Vol. 30, No. 8

JOURNAL OF CLINICAL MICROBIOLOGY, Aug. 1992, p. 2181-2183

0095-1137/92/082181-03$02.00/0 Copyright © 1992, American Society for Microbiology

Evaluation of a Urease-Based Confirmatory Enzyme-Linked Immunosorbent Assay for Diagnosis of Neisseria gonorrhoeae M. CARBALLO,1 J. R. DILLON,`* M. LUSSIER,2'3 P. MILTHORP,3t S. WINSTON,3t AND B. BRODEUR2 National Laboratory for Sexually Transmitted Diseases' and National Laboratory for Immunology, 2 Laboratory Centre for Disease Control, Ottawa, Ontario KIA OL2, and ADI Diagnostics Inc., Rexdale, Ontanio M9W 4Z7, Canada Received 23 October 1991/Accepted 14 May 1992

A new urease-based enzyme-linked immunosorbent assay utilizing novel monoclonal antibodies was evaluated for the culture confirmation ofNeisseria gonorrhoeae, with 270 isolates of N. gonorrhoeae, 56 isolates of diverse Neisseria spp., and 29 MoraxeUa isolates. The test was highly specific (100.00o) and sensitive (97.83%). No cross-reactions were observed with any of the Neisseria or MoraxeUa isolates tested. Fifty percent (3 of 6) of the false-negative results were obtained with isolates of serovar IA4, a serovar rarely encountered in North America.

Neisseria gonorrhoeae is one of the leading causes of sexually transmitted diseases (10, 20, 21). The accurate and expedient diagnosis of gonorrhea requires laboratory tests that are highly sensitive and specific. The traditional method for identifying Neisseria species was a carbohydrate degradation test using cystine-tryptic digest agar base medium (CTA sugars) (23, 27); this test requires 24 to 48 h of incubation. Recently, alternative rapid methods, including substrate (carbohydrate and chromogenic) utilization tests, various immunological tests (1, 2, 5, 11, 12, 16, 17, 22, 25, 28), and DNA probe tests (14, 19, 20, 24, 26), have been evaluated. Several immunological methods, such as coagglutination and fluorescent-antibody tests, have been used as confirmatory tests after primary culture (2, 11, 12, 17). Enzyme immunoassays, such as Gonozyme, directly detect gonococcal antigens from clinical specimens, eliminating the need for primary culture but compromising the sensitivity and specificity of the test (16, 25). In order to maximize automation without hindering the sensitivity and specificity of the test, a unique urease-based sandwich enzyme-linked immunosorbent assay (ELISA) system (the ELISA G.C. kit) using previously developed monoclonal antibodies was developed (ADI Diagnostics Inc., Rexdale, Ontario, Canada). We report on the evaluation of this system for its ability to confirm the presence of N. gonorrhoeae. A total of 355 strains were evaluated with the ELISA G.C. kit. These included 270 N. gonorrhoeae isolates, consisting of nine auxotypes and 14 serovars and 56 Neisseria isolates (33 N. meningitidis isolates belonging to nine serogroups, 9 N. subflava isolates, 5 N. cinerea isolates, 4 N. lactamica isolates, 3 N. sicca isolates, 1 N. mucosa isolate, and 1 N. flavescens isolate), and 29 Moraxella isolates (26 Branhamella catarrhalis isolates and 3 Moraxella phenylpyruvica isolates). The isolates were selected from the culture collection of the National Laboratory for Sexually Transmitted * Corresponding author. t Present address: Cyberfluor, Toronto, Ontario M5T 1X4, Canada. t Present address: Univax Biologics Inc., Rockville, MD 20852.

Diseases, Laboratory Centre for Disease Control, Ottawa, Ontario, Canada. This collection included 88 isolates collected by the National Laboratory for Sexually Transmitted Diseases during a 1988 to 1989 National Surveillance Study (8) and 182 isolates previously selected for the evaluation of different diagnostic kits and serotyping methods for N. gonorrhoeae (5, 11). All of the isolates had been stored either frozen at -70°C in brain heart infusion broth (Difco Laboratories, Detroit, Mich.) plus 15% glycerol or lyophilized in 2% skim milk (9). Before being tested, isolates were subcultured on GC medium base (Difco) supplemented with 1% Kellogg defined supplement and incubated overnight in a humid atmosphere with 5% CO2. Isolates were initially identified or reconfirmed by Gram stain, oxidase reaction, colony morphology, and sugar utilization tests. The identities of isolates giving anomalous results were reconfirmed by immunological tests, GonoGen, Syva MicroTrak, and a carbohydrate utilization test, Minitek. The two monoclonal antibodies (MAbs), NG-38 and NG78, used in this capture assay (22) were obtained following the immunization of mice with lithium acetate-extracted outer membrane preparations (3, 22). Both MAbs are immunoglobulin G2a; NG-38 and NG-78 recognized only protein IB (PIB) and protein IA (PIA) strains of N. gonorrhoeae, respectively, as ascertained by dot immunoassay (22). These MAbs did not react with other Neissenia spp. or more distantly related bacterial species. The sandwich ELISA kit, prepared at ADI, included antibody-coated microwell strips (12 x 1 mm), extraction buffer, monoclonal antibody conjugate, substrate solution, and positive and negative control solutions. The 12-well strips were precoated with both purified rabbit antibodies against N. gonorrhoeae PIA and PIB strains. Bacterial suspensions were prepared in 0.1 M phosphate-buffered saline (PBS), and their optical densities were measured at 470 nm. The suspensions were diluted in extraction buffer to a final concentration of approximately 5 x 107 cells per ml or within a 10-fold range of this value. Antibody-coated strips were inoculated with the extracted samples (final concentration, 5 x 105 cells per well) followed by 50 ,ul of ureaseconjugated monoclonal antibody solution (6, 15). The posi2181

2182

J. CLIN. MICROBIOL.

NOTES

tive and negative controls consisted of individual outer membrane extracts (5 jig/ml) of PIA strain NLI-S2, PIB strain NLI-S3 of N. gonorrhoeae, and strain 604A of N. meningitidis. After the extracted bacteria and MAb conjugates were incubated for 30 min at 37°C, the plate was washed six times with 0.85% saline, the urease substrate solution was added, and the plate was incubated for 30 min at room temperature. The A620 of the product of the reaction was read. A result was defined as positive when the absorbance of the sample minus the mean absorbance of two negative controls was greater than the control absorbance plus a factor of 0.05. All testing was performed in a singleblind study format. Of 276 isolates identified as N. gonorrhoeae by biochemical and immunological identification tests, the ELISA identified 270 and failed to identify 6. Of the 85 nongonococcal isolates identified by biochemical identification tests, the ELISA identified as negative all 85. The ELISA was 100% specific, as no other species were misidentified as N. gonorrhoeae. However, even after retesting, the immunoassay failed to identify six isolates of N. gonorrhoeae, thus contributing to a sensitivity of 97.83%. The positive predictive value was 100.00%, while the negative predictive value was 93.40%. All of the false-negative isolates were prototrophic; two belonged to serovar IB-1, one belonged to serovar IB-2, and three belonged to serovar IA-4. Difficulties in detecting isolates belonging to serovar IA-4 have been previously reported when the coagglutination method was used and when an enzyme immunoassay developed for the identification and serotyping of N. gonorrhoeae isolates was used (5, 22). Isolates belonging to serovar IA-4 are rare in North America (8, 18). In a recent Canadian national survey (1988 to 1989), only 0.2% of isolates belonging to IA serovars were typed as IA-4 (8). Other groups have reported this serovar to be common in isolates from geographical areas such as Sweden, where 45% of the penicillinase-producing N. gonorrhoeae isolates were serotyped as IA-4, and Scotland, which has 6% of all the PIA isolates belonging to this serovar (4, 7). The detection of such isolates and those reported to be nontypeable (13) may require the addition of an extra or alternate monoclonal antibody. With the ELISA kit, no false positives were detected with any of the nongonococcal isolates tested. By contrast, polyclonal antibody-based ELISAs for the direct screening of endocervical and urethral specimens, such as GONOZYME (Abbott), have a specificity ranging from 87.2 to 96.8% (16, 25). Periodic upgrading of ELISA kits to account for the genetic variation of protein I will ensure their reliability as confirmatory tests. Additional evaluations will be required to demonstrate the utility of this ELISA kit for the diagnosis of N. gonorrhoeae directly from urogenital samples. In conclusion, the present ELISA satisfied the criteria one looks for in a diagnostic kit: it was highly sensitive (97.83%) and specific (100.00%), easy to run, and economical (less than $2 per sample), allowing for the testing of a large number of samples, compared with coagglutination and fluorescence tests. REFERENCES 1. Arko, R. J., K. G. Finley-Price, K.-H. Wong, S. R. Johnson, and G. Reising. 1982. Identification of problem Neissena gonorrhoeae cultures by standard and experimental tests. J. Clin. Microbiol. 15:435-438. 2. Boehm, D. M., M. Bernhardt, T. A. Kurzynski, D. R. Pennell, and R. F. Schell. 1990. Evaluation of two commercial procedures for rapid identification of Neissena gonorrhoeae using a reference panel of antigenically diverse gonococci. J. Clin.

Microbiol. 28:2099-2100. 3. Brodeur, B. R., F. E. Ashton, and B. B. Diena. 1978. Enzymelinked immunoabsorbent assays for the detection of Neisseria gonorrhoeae specific antibodies. Can. J. Microbiol. 24:13001305. 4. Bygdeman, S. M. 1987. Polyclonal and monoclonal antibodies applied to the epidemiology of gonococcal infections, p. 117165. In H. Young and A. McMillan (ed.), Immunological diagnosis of sexually transmitted diseases. Marcel Dekker, Inc., New York. 5. Carballo, M., and J. R. Dillon. Evaluation of an enzyme immunoassay and a modified coagglutination assay for typing gonococcal isolates with monoclonal antibodies. Sex. Transm. Dis., in press. 6. Chandler, H. M., A. R. Coulter, K. Healy, M. Kornitschuk, A. MacGregor, and J. G. R. Hurrell. 1983. Monoclonal hybridoma antibodies against human IgE and their use in a rapid and sensitive enzyme immunoassay for the semiquantitative assessment of total IgE levels in human blood. Int. Arch. Allergy Appl. Immunol. 72:267-272. 7. Coghill, D. V., and H. Young. 1987. Serological classification of Neissena gonorrhoeae with monoclonal antibody coagglutination reagents. Genitourin. Med. 63:225-232. 8. Dillon, J. R. Unpublished data. 9. Dillon, J. R. 1983. Laboratory methods for Neissenia gonorrhoeae (MIC, auxotype, screening for PPNG). Catalog no. H47-58/1983E. National Health and Welfare, Ottawa, Ontario, Canada. 10. Dillon, J. R., S. M. Bygdeman, and E. G. Sandstrom. 1987. Serological ecology of Neisseria gonorrhoeae (PPNG and nonPPNG) strains: Canadian perspective. Genitourin. Med. 63:160168. 11. Dillon, J. R., M. Carballo, and M. Pauze. 1988. Evaluation of eight methods for identification of pathogenic Neisseria species:

12. 13.

14. 15. 16. 17. 18.

19.

20. 21.

22. 23.

Neisseria-Kwik, RIM-N, Gonobio-Test, Minitek, Gonochek II, GonoGen, Phadebact Monoclonal GC OMNI Test, and Syva MicroTrak Test. J. Clin. Microbiol. 26:493-497. Dolter, J., L. Bryant, and J. M. Janda. 1980. Evaluation of five rapid systems for the identification of Neisseria gonorrhoeae. Diagn. Microbiol. Infect. Dis. 13:265-267. Gill, M. J. 1991. Serotyping Neisseria gonorrhoeae: a report of the Fourth International Workshop. Genitourin. Med. 67:53-57. Granato, P. A., and M. R. Franz. 1990. Use of the Gen-Probe PACE system for the detection of Neissenia gonorrhoeae in urogenital samples. Diagn. Microbiol. Infect. Dis. 13:217-221. Healy, K., H. M. Chandler, J. C. Cox, and J. G. R. Hurrell. 1983. A rapid semiquantitative capillary enzyme immunoassay for digoxin. Clin. Chim. Acta 134:51-58. Hossain, A., T. M. F. Baldr, M. Siddiqui, and S. de Silva. 1988. Enzyme immunoassay (EIA) in the rapid diagnosis of gonorrhoea. J. Hyg. Epidemiol. Microbiol. Immunol. 32:425-431. Ison, C. A., A. Tanna, and C. S. F. Easmon. 1988. Evaluation of a fluorescent monoclonal antibody reagent for identification of cultured Neissena gonorrhoeae. J. Med. Microbiol. 26:121-123. Knapp, J. S., K. K. Holmes, P. Bonin, and E. W. Hook III. 1987. Epidemiology of gonorrhea: distribution and temporal changes in auxotype/serovar classes of Neisseria gonorrhoeae. Sex. Transm. Dis. 14:26-32. Kuritza, A. P., S. C. Edberg, C. Chapis, and P. Gallo. 1989. Identification of Neisseria gonorrhoeae with the ORTHOProbe DNA probe test. Diagn. Microbiol. Infect. Dis. 12:129-132. Lewis, J. S., D. Kraning-Brown, and D. A. Trainor. 1990. DNA probe confirmatory test for Neisseria gonorrhoeae. J. Clin. Microbiol. 28:2349-2350. Lind, I. 1990. Epidemiology of antibiotic resistant Neisseria gonorrhoeae in industrialized and developing countries. Scand. J. Infect. Dis. 69:77-82. Lussier, M., B. R. Brodeur, and S. Winston. 1989. Detection of Neisseria gonorrhoeae by dot-enzyme immunoassay using monoclonal antibodies. J. Immunoassay 10:373-394. Morello, J. A., W. M. Janda, and M. Bohnhoff. 1985. Neisseria and Branhamella, p. 176-192. In E. H. Lennette, A. Balows, W. J. Hausler, Jr., and H. J. Shadomy (ed.), Manual of clinical

VOL. 30, 1992

microbiology, 4th ed. American Society for Microbiology, Washington, D.C. 24. Panke, E. S., L. I. Yang, P. A. Leist, P. Magevney, R. J. Fry, and R. F. Lee. 1991. Comparison of Gen-Probe DNA probe test and culture for the detection of Neissena gonorrhoeae in endocervical specimens. J. Clin. Microbiol. 29:883-888. 25. Papasian, C. J., W. R. Bartholomew, and D. Amsterdam. 1984. Validity of an enzyme immunoassay for detection of Neisseria gonorrhoeae antigens. J. Clin. Microbiol. 19:347-350. 26. Rossau, R., M. Duhamel, E. Van Dyck, P. Piot, and H. Van

NOTES

2183

Heuverswyn. 1990. Evaluation of an rRNA-derived oligonucleotide probe for culture confirmation of Neisseria gonorrhoeae. J. Clin. Microbiol. 28:944-948. 27. Vera, H. D. 1948. A simple medium for identification and maintenance of the gonococcus and other bacteria. J. Bacteriol. 55:531-536. 28. Yong, D. C. T., and A. Prytula. 1978. Rapid micro-carbohydrate test for confirmation of Neisseria gonorrhoeae. J. Clin. Microbiol. 8:643-647.

Evaluation of a urease-based confirmatory enzyme-linked immunosorbent assay for diagnosis of Neisseria gonorrhoeae.

A new urease-based enzyme-linked immunosorbent assay utilizing novel monoclonal antibodies was evaluated for the culture confirmation of Neisseria gon...
498KB Sizes 0 Downloads 0 Views