due to Mycobacterium avium-intracellulare Complex

Transcription

1 F Original Article Enzyme-Linked Immunosorbent Assay for the Differential iagnosis of Pulmonary Tuberculosis and Pulmonary Diseases due to Mycobacterium avium-intracellulare Complex iroshi AMANO, Kenji MlZOGUCHI, Michio TSUKAMURA*, Takanao MURATE, Tadao HASEGAWAand Kaoru SHIMOKATA Westudied the usefulness of antibody estimation by the enzyme-linked immunosorbent assay (ELISA) technique in the differential diagnosis of pulmonary mycobacterial diseases. Sera from patients with active pulmonary diseases caused by Mycobacterium tuberculosis (n = 42) or Mycobacterium avium-intracellulare complex (MAI) (n = 26) and from healthy controls (n = 75) were assayed for IgG and IgA antibody titers to antigen of M. tuberculosis and MAI. The IgG and IgA antibody titers to M. tuberculosis antigen were significantly higher in patients with mycobacterial diseases than in healthy controls, however, there was no significant difference between patients with pulmonary tuberculosis and those with pulmonary diseases due to M. avium-intracellulare complex (MAID). The IgG and IgA antibody titers to MAI antigen in MAID patients were significantly higher than in healthy controls and in patients with pulmonary tuberculosis. ELISA for IgG antibody against M. tuberculosis antigen had a sensitivity of 54.8% and a specificity of 73.1% in the differential diagnosis of pulmonary tuberculosis. The estimation of the IgA antibody against MAI antigen using ELISA is more useful for the differential diagnosis of MAID because of its sensitivity (69.2%) and specificity (90.5%). Key Words: Enzyme-linked immunosorbent assay (ELISA), Mycobacterium avium-intracellulare complex disease, Pulmonary tuberculosis The growth of mycobacterial species is generally very slow, and about four to eight weeks are required for identification of the organism. It has been hoped that a method would be developed for the rapid differential diagnosis of pulmonary atypical mycobacteriosis and tuberculosis. IgG, IgA or IgM antibody to antigens derived from M. tuberculosis, bacille de Calmette-Guerin (BCG) or M. kansasii with ELISA has been estimated for the diagnosis of pulmonary tuberculosis (1-22). Moreover, ELISA and tuberculin reactivity complement each other diagnostically (8). Increased recognition of the prevalence of pulmonary disease caused by mycobacteria other than M. tuberculosis has focused on the diagnosis of the pulmonary diseases caused by MAI and M. kansasii (23, 24). In Japan, MAID is the most frequent in atypical mycobacteriosis. But there hae been few reports concerning estimation of antibodies against MAI with ELISA except a report of disseminated infection by MAI in acquired immunodeficiency syndrome (AIDS) and hairy cell leukemia (25). Therefore, we studied the usefulness rom the First Department of Internal Medicine, Nagoya University School of Medicine, Nagoya and *the National Chubu Hospital, Ohbu Received for publication September 26, eprint requests to: Kaoru Shimokata, MD, the First Department of Medicine, Nagoya University School of Medicine, 65, Tsurumai-cho, Showa-ku, Nagoya 466, Japan. 196 Jpn J Med Vol 28, No 2 (March, April 1989)

2 ELISA for Differential Diagnosis of Mycobacterial Diseases of ELISA for measuring the IgG and IgA antibodies against antigens derived from MAI and M. tuberculosis so as to enable differential diagnosis of pulmonary MAID and tuberculosis. Because of the less significance of IgM antibodies (2, 7, 1 1), we did not measure IgM antibodies. MATERIALS AND METHODS Patients and healthy controls. Patients with pulmonary MAID were diagnosed according to the 1985 edition of Diagnostic Standard of Atypical MycobacterialDiseases (26). Forty two patients with active pulmonary tuberculosis had positive smear or positive culture growth of M. tuberculosis from sputum specimens and infiltrates on chest radiograph. Their mean age was 54.5 (SE=3.0) years old. Twenty six patients with MAID were diagnosed by frequent detection and many times of isolation of the organism from sputum by culture in the presence of an infiltrate on chest radiograph. Their mean age was 63.0 (SE =2.5) years old. Thirteen had a clear history of pulmonary tuberculosis, while the rest had no such history. They had no malignancy and were not immunosuppressed. Seventy five healthy persons who showed positive tuberculin skin test were also examined. Their mean age was 37.4 (SE=3.0) years old. Antigens. A purified protein derivative (PPD) sample of M. tuberculosis H37Ra was a product- of Mitsui Seiyaku Co., Ltd., Tokyo. It was dissolved at 30 /ig/ml in phosphate-buffered saline (PBS) and used as M.tuberculosis antigen. MAI antigen was prepared as follows. A modified Sauton medium was inoculated with one loopful of MAI (Strain 13011, Gamoh) and incubated at 37 C for 28 days. The culture fluid was sterilized by heating at 100 C for 30 min and then centrifuged at 3,000 rpm for 20 min. The supernatant thus obtained was stored in a refrigerator until use. The protein concentration in the supernatant was 46 ^g/ml. According to Tsukamura (27), this concentration of antigen solution was equivalent to 15 /*g/ml of the tt-antigen prepared by Takeya et al. (28) by the skin test in humans. The modified Sauton medium was composed of 10 g glucose, 4.0 g sodium glutamate, 0.5 g KH2PO4, 0.5 g MgSO4-7H2O and 1000 ml of distilled water.the ph was adjusted to 7.0 by adding 10% KOH solution. The medium was sterilized by autoclaving at 120 C for 20 min. Antisera. Peroxidase-conjugated goat antisera specific for human IgG and IgA were obtained rom Cappel Laboratories Inc., Penn. U.S.A. ELISA for antibody to antigen solution of M. tuberculosis or culture filtrate of MAI. Each well of disposable microtiter flexible assay plates (Falcon 3912; Becton Dickinson and Co., Oxnard, Ca. U.S.A.) was filled with 100^1 of antigen solution and incubated for 24 hr at 4 C. The plates were then washed with PBS and flooded with 2% bovine serum albumin (BSA) for 24 hr at 4 C to block the free absorption sites. In a preliminary study in which sera were serially diluted from 1:10 to 1:2560, dilution from 1:10 to 1:40 was appropriate for the optical density (OD) reading in our series, therefore, we used 1 :10 diluted serum. Each serum sample was diluted with PBS, and 50[A of this diluted serum was added to the antigen-coated wells. After 30 min incubation at 37 C and washing five times with PBS containing 0.05% Tween, 50 f.t\ of peroxidaseonjugated antibody to human IgG or IgA diluted 1 :4000 in PBS was added to each well. After incubation for 30 min at 37 C and washing five times with PBS containing 0.05% Tween, the peroxidase enzymatic activity was assayed by adding 100 //I of substrate, o-phenylenediamine in a phosphate - citrate buffered saline with 0.01 % H2O2. The reaction was stopped by the addition of 50/^1 of 4N H2SO4 after 15 min. The OD was determined with a spectrophotometer (Minireader II; Dynatech Laboratories, Alexandria, Va. U.S.A.). The machine was calibrated using the blank substrate, and the OD was read at 490 nm (10). Because the antibody level of a single specimen as reflected by the OD value varied from day to day, the assay was standardized by comparing the OD values of the test sera to the standard reference serum of high antibody content. The standard serum was included in each plates. An optical density index (ODI) for each serum was determined with the standard reference serum of high antibody content assigned a value of 100%. Statistical methods. The differences in the mean antibody concentrations between the groups were analyzed using the analysis of variance. The incidence of a positive rate in the antibody titer was analyzed using the chi square test. The results were Jpn J Med Vol 28, No 2 (March, April 1989) 197

3 Amanoet al considered statistically significant with p values of <0.05. RESULT The mean ODI± SE levels of IgG and IgA antibodies to M. tuberculosis and MAI antigens are presented in Figures 1-4. The mean ODI levels of IgG and IgA antibodies to M. tuberculosis antigen in the pulmonary tuberculosis and MAID groups were significantly higher than those in the healthy controls (Figures 1-2). The mean ODI levels of both IgG and IgA antibodies to MAI antigen in the MAID group were significantly higher than those in the healthy controls and the tuberculosis group (Figures 3-4). The cutoff points for the ODI levels of IgG and IgA antibodies to M. tuberculosis antigen were determined to be and 99.2, whose value was S mean + 2SD in the healthy controls. As for positive rate in IgG, there were significant differences between the healthy controls and mycobacterial diseases, and between pulmonary tuberculosis group and MAID group as shown in Table 1. When MAID patients were divided into two groups, that is, with and without a history of tuberculosis, there was a significant difference of positive rate for IgG antibody against PPD antigen between pulmonary tuberculosis group (23/42, 54.8%) and MAID group without a history of tuberculosis (2/13, 15.4%). The sensitivity (percentage of positive tests among tuberculous patients) and specificity (percentage of negative tests among MAID patients) of ELISA for IgG antibody to PPD were 54.8% and 73.1% (Table 1). When patients with anamnesis of pulmonary tuberculosis were excluded from MAID patient group, the specificity increased from 73.1% to 84.6%. Similarly, the cutoff points for the ODI Fig. 1. The ELISA IgG titers to PPD antigen in sera MAID, and from tuberculin-positive healthy controls. Open circles represent pulmonary MAID patients with a history of pulmonary tuberculosis. Vertical bars represent mean± SE values. A horizontal line represents a cutoff level determined by mean + 2SD of healthy controls. There are significant differences of ODI between patients with pulmonary tuberculosis (101.1±3.6) and healthy controls (72.7±1.8) (p <0.001), and between patients with MAID (98.6 ± 2.5) and healthy controls (p< 0.001). Fig. 2. The ELISA IgA titers to PPD antigen in sera MAID, and from tuberculin-positive healthy controls. Open circles represent pulmonary MAID patients with a history of pulmonary tuberculosis. There are significant differences of ODI between patients with pulmonary tuberculosis (98.9± 3.8) and healthy controls (73.0± 1.5) (p<0.001), and between patients with MAID (93.2±4.1) and healthy controls (p<0.001). 198 Jpn J Med Vol 28, No 2 (March, April 1989)

4 ELISA for Differential Diagnosis of Mycobacterial Diseases Fig. 3. The ELISA IgG titers to MAI antigen in sera MAID, and from tuberculin-positive healthy controls. There are significant differences of ODI between patients with pulmonary tuberculosis (87.1 ± 3.0) and those with MAID (96.9± 3.2) (p <0.05), and between patients with MAID and healthy controls (69.0± 1.9) (p <0.001). Fig. 4. The ELISA IgA titers to MAI antigen in sera MAID, and from healthy controls. There are significant differences of ODI between patients with pulmonary tuberculosis (78.1±4.4) and those with MAID (114.9±5.4) (p <0.001), and between patients with MAID and healthy controls (65.4±2.5) (p <0.001). able 1. Positive rate for antibody titer to a PPD of M. tuberculosis using the optical density indexa) Table 2. Positive rate for antibody titer to the culture filtrate of M. avium-intracellulare complex using the optical density indexa) The cutoff points for the ODI levels of IgG and IgA antibodies were determined to be and 99.2 whose values meant mean +2SD in the healthy controls. Significant differences between 1-2 (p<0.05), 1-3 (p<0.001) and 2-3 (p<0.001). Significant differences between 1-3 (p<0.001) and 2-3 (p<0.001) levelsof IgG and IgA antibodies to MAI antigen weredetermined to be and 108.5, respectively. There were statistically significant differences between MAID and healthy controls and between The cutoff points for the ODI levels of IgG and IgA antibodies were determined to be and whose values meant mean +2SD in the healthy controls. Significant differences between 1-3 (p <0.001) and 2-3 (p<0.001) Significant differences between 1-2 (p<0.001) and 2-3 (p<0.001) pulmonarytuberculosis and healthy controls in relation to IgG antibody, and between MAID and the other two groups in relation to IgA antibody (Table 2). The sensitivity and specificity of ELISA for IgA Jpn J Med Vol 28, No 2 (March, April 1989) 199

5 Amanoet al antibody to MAI antigen were 18/26 (69.2%) (percentage of positive tests among MAID) and 38/42 (90.5%) (percentage of negative tests among pulmonary tuberculosis). These data were shown in Table 2. DISCUSSIO Some studies have been concerned with the antibody titer to M. tuberculosis or M. kansasii derived antigens using ELISA in pulmonary atypical myocbacterial diseases (AMD) (6, 10, 13, 14). However, these reports contain a variety of AMD and each number of patients with AMD is small. The most frequent pulmonary AMD in our experience is MAID. Our study includes a larger number of patients with MAID than has been reported in any other similar studies. In the estimation of IgG antibody to PPD antigen, the sensitivity for pulmonary tuberculosis was 54.8% and the specificity was 73.1%. Therefore, 26.9% patients with MAID showed positive test. There may be two reasons for the high positive test inmaid. One is the cross antigenicity between M. tuberculosis and MAI. The other is the past history of pulmonary tuberculosis in MAID patients. When MAID patients were divided into two groups, those with or without a history of pulmonary tuberculosis, the positive test was observed in 5 of 13 (38.5%) patients in the former and 2 of 13 (15.4%) patients in the latter. The most interesting finding in the present study is that ELISA using the MAI antigen was more specific in MAI patients than PPD was in tuberculosis patients. In the estimation of IgA antibodies to MAI antigen, the sensitivity for MAID was 69.2% and the specificity was 90.5%, and the positive predictive value for MAID was 81.8% and the negative predictive value 82.6%. Therefore, we thought that the estimation of the IgA antibody to MAI antigen is useful for the differential diagnosis between MAID and pulmonary tuberculosis. In a histopathologic study of lung lesions caused by infection with unclassified acid-fast bacilli in 19 cases (29), definite endobronchitis was observed in 4 of 8 lung specimens of MAID patients, and tissue reaction characterized by the presence of lymphocytes, plasma cells, monocytes, and histiocytes, was observed in all specimens in the tissues surrounding N the lesion itself. From these observations it might be possible that IgA antibody is highly produced in response to MAI antigen in MAID patients. ELISA is now applied not only for detection of serum antibodies and antigens derived from mycobacterium in sera, but also for other specimens, for example, bronchoalveolar lavage fluid (30). It is expected that more highly purified antigens will be available. Then, ELISA will be more useful tool for the ifferential diagno sis among mycobacterial diseases. ACKNOWLEDGMENT:The writers wish to thank Professor Hidehiko Saito, the First Department of Medicine, and Professor Izumi Nakashima, Department of Immunology, Nagoya University School of Medicine, for their support and encouragement throughout the course of this study. REFERENCES 1) Nassau E, Parsons ER, Johnson GD: The detection of antibodies to Mycobacterium tuberculosis by microplate enzyme-linked immunosorbent assay (ELISA). Tubercle 57: 67, ) Grange JM, Gibson J, Nassau E, et al: Enzyme-linked immunosorbent assay (ELISA): a study of antibodies to mycobacterium tuberculosis in the IgG, IgA and IgM classes in tuberculosis, sarcoidosis and Crohn's disease. Tubercle 61: 145, ) Reggiardo Z, Vazquez E, Schnaper L: ELISA tests for antibodies against mycobacterial glycolipids. J Immunol Methods 34: 55, ) Tandon A, Sexena RP, Sexena KC, et al: Diagnostic potentialities of enzyme-linked immunosorbent assay in tuberculosis using purified tuberculin antigen. Tubercle 61: 87, ) Daniel TM, Oxtoby MJ, Pinto E, et al: The immune spectrum in patients with pulmonary tuberculosis. Am Rev Respir Dis 123: 556, ) Benjamin RG, Daniel TM: Serodiagnosis of tuberculosis using the enzyme-linked immunoabsorbent assay (ELISA) of antibody to Mycobacterium tuberculosis antigen 5. Am Rev Respir Dis 126: 1013, ) Kardjito T, Handoyo I, Grange JM: Diagnosis of active tuberculosis by immunological methods. 1. The effect of tuberculin reactivity and previous BCG vaccination on the antibody levels determined by ELISA. Tubercle 63: 269, ) Karjito T, Grange JM: Diagnosis of active tuberculosis by immunological methods. 2. Qualitative differences in the dermal response to tuberculin in patients with active pulmonary disease and healthy tuberculin-positive individuals. Tubercle 63: 275, ) Viljanen MK, Eskola J, Tala E: Enzyme-linked immunosorbent assay (ELISA) for antibodies to purified protein derivative of tuberculin (PPD). Eur J Respir Dis 63: 257, ) Kalish SB, Radin RC, Phair JP, et al: Use of an enzyme- 200 Jpn J Med Vol 28, No 2 (March, April 1989)

6 ELISA for Differential Diagnosis of Mycobacterial Diseases linked immunosorbent assay technique in the differential diagnosis of active pulmonary tuberculosis in humans. J Infect Dis 147: 523, Radin RC, Zeiss CR, Phair JP: Antibodies to purified protein derivative in different immunoglobulin classes in the diagnosis of tuberculosis in man. Int Arch Allergy Appl Immunol 70: 25, Balestrino EA, Daniel TM, De Latini MDS, et al: Serodiagnosis of pulmonary tuberculosis in Argentina by enzyme-linked immunosorbent assay (ELISA) of IgG antibody to Mycobacterium tuberculosis antigen 5 and tuberculin purified protein derivative. Bull World Hlth Org 62: 755, Benjamin RG, Debanne SM, Ma Y, et al: Evaluation of mycobacterial antigens in an enzyme-linked immunosorbent assay (ELISA) for the serodiagnosis of tuberulosis. J Med Microbiol 18: 309, Zeiss CR, Kalish SB, Erlich KS, et al: IgG antibody to purified protein derivative by enzyme-linked immunosorbent assay in the diagnosis of pulmonary tuberculosis. Am Rev Respir Dis 130: 845, onishi K, Kokubu K, Murakami S, et al: Diagnostic potentialities of ELISA (enzyme-linked immunosorbent assay) in pulmonary tuberculosis using purified protein derivative. Respir Res 3: 700, oshino T, Nishioka S, Fujimura M, et al: ELISA for IgG antibody to purified protein derivative (PPD) of patients with pulmonary tuberculosis. Kekkaku 59: 621, aniel TM, Debanne SM, van der Kuyp F: Enzymelinked immunosorbent assay using Mycobacterium tuberculosis antigen 5 and PPD for the serodiagnosis of tuberculosis. Chest 88: 388, MaY,WangYM,Daniel TM: Enzyme-linked immunosorbent assay using Mycobacterium tuberculosis antigen 5 for the diagnosis of pulmonary tuberculosis in China. AmRev Respir Dis 134: 1273, Daniel TM, Dsbanne SM: The serodiagnosis of tuberculosis and other mycobacterial diseases by enzymelinked immunosorbent assay. AmRev Respir Dis 135: 1137, Kusano N: Evaluation of serodiagnosis detecting IgG antibodies against purified protein derivative and alpha antigen in patients with active pulmonary tuberculosis by enzyme-linked immunosorbent assay. Kekkaku 62: 211, Nagae H, Miura T, Liu C, et al: Improvement of enzyme-linked immunosorbent assay for the serodiagnosis of active pulmonary tuberculosis. Kekkaku 62: 503, Yamamoto S, Tabata K, Toida I, et al: Determination of specific antibody value in patients with tuberculosis and atypical mycobacteriosis. Kekkaku 62: 549, Ahn CH, Lowell JR, Onstad GD, et al: A demographic study of disease due to Mycobacterium kansasii or M. intracellulare-avium in Texas. Chest 75: 120, Ahn CH, McLarty JW, Ahn SS, et al: Diagnostic criteria for pulmonary disease caused by Mycobacterium kansasii and Mycobacterium intracellulare. Am Rev Respir Dis 125: 388, Winter SM, Bernard EM, Gold JWM, et al: Humoral response to disseminated infection by Mycobacterium avium-mycobacterium intracellulare in acquired immunodeficiency syndrome and hairy cell leukemia. J Infect Dis 151: 523, Tsukamura M: Diagnostic standard of atypical mycobacterial diseases (pulmonary infectious diseases). Kekkaku 60: 51, Tsukamura M: Intermediate between Mycobacterium scrofulaceum and Mycobacterium intracellulare. Iryo 27: 232, Takeya K, Zinnaka Y, Yamaura K, et al: Bacteriophage susceptibility and tuberculin specificity of unclassified mycobacteria. Am Rev Respir Dis 81: 674, Merckx JJ, Soule EH, Karlson AG: The histopathology of lesions caused by infection with unclassified acid-fast bacteria in man. AmJ Clin Pathol 41: 244, Raja A, Baughman PR, Daniel TM: The detection by immunoassay of antibody to mycobacterial antigens and mycobacterial antigens in bronchoalveolar lavage fluid from patients with tuberculosis and control subjects. Chest 94: 133, Jpn J Med Vol 28, No 2 (March, April 1989) 201