Vol. 28, No. 8
JOURNAL OF CLINICAL MICROBIOLOGY, Aug. 1990, p. 1694-1697
0095-1137/90/081694-04$02.00/0 Copyright © 1990, American Society for Microbiology
Identification of Various Serovar Strains of Mycobacterium avium Complex by Using DNA Probes Specific for Mycobacterium avium and Mycobacterium intracellulare HAJIME SAITO,1* HARUAKI TOMIOKA,' KATSUMASA SATO,' HIROMICHI TASAKA,2 AND DAVID J. DAWSON3 Department of Microbiology and Immunology, Shimane Medical University, Izumo 693,1 and Department of Bacteriology, Hiroshima University School of Medicine, Hiroshima 734,2 Japan, and Laboratory of Microbiology and Pathology, Queensland Department of Health, Brisbane 4001, Queensland, Australia3 Received 12 March 1990/Accepted 1 May 1990
Reference strains of the Mycobacterium avium complex (MAC) belonging to serovars 1 to 28 were examined with DNA probes (Gen Probe Rapid Diagnostic System for the MAC; Gen Probe Inc., San Diego, Calif.) specific for either M. avium or Mycobacterium intracellulare. This study revealed that the earlier designations of the MAC serovars, in which serovars 1 to 3 and 4 to 28 were regarded as M. avium and M. intracellulare, respectively, should be revised as follows. First, M. avium includes serovars 1 to 6, 8 to 11, and 21. Second, M. intracellulare includes serovars 7, 12 to 20, and 25. However, other serovars, such as serovars 22 to 24 and 26 to 28, involve M. intracelulare, Mycobacterium scrofulaceum, and MAC that are lacking in the reactivity with either DNA probe and that are too disordered to enable a conclusive description here, particularly concerning their taxonomic positions in the MAC. Mycobacterium avium and Mycobacterium intracellulare closely related and difficult to distinguish from one another by conventional cultural and biochemical tests. They are often referred to as the M. avium complex (MAC). Recently, a DNA probe kit (Gen Probe Rapid Diagnostic System for the MAC; Gen Probe Inc., San Diego, Calif.) was developed to identify the two MAC species, M. avium and M. intracellulare (5, 10). This diagnostic kit has proved to be rapid, sensitive, and specific for identification of the two MAC species (3, 4, 9). We have also confirmed the value of this assay system (15), but in a few instances, we encountered mycobacterial strains that did not react with either the M. avium or the M. intracellulare probe, even though they possessed MAC-specific a-antigen and exhibited cultural and biochemical characteristics of the MAC (15). Similar observations were also reported by Enns (4) and Sherman et al. (17). At present, MAC consists of 28 serovars, and until recently, serovars 1 to 3 and 4 to 28 were designated M. avium and M. intracellulare, respectively (6, 16, 19). However, Baess (1, 2) and Magnusson (12) recently reported that strains belonging to serovars 4 to 6 and 8 are more closely related to M. avium than they are to M. intracellulare. Therefore, we decided to test various standard strains of MAC (6) belonging to serovars 1 to 28 using the MAC DNA probes. We examined 105 strains of the MAC, involving 69 serovar reference strains (6), for their reactivities with either MAC DNA probe, conventional cultural and biochemical properties, and a-antigens (18).
tained from G. P. Kubica, Centers for Disease Control (CDC), Atlanta, Ga., represented serovar reference strains (6). Thirty-six strains were isolates from patients infected with MAC, i.e., 29 strains from Queensland Department of Health (QDH), Brisbane, Australia, and 7 strains from Shimane Medical University (SMU), Izumo, Japan. All strains were passed once or twice on 7H10 agar plates to check their purity. DNA probe test. The DNA probe test for the MAC was performed according to the directions in the manual for the Gen Probe Rapid Diagnostic System for the MAC. Briefly, 0.1 ml of bacterial suspension prepared from cultures grown for about 3 to 4 weeks on 1% Ogawa egg medium (14) was sonicated in the tubes containing glass beads and lysing reagents at 59°C in a sonicator bath for 15 min. One milliliter each of '251-labeled DNA probe solution for M. avium or M. intracellulare was added to the resultant cell lysate containing rRNA and was subjected to heating at 72°C for 1 h for hybridization. The incubation mixture was mixed with 4 ml of separation suspension, which consisted of hydroxyapatite and buffer solution; incubated at 72°C for 5 min; and centrifuged. The precipitate was washed once with wash solution, recentrifuged, and counted for radioactivity by using a gamma counter. Percent hybridization was calculated by the following equation: percent hybridization = [(sample cpm background cpm)/(total counts cpm - background cpm)] 100. A percent hybridization greater than or equal to 10% was regarded as positive. a-Antigen analysis. The a-antigen of test mycobacteria was analyzed by an agar gel precipitation test by using anti-a-antigen sera, as described previously (18). Briefly, growth from 1% Ogawa egg medium was collected into a chilled mortar and was ground with glass beads by adding 20 mM Tris hydrochloride buffer (pH 8.0). The supernatant was collected by centrifugation. The resultant cell extract was used as antigen solution in an agar gel precipitation test by using anti-MAC or anti-Mycobacterium scrofulaceum rabbit
are
X
MATERIALS AND METHODS
Organisms. A total of 105 strains of the MAC belonging to serovars 1 to 28 (2 to 6 strains for each serovar, except for serovar 17) were used in this study. A total of 48 strains obtained from A. Y. Tsang, National Jewish Hospital and Research Center (NJC), Denver, Colo., and 21 strains ob*
Corresponding author.
serum.
1694
VOL. 28, 1990
Conventional biochemical identification. Culture and biochemical identification of the test strains were performed by standard methods (8, 11). Organisms were identified as MAC when they were nonpigmented slow growers; resistant to hydroxylamine hydrochloride (500 ,ug/ml), ethambutol (5 ,u.g/ml), and p-nitrobenzoic acid (2 mg/ml); and positive for tellurite reduction and heat-stable catalase but negative for niacin production, nitrate reduction, Tween 80 hydrolysis (7 and 14 days), and urease. RESULTS Reactivities of MAC strains belonging to serovars 1 to 21 with MAC DNA probes. AUl 29 strains belonging to serovars 1 to 6, 8 to 11, and 21 (18 strains from NJC, 4 strains from CDC, and 7 strains from SMU) reacted with the M. avium probe with positive percent hybridization values, 34 to 60% (Table 1). These same strains hybridized with the M. intracellulare probe only to 0.7 to 3.3%; i.e., it was lower than the negative limit (5%). Conversely, all 19 strains belonging to serovars 7 and 12 to 20 (18 strains from NJC, 1 strain from CDC) reacted positively with the M. intracellulare probe with hybridization values of 30 to 43%, while their binding with M. avium probe, with hybridization values ranging from 0.8 to 3.6%, was much lower than the negative limit. Therefore, it can be regarded that serovars 1 to 6, 8 to 11, and 21 are M. avium, while serovars 7 and 12 to 20 are M. intracellulare. A somewhat confused situation was seen for serovar 18. In separate experiments (detailed data not shown), strains Melnick and Melnick Altman obtained from NJC and CDC, respectively, reacted positively with the M. avium DNA probe at 11 to 30% hybridization and the M. intracellulare probe reacted at 5 to 12% hybridization. In our serotyping test, these strains gave weak agglutination in anti-serovar 18 serum. However, six of seven strains belonging to serovar 18 isolated from patients in Australia reacted with the M. intracellulare probe with the following hybridization values:
S72/62, 18.3%; S72/215, 18.6%; S73/19, 17.4%; S73/56, 18.0%; S73/118, 17.4%; and S73/150, 18.0% (hybridization with the M. avium probe ranged from 0.7 to 0.8%). The other strain, S72/83, did not react with either probe (hybridization to the M. avium and M. intracellulare probes was 4.7 and 1.4%, respectively). Therefore, serovar 18 substantially belongs to M. intracellulare, although the possibility cannot be entirely ruled out that this serovar involves not only M. intracellulare but also M. avium. Similar mixups are known for Mycobacterium tuberculosis (7). The same result mentioned above was obtained for the three strains from CDC with the same strain names as those from NJC (strain 34540, 14186-1424, and P54) (data not shown). Reactivities with DNA probes of MAC strains belonging to serovars 22 to 28. As shown in Table 2, 10 strains of serovar 25 (strains 72-888 and 1195 CDC from NJC and strains S88/21, S88/55, S88/90, S88/204, S88/259, S88/334, S88/425, and S88/465 from QDH) reacted positively with the M. intracellulare probe but not with the M. avium probe, indicating that these serovar strains also belong to M. intracellulare. Except for serovar 25, the reactivities of MAC strains of serovars 22 to 28 to MAC DNA probes were very confused (Table 2). Two strains of serovar 22 (S88/142 and S88/381) reacted positively with the M. intracellulare DNA probe. These strains had MAC-specific biological characteristics and a-antigen. The other two strains of serovars 22 and five strains of serovar 27 lacked reactivity not only with the M.
MAC SEROVARS AND DNA PROBE TEST
1695
TABLE 1. Reactivities of various serovar strains from NJC and CDC and some clinical isolates in SMU with the MAC DNA probes % Hybridization
Serovar
1 1 2A 2 3A 3 4A 4 SA 5B 6A 6B 7A 7 8A 8B 8 8 8
Reference strain
11907-300 16741 Cardiff 6194 B-92 6195 128 Germany 13528-1079 TMC 1463 25546-759 4443-1237 12315/67 34540 157 Manten 158 Leonard ATCC 23435 SJB #2 N-307 N-339 N-361 N-370 N-372 N-444 N-463 1784-286 204-6450 TMC 1461 1602-1965 VITOCH 14186-1424 Wood Duck P-42 Chance ATCC 25122 Edgar Boone P-39 Dent
Source
NJC NJC NJC NJC NJC NJC NJC NJC NJC NJC CDC NJC
NJC NJC CDC NJC SMU SMU SMU SMU 8 SMU 8 SMU 8 SMU 8 NJC 9A NJC 9B NJC 10 NJC 10B CDC 11A 11B NJC 12A NJC NJC 12B NJC 13A NJC 13B NJC 14A NJC 14B NJC 15A NJC 15B Simpson NJC 16A Yandle NJC 16B ATCC 15987 17A P-54 NJC CDC 18B 2219 Altman 18 4990 O'Connorb NJC NJC 19A Darden NJC 19B W-552 NJC 20 TMC 1419 20 AT 545 Findley NJC NJC 21A 2993 CDC 21B T77
avium
M.obe probe 43.3 46.1 42.2 40.0 48.8 39.2
35.5 47.6 49.7 35.8 52.0 33.7 1.1 1.4 50.0 46.1 59.4
52.3 48.6 49.0 52.1 60.0 52.7 42.2 48.8 48.5 36.0 48.9 41.7 1.1 2.8 1.0 0.8 2.1 1.8 1.1 3.5 1.2 1.5 0.9 1.4 3.6 1.3 1.7 1.0 1.8 36.6 53.0
M. intracellulare probe
1.0 1.4 1.1 1.5 1.3 0.9 0.7 1.5 2.1
0.8 1.3 0.7 35.8 42.2 2.2 1.0 2.0 1.1 1.4 1.1 1.4 3.3
1.9 1.1 1.3 1.6 0.8 1.1 1.1 40.3 35.6 33.5 34.2 38.2 35.9 31.1 42.5 39.5 34.0 30.3 35.3 37.6 37.4 34.9 32.5 38.9 1.0 1.7
Identification
M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M.
avium avium avium avium avium avium avium avium avium avium avium avium intracellulare intracellulare avium avium avium avium avium avium avium avium avium avium avium avium avium avium avium
intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare avium
avium
Serovar typing was done by W. B. Schaefer for NJC and CDC strains and by D. J. Dawson for SMU and QDH (see the text) strains. b In our seroagglutination test, this reference strain did not give specific agglutination with any of the antisera. a
avium probe but also with the M. intracellulare probe (hybridization values with M. avium and M. intracellulare were 0.5 to 8.3% [average, 2.2%] and 0.9 to 2.0% [average, 1.5%], respectively). Some of these strains (strains 5154, 10409, Harrison, Lane 3081) exhibited cultural, biochemical, and serological properties characteristic of M. scrofulaceum: scotochromogenic colonial morphology, resistance to ethambutol (5 ,ug/ml), negative activity for nitrate reduction and Tween 80 hydrolysis, and presence of M. scrofulaccum-
1696
SAITO ET AL.
J. CLIN. MICROBIOL.
TABLE 2. Reactivities of MAC strains belonging to serovars 22 to 28 with MAC DNA probes % Hybridization' Serovar
Strain
Source
M. avium
M.
intra-
Biological characteristics and
probe
cellular probe
a-antigen
8.3 0.8 1.1 1.1
2.0 1.0 31.6 23.9
M. scrofulaceum M. scrofulaceum MAC MAC
2.0 6.7 6.5 1.1
2.3 2.7 3.4 1.5
MAC MAC MAC MAC
8.4 1.0
MAC MAC MAC MAC MAC
22A 22B 22 22
5154 O'Connor NJC 10409 NJC
S88/142 S88/381
QDH QDH
23A 23B 23 23
CDC841B 23393 CDC 1214
CDC NJC
S88/420
QDH
24A 24B 24 24 24
12645 2154
NJC
S88/87 S88/294 S88/543
QDH QDH QDH
2.1 2.4 2.0
2.5 33.9 2.3 2.1 1.8
25A 25B 25 25 25 25 25 25 25 25
72-888 1195 CDC
NJC NJC
S88/21 S88/55 S88/90 S88/204 S88/259 S88/334
QDH QDH QDH QDH QDH QDH QDH QDH
1.5 0.7 1.1 1.4 1.1 1.4 1.6 1.2 1.3 1.2
34.1 32.8 27.5 27.3 30.9 33.1 32.6 29.3 27.4 30.2
MAC MAC MAC MAC MAC MAC MAC MAC MAC MAC
26A 26B
Cox 1944
0.9 0.7 1.4 6.7 1.4 1.5
1.0 1.0 35.2 1.8 2.8 1.9
M. M.
0.5 0.8 1.7 1.3 1.4
0.9 1.2 1.8 1.5 2.0
2.2 3.2 1.2 1.5
37.3 1.9 29.7 1.2
S88/425
S88/465
NJC
CDC
26
Mackenzie Hillberry 1244
CDC NJC NJC
26 26 26
S87/210 S88/155 S88/175
QDH QDH QDH
27A 27B
Harrison Lane 3081
NJC NJC
27 27 27
S88/59 S88/234 S88/237
QDH QDH QDH
28A 28B 28 28
6845 9055 Matthews
NJC
S87/69 S88/101
QDH QDH
CDC
scrofulaceum scrofulaceum
MAC MACb
MAC MAC
M. scrofulaceum M. scrofulaceum MAC MAC MAC
MAC MAC MAC MAC
a Values are means of two to five tests, except for Harrison (serovar 27A) and Lane 3081 (serovar 27B).
b M. intracellulare ATCC
19583-specific a-antigen.
specific a-antigen. Therefore, we identified them as M. scrofulaceum. Strains 5154 and 10409 (originally serotyped as serovar 22) and strain Harrison (originally serotyped as serovar 27) were found to react with anti-serovar 22 and anti-serovar 27 sera, respectively, by our serotyping test, but strain Lane 3081 (originally identified as serovar 27) did not give specific agglutination with any of the antisera. The remaining three strains of serovar 27 (S88/59, S88/234, S88/237) had characteristics typical of MAC. Strains of serovars 23, 24, 26, and 28 possessed MACspecific a-antigen and cultural and biochemical properties characteristic of the MAC (Table 2). Fourteen strains failed to react (or reacted weakly) with either DNA probe (hybridization, 0.7 to 6.7%), and four strains reacted with the M. intracellulare DNA probe (hybridization, 29.7 to 37.3%).
Strains Cox 1944 and Mackenzie had a-antigen specific for M. scrofulaceum but not for MAC. These strains also failed to react with either MAC DNA probe. Similarly, the serovar reference strains from CDC, strains 5154 and 10409 (serovar 22), 23393 (serovar 23), 12645 (serovar 24A), Harrison (serovar 27A), and Lane 3081 (serovar 27B), failed to react (or reacted weakly) with either probe (hybridization, 0.9 to 9.3%); and strain 6845 Koukout (serovar 28A) reacted with M. intracellulare DNA probe (hybridization, 22.3%). DISCUSSION In the present study, we applied the DNA probe test to various MAC serovar strains, including reference strains (6) and isolates from patients. First, we found that serovars 1 to 6, 8 to 11, and 21 showed clear positive reactivities with the M. avium probe but not with the M. intracellulare probe; all other features corroborated their identity as M. avium. Second, serovars 7, 12 to 20, and 25 were identified as M. intracellulare, although the position with serovar 18 needs clarification. This grouping is consistent with the results of DNA-DNA hybridization tests by Baess (1, 2) and sensitin tests on guinea pigs by Magnusson (12), in which strains belonging to serovars 1, 2, 4, 6, and 9 were identified as M. avium and serovars 12, 14, 16, 17, and 20 were identified as M. intracellulare. In cases of serovars 22 to 28 (except for serovar 25), the strains tested were largely divided into three groups. First, serovars 22, 26, and 27 involved some strains which had no reactivity with MAC probes but which we would identify as M. scrofulaceum, because they possessed M. scrofulaceumspecific a-antigen (18) and exhibited characteristic cultural and biochemical properties of M. scrofulaceum (8, 11). In relation to this, it may be noteworthy that Schaefer (16) and McClatchy (13) observed the existence of a number of isolates that were identified biochemically as M. scrofulaceum but serotyped as MAC and vice versa. Second, some strains of serovars 22, 24, 26, and 28 had positive reactivities with the DNA probe of the M. intracellulare- and MACspecific a-antigen (18) (see Table 2) and had properties characteristic of the MAC (8, 11). Third, serovars 23 and the remaining strains of serovars 24, 26, and 28 did not react with either the M. avium or the M. intracellulare probe; but they possessed MAC-specific a-antigen and cultural and biochemical properties of MAC. Similar MAC strains with low reactivities with the MAC DNA probes have been reported anecdotally by CDC and by Enns (4), in the case of clinical isolates of MAC. Enns (4) reported that 8 of 60 strains (13%) identified as MAC by conventional cultural and biochemical methods had no positive hybridization values (
JOURNAL OF CLINICAL MICROBIOLOGY, Aug. 1990, p. 1694-1697
0095-1137/90/081694-04$02.00/0 Copyright © 1990, American Society for Microbiology
Identification of Various Serovar Strains of Mycobacterium avium Complex by Using DNA Probes Specific for Mycobacterium avium and Mycobacterium intracellulare HAJIME SAITO,1* HARUAKI TOMIOKA,' KATSUMASA SATO,' HIROMICHI TASAKA,2 AND DAVID J. DAWSON3 Department of Microbiology and Immunology, Shimane Medical University, Izumo 693,1 and Department of Bacteriology, Hiroshima University School of Medicine, Hiroshima 734,2 Japan, and Laboratory of Microbiology and Pathology, Queensland Department of Health, Brisbane 4001, Queensland, Australia3 Received 12 March 1990/Accepted 1 May 1990
Reference strains of the Mycobacterium avium complex (MAC) belonging to serovars 1 to 28 were examined with DNA probes (Gen Probe Rapid Diagnostic System for the MAC; Gen Probe Inc., San Diego, Calif.) specific for either M. avium or Mycobacterium intracellulare. This study revealed that the earlier designations of the MAC serovars, in which serovars 1 to 3 and 4 to 28 were regarded as M. avium and M. intracellulare, respectively, should be revised as follows. First, M. avium includes serovars 1 to 6, 8 to 11, and 21. Second, M. intracellulare includes serovars 7, 12 to 20, and 25. However, other serovars, such as serovars 22 to 24 and 26 to 28, involve M. intracelulare, Mycobacterium scrofulaceum, and MAC that are lacking in the reactivity with either DNA probe and that are too disordered to enable a conclusive description here, particularly concerning their taxonomic positions in the MAC. Mycobacterium avium and Mycobacterium intracellulare closely related and difficult to distinguish from one another by conventional cultural and biochemical tests. They are often referred to as the M. avium complex (MAC). Recently, a DNA probe kit (Gen Probe Rapid Diagnostic System for the MAC; Gen Probe Inc., San Diego, Calif.) was developed to identify the two MAC species, M. avium and M. intracellulare (5, 10). This diagnostic kit has proved to be rapid, sensitive, and specific for identification of the two MAC species (3, 4, 9). We have also confirmed the value of this assay system (15), but in a few instances, we encountered mycobacterial strains that did not react with either the M. avium or the M. intracellulare probe, even though they possessed MAC-specific a-antigen and exhibited cultural and biochemical characteristics of the MAC (15). Similar observations were also reported by Enns (4) and Sherman et al. (17). At present, MAC consists of 28 serovars, and until recently, serovars 1 to 3 and 4 to 28 were designated M. avium and M. intracellulare, respectively (6, 16, 19). However, Baess (1, 2) and Magnusson (12) recently reported that strains belonging to serovars 4 to 6 and 8 are more closely related to M. avium than they are to M. intracellulare. Therefore, we decided to test various standard strains of MAC (6) belonging to serovars 1 to 28 using the MAC DNA probes. We examined 105 strains of the MAC, involving 69 serovar reference strains (6), for their reactivities with either MAC DNA probe, conventional cultural and biochemical properties, and a-antigens (18).
tained from G. P. Kubica, Centers for Disease Control (CDC), Atlanta, Ga., represented serovar reference strains (6). Thirty-six strains were isolates from patients infected with MAC, i.e., 29 strains from Queensland Department of Health (QDH), Brisbane, Australia, and 7 strains from Shimane Medical University (SMU), Izumo, Japan. All strains were passed once or twice on 7H10 agar plates to check their purity. DNA probe test. The DNA probe test for the MAC was performed according to the directions in the manual for the Gen Probe Rapid Diagnostic System for the MAC. Briefly, 0.1 ml of bacterial suspension prepared from cultures grown for about 3 to 4 weeks on 1% Ogawa egg medium (14) was sonicated in the tubes containing glass beads and lysing reagents at 59°C in a sonicator bath for 15 min. One milliliter each of '251-labeled DNA probe solution for M. avium or M. intracellulare was added to the resultant cell lysate containing rRNA and was subjected to heating at 72°C for 1 h for hybridization. The incubation mixture was mixed with 4 ml of separation suspension, which consisted of hydroxyapatite and buffer solution; incubated at 72°C for 5 min; and centrifuged. The precipitate was washed once with wash solution, recentrifuged, and counted for radioactivity by using a gamma counter. Percent hybridization was calculated by the following equation: percent hybridization = [(sample cpm background cpm)/(total counts cpm - background cpm)] 100. A percent hybridization greater than or equal to 10% was regarded as positive. a-Antigen analysis. The a-antigen of test mycobacteria was analyzed by an agar gel precipitation test by using anti-a-antigen sera, as described previously (18). Briefly, growth from 1% Ogawa egg medium was collected into a chilled mortar and was ground with glass beads by adding 20 mM Tris hydrochloride buffer (pH 8.0). The supernatant was collected by centrifugation. The resultant cell extract was used as antigen solution in an agar gel precipitation test by using anti-MAC or anti-Mycobacterium scrofulaceum rabbit
are
X
MATERIALS AND METHODS
Organisms. A total of 105 strains of the MAC belonging to serovars 1 to 28 (2 to 6 strains for each serovar, except for serovar 17) were used in this study. A total of 48 strains obtained from A. Y. Tsang, National Jewish Hospital and Research Center (NJC), Denver, Colo., and 21 strains ob*
Corresponding author.
serum.
1694
VOL. 28, 1990
Conventional biochemical identification. Culture and biochemical identification of the test strains were performed by standard methods (8, 11). Organisms were identified as MAC when they were nonpigmented slow growers; resistant to hydroxylamine hydrochloride (500 ,ug/ml), ethambutol (5 ,u.g/ml), and p-nitrobenzoic acid (2 mg/ml); and positive for tellurite reduction and heat-stable catalase but negative for niacin production, nitrate reduction, Tween 80 hydrolysis (7 and 14 days), and urease. RESULTS Reactivities of MAC strains belonging to serovars 1 to 21 with MAC DNA probes. AUl 29 strains belonging to serovars 1 to 6, 8 to 11, and 21 (18 strains from NJC, 4 strains from CDC, and 7 strains from SMU) reacted with the M. avium probe with positive percent hybridization values, 34 to 60% (Table 1). These same strains hybridized with the M. intracellulare probe only to 0.7 to 3.3%; i.e., it was lower than the negative limit (5%). Conversely, all 19 strains belonging to serovars 7 and 12 to 20 (18 strains from NJC, 1 strain from CDC) reacted positively with the M. intracellulare probe with hybridization values of 30 to 43%, while their binding with M. avium probe, with hybridization values ranging from 0.8 to 3.6%, was much lower than the negative limit. Therefore, it can be regarded that serovars 1 to 6, 8 to 11, and 21 are M. avium, while serovars 7 and 12 to 20 are M. intracellulare. A somewhat confused situation was seen for serovar 18. In separate experiments (detailed data not shown), strains Melnick and Melnick Altman obtained from NJC and CDC, respectively, reacted positively with the M. avium DNA probe at 11 to 30% hybridization and the M. intracellulare probe reacted at 5 to 12% hybridization. In our serotyping test, these strains gave weak agglutination in anti-serovar 18 serum. However, six of seven strains belonging to serovar 18 isolated from patients in Australia reacted with the M. intracellulare probe with the following hybridization values:
S72/62, 18.3%; S72/215, 18.6%; S73/19, 17.4%; S73/56, 18.0%; S73/118, 17.4%; and S73/150, 18.0% (hybridization with the M. avium probe ranged from 0.7 to 0.8%). The other strain, S72/83, did not react with either probe (hybridization to the M. avium and M. intracellulare probes was 4.7 and 1.4%, respectively). Therefore, serovar 18 substantially belongs to M. intracellulare, although the possibility cannot be entirely ruled out that this serovar involves not only M. intracellulare but also M. avium. Similar mixups are known for Mycobacterium tuberculosis (7). The same result mentioned above was obtained for the three strains from CDC with the same strain names as those from NJC (strain 34540, 14186-1424, and P54) (data not shown). Reactivities with DNA probes of MAC strains belonging to serovars 22 to 28. As shown in Table 2, 10 strains of serovar 25 (strains 72-888 and 1195 CDC from NJC and strains S88/21, S88/55, S88/90, S88/204, S88/259, S88/334, S88/425, and S88/465 from QDH) reacted positively with the M. intracellulare probe but not with the M. avium probe, indicating that these serovar strains also belong to M. intracellulare. Except for serovar 25, the reactivities of MAC strains of serovars 22 to 28 to MAC DNA probes were very confused (Table 2). Two strains of serovar 22 (S88/142 and S88/381) reacted positively with the M. intracellulare DNA probe. These strains had MAC-specific biological characteristics and a-antigen. The other two strains of serovars 22 and five strains of serovar 27 lacked reactivity not only with the M.
MAC SEROVARS AND DNA PROBE TEST
1695
TABLE 1. Reactivities of various serovar strains from NJC and CDC and some clinical isolates in SMU with the MAC DNA probes % Hybridization
Serovar
1 1 2A 2 3A 3 4A 4 SA 5B 6A 6B 7A 7 8A 8B 8 8 8
Reference strain
11907-300 16741 Cardiff 6194 B-92 6195 128 Germany 13528-1079 TMC 1463 25546-759 4443-1237 12315/67 34540 157 Manten 158 Leonard ATCC 23435 SJB #2 N-307 N-339 N-361 N-370 N-372 N-444 N-463 1784-286 204-6450 TMC 1461 1602-1965 VITOCH 14186-1424 Wood Duck P-42 Chance ATCC 25122 Edgar Boone P-39 Dent
Source
NJC NJC NJC NJC NJC NJC NJC NJC NJC NJC CDC NJC
NJC NJC CDC NJC SMU SMU SMU SMU 8 SMU 8 SMU 8 SMU 8 NJC 9A NJC 9B NJC 10 NJC 10B CDC 11A 11B NJC 12A NJC NJC 12B NJC 13A NJC 13B NJC 14A NJC 14B NJC 15A NJC 15B Simpson NJC 16A Yandle NJC 16B ATCC 15987 17A P-54 NJC CDC 18B 2219 Altman 18 4990 O'Connorb NJC NJC 19A Darden NJC 19B W-552 NJC 20 TMC 1419 20 AT 545 Findley NJC NJC 21A 2993 CDC 21B T77
avium
M.obe probe 43.3 46.1 42.2 40.0 48.8 39.2
35.5 47.6 49.7 35.8 52.0 33.7 1.1 1.4 50.0 46.1 59.4
52.3 48.6 49.0 52.1 60.0 52.7 42.2 48.8 48.5 36.0 48.9 41.7 1.1 2.8 1.0 0.8 2.1 1.8 1.1 3.5 1.2 1.5 0.9 1.4 3.6 1.3 1.7 1.0 1.8 36.6 53.0
M. intracellulare probe
1.0 1.4 1.1 1.5 1.3 0.9 0.7 1.5 2.1
0.8 1.3 0.7 35.8 42.2 2.2 1.0 2.0 1.1 1.4 1.1 1.4 3.3
1.9 1.1 1.3 1.6 0.8 1.1 1.1 40.3 35.6 33.5 34.2 38.2 35.9 31.1 42.5 39.5 34.0 30.3 35.3 37.6 37.4 34.9 32.5 38.9 1.0 1.7
Identification
M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M. M.
avium avium avium avium avium avium avium avium avium avium avium avium intracellulare intracellulare avium avium avium avium avium avium avium avium avium avium avium avium avium avium avium
intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare intracellulare avium
avium
Serovar typing was done by W. B. Schaefer for NJC and CDC strains and by D. J. Dawson for SMU and QDH (see the text) strains. b In our seroagglutination test, this reference strain did not give specific agglutination with any of the antisera. a
avium probe but also with the M. intracellulare probe (hybridization values with M. avium and M. intracellulare were 0.5 to 8.3% [average, 2.2%] and 0.9 to 2.0% [average, 1.5%], respectively). Some of these strains (strains 5154, 10409, Harrison, Lane 3081) exhibited cultural, biochemical, and serological properties characteristic of M. scrofulaceum: scotochromogenic colonial morphology, resistance to ethambutol (5 ,ug/ml), negative activity for nitrate reduction and Tween 80 hydrolysis, and presence of M. scrofulaccum-
1696
SAITO ET AL.
J. CLIN. MICROBIOL.
TABLE 2. Reactivities of MAC strains belonging to serovars 22 to 28 with MAC DNA probes % Hybridization' Serovar
Strain
Source
M. avium
M.
intra-
Biological characteristics and
probe
cellular probe
a-antigen
8.3 0.8 1.1 1.1
2.0 1.0 31.6 23.9
M. scrofulaceum M. scrofulaceum MAC MAC
2.0 6.7 6.5 1.1
2.3 2.7 3.4 1.5
MAC MAC MAC MAC
8.4 1.0
MAC MAC MAC MAC MAC
22A 22B 22 22
5154 O'Connor NJC 10409 NJC
S88/142 S88/381
QDH QDH
23A 23B 23 23
CDC841B 23393 CDC 1214
CDC NJC
S88/420
QDH
24A 24B 24 24 24
12645 2154
NJC
S88/87 S88/294 S88/543
QDH QDH QDH
2.1 2.4 2.0
2.5 33.9 2.3 2.1 1.8
25A 25B 25 25 25 25 25 25 25 25
72-888 1195 CDC
NJC NJC
S88/21 S88/55 S88/90 S88/204 S88/259 S88/334
QDH QDH QDH QDH QDH QDH QDH QDH
1.5 0.7 1.1 1.4 1.1 1.4 1.6 1.2 1.3 1.2
34.1 32.8 27.5 27.3 30.9 33.1 32.6 29.3 27.4 30.2
MAC MAC MAC MAC MAC MAC MAC MAC MAC MAC
26A 26B
Cox 1944
0.9 0.7 1.4 6.7 1.4 1.5
1.0 1.0 35.2 1.8 2.8 1.9
M. M.
0.5 0.8 1.7 1.3 1.4
0.9 1.2 1.8 1.5 2.0
2.2 3.2 1.2 1.5
37.3 1.9 29.7 1.2
S88/425
S88/465
NJC
CDC
26
Mackenzie Hillberry 1244
CDC NJC NJC
26 26 26
S87/210 S88/155 S88/175
QDH QDH QDH
27A 27B
Harrison Lane 3081
NJC NJC
27 27 27
S88/59 S88/234 S88/237
QDH QDH QDH
28A 28B 28 28
6845 9055 Matthews
NJC
S87/69 S88/101
QDH QDH
CDC
scrofulaceum scrofulaceum
MAC MACb
MAC MAC
M. scrofulaceum M. scrofulaceum MAC MAC MAC
MAC MAC MAC MAC
a Values are means of two to five tests, except for Harrison (serovar 27A) and Lane 3081 (serovar 27B).
b M. intracellulare ATCC
19583-specific a-antigen.
specific a-antigen. Therefore, we identified them as M. scrofulaceum. Strains 5154 and 10409 (originally serotyped as serovar 22) and strain Harrison (originally serotyped as serovar 27) were found to react with anti-serovar 22 and anti-serovar 27 sera, respectively, by our serotyping test, but strain Lane 3081 (originally identified as serovar 27) did not give specific agglutination with any of the antisera. The remaining three strains of serovar 27 (S88/59, S88/234, S88/237) had characteristics typical of MAC. Strains of serovars 23, 24, 26, and 28 possessed MACspecific a-antigen and cultural and biochemical properties characteristic of the MAC (Table 2). Fourteen strains failed to react (or reacted weakly) with either DNA probe (hybridization, 0.7 to 6.7%), and four strains reacted with the M. intracellulare DNA probe (hybridization, 29.7 to 37.3%).
Strains Cox 1944 and Mackenzie had a-antigen specific for M. scrofulaceum but not for MAC. These strains also failed to react with either MAC DNA probe. Similarly, the serovar reference strains from CDC, strains 5154 and 10409 (serovar 22), 23393 (serovar 23), 12645 (serovar 24A), Harrison (serovar 27A), and Lane 3081 (serovar 27B), failed to react (or reacted weakly) with either probe (hybridization, 0.9 to 9.3%); and strain 6845 Koukout (serovar 28A) reacted with M. intracellulare DNA probe (hybridization, 22.3%). DISCUSSION In the present study, we applied the DNA probe test to various MAC serovar strains, including reference strains (6) and isolates from patients. First, we found that serovars 1 to 6, 8 to 11, and 21 showed clear positive reactivities with the M. avium probe but not with the M. intracellulare probe; all other features corroborated their identity as M. avium. Second, serovars 7, 12 to 20, and 25 were identified as M. intracellulare, although the position with serovar 18 needs clarification. This grouping is consistent with the results of DNA-DNA hybridization tests by Baess (1, 2) and sensitin tests on guinea pigs by Magnusson (12), in which strains belonging to serovars 1, 2, 4, 6, and 9 were identified as M. avium and serovars 12, 14, 16, 17, and 20 were identified as M. intracellulare. In cases of serovars 22 to 28 (except for serovar 25), the strains tested were largely divided into three groups. First, serovars 22, 26, and 27 involved some strains which had no reactivity with MAC probes but which we would identify as M. scrofulaceum, because they possessed M. scrofulaceumspecific a-antigen (18) and exhibited characteristic cultural and biochemical properties of M. scrofulaceum (8, 11). In relation to this, it may be noteworthy that Schaefer (16) and McClatchy (13) observed the existence of a number of isolates that were identified biochemically as M. scrofulaceum but serotyped as MAC and vice versa. Second, some strains of serovars 22, 24, 26, and 28 had positive reactivities with the DNA probe of the M. intracellulare- and MACspecific a-antigen (18) (see Table 2) and had properties characteristic of the MAC (8, 11). Third, serovars 23 and the remaining strains of serovars 24, 26, and 28 did not react with either the M. avium or the M. intracellulare probe; but they possessed MAC-specific a-antigen and cultural and biochemical properties of MAC. Similar MAC strains with low reactivities with the MAC DNA probes have been reported anecdotally by CDC and by Enns (4), in the case of clinical isolates of MAC. Enns (4) reported that 8 of 60 strains (13%) identified as MAC by conventional cultural and biochemical methods had no positive hybridization values (
