The Microimmunofluorescence Test for Chlamydia pneumoniae Infection: Technique and Interpretation

Transcription

1 S421 The Microimmunofluorescence Test for Chlamydia pneumoniae Infection: Technique and Interpretation San-pin Wang Department of Pathobiology, University of Washington, Seattle A brief description of current procedures for the Chlamydia microimmunofluorescence () test is presented. To date, use of serology with Chlamydia pneumoniae (TWAR) antigen has provided the most sensitive and specific method for diagnosis of acute TWAR infection. In primary infections, the TWAR IgM antibody response is longer lasting and IgG antibody is slower to develop compared with Chlamydia trachomatis infection. Unlike other Chlamydia species, only a single serovar for C. pneumoniae has been recognized in the system and cross-reaction with other species is negligible. While IgM antibody response is an important marker for serodiagnosis of acute infection, rheumatoid factor often causes falsepositive reactions. Persistent TWAR IgG antibody has been useful for seroepidemiologic studies and an association of TWAR IgG antibody and atherosclerotic diseases has been observed. IgA antibody may not be a useful marker for chronic TWAR infection or for acute infection. The complement-fixation (CF) test is the time-honored serologic method for diagnosis of psittacosis and lymphogranuloma venereum [1]. The CF test measures Chlamydia genus antibody. It is also useful for primary infection with Chlamydia pneumoniae (TWAR) infection. It was for use with Chlamydia trachomatis infection that we developed the microimmunofluorescence () test [2, 3], which works well with TWAR as antigen. It was essential to the discovery of the clinical disease caused by TWAR [4 7] and to the description of its epidemiology [8 12]. Here I briefly describe the method [13] and some characteristic features of TWAR serology. Materials and Methods Human sera. All human sera reported here were derived from past studies including sera from patients with C. trachomatis or TWAR infections [3, 6] and some sera from C. Mordhorst, Ornithosis Laboratory, Serum Institute, Copenhagen [10]. The background histories of these patients were known and many of the infections were confirmed by isolation of the organisms. Antigens. The antigens used in tests were from strains of all C. trachomatis serotypes and from Chlamydia psittaci (CP) strains of avian origin (GR-9 from duck, 6BC and PS-1 from parrot, CP-3 from pigeon, and TT-3 from turkey) obtained from Julius Schachter some years ago. TWAR strains used include TW-183 (Taiwan) and AR-39 (Seattle). These strains were grown in HeLa- 229 or HL cell culture, purified, concentrated, and formalinized. serologic method. is an indirect fluorescent antibody (FA) technique that enables observation of the binding of antigen Reprints or correspondence: Dr. San-pin Wang, Dept. of Pathobiology, University of Washington, Box , Seattle, WA (spwang@ u.washington.edu). Journal of Infectious Diseases 2000;181(Suppl 3):S by the Infectious Diseases Society of America. All rights reserved /2000/18106S-0007$03.00 antibody that is detected secondarily by fluorescein-conjugated anti-globulin to the corresponding antibody molecules. The antigens used in the test are whole elementary body Chlamydia organisms that are grown in cell culture, purified, and treated with 0.05% formalin. The formalin-fixed organisms can be stored for many years in the refrigerator. The antigen as used is 10 9 particles per milliliter. It is important that it appears homogenous and dense under 400 magnification. The antigen is mixed with an equal amount of 2% 3% normal yolk sac suspension (normally 0.1 ml each) in order to enhance adhesion of antigen on the microscope slide and to facilitate location of the antigen spot during microscopic examination. In order to differentiate the immunoglobulin class of antibody, fluorescein conjugates of anti-igm or anti-igg are used. In order to measure antibody titer, microprocedures are utilized. Dip pen points are used to place antigen on the slide. A template with prior markings is placed under the slide to guide the placements. The template has a central 15-mm square area with 16 dots arranged 4 4 with each 5 mm from adjacent dots. The slides are dried and fixed with pure acetone for 15 min at room temperature. Serial two-fold dilutions of serum are made and applied to antigens with a bacteriologic loop. The small drops of serial serum dilutions are kept on the surface of the slide by surface tension while reacting with test antigens. The slides are placed in a covered plate with a wet paper towel underneath, incubated for 30 min, and gently rinsed by dips and drains in four changes of PBS and two changes of distilled water. After the slides are dried, fluorescein-conjugate of anti-immunoglobulin mixed with an appropriate counter stain (rhodamine-bovine albumin or Evans blue) is applied with a bacteriologic loop. They are again incubated, rinsed, and dried as before. A coverslip is mounted with appropriate mounting fluid over the antigen spots. Microscopic examination. An FA scope such as a Zeiss Axioskop with epi-illumination works well. Preliminary markings on the edges of the microscope stage are made according to the 16 dot positions of the template. This helps quickly locate each group of antigens. Binocular observations are made using 10 ocular lenses in combination with 10 and 40 objective lenses. Oil im-

2 S422 Wang JID 2000;181 (Suppl 3) Table 1. Age, sex serology for chlamydia CF-positive samples. Serum dates CF IgM IgG GR PS CP TT TWAR GR PS CP TT TWAR 27, M 12/20/ /13/ , F 03/20/ /17/ , F 03/20/ /17/ , M 01/13/ /14/ , F 11/30/ /19/ , M 04/01/ /22/ , M 01/08/ /04/ NOTE. Sample from Danish patients; Serum Institut, Copenhagen. GR, GR-9 duck; PS, PS-1 parrot; CP, CP- 3 pigeon; TT, TT-3 turkey; TWAR, AR-39 human. mersion is not used. Only typical fluorescence associated with evenly distributed Chlamydia organisms is considered a positive reaction. Anti-human immunoglobulin conjugated with fluorescein isothiocyanate (FITC). FITC-conjugated antihuman IgM and IgG were obtained from commercial sources. Appropriate dilutions were determined for each lot of products obtained; normally 1:20 1:30 dilutions were used for indirect FA staining. Gullsorb test. For removal of rheumatoid factor from human serum [14], Gullsorb reagent (anti-human IgG) was obtained from commercial sources. Undiluted reagent was added to serum to make a 1:8 dilution of serum, and serial dilutions were made from the mixture (without further incubation) for routine of IgM antibody. As a control, PBS was added to serum (1:8 dilution) and tested concurrently. No titer decrease of IgM antibody titer in the treated serum compared with the control is considered a real IgM antibody. Results Earlier experiences. Following development of the test, TW-183, the prototype isolate of TWAR organism, was used as antigen with groups of human sera. These tests showed Figure 1. Comparison of IgM and IgG antibody responses in 23 patients with nongonococcal urethritis (NGU) and in 32 with acute respiratory distress syndrome (ARD).

3 JID 2000;181 (Suppl 3) Test for C. pneumoniae S423 Figure 2. responses. Comparison of IgM and IgG TWAR antibody high population antibody prevalence and no relation to eye disease (Grayston JT, elsewhere this symposium). Classification studies failed to find any other isolates similar to TW-183. It was 10 years later that TW-183 was associated with pneumonia [5]. In retrospect, many patients with respiratory tract infection diagnosed as psittacosis on the basis of the chlamydia CF test were really infected with TWAR. Table 1 shows examples of tests with paired sera from 7 Danish patients with acute respiratory tract infections. All had chlamydia CF antibody: 4 had evidence of infection with CP and 3 had TWAR infection. Each of the psittacosis infections was with a different strain. One patient s sera reacted with each of the following antigens: PS-1, GR-9, CP-3, and TT-3. That CF antibody often appears sooner than antibody can be seen in 3 of the CP infections. Each of the 3 acute TWAR infections had a different antibody pattern in which the IgM antibody was important. Four cases reacted to psittacosis strains by titer rise in IgM and/or IgG fractions against PS-1 antigen (derived from a parrot), GR-9 antigen (from a duck), CP-3 antigen (from a pigeon), and TT- 3 antigen (from a turkey). All of these cases had a history of exposure to avian psittacosis infections. However, the remaining examples of respiratory tract infections were only against TWAR antigen and are considered typical of acute TWAR infection. TWAR serotype. Only a single serotype for TWAR organism has been recognized, although some antigenic variation has been proposed. We made antigens from 8 TWAR isolates from different parts of the world. The antigens were tested simultaneously in with paired sera from several patients with acute TWAR infection. Each serum sample reacted similarly with almost identical titers (both in IgM and IgG) to each of these 8 isolates. When the 8 antigens were tested against 55 patient sera, most of the IgM and IgG antibody titers were similar within a two-tube variation. We concluded that the 8 TWAR strains from around the world were similar as antigens in the test [15]. IgM/IgG antibody responses. Figure 1 shows a comparison of antibody responses over time of 23 C. trachomatis culturepositive, nongonococcal urethritis (NGU) patients and 32 acute respiratory disease (ARD) patients (due to TWAR infection). While IgM antibodies were quickly replaced by IgG antibodies in these NGU cases, IgM antibodies rose more slowly but persisted longer in TWAR infections. Figure 2 shows a comparison of geometric mean titers for IgM and IgG antibodies from these 32 ARD cases. For the first 30 days, antibody responses were characterized by relatively high-titered IgM antibody with slowdeveloping lower-titered IgG antibody. This is a typical pattern for acute TWAR infections. IgM antibody titers lasted for more than 60 days, while IgG antibody titers rose after 30 days and persisted beyond the 3 months shown. False IgM reactions. False-positive IgM reactions in the test are not uncommon, especially in older patients. They are due to an autoimmune reaction by IgG to self-immunoglobulin, particularly IgM. This has been called rheumatoid factor [14]. To determine if IgM antibody is true antibody, the serum is absorbed with anti-human IgG antibody (Gullsorb test) to remove IgG prior to testing for IgM. A concurrent test with PBS in place of Gullsorb is the control. Table 2 shows examples of tests with IgM-positive patient sera. The first 2 paired sera are from patients from whom the TWAR organism was isolated. Those 2 still had IgM antibody after treatment with anti-igg antibody, but the other 3 pairs did not. These are examples of false IgM antibody. Interpretation of serologic results. Table 3 shows six types of test results with TWAR antigen. We first screen each serum sample for IgM and IgG antibody at a 1:8 dilution. Positive reactions are then tested at two-fold dilutions to 1:1024. The first pair was negative in the screening test and not tested further. For the second pair, IgG was tested at dilutions and both the first and second serum had antibody at 1:128. This is interpreted as preexisting antibody, like that seen in most adults without acute infection. The last four pairs show different types of antibody patterns that are considered to indicate acute TWAR infection. On the basis of our experience, we consider serologic positives for acute infection to include those with 4-fold antibody titer rise, IgM antibody titer 16, and IgG titer 512 [7]. High antibody titer alone provides Table 2. Gullsorb tests for confirmation of -positive IgM sera from patients with acute respiratory distress syndrome. Age, sex TWAR isolation Serum date IgG IgM IgM after treatment with Anti-IgG PBS 19, M 05/10/ /19/ , F 01/29/ /24/ , F 12/12/ /01/ , F 08/19/ /23/ , M 06/15/ /28/

4 S424 Wang JID 2000;181 (Suppl 3) Table 3. Study test results for detection of TWAR infections. Patients by Antibody screening (1:8) Antibody titers (1:8 1024) Age Sex Date IgM IgG IgM IgG Interpretation 1 55 F Negative 2 53 F Chronic 3 16 M Acute 4 11 F Acute 5 31 F Acute 6 50 M Acute less definite proof of current infection as it may instead be associated with recent infection. While single serum samples may provide presumptive diagnostic information, paired or serial sera allow determination of antibody changes that offer more precise serologic diagnosis. Sensitivity of test for TWAR laboratory diagnosis. Table 4 shows results of laboratory diagnoses from our four major studies of respiratory infections. Two were done during TWAR epidemics and the other two in nonepidemic periods. We had complete control of specimen collection and patient data. The serologic positives include four-fold antibody titer rise, IgM antibody 116, and high titer IgG antibody 512. Of the positive patients, 63% 75% had the organism demonstrated by isolation, polymerase chain reaction (PCR), or both. The organism was found more frequently in the two studies that used PCR in addition to isolation. Only 1 isolation- or PCR-positive subject failed to show an acute antibody response and that patient had an IgG titer of 128. Isolation and, to a lesser extent, PCR have been dependent on careful maintenance of the cold chain from patients to laboratory. Even in the best circumstances, demonstration of the organism has been less successful than serology in identifying acute infection [12]. Seroepidemiology. Soon after the discovery of the significance of the TWAR organism in human infection, extensive epidemiologic studies were done by serology using TWAR antigens to extend knowledge of the epidemiology of TWAR respiratory infections [10, 11] and to describe its population antibody prevalence [8 12]. Seroepidemiologic studies also suggested an association of C. pneumoniae infection and coronary artery disease, the major subject of discussion of this symposium. The initial study was reported from Finland by Saikku et al. in 1988 [16]. My colleagues and I confirmed the finding [17]. Many investigators in different parts of the world have performed similar studies [18]. Discussion If serology for chlamydial infections is reviewed as a whole, some features appear to be characteristic for TWAR infections. While cross-reactions among closely related serotypes of C. trachomatis are well known [3], cross-reactions between TWAR and other species (e.g., C. trachomatis and C. psittaci) have not been seen. While multiple serotypes have been recognized in C. trachomatis and C. psittaci, only a single serotype has been recognized for TWAR organism [15]. Despite great efforts to determine dominant antigens responsible for the specificity, a clear answer has not been obtained [7]. The development of an effective ELISA test might have been hampered by the lack of such information. For laboratory diagnosis of C. trachomatis infections, detection of the organism by isolation or PCR is considered the reference standard. Our results with TWAR infections suggest that improvement in organism detection is needed. In primary TWAR infection IgM antibody is uniquely useful for serodiagnosis. False IgM antibody reactions occur because of rheumatoid factor. It is necessary to routinely test IgM positives by prior absorption with anti-human IgG antibody. IgA antibody has been proposed as a useful marker for chronic TWAR infection [19]. The drawback (unpublished data) is the considerable variation in the commercial fluorescein conjugates of anti-human IgA, which creates a disadvantage for accurate detection of IgA antibody in humans. Even when one of the best conjugates is used, IgA antibody has not been detected in a long-term and frequent follow-up of the majority of patients who have laboratory-confirmed acute TWAR infections. In a follow-up study of persons with long-lasting IgG antibodies, a large portion of the subjects were also negative for IgA antibody. The presence of IgA antibody is always accompanied by IgG antibody in the same serum sample. It is obvious that many people never produce IgA antibody. Those with IgA antibody remain negative despite having long-lasting IgG antibody. We do not know why the production of IgA antibody is limited to certain persons. Positive IgA in addition to positive IgG antibodies occur more frequently in old than in young adults. However, there is no evidence to suggest that IgA antibody is associated with chronic TWAR infection. There appears to be no reason to consider IgA antibody as a useful marker for either acute or chronic infection. Table 4. Results of C. pneumoniae laboratory diagnostic tests in four studies of acute respiratory tract infections. Study No. positive Isolation or PCR positive positive negative UW students (epidemic) UW students (endemic) Group Health Cooperative of Puget Sound (endemic) Finnish military (epidemic) NOTE. PCR, polymerase chain reaction; UW, University of Washington.

5 JID 2000;181 (Suppl 3) Test for C. pneumoniae S425 Acknowledgments I thank my countless colleagues for collaboration, in particular J. Thomas Grayston, for meticulous and invaluable contributions in the preparation of this article, and Cho-chou Kuo, Lee Ann Campbell, and Carl H. Mordhorst, for intellectual support. Special thanks to my long-time laboratory assistant, Elsie Lee, and the technical and clerical assistance of David Grayston and Ingrid Noé. References 1. Meyer KF, Eddie B. Psittacosis. In: Lennnet EH, Schmidt NJ, eds. Diagnosis procedures for virus and rickettsial diseases. 2nd ed. New York: American Public Health Association, 1956: Wang SP, Grayston JT. Immunological relationship between genital TRIC and related organisms in a new microtiter indirect immunofluorescence test. Am J Ophthalmol 1970;70: Wang SP, Grayston JT. Micro-immunofluorescence serology of Chlamydia trachomatis. In: de la Maza LM, Peterson EM, eds. Medical Virology III. New York: Elsevier Science, 1984: Saikku P, Wang SP, Kleemola M, et al. An epidemic of mild pneumonia due to an unusual strain of Chlamydia psittaci. J Infect Dis 1985;151: Grayston JT, Wang SP. History of Chlamydia pneumoniae (TWAR). In: Allegra L, Blasi F, eds. Chlamydia pneumoniae: the lung and heart. Milan, Italy: Springer-Verlag Italia, 1999: Grayston JT, Kuo CC, Wang SP, et al. A new Chlamydia psittaci strain called TWAR from acute respiratory tract infections. N Engl J Med 1986;315: Kuo CC, Jackson LA, Campbell LA, Grayston JT. Chlamydia pneumoniae (TWAR) [review]. Clin Microbiol Rev 1995;8: Wang SP, Grayston JT. Microimmunofluorescence serological studies with the TWAR organism. In: Oriel JD, ed. Chlamydial infections. Cambridge, UK: Cambridge University Press, 1986: Wang SP, Grayston JT. Population prevalence antibody to Chlamydia pneumoniae, strain TWAR. In: Bowie WR, ed. Chlamydial infections: proceedings of the Seventh International Symposium on Human Chlamydial Infections. Cambridge, UK: Cambridge University Press, 1990: Mordhorst CH, Wang SP, Grayston JT. Chlamydia pneumoniae, strain TWAR, infections in Denmark In: Bowie WR, ed. Chlamydial infections: proceedings of the Seventh International Symposium on Human Chlamydial Infections. Cambridge, UK: Cambridge UniversityPress, 1990: Aldous MB, Grayston JT, Wang SP, et al. Seroepidemiology of Chlamydia pneumoniae (TWAR) infections in Seattle families J Infect Dis 1992;166: Wang SP, Grayston JT. Chlamydia pneumoniae (TWAR) micro-immunofluorescence antibody studies 1998 update. In: Stephens RS, ed. Chlamydia infections. Berkeley, CA: University of California Printing Services, 1998: Wang SP. Serology for Chlamydia pneumoniae (TWAR). In: Allegra L, Blasi F, eds. Chlamydia pneumoniae: the lung and heart. Milan, Italy: Springer- Verlag Italia, 1999: Verkooyen RP, Hazenberg MA, van Haaren GH, et al. Age-related interference with Chlamydia pneumoniae micro-immunofluorescence serology due to circulating rheumatoid factor. J Clin Microbiol 1992;30: Wang SP, Grayston JT. The similarity of Chlamydia pneumoniae (TWAR) isolates as antigen in the microimmunofluorescence () test. In: Orfila J, ed. Chlamydial infections. Bologna, Italy: Societa Editrice Esculapio, 1993: Saikku P, Leinonen M, Mattila K, et al. Serological evidence of an association of novel Chlamydia, TWAR, with chronic coronary heart disease and acute myocardial infarction. Lancet 1988;2: Thom DH, Grayston JRT, Siscovick DS, et al. Association of prior infection with Chlamydia pneumoniae and angiographically demonstrated coronary artery disease. JAMA 1992;268: Grayston JT. Does Chlamydia pneumoniae cause atherosclerosis? Arch Surg 1999;134: Saikku P, Leinonen M, Tenkanen L, et al. Chronic Chlamydia pneumoniae infection as a risk factor for coronary heart disease in the Helsinki Heart Study. Ann Intern Med 1992;116:273 8.