Molecular Characterization of Mycobacterium avium Complex Isolates ...

CURRENT MICROBIOLOGY Vol. 33 (1996), pp. 352–358

An International Journal

R Springer-Verlag New York Inc. 1996

Molecular Characterization of Mycobacterium avium Complex Isolates from Caribbean Patients by DT1/DT6-PCR, Nonradioactive Southern Hybridization, and the Accuprobe System Christophe Sola, Anne Devallois, Khye Seng Goh, Eric Legrand, Nalin Rastogi Unite´ de la Tuberculose et des Mycobacte´ries, Institut Pasteur, Morne Jolivie`re B.P. 484, 97165 Pointe a` Pitre Cedex, Guadeloupe, French West Indies Received: 9 May 1996 / Accepted: 10 June 1996

Abstract. A genetic fingerprinting analysis of Caribbean isolates of M. avium complex (MAC) from AIDS patients by a Southern blotting technique after Pstl digestion with nonradioactive DNA probes coding for single-copy sequences DT1 and DT6 was performed. In parallel, a selective amplification of a 187-bp fragment within the DT6 sequence with AV6/AV7 primers for Mycobacterium avium and of a 666-bp fragment within the DT1 sequence of M. intracellulare with the IN38/IN41 primers was also performed, and the molecular speciation with these two methods was compared with results obtained with DNA probes of the Accuprobe system. 66 strains investigated comprised 31 international reference isolates of MAC belonging to serovars 1–28 and 42–44, and 35 clinical isolates including 24 strains from Caribbean AIDS patients. 91.43% of the clinical isolates tested gave concordant data with the DT1/DT6 Southern hybridization and PCR as compared with 74.28% for PCR and Accuprobe, and 71.43% for Accuprobe and Southern hybridization. Our results corroborated previous findings showing that the DT1 probe was specific for M. intracellulare, whereas the DT6 probe was specific for M. avium (reference serovars 2 and 3 probed positive both with DT1 and DT6 probes). Contrary to DT1 probe, which did not reveal sufficient polymorphism to discriminate between MAC isolates, DT6 probe showed an interesting polymorphism giving four distinct clusters. Three clusters corresponded to profiles previously reported for reference and/or clinical isolates; however, a fourth cluster was discovered in five Caribbean isolates from four AIDS patients that did not correspond to previously published genetic patterns. When probed with the insertion sequence IS1245, this cluster retained its homogeneity.

Mycobacterium avium complex (MAC) comprises two genetically distinct but difficult to discriminate species: M. avium, which predominates (87–98% of isolates) in AIDS patients, and M. intracellulare, which is more frequent among non-AIDS patients [6]. Because of poor phenotypic differences, conventional cultural and biochemical tests give little information to help separate these two closely related and nearly indistinguishable species in a clinical microbiology laboratory [10]; therefore, they are commonly referred to as the Mycobacterium avium-intracellulare complex or MAC. A combination of results acquired by seroagglutination [20], peptidoglycolipid typing [4], and molecular Correspondence to: N. Rastogi

biology methods [15] has proven useful to define 31 serotypes (serovars) among the MAC complex. From the correlation of seroagglutination and of animal virulence studies, it was first recognized that the serovars 1–3 belong to M. avium, serovars 12–28 belong to M. intracellulare, and there were 8 intermediate serovars. From previously published work on hybridization studies [1], the systematics of these bacteria was changed: the serotypes 1–6, 8–11, and 21 are found within the M. avium species, while the serotypes 7, 12–20, 23–25 are found within M. intracellulare. The serovars 41–43 belong to M. scrofulaceum. The status of serovars 26–28 is not clear at present [18]. Irrespective of various IS sequences described recently, only few targets specific for M. avium or for M.

C. Sola et al.: Molecular Characterization of MAC in the Caribbeans

intracellulare have been reported [2]. Amplification of conserved mycobacterial sequences followed by a restriction enzyme analysis was also described [17]. DT1 and DT6 sequences, however, provide a completely independent identification system from those previously reported [18]. We selected these single-copy sequences DT1 and DT6 (which do not belong to the insertion sequences family), as they not only result in simplified RFLP profiles but also as the information obtained simultaneously helps to speciate the organisms as M. avium or M. intracellulare [18, 19]. In the present investigation, a genetic fingerprinting analysis of M. avium complex (MAC) from Caribbean AIDS patients was performed by Southern blotting with nonradioactive DNA probes derived from DT1 and DT6, and the results were compared with those obtained with DT1/DT6-PCR and nucleic acid hybridization with the Accuprobe system (Gen-Probe Inc., San Diego, Calif.). Materials and Methods Bacteria and growth. In total, 66 MAC strains were studied and comprised 31 international reference isolates of the standard serotype collection [20] and 35 clinical isolates, including 24 Caribbean and 11 European patients. The Caribbean isolates were from patients residing in the French Caribbean islands, and were isolated at the Institut Pasteur of Guadeloupe, whereas the European clinical isolates were addressed to the National Reference Center for Mycobacteria at the Institut Pasteur, Paris. Strain identification was performed on the basis of biochemical and cultural characteristics including mycolic acid analysis [3] and nucleic acid hybridization with the Accuprobe system [7] (Gen-Probe Inc.). Preparation of genomic DNA. For PCR tests, the bacterial DNA was prepared with Chelex-100 (BioRad, Richmond, California) essentially as previously described [11]. Briefly, a loop of bacteria from Lowenstein Jensen medium (LJ) slants was scraped, suspended in 300 µl of a suspension of 10% wt/wt Chelex-100 (containing 0.1%, wt/vol, SDS; 1%, vol/vol, Nonidet-P40; 1%, vol/vol, Tween-20), incubated for 20 min at 95°C, centrifuged (5 min, 15,000 g), and the DNA from the supernatant was extracted by phenol-chloroform and precipitated by ethanol. The AccuProbe system. This method is based on nucleic acid hybridization for the identification of M. avium or M. intracellulare from cultures, and uses acridinium ester-labeled, single-stranded DNA probes that are complementary to the rRNA of target organisms [7]. One loopful of bacteria from fresh LJ slants was lysed by sonication in a tube containing 100 µl each of lysis and hybridization reagents for 15 min, followed by incubation of 100 µl of lysate with lyophilized DNA probe at 60°C for 15 min. The contents were mixed well after the addition of 300 µl of selection reagent, incubated further at 60°C for 5 min, kept at room temperature for at least 5 min, and the results were expressed as relative light units (RLUs) with a Leader luminometer. A positive reaction was above the cut-off value of 30,000 RLUs with a repeat range of 20,000–29,999 RLUs. Parallel positive controls included M. avium ATCC 25291 or M. intracellulare ATCC 13950 in agreement with the probes tested. M. tuberculosis ATCC 25177 tested negative with all three probes used.

353 PCR assays. The method used was essentially similar to that described by Thierry et al. [18]; AV6/AV7 primers (58-ATGGCCGGGAGACGATCTATGCCGGCGTAC-38 and 58-CGTTCGATCGCAGTTTGTGCAGCGCGTACA-38, respectively) directed the amplification of a 187-bp fragment within the DT6 sequence, whereas the IN38/IN41 primers (58-GAACGCCCGTTGGCTGGCCATTCACGAAGGAG-38 and 58-GCGCAACACGGTCGGACAGGCCTTCCTCGA-38) directed the amplification of a 666-bp fragment within the DT1 sequence. Briefly, amplification reactions were performed in 50-µl volume reactions containing 50 mM Tris-HCl (pH 8.5), 2 mM MgCl2, 100 µg of bovine serum albumin/ml, 100 pmol of each primer, 200 µM each of the four deoxyribonucleoside triphosphates (dATP, dGTP, dTTP, and dCTP), 2 ng of template DNA, and 2 units of Thermus aquaticus DNA polymerase (Gibco-BRL Life Technologies, Cergy-Pontoise, France). The amplification mixture was overlaid with 50 µl of mineral oil and was subjected to 30 cycles of amplification (Perkin Elmer Corp., Norwalk, Connecticut) as follows. Samples were incubated at 94°C for 1 min to denature the DNA, 60°C for 1 min to anneal the primers, and 72°C for 1 min to extend the annealed primers. Each amplification experiment included a negative control sample without DNA and a positive control sample with 2 ng of M. avium ATCC 25291 (serotype 2) for AV6/AV7 primers, and 2 ng of M. intracellulare (serotype 23) for the IN38/IN41 primers, as the latter serotype is known to react uniquely with DT1 probe [18]. 50% of the amplification reaction was analyzed by electrophoresis on a 3% NuSieveR, 3:1 agarose gel (FMC BioProducts, Rockland, Maine) by using the 100-bp ladder (Pharmacia Biotech, Uppsala, Sweden) as a marker. Gels were stained with ethidium bromide and photographed on a UV transilluminator. Bacterial DNA preparation for Southern hybridization. Bacterial DNA was prepared by lysing cell wall-deficient forms (CWDF) essentially as described earlier [13]. Briefly, organisms were grown under agitation (100 strokes/min) for 4 days at 37°C in 10 ml of ADC-enriched complete Middlebrook 7H9 medium (Difco, Detroit, Michigan) containing 0.05%, vol/vol, Tween 80. Exponential growth at this step was resumed by further adding 20 ml of fresh medium for one more day, followed by supplementation with D-cycloserine (100 µg/ml), glycine (1.4% wt/vol), EDTA (5 mM), and lysozyme (200 µg/ml). The cultures were incubated for another 48 h at 37°C under mild agitation (50 strokes/mn), centrifuged, washed with 3 ml of TE (Tris 10 mM-EDTA 1 mM, pH 8) and incubated under mild agitation for 18 h at 37°C in 3 ml of lysis solution (Tris-EDTA 50 mM, pH 8; NaCl 100 mM; proteinase K, 100 µg/ml; sodium dodecyl sulfate, 0.5%, vol/vol). DNA was extracted with phenol-chloroform and precipitated with ethanol according to standard procedures [16]. Preparation of DT1/DT6 and IS1245 probes. Escherichia coli (XL1-Blue strain, Stratagene, Calif., USA) was transformed with the plasmids pMA01 and pMA02 harboring DT1 and DT6 fragments [18]. After culture, plasmids were extracted by standard procedures [16], and 876 bp SalIBamHI (DT1) and a 719-bp SalI-EcoRI (DT6) fragments were purified with the Genclean kit (Bio 101, La Jolla, California) and labeled with digoxigenin (Dig DNA labeling kit, BoehringerMannheim, Mannheim, Germany) by the random priming method. The 427-bp IS1245 probe was digoxigenin labeled (Dig-system, BoehringerMannheim) during synthesis by PCR according to Guerrero et al. [8]. Southern hybridization. For Southern hybridization analysis of mycobacterial genomic DNA, 5 µg of total DNA was digested with 40–60 units of PstI (Pharmacia, Uppsala, Sweden) in appropriate buffer for 10–16 h at 37°C; fragments were separated by electrophoresis on 0.8%, wt/vol, agarose gel in Tris-EDTA-sodium acetate buffer and transferred to a Hybond N1 membrane (Amersham, Buckinghamshire, England). After prior DNA fixation for 2 h at 80°C, the membranes were

354

CURRENT MICROBIOLOGY Vol. 33 (1996)

Table 1. Nonradioactive DT1 and DT6 Southern hybridization and RFLP clustering of the international serovar reference catalog (31 isolates) compared with DT1/DT6 PCR and Accuprobe results Hybridization and number of bands (size in kb) Serovara

PCRb

Accuprobec

DT1

1, 5, 6, 8, 9, 10, 11, 21

M. avium

M. avium

2

2, 3

M. avium

M. avium

1 (1 band) (4.5 kb)

4 7, 12–14 15 16, 17, 20, 23–25 18–19 26 22, 27, 28 42–44

M. avium M. intracellulare M. intracellulare M. intracellulare M. intracellulare — — —

M. avium M. intracellulare M. intracellulare M. intracellulare M. intracellulare M. intracellulare — —

2 1 (1 band) (4.5 kb) 1 (1 band) (20 kb) 1 (1 band) (5 kb) 1 (1 band) (.15 kb) 2 2 2

DT6 1 (2 bands) (4.2 kb 1 1.8 kb) 1 (2 bands) (5 kb 1 3 kb) 1 (1 band) (.5 kb) 2 2 2 2 2 2 2

RFLP cluster 1 2 4 None None None None None None None

a

The serovars 42–44 corresponded to M. scrofulaceum. PCR (summarized from reference [18]) for M. avium was performed with AV6–AV7 primers, and for M. intracellulare with IN38–IN41 primers. c M. avium and M. intracellulare probes of the Accuprobe system (summarized from [21]). b

hybridized overnight at 65°C with DT1 or DT6 probes, washed finally in SSC, 0.13, for 20 min at 65°C, followed by colorimetric or chemiluminescent detection (Boehringer-Mannheim). For probing with IS1245, hybridization was performed in 50%, vol/vol, formamide hybridization buffer (Boehringer-Mannheim) at 42°C; final washings were performed at 20°C in SSC, 0.23, for 20 min; and detection was performed as above.

Results Typing of the reference catalog. In the first phase of this study, reference catalog of MAC serovars was probed with DT1 and DT6 to corroborate previous findings [18] and to validate the methodology as applied in the Caribbean setting. The results obtained were compared with speciation results furnished by DT1/DT6 PCR and the Accuprobe system. The results on Southern hybridization of the international reference collection of the 31 serovars are illustrated in Table 1. Our data confirmed the specificity of the DT1 probe for M. intracellulare (with the exception of serovars 2 and 3 of M. avium, which contained both DT1 and DT6 sequences) [18], with a single-band pattern devoid of a significant polymorphism. DT6 was specific for M. avium with a one- to two band pattern, showing an interesting polymorphism; three main patterns observed included a two-band pattern of 4.2 kb and 1.8 kb (serovars 1, 5, 6, 8–11, 21, called group 1), a two-banded pattern of about 3 kb and 5 kb (serovars 2 and 3; called group 2), and a single band with the serovar 4 (called group 4). When our results are compared with published data on DT1/DT6-PCR [18] and Accuprobe [21] on international reference strains, a perfect correlation among

these three methods was observed (Table 1). The present evaluation corroborates recent observations of Thierry et al. [18] and further extends the DT1/DT6 hybridization methodology to nonradioactive detection formats, easily applicable in smaller research facilities not having access to cumbersome and expensive radioactive waste-handling facilities. Correlation of DT1/DT6-PCR and Southern hybridization. When a selective amplification of a 187-bp fragment within the DT6 sequence using AV6/AV7 primers for Mycobacterium avium and of a 666-bp fragment within the DT1 sequence of M. intracellulare with the IN38/IN41 primers was performed, 32 of 35 clinical isolates tested (91.43%) gave concordant data with the DT1/DT6 Southern hybridization (Table 2). Identifiable DT6 banding patterns were not observed for four Caribbean isolates by Southern hybridization; two of these isolates (069P40 and 950003), however, gave identical results with DT6-PCR and Accuprobe system, and were correctly identified as M. avium, confirming the presence of DT6 fragment within their DNA (Table 2). In our opinion, the lack of seeing DT6 bands by Southern hybridization simply indicates the lesser sensitivity of this method compared with the PCR, which is able to amplify even scant proportions of specific DNA. Two other isolates (940070 and 940088) were simultaneously PCR negative and devoid of identifiable banding patterns upon Southern hybridization, corroborating the absence of DT1/DT6 fragments within their DNA; they were speciated as M. intracellulare and MAC, respectively, by use of the Accuprobe system

355

C. Sola et al.: Molecular Characterization of MAC in the Caribbeans Table 2. Results of MAC speciation by DT1/DT6-PCR and Accuprobe compared with nonradioactive DT1 and DT6 Southern hybridization methodology on 24 Caribbean and 11 European clinical isolates Hybridization with Clinical isolatea

PCR

Accuprobe

DT1

DT6

RFLP cluster and speciation

333P40 069P40 940013 940015 940025 940050 940052 940061 940062 940063 940070 940079 940086 940087 940088 940093 940108 940040 940064 940089 940138d 940141 940145 950003

M. avium M. avium M. avium M. avium — M. avium M. intracellulare M. avium M. avium M. avium — M. avium M. avium M. avium — M. avium M. avium M. avium M. intracellulare M. avium avium/intrad M. avium M. intracellulare M. avium

M. avium M. avium M. avium MACc M. intracellulare M. avium MAC M. avium M. avium M. avium M. intracellulare M. avium M. avium M. avium MAC M. avium M. avium M. avium MAC M. avium MAC M. avium M. intracellulare M. avium

2 2 2 2 1 2 1 2 2 2 2 1 2 2 2 2 2 2 1 2 1 2 1 2

1 2 1 1 2 1 2 1 1 1 2 1 1 1 2 1 1 1 2 1 1 1 2 2

Hb (M. avium) 2 (untypable) H (M. avium) H (M. avium) M. intracellulare other (M. avium) M. intracellulare 4 (M. avium) 1 (M. avium) H (M. avium) 2 (untypable) 2 (M. avium) H (M. avium) 1 (M. avium) 2 (untypable) 1 (M. avium) other (M. avium) 1 (M. avium) M. intracellulare 4 (M. avium) other (mixed serovar) 4 (M. avium) M. intracellulare 2 (untypable)

A 002 A 551 A 733 A 969 A 1257 A 711 A 1423 A 804 A 827 A 1110 A 1295

M. avium M. avium M. avium M. avium M. avium M. avium M. avium M. avium M. avium M. avium M. avium

M. avium MAC M. avium MAC M. avium M. avium M. avium M. avium M. avium MAC M. avium

2 2 2 2 2 2 2 2 2 2 2

1 1 1 1 1 1 1 1 1 1 1

other (M. avium) 1 (M. avium) 2 (M. avium) 2 (M. avium) 1 (M. avium) 2 (M. avium) 1 (M. avium) 1 (M. avium) 1 (M. avium) 1 (M. avium) 1 (M. avium)

a

The first part shows 24 Caribbean isolates and the second part 11 European isolates. H or ‘‘hybrid’’ pattern with two bands of 4.2 and 3 kb. c Isolates that were MAC probe positive but were negative for both for M. avium and M. intracellulare probes of the Accuprobe system. d Mixed serovar (1 1 20) isolate, which was DT1/DT6-PCR and Southern hybridization positive. b

because of the difference in target used for hybridization by this system (Table 2). DT1/DT6-PCR and Southern hybridization as compared with the Accuprobe system. When the results were compared with those obtained with the Accuprobe system, 74.28% (26 of 35 isolates) concordant data were obtained for PCR1Acuprobe, and 71.43% (25 of 35 isolates) for Accuprobe1Southern hybridization (Table 2). The main reason is that, despite detection of 100% of the isolates by one of the three probes of the Accuprobe system, it was less able to speciate the MAC isolates at the species level than the DT1/DT6-PCR or Southern

hybridization. For example, 8 of 35 isolates (5 of 24 Caribbean, and 3 of 11 European isolates) were typed as MAC by Accuprobe and could not be discriminated as M. avium or M. intracellulare; however, 7 of these MAC isolates were typable (4 as M. avium, 2 as M. intracellulare, and 1 as a mixed serovar) with PCR and/or Southern hybridization, both of which gave concordant results (Table 2). DT6-RFLP patterns of the clinical isolates. Molecular typing of clinical isolates was performed with Southern blotting after PstI digestion with nonradioactive digoxigenin-labeled DT6 probe, and these results are summa-

356


Fig. 1. The schematized representation of various banding patterns obtained by the video-copied images of Southern blots (Gel-Analyst and Video-copy, Bioprobet systems, Montreuil, France), followed by analysis with the Taxotront software (P.A.D. Grimont, Institut Pasteur, Paris). Figure 1A illustrates the four PstI-DT6-RFLP patterns generated: Lane 1, group 2; lanes 2 and 6, group 1; lanes 3, 5, and 7, ‘‘H’’ cluster isolates; and lane 4, group 4. Figure 1B (lanes 1 to 5) shows the five ‘‘H’’ cluster isolates (two-banded pattern of 4.2 and 3 kb) from four patients upon PstIDT6-RFLP. Please refer to Tables 1 and 2 for precise molecular weights and cluster designation. Figure 1C illustrates the PstI-IS1245 RFLP: lanes 1 and 8 show the molecular weight ladder (Lambda-HindIII), lanes 2 to 6 represent the ‘‘H’’ cluster isolates, whereas lane 7 shows an unrelated strain (notice a single band difference for the isolate 94-0015 shown in lane 4).

rized in Table 2. The representative banding patterns are illustrated in a schematized form (Fig. 1A) upon videocopied image of the various Southern blots with the Gel-Analyst software (Gel-Analyst and Video-copy, Bioprobet systems, Montreuil, France), followed by further analysis with the Taxotront software (P.A.D. Grimont, Institut Pasteur, Paris). In addition to the three profiles initially found within the standard serovar collection of MAC isolates, a previously unreported two-banded pattern of 4.2 and 3 kb was also observed (Fig. 1B and Table 2). This cluster was unique to the five Caribbean isolates, corresponding to four terminally ill AIDS patients (CD4 ,50 mm3) from the same hospital, and called group H (H for ‘‘hybrid,’’ as the profile obtained was initially thought to be a hybrid of groups 1 and 2). However, as the group H isolates did not cross-react with the DT1 probe, which is able to pick M. avium serovar 2 and 3 strains, these strains did not represent a polyclonal infection [22]. Further identification of the ‘‘H’’ cluster isolates. In order to investigate the homogeneity of ‘‘H’’ cluster isolates, these were probed with the recently described IS1245 probe, which has been reported to be highly discriminatory for human and animal MAC isolates [8]. The results obtained showed that all the isolates were identical except for a single band difference for the isolate 94-0015 (Figs 1C and 2).

Discussion Mycobacterium avium complex (MAC) is a major opportunistic pathogen afflicting acquired immunodeficiency syndrome (AIDS) patients. Because of slow growth, lengthy identification procedures, and scant biochemical

Fig. 2. A photographic representation of the autoradiography showing the IS1245 (427bp, digoxigenin-labeled, PCR probe) reprobing of the original PstI-DT6 blot. For further comments, molecular weight ladder, and lanes represented, please refer to Fig. 1C and its legend showing the video-copied image of this autoradiograph (two intermediate, nonspecific hybridization bands are not shown in the video copy).

differences between M. avium and M. intracellulare, molecular techniques allowing for rapid identification and speciation applicable in routine clinical microbiology laboratories need to be developed. In this context, combined application of DT1/DT6-PCR and Southern hybridization may be important tools in patient management. Considering the reported heterogeneity of MAC

C. Sola et al.: Molecular Characterization of MAC in the Caribbeans

organisms, such an approach may also help to better characterize possible geographical variations among M. avium isolates, such as the presence of a previously unreported ‘‘H’’ cluster strains among Caribbean patients in this study. This study also showed that 91.43% of the clinical isolates tested gave concordant data with the DT1/DT6 Southern hybridization and PCR compared with 74.28% for PCR and Accuprobe, and 71.43% for Accuprobe and Southern hybridization. It is noteworthy that 91.43% of isolates in this study were correctly identified as MAC by DT1/DT6-PCR, with 80% being correctly speciated as M. avium with DT6-PCR alone, and 11.43% as M. intracellulare (or mixed serotypes) with DT1-PCR as a secondary test. We may, therefore, conclude that, in addition to the Accuprobe test in our setting, DT1/DT6PCR is also a useful in-house option for rapid detection and speciation of MAC clinical isolates. The molecular epidemiology of MAC organisms is of great interest because the origin of these infections is poorly understood. It has been suggested that some serovars may be more virulent than others, which, however, does not explain the reasons that distinct serovars have been shown to dominate in distinct geographical areas [9]. It can be postulated that specific environmental conditions may be a determining factor influencing the occurrence of various subtypes, and the molecular characterization of MAC clinical isolates from distant parts of the world may further help to understand the basis of genetic diversity of these strains [14]. In the above context, the results obtained not only corroborated previous findings [18, 19], but also developed a nonradioactive hybridization detection system, an application projected as a potential future development by its original contributors [18]. The finding of previously unreported ‘‘H cluster’’ isolates from Caribbean patients was unexpected, since the typing of the collection of serovars had revealed only three common patterns with the PstI-DT6 combination [18]. Although we may reasonably hypothesize that the source of this infection was common to all four patients (for example, tap water, food, soil from potted plants etc. have been suggested as a potential environmental source of infection in AIDS patients) [5, 23], experimental evidence showing specificity of these isolates in the Caribbean environment and their ecological reservoir is still lacking. Although the banding pattern on the serovar collection initially tended to indicate a poor polymorphism within the DT1 and DT6 sequences, these probes were tested on clinical isolates because of the presence of a PstI site inside the DT6 fragment [18], and we decided to use this enzyme to search for (RFLPs). Alternatively,

357 when using IS elements as probes, the RFLP analysis might be more difficult because of complex patterns needing complementary software and information systems. Although a general concern with the DT1/DT6 procedure may be the somewhat limited amount of information generated from assays that use a single enzyme to analyze variability in a single-copy gene, we intentionally chose to strictly follow the methodology by the original contributors [18], since ours is the only independent assay corroborating previous observations and extending them to clinical isolates. Results on DT6 polymorphism including the reproducibility of clusters generated may be further investigated with a battery of enzymes in future studies. Although we did not intend to extensively study the M. intracellulare polymorphism at this step (because of the absence of Pst1 site in DT1 probe), the variability in DT1 hybridization patterns was significant (Table 1; single bands of 4.5–20 kb) and suggested the potential development of a system based on DT1-polymorphism with enzymes cleaving within the DT1-probe. For the time being, such a study is hampered by extremely rare M. intracellulare infections among AIDS patients in Guadeloupe, and would be feasible in the future when enough isolates are accumulated. By classical biochemical criteria, all the strains tested in this study were unambiguously assigned as MAC. Our results, however, open new questions about the ‘‘H’’-cluster strains described. We are currently working on a restriction mapping project to analyze the genetic relatedness between these strains. The Accuprobe analysis of these samples clearly established that ‘‘H’’ cluster unequivocally consisted of M. avium strains (Table 2). Further analysis of M. avium clinical isolates from the Caribbean belt and Latin America would not only allow a precise definition of the genetic relationships between the Caribbean strains, but would also broaden our knowledge of the environmental source of this organism. The heterogeneity of MAC strains from different regions of the world and the genetic variability of isolates from different hosts have been discussed in detail previously with IS 900 and IS 901 RFLP-typing [12]. The results of the present investigation, therefore, corroborte previous findings about the genetic and geographic variability of MAC organisms, and further show that DT1/DT6-PCR is an inexpensive, rapid, and equally sensitive in-house option to the Accuprobe system for identification and speciation of MAC organisms in routine clinical microbiology laboratories.

358


ACKNOWLEDGMENTS The mycobacteria and TB project at Guadeloupe received financial support through Project CORDET, Ministry of Overseas Departments and Territories, Paris, France and Pharmacia S.A., St. Quentin-Yvelines, France. N. Rastogi is grateful to Gen Probe-France (Chatillon, France) for lending a Leader-50 luminometer; to the Fogarty International Center, National Institutes of Health, Bethesda, Maryland for an ‘‘HIV, AIDS and Related Illness Award (AIDS-FIRCA #TWOO533: awarded jointly with W.W. Barrow, Southern Research Institute, Birmingham, Alabama); and to J.L. Guesdon (Institut Pasteur, Paris, France) for kindly providing the pMA01 and pMA02 plasmids. A. Devallois was awarded a Ph.D. fellowship from the Ministry of Education and Research, Paris, France.

Literature Cited 1. Baess I (1983) Deoxyribonucleic acid relationships between different serovars of Mycobacterium avium, M. intracellulare, and M. scrofulaceum. Acta Pathol Microbiol Scand 91:201–203 2. Crawford JT (1994) Development of rapid techniques for identification of M. avium infections. Res Microbiol 145:177–181 3. David H, Levy-Fre´bault V, Thorel MF (1989) Me´thodes de laboratoire pour mycobacte´riologie clinique. Paris: Institut Pasteur 4. Denner JC, Tsang AY, Chatterjee D, Brennan P (1992) Comprehensive approach to identification of serovars of Mycobacterium avium complex. J Clin Microbiol 30:473–478 5. Du Moulin GC, Stottmeier KD, Pelletier PA, Tsang AY, HedleyWhite J (1988) Concentration of Mycobacterium avium in hospital hot water systems. J Am Med Assoc 260:1599–1601 6. Falkinham JO III (1994) Epidemiology of Mycobacterium avium infections in the pre- and post-HIV era. Res Microbiol 145:169– 172 7. Gen-Probe Inc. (1992) AccuProbe@, Mycobacterium avium, Mycobacterium intracellulare culture identification test. In: Manual for in vitro diagnostic use. San Diego, CA: Gen-Probe Inc, pp 1–6 8. Guerrero C, Bernasconi C, Burki D, Bodmer T, Telenti A (1995) A novel insertion sequence element from Mycobacterium avium IS 1245 is a specific target for analysis of strain relatedness. J Clin Microbiol 33:304–307 9. Hoffner SE, Ka¨llenius G, Petrini B, Brennan PJ, Tsang AY (1990) Serovars of Mycobacterium avium complex isolated from patients in Sweden. J Clin Microbiol 28:1105–1107 10. Inderlied CB, Kemper CA, Bermudez LEM (1993) The Mycobacterium avium complex. Clin Microbiol Rev 6:266–310 11. Lamballerie XD, Zandotti C, Vignoli C, Bollet C, de Micco P (1992) A one-step microbial DNA extraction method using ‘‘Chelex 100’’ suitable for gene amplification. Res Microbiol 143:785– 790 12. McFadden JJ, Kunze ZM, Portaels F, Labrousse V, Rastogi N

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

(1992) Epidemiological and genetic markers, virulence factors and intracellular growth of Mycobacterium avium in AIDS. Res Microbiol 143:423–430. Rastogi N, David HL (1981) Ultrastructural and chemical studies on wall-deficient forms, spheroplasts and membrane vesicles from Mycobacterium aurum. J Gen Microbiol 124:71–79 Rastogi N, Barrow WW, Falkinham III JO, Thoen CO, Crawford JT, Mangura BT, Reichman LB, Heifets LB, Dautzenberg B, Young LS, Bermudez LEM, Inderlied CB, Suzuki AE, Inamine JM, Gangadharam PRJ, Reddy MV, Denis M, Shiratsuchi H, Johnson JL, Ellner JJ, Belisle JT, Brennan PJ (1994) Laboratory and clinical aspects of the Mycobacterium avium epidemic: contributing factors associated with variability of drug susceptibility and immune responsiveness, and the multifaceted nature of pathogenicity. Res Microbiol 145:167–261 Saito H, Tomioka H, Sato K, Tasaka H, Dawson DJ (1990) Identification of various serovar strains of Mycobacterium avium complex by using DNA probes specific for Mycobacterium avium and Mycobacterium intracellulare. J Clin Microbiol 28:1694–1697 Sambrook J, Fritsch EF, Maniatis T (1989) Molecular cloning, a laboratory manual, second ed. Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory Press Telenti A, Marchesi F, Balz M, Bally F, Bo¨ttger EC, Bodmer T (1993) Rapid identification of mycobacteria to the species level by polymerase chain reaction and restriction enzyme analysis. J Clin Microbiol 31:175–178 Thierry D, Vincent V, Cle´ment F, Guesdon J-L (1993a) Isolation of specific DNA fragments of Mycobacterium avium and their possible use in diagnosis. J Clin Microbiol 31:1048–1054 Thierry D, Bauge´ S, Poveda J-D, Vincent V, Guesdon J-L (1993b) Rapid identification of Mycobacterium avium-intracellulare complex strains: clinical practice evaluation of DT6 and DT1 probes. J Infect Dis 168:1337–1338 Tsang AY, Drupa I, Goldberg M, McClatchy JK, Brennan PJ (1983) Use of serology and thin-layer chromatography for the assembly of an authenticated collection of serovars within the Mycobacterium avium-Mycobacterium intracellulare-Mycobacterium scrofulaceum complex. Int J Syst Bacteriol 33:285–292 Vamerot A, Cle´ment F, Vincent Le´vy-Fre´bault V (1992) Evaluation of nonradioactive DNA probes for the identification of Mycobacterium avium complex. Methods Mol Cell Biol 3:198–203 Von Reyn CF, Pestel M, Arbeit RD (1996) Clinical and epidemiological implications of polyclonal infection due to Mycobacterium avium complex. Res Microbiol 147:24–30 Yajko DM, Chin DP, Gonzalez PC, Nassos PS, Hopewell PC, Reingold AL, Horsburgh Jr. CR, Yakrus MA, Ostroff SM, Hadley WK (1994) Mycobacterium avium complex in water, food, and soil samples collected from the environment of HIV-infected individuals. J AIDS Hum Retrovirol 9:176–182