about acute infection. An IgM response to H. pylori has been seen by day 18 of acquisition, but IgG and IgA seroconversions occur together after 60 days, at which time IgM titers decline (16). This shows that the IgM titer rises early in H. pylori infection and that the duration of its elevated level is brief. In other words, the immunologic response to H. pylori infection may be similar to that observed in other chronic infections. Due to the nature of H. pylori infection, the initial infection by this organism may not necessarily be symptomatic and is not clearly different from those of an “upset stomach;” therefore, measurement of IgM antibodies to H. pylori may help early detection of the infection. Early detection of H. pylori infection may be useful for early treatment of individuals with acute infection. In this study we used a relatively large population of patients and evaluated the usefulness of anti-H. pylori IgM detection. By using specimens representing different groups of individuals, we could show that detection of IgM may be used as a tool for establishing recent infection in those individuals who are only IgM-positive. Based on our data, in order to improve diagnostic values of serological tests for H. pylori detection, concurrent measurements of IgG, IgM, and IgA are recommended. Our Western blot studies with IgM-positive specimens demonstrated that IgM antibody responses to H. pylori do not show consistent patterns from one patient to another. This finding confirms other reports and shows that immune responses to H. pylori antigenic fractions demonstrate a remarkable complexity and diversity (15, 17–19). Our Western blot data showed that none of the H. pylori-positive sera

reacted with the 30-kDa fraction. This fraction is considered to be one of the main antigenic fractions of H. pylori (16). These findings suggest that for detection of IgM antibodies to H. pylori, a complete mixture of H. pylori antigens should be used. In addition to Western blot data, the results of linearity studies and inhibition assays confirmed that the IgM antibody to H. pylori detected in this study is specific to H. pylori alone. Destruction of IgM in the tested sera and serial dilution of positive patient sera both resulted in significant decreases in ELISA values of tested-positive sera. Our ELISA assays used for detection of IgG, IgA, and IgM antibodies to H. pylori generated considerable information regarding the type of antibody response and the distribution of infection throughout the population. In addition to having diagnostic values, antibody detection methods contribute a great deal to epidemiological research, in which typically large numbers of subjects are studied from easily obtained blood samples. Since the establishment of the acute stage of infection has a particular role in routine medical practice and epidemiological research, we showed that detection of IgM to H. pylori may also be used for epidemiological purposes, specifically because our findings show that the IgM antibody to H. pylori can be measured in individuals of any race and that age and gender have no detectable influences on IgM assay performance. Reports suggest that detection of IgG against H. pylori in conjunction with IgA might indicate a more acute infection, compared with cases in which an IgA or IgG alone is present (1, 5, 20–22). Our findings show that not all cases of acute H. pylori infection necessarily involve IgG and IgA positivity. Our data show that in symptomatic individuals, nearly 4% of sera are positive only for IgM. This means that detection of IgM in

82 conjunction with IgG antibodies to H. pylori may better establish the existence of an acute infection. Our study on asymptomatic individuals whose sera were randomly selected provides additional evidence of this observation. Our data show that the number of IgM positives in the general asymptomatic population is nearly 80% less than that of symptomatic patients (2.2 vs 10.2%). Studies show that all IgG-positive individuals are not also IgA-positive, rather a small percentage of patients infected with H. pylori does not show systemic IgG responses and only demonstrates IgA responses (11, 23). Our data confirmed previous studies and showed that about 5% of H. pylori-infected individuals manifest only IgA response. In contrast to a number of published reports which suggest IgM to H. pylori does not have significant specificity, in our study, by Western blot, cross-reactivity, inhibition, and linearity studies, we clearly determined that the detected IgM is specific to H. pylori. We also determined the percentage of precision of the IgM assay to be 2.4–4.2 (% intra-CV) and 3.6–4.4 (% inter-CV). The raw data related to precision studies are not presented. In this study, the percentage of H. pylori antibody-positive specimens selected randomly from patients being blood-tested for reasons unrelated to H. pylori infection was shown to be 42.1% (mostly positive for IgG). In contrast, the percentage of H. pylori antibody-positives in individuals who had complaints of various GI disorders and whose sera were evaluated for the presence of antibodies to H. pylori was lower (33.7%). The random specimens were mostly from adult Hispanic and Asian individuals. A high prevalence of H. pylori infection in Hispanics has been previously reported (24); therefore, we speculate that higher incidences of H. pylori-positive sera reported for this group may be related to the origin of specimens. The ethnicity of subjects in groups three and four were not accurately established, but overall they represented all races. In this study we could not establish any association between age and the presence of IgM in infected individuals. Anderson et al. (25) reported that IgG antibodies to H. pylori antigens with a molecular mass between 15 and 30 kDa are more frequently found in infected individuals. In their study they showed that antibodies to low-molecularweight antigens of H. pylori are reliable markers for the detection of H. pylori infection in children. In this study, we found that a significant number of H. pylori IgM-positive sera reacted with an antigenic fraction of H. pylori in the range of 15 kDa. In contrast, none of the tested sera reacted with the 30-kDa fraction, which is considered to be one of the main antigenic fractions of H. pylori (26). This finding

ALEM ET AL.

correlates with our previous studies in which we demonstrated inconsistant reactivity of IgG antibodies to 30-kDa fractions of H. pylori (15, 19). In summary, the primary goal of the present study was to determine the feasibility and accuracy of H. pylori IgM antibody detection and its diagnostic value in establishing an acute stage of infection. By using a relatively large number of patient specimens we found that IgM antibodies to H. pylori are present in 10.4% of clinically confirmed positive specimens, 1.1% of a random population, and 8.6% of sera suspected for H. pylori infection. In this study we could clearly show the specificity of IgM antibodies to H. pylori. Since the value of detection of IgA and IgG antibodies to H. pylori is clinically established, we think detection of IgM to this microorganism has great merit for a better and more accurate evaluation of acute H. pylori infection.

REFERENCES

1. Perez-Perez, G. I., Dworkin B. M., Chodos, J. E., et al. (1998). Campylobacter pylori antibodies in humans. Ann. Intern. Med. 109, 11–17. 2. Hirschl A. M. (1987) Frequency of occurrence of Campylobacter pylori and analysis of the systemic and local immune response. Zbl. Bakt. A 226, 526–542. 3. Apel, I., Jacobs, E., Kist, M., et al. (1988). Antibody response of patients against a 120 kDa surface protein of Campylobacter pylori. Zbl. Bakt. A 268, 271–176. 4. von Wulffen, H., and Grote, H. J. (1988). Enzyme-linked immunosorbent assay for detection of immunoglobulin A and G to Campylobacter pylori. Eur. J. Clin. Microbiol. Infect. Dis. 7, 559–565. 5. Bolton, F. J., Hutchinson, D. N., Hinchliffe, P. M., et al. (1989). Distribution in various clinical groups of antibody to Campylobacter pylori detected by enzyme-linked immunosorbent assay, complement fixation and microagglutination tests. Serodiagn. Immunother. Infect. Dis. 3, 41–50. 6. Jones, M. D., Eldridge, J., Fox, A. J., et al. (1986). Antibody to the gastric campylobacter-like organism (“Campylobacter pylori”)—Clinical correlation and distribution in the normal population. J. Med. Microbiol. 22, 57–62. 7. Strickland, R. G., and Mackay, I. R. (1973). A reappraisal of the nature and significance of chronic atrophic gastritis. Am. J. Dig. Dis. 18, 426–440. 8. Morris, A., Ali, M. R., Nicholson, G. I., et al. (1991). Long-term follow-up of voluntary ingestion of H. pylori. Ann. Intern. Med. 114, 662–663. 9. Newell, D. G., and Rathbone, B. J. (1989). The serodiagnosis of Campylobacter pylori infection: Serodiagn. Immunother. Infect. Dis. 3, 1–6. 10. Easthan, E. J., Elliott, T. S. J., Berkeley, D., et al. (1988). Campylobacter pylori infection in children. J. Infect. 16, 77–79. 11. Kosunen, T. U., Seppala, K., Sarna, S., et al. (1992). Diagnostic

DETECTION OF IgM ANTIBODIES TO H. pylori

12.

13.

14.

15.

16.

17.

18.

value of decreasing IgG, IgA and IgM antibody titers after eradication of H. pylori. Lancet 339, 893–895. Rosenstock, S., Kay, L., Rosenstock, C., et al. (1997). Relation between H. pylori infection and gastrointestinal symptoms and syndromes. Gut 41(2), 169–176. Dobbs, R. J., Dobbs, S. M., Charlett, A., et al. (2000). Downward shift in serum IgM with H. pylori seropositivity. J. Infect. 41(3), 240–244. Rosenstock, S., Jorgensen, T., Andersen, L., et al. (2000). Seroconversion and seroreversion in IgG antibodies to H. pylori: a serology based prospective cohort study. J. Epidermiol. Community Health 54(6), 444–450. Alem, M., Foley, T. J., and Cohen, H. (1993). Detection of immunoglobulin G antibodies to H. pylori in urine by an enzyme immunoassay method. J. Clin. Microbiol. 31, 2174–2177. Kist, M. (1991). Immunology of H. pylori. In “Helicobacter pylori in Peptic Ulceration and Gastritis,” (B. J. Marshall, et al., Eds.), pp. 92–110, Blackwell Sci., Boston. Newell, D. G. (1987). Human antibody response to the surface antigens of Campylobacter pyloridis. Serodiagn. Immunother. 1, 209–217. Newell, D. G., and Stacy, A. (1989). Antigen for serodiagnosis of Campylobacter pyloridis infections. Gastroentrol. Clin. Biol. 13, 37B–41B.

83 19. Alem, M., Foley, T. J., and Dooley, C. P. (1993). Antibody response to H. pylori antigen fractions. In “93rd General Meeting of ASM” (Abstract V-13), Atlanta, GA. 20. Rathbone, B. J., Wyatt, J. I., Worsely, B. W., et al. (1986). Systemic and local antibody response to gastric Campylobacter pyloridis in non-ulcer dyspepsia. Gut 27, 642–647. 21. von Wulffen, H., Grote, H. J., and Kramer-Hansen, H. (1987). Serological screening for Campylobacter pylori in candidates for renal transplantation. Lancet I, 1140–1141. 22. Reiff, A., Jacobs, E., and Kist, M. (1989). Seroepidemiological study of the immune response to Campylobacter pylori in potential risk group. Eur. J. Clin. Microbiol. Infect. Dis. 8, 592–596. 23. Jaskowski, T. D., Martin, T. B., Hill, H. R., et al. (1997). Immunoglobulin A antibodies to H. pylori. J. Clin. Microbiol. 35, 2999– 3000. 24. Dehesa, M., Dooley, C. P., Cohen, H., et al. (1991). High prevalence of H. pylori infection and histologic gastritis in asymptomatic Hispanics. J. Clin. Microbiol. 29, 1128–1131. 25. Anderson, L. P., Espersen, F., Souckova, A., et al. (1995). Isolation and preliminary evaluation of a low-molecular-mass antigen preparation for improved detection of H. pylori immunoglobulin G antibodies. Clin. Diagn. Lab. Immunol. 2, 156–159. 26. Blaser, M. (1989). “Campylobacter pylori in Gastritis and Peptic Ulcer Diseases.” Igaku-Shion Ltd., New York/Tokyo.