Use of Specific rRNA Oligonucleotide Probes for Microscopic Detection of Mycobacterium avium Complex Organisms in Tissue
http://www.100md.com
微生物临床杂志 2005年第4期
Department of Molecular, Cellular, and Developmental Biology, University of Colorado, Boulder, Colorado
Department of Medicine, Division of Infectious Diseases, University of Colorado Health Sciences Center, Denver, Colorado
ABSTRACT
Members of the Mycobacterium avium complex (MAC) are important environmental pathogens that are implicated in several chronic, idiopathic diseases. Diagnosis of MAC-based diseases is compromised by the need to cultivate these fastidious and slowly growing organisms in order to identify which mycobacterial species are present. Detection is particularly difficult when MAC is intracellular or embedded within mammalian tissues. We report on the development of culture-independent, in situ hybridization (ISH) assays for the detection of MAC in culture, sputum, and tissue. This assay includes a highly reliable technique for the permeabilization of mycobacterial cells within culture and tissues. We describe a set of rRNA-based oligonucleotide probes that specifically detect either M. intracellulare, the two M. avium subspecies associated with human disease, or all members of MAC. The results call into question the validity of ISH results derived by the use of other gene loci, such as IS900.
INTRODUCTION
The Mycobacterium avium complex (MAC) consists of genetically similar, slowly growing bacteria that include M. avium subsp. avium, M. avium subsp. paratuberculosis, M. avium subsp. silvaticum, M. avium subsp. hominis, and M. intracellulare (16). MAC organisms are opportunistic pathogens that can be isolated from multiple environmental sources, including drinking water, soil, plants, dairy products, bioaerosols, and animals (25, 52). Although human exposure to MAC is ubiquitous, most individuals rarely develop infections. Immunocompromised individuals, such as those with AIDS or individuals who have had organ transplants, are at the greatest risk for MAC infection (3, 33). Before the emergence of AIDS, most MAC infections were pulmonary in nature and typically affected patients with preexisting lung diseases, such as emphysema or cystic fibrosis (13, 39, 51). MAC is also the most frequent cause of pediatric cervical lymphadenitis (7, 26). In recent years the emergence of cases of hot tub lung and lifeguard lung has been noted. In these cases disease occurs secondary to a hypersensitivity that develops due to aerosolized MAC bacteria in therapeutic pools, hot tubs, or indoor swimming pools (25, 37).
Although M. avium subsp. avium and M. intracellulare are the main causative agents of human MAC diseases (14), M. avium subsp. paratuberculosis has been suggested to be an emerging human pathogen (29). In addition to the well-known MAC infections, members of MAC are implicated as causes of human granulomatous diseases, such as sarcoidosis and Crohn's disease (CD) (45). Sarcoidosis is an idiopathic, multisystem disease that has many features in common with mycobacterial infections (30, 40, 55). Although acid-fast bacilli have not been detected in sarcoid granulomas, several studies (18, 21, 44, 47) have reported on the detection of mycobacteria, including M. avium subsp. paratuberculosis, in granulomatous tissue by molecular biology-based techniques. Two recent medical case reports demonstrate that M. avium subsp. paratuberculosis is able to infect humans. First, in 2002, a German AIDS patient developed a disseminated mycobacterial infection, and M. avium subsp. paratuberculosis was isolated from the patient (53). Second, Hermon-Taylor et al. (31) reported on a case of pediatric cervical lymphadenitis in which M. avium subsp. paratuberculosis was the primary infectious agent. It is perhaps noteworthy that this patient later developed CD.
M. avium subsp. paratuberculosis has long been suspected as a trigger of CD, a chronic human inflammatory bowel disease of unknown etiology (2, 8, 22, 28, 32). CD clinically resembles Johne's disease, which occurs in ruminants and nonhuman primates and which is characterized by severe gastroenteritis caused by M. avium subsp. paratuberculosis (6, 28, 62). Detection of M. avium subsp. paratuberculosis by either culture or molecular biology-based techniques in a subset of patients with CD has been reported (5, 11, 34, 50, 54, 56, 58). However, unambiguous identification of M. avium subsp. paratuberculosis as the etiologic agent of CD is hampered by the difficult and time-consuming methods necessary to detect the organism. M. avium subsp. paratuberculosis is extremely difficult to culture and requires prolonged incubation for weeks to months (65). At present, most molecular analyses of Johne's disease and CD rely on detection of the M. avium subsp. paratuberculosis-specific insertion element IS900 by nested and/or quantitative PCR (4, 12, 23, 63). However, as is the case with culture studies, the application of these molecular biology-based techniques to CD has led to contradictory results (10, 38, 48, 64). An alternative approach, in situ hybridization (ISH), is an attractive technique for the detection of M. avium subsp. paratuberculosis and other mycobacteria. Not only does ISH allow a more rapid diagnosis than culture methods, but also fewer bacilli can be detected by ISH than by current PCR methods. Furthermore, unlike PCR, ISH provides detailed, tissue-based morphological information.
One of the major obstacles to the use of ISH to identify mycobacteria is the tough, lipid-containing cell walls of these organisms, which renders cells impermeable to oligonucleotide probes (41). However, Hulten et al. (35, 36) developed an ISH methodology to identify M. avium subsp. paratuberculosis by microscopic detection of IS900. The IS900-based ISH (IS900 ISH) procedure uses a digoxigenin- or biotin-labeled double-stranded DNA probe and enzymatic amplification of the signal with alkaline phosphatase (AP)-conjugated antibodies. However, the IS900 ISH technique can detect only cell wall-deficient forms of M. avium subsp. paratuberculosis, which are hypothesized to play a role in colonization or persistence in CD (35, 64). By this method, both Hulten et al. (34) and Sechi et al. (58, 59) reported on the detection of M. avium subsp. paratuberculosis in tissues obtained from patients with CD. However, permeabilization issues due to the large size of the double-stranded DNA probe (241 bp) and the low copy number of IS900 targets (1 to 14 copies per M. avium subsp. paratuberculosis genome) render this technique and the results questionable. It seems unlikely that a few hybridization targets per cell, even including potential IS900 transcripts, would be able to sequester sufficient amounts of probe to visualize a single bacterial cell.
An alternative hybridization approach to IS900 ISH is the use of oligonucleotide probes that target rRNA (1, 27). rRNA is typically present in cells at high copy numbers (thousands of copies per cell) and thus provides a far greater number of potential targets than the IS900 genes. The abundance of rRNA in cells allows the ready visualization of individual organisms when fluorescent probes bind to their rRNA targets. Phylogenetic comparisons of rRNA gene sequences can be used to design oligonucleotide probes that can distinguish between very closely related organisms, such as species of mycobacteria (42). Several commercial products are approved for the molecular biology-based detection of mycobacteria, and most use 16S rRNA sequences as probes (e.g., AccuProbe rRNA hybridization [Gen-Probe, Inc., San Diego, Calif.]) or amplification primers (e.g., Amplified Mycobacterium Tuberculosis Direct Test [Gen-Probe, Inc.]) (19, 61, 65, 66). All of these commercial assays either require cultivation of the organism or are approved only for the assay of sputum samples. There is a clear need for rapid diagnostic methodologies that are applicable to many different tissue types, in addition to sputum. In this paper we report on the development of several different MAC rRNA-specific probes and techniques for the reproducible and efficient permeabilization of mycobacteria in tissue and culture. Using bovine Johne's disease as a model system, we compare our methods to previously published IS900-based protocols. Finally, we demonstrate the general applicability of rRNA-specific probes to the diagnosis of a variety of proven and suspected human cases of MAC disease.
MATERIALS AND METHODS
Culture preparations. Mycobacterial cultures were acquired from Leonid Heifets at The National Jewish Medical and Research Center, Denver, Colo. (M. avium subsp. paratuberculosis, M. intracellulare, M. kansasii, and M. tuberculosis) and Mark Hernandez at the University of Colorado, Boulder (M. avium subsp. avium, M. phlei, M. parafortuitum, Bacillus subtilis, Escherichia coli, Rhodococcus sp., and Corynebacterium sp.). All cultures were fixed in 10% formalin for 4 to 10 h, washed with phosphate-buffered saline (PBS; pH 8.0) (57), and stored in 70% ethanol-30% PBS at –20°C. Fixed cultures were applied to slides prepared with silane (Sigma-Aldrich). The slides were dipped in 100% xylene for 5 min and rehydrated in an ethanol series (100, 70, 30, and 0% ethanol in 10 mM Tris [pH 7.5]). The cultures were then treated with lysozyme (1 mg/ml; Sigma-Aldrich) and achromopeptidase (30 U/ml; Sigma-Aldrich) in 10 mM Tris (pH 7.5) at 37°C for 25 min. The cultures were washed for 5 min in 10 mM Tris (pH 7.5) and then dehydrated in an ethanol series (30, 70, and 100% ethanol in 10 mM Tris [pH 7.5]). The slides were allowed to air dry before hybridization.
Tissue preparation. Mesenteric lymph nodes and ileal tissues from diseased and healthy dairy cows were obtained from Randall Basaraba at Colorado State University, Fort Collins. Mediastinal lymph node sections from M. tuberculosis-infected guinea pigs were also obtained from Randall Basaraba. The presence of M. avium subsp. paratuberculosis in tissues from cows with Johne's disease was confirmed by Ziehl-Neelsen staining (49), IS900-based quantitative PCR (5), and M. avium subsp. paratuberculosis rRNA-specific quantitative PCR (15). M. avium subsp. paratuberculosis was not detected in negative control samples from healthy cattle by either acid-fast staining and microscopy or PCR. The National Jewish Medical and Research Center provided archived small-bowel sections from a pediatric CD patient and lung tissues from adult patients infected with M. avium and/or M. intracellulare. The Swedish Medical Center (Denver, Colo.) provided an archived tissue biopsy specimen from the hand of a patient with tenosynovitis secondary to M. avium infection. The Denver Health Medical Center (Denver, Colo.) provided acid-fast-positive sputum specimens from an elderly smoker with a pulmonary mycobacterial infection. All tissue samples were deidentified. This research was conducted under the University of Colorado at Boulder Human Research Committee exemption of protocol 0799.14 (Molecular analysis of microbes in human and animal diseased tissues) for analysis of deidentified human and animal tissues.
The tissue specimens were fixed in 10% formalin for 24 h, washed with PBS, and stored in 70% ethanol-30% PBS at –20°C. Tissue was embedded in paraffin wax, and 4-μm sections were placed on slides prepared with silane (Sigma-Aldrich). The tissue specimens were dewaxed in 100% xylene for 20 min and then rehydrated in an ethanol series (100, 70, 30, and 0% ethanol in 10 mM Tris [pH 7.5]). The sections were then treated with lysozyme (1 mg/ml; Sigma-Aldrich) and achromopeptidase (30 U/ml; Sigma-Aldrich) in 10 mM Tris (pH 7.5) at 37°C for 25 min. The tissue specimens were washed for 5 min in 10 mM Tris (pH 7.5) and dehydrated in an ethanol series (30, 70, and 100% ethanol in 10 mM Tris [pH 7.5]). All sections were allowed to air dry before hybridization.
16S and 23S rRNA-specific oligonucleotide probe hybridization. The probes used in this study were EUB338 (5'-CTG CTG CCT CCC GTA GGA GT-3') (1, 17, 27), MAVP187ssu (5'-TGC GTC TTG AGG TCC TAT CC-3'), MAVP515lsu (5'-TGT CCA TGC ATG CGG TTT-3'), MAC2543lsu (5'-ACG CCA CTA CAC CCC AAA-3'), MIN351ssu (5'-AGG TAG AGC TGA GAT GTA TCC T-3'), and MIN1586lsu (5'-CCC CGA AAC TCC ATG CCC-3'). All oligonucleotides were obtained from Integrated DNA Technologies, Inc., and were labeled at the 5' and 3' ends with either 6'-carboxyfluorescein (FAM) or Cy3. All probes described above except EUB338 were designed by using the probe design function of ARB software (43) and small- and large-subunit rRNA gene sequences obtained from the GenBank database. The hybridization buffer consisted of 900 mM NaCl, 20 mM Tris (pH 8.0), 0.01% sodium dodecyl sulfate, and 20% formamide. Probes were added to the hybridization buffer at a final concentration of 1 ng/μl, and the probe-hybridization buffer was applied to the specimens. Frame-Seal incubation chambers (MJ Research, Inc.) for slides were used to prevent evaporation. The hybridization slides were placed in an MJ Research, Inc., slide thermocycler and were heated at 94°C for 5 min. The slides were incubated at either 36°C (probes MIN351lsu, MIN1586lsu, and MAC2543lsu) or 40°C (probes MAVP187ssu, MAVP515lsu, and EUB338) for 6 to 18 h. Following hybridization, the slides were washed with 225 mM NaCl-5 mM EDTA-0.01% sodium dodecyl sulfate-20 mM Tris (pH 8.0) for 20 min at either 37 or 41°C. The slides were then dipped briefly in cold 20 mM Tris (pH 8.0) in order to remove excess salt and were finally mounted with Citifluor (Electron Microscopy Sciences) antifading reagent. The slides were examined under epifluorescence with a Nikon Eclipse E600 microscope or a Leica confocal laser scanning microscope.
Antifluorescein-AP visualization. Hybridization of the FAM-labeled oligonucleotide probes was carried out as described above. Hybridization was followed by incubation of the slides in blocking buffer (3% bovine serum albumin, 100 mM Tris [pH 7.5], 150 mM NaCl, 0.3% Triton X-100) at 4°C for 30 min. Fresh blocking buffer that contained 2 mU of antifluorescein-labeled AP (antifluorescein-AP) Fab fragments (Roche) per μl was then added to the sections. The slides were incubated at 4°C for 2 to 3 h. The slides were then washed for 15 min with buffer 1 (100 mM Tris [pH 7.5], 150 mM NaCl, 0.3% Triton X-100), followed by another wash with buffer 2 (100 mM Tris [pH 9.5], 150 mM NaCl, 50 mM MgCl2) for 15 min. The slides were placed in AP buffer (100 mM Tris [pH 9.5], 150 mM NaCl, 50 mM MgCl2, 0.2 mg of levamisol [Sigma-Aldrich] per ml, 0.1% Tween) for 10 min. 5-Bromo-4-chloro-3-indolylphosphate (BCIP)-2-(4-iodophenyl)-3-(4-nitrophenyl)-5-phenyltetrazolium chloride (INT) (Roche) was then added to fresh AP buffer (75 μl of INT-BCIP per 10 ml of AP buffer) and applied to the slides, and the slides were then incubated in the dark at 25°C for 1 to 2 h. The reaction was stopped by a brief wash in water. The sections were counterstained with methylene blue, mounted with Faramount (DakoCytomation), and observed under a bright-field microscope.
IS900-biotin and IS900-fluorescein hybridization. The IS900-biotin- and IS900-fluorescein-labeled probes were prepared as described by Sechi et al. (58), with the exception that fluorescein-12-dUTP (Roche) was incorporated into the IS900-fluorescein probe. Hybridization was performed as described by Hulten et al. (36), with the exception that biotin rather than digoxigenin was used to label the 241-bp double-stranded DNA probe. The tissue samples were deparaffinized in 100% xylene, rehydrated through a graded ethanol series (100, 70, and 30% ethanol in PBS), and washed with PBS (pH 7.4). The tissue specimens were then incubated with proteinase K (Sigma-Aldrich) in PBS at concentrations ranging from 0 to 1 mg/ml for 20 min at 37°C. Proteinase K was inactivated with 0.2% glycine in PBS. The sections were then dehydrated through a graded ethanol series and allowed to air dry. The sections were hybridized with 1 ng of labeled probe per μl in hybridization buffer (50% deionized formamide, 2x SSC [1x SSC is 0.15 M NaCl plus 0.015 M sodium citrate] [57], 10% dextran sulfate, 0.25 μg of yeast tRNA [Sigma-Aldrich] per μl, 0.5 μg of denatured salmon sperm DNA [Sigma-Aldrich] per μl, 1x Denhardt's solution [Sigma-Aldrich]). The probe was boiled for 10 min and cooled on ice for 10 min before application. The sections with the probe were heated for 10 min at 95°C and chilled on ice for 10 min. Hybridization was performed overnight at 37°C. The washing steps included washes with 2x SSC for 15 min at room temperature (RT), 1x SSC for 15 min at RT, 0.3x SSC for 15 min at 40°C, and 0.3x SSC for another 15 min at RT. The washes were followed by incubation of the slides in blocking buffer at room temperature for 30 min. Fresh blocking buffer that contained 2 mU of antibiotin-labeled AP (antibiotin-AP) Fab fragments (Roche) per μl was added to the slides, and the slides were incubated at room temperature for 2 to 3 h. The slides were then washed with buffer 1 for 15 min, followed by another wash with buffer 2 for 15 min. The slides were then placed in AP buffer for 10 min. INT-BCIP (75 μl per 10 ml of AP buffer) was added to fresh AP buffer, and the mixture was applied to the slides, which were incubated in the dark at RT for 4 to 12 h. The reaction was stopped by washing the slides in distilled water. The slides were mounted with Faramount (DakoCytomation) and observed under a bright-field microscope. IS900-fluorescein hybridizations were performed as described above for IS900-biotin, except that epifluorescence microscopy was used for direct visualization of the specimens.
Dot blot hybridization was used to confirm that the IS900-biotin and IS900-fluorescein probes were synthesized correctly. Briefly, M. avium subsp. paratuberculosis genomic DNA, E. coli genomic DNA, the IS900 PCR product, and the E. coli 16S rRNA PCR product were blotted onto nitrocellulose (Hybond+; Amersham Biosciences), according to the procedures of the manufacturer. The dot blot was hybridized with either probe by the procedures described above. The blots were visualized with either antibiotin-AP Fab fragments (Roche) or antifluorescein-AP Fab fragments (Roche). The AP and INT-BCIP colorimetric reaction mixture was incubated for 20 min at RT. The probes hybridized only to M. avium subsp. paratuberculosis genomic DNA and the IS900 PCR product (data not shown).
RESULTS
Development of ISH protocols and characterization of MAC rRNA ISH probes. A variety of potential MAC-specific ISH oligonucleotide probe sequences were chosen by using the Probe Design function of the ARB software package (43) and an alignment of small- and large-subunit rRNA sequences obtained from GenBank. M. avium subsp. paratuberculosis and M. avium subsp. avium rRNA sequences are identical, so it was impossible to design probes that would differentiate between the two subspecies. As a first test of specificity, fluorescently labeled probes were applied to pure cultures of cognate and noncognate species, including M. avium subsp. paratuberculosis, M. avium subsp. avium, M. intracellulare, M. tuberculosis, M. kansasii, M. parafortuitum, M. phlei, E. coli, B. subtilis, a Rhodococcus sp., and a Corynebacterium sp. A broad-range bacterial 16S probe, EUB338-FAM, was included in every hybridization experiment in order to test the effectiveness of the permeabilization and the hybridization protocols.
In initial experiments, none of the ISH probes, including EUB338, successfully hybridized to any of the mycobacterial cells. The complex, lipid-rich cell walls of mycobacteria are notoriously refractory to the lysis steps of standard ISH protocols (17). Consequently, we tested a number of techniques to develop a protocol that would enhance the permeabilization of mycobacterial cells. Enzymatic digestion of fixed cells with proteinase K (58, 59) failed to improve the fluorescent signal upon hybridization with EUB338 or the MAC-specific probes. Enzymatic digestion with lysozyme and achromopeptidase (60) resulted in a fluorescent signal for only a small fraction of the fixed cells from culture. We found, however, that a brief incubation of the fixed mycobacterial cells in 100% xylene, combined with subsequent lysozyme and achromopeptidase digestions, resulted in strong fluorescent hybridization signals. Treatment with all three reagents (xylene, lysozyme, and achromopeptidase) was absolutely required for successful ISH of mycobacterial cells. Although each oligonucleotide probe was tested individually (Table 1), we found that the combination of the species-specific probes for ISH resulted in a stronger fluorescent signal, even though the final probe concentration remained the same (data not shown).
By this enhanced ISH protocol, Cy3-conjugated probes MAVP187ssu and MAVP515lsu (Table 1) hybridized well to both M. avium subsp. avium and M. avium subsp. paratuberculosis (Fig. 1A and B, respectively) but not to M. intracellulare (Fig. 1D), M. tuberculosis (Fig. 1F), or other related and unrelated bacterial species (listed above; data not shown). Hybridization with probe EUB338-FAM clearly labeled the M. intracellulare and M. tuberculosis cells (Fig. 1C and E, respectively), indicating that cell wall permeabilization was effective. As expected, probe MAC2543lsu hybridized to all three members of the MAC complex, M. intracellulare, M. avium subsp. avium, and M. avium subsp. paratuberculosis, but failed to hybridize to the other species tested (Fig. 1G to L and Table 1). Finally, probes MIN351lsu and MIN1586lsu specifically hybridized to M. intracellulare cells but not to the cells of the other species tested, including other members of MAC, M. avium subsp. avium and M. avium subsp. paratuberculosis (Fig. 1M to R and Table 1).
Application of MAC-specific ISH probes to tissue samples. Tissues from dairy cows with well-characterized Johne's disease and healthy animals were used in order to establish the ability of the MAC-specific probes to stain the MAC cells embedded within tissues. The presence or absence of M. avium subsp. paratuberculosis in infected and uninfected mesenteric lymph nodes (MLNs) was first determined by Ziehl-Neelsen staining and quantitative PCR (data not shown) (5, 15). Both negative and positive control MLN samples were then probed with a combination of Cy3-labeled MAVP187ssu and MAVP515lsu fluorescent oligonucleotides. The negative control MLNs showed no sign of fluorescent bacilli when they were probed with the MAVP187ssu and MAVP515lsu oligonucleotides (Fig. 2A). In contrast, hybridization of the positive control MLNs resulted in an intense and abundant fluorescent signal (Fig. 2B) in an aggregate pattern similar to the acid-fast staining pattern (see Fig. 4B). The MIN351ssu and MIN1586lsu probes did not hybridize to either the bacillus-positive or the bacillus-negative MLN specimens (data not shown). The MAC2543lsu probe labeled the M. avium subsp. paratuberculosis-positive MLN specimen but not the M. avium subsp. paratuberculosis-negative MLN specimen (data not shown). Because the ISH protocol requires extensive digestion of the mycobacterial cell wall for probe accessibility, we were unable to find conditions that would allow dual staining with Ziehl-Neelsen and rRNA-specific probes.
Figure 2C demonstrates the ability of confocal laser scanning microscopy to detect clearly defined bacilli that hybridized with the MAVP187ssu-Cy3 and MAVP515lsu-Cy3 probes in M. avium subsp. paratuberculosis-infected bovine MLN tissue. The nuclei of lymphatic cells are stained with 4',6'-diamidino-2-phenylindole (DAPI) and appear blue. The confocal laser scanning microscope was able to resolve bacilli in different focal planes within the tissue, indicating that the probe had penetrated the entire tissue section. As expected for M. avium subsp. paratuberculosis, the labeled cells are rods approximately 1 μm in length. In general, tissue autofluorescence did not greatly affect the labeling or detection of M. avium subsp. paratuberculosis cells in most bovine or human tissues. However, high levels of autofluorescence were observed when erythrocytes, collagen, and/or elastic fibers were abundant, such as in lung tissue samples. We consequently developed a nonfluorescent approach to probe detection that uses AP conjugated to antifluorescein antibodies to enzymatically amplify a colorimetric dye (INT-BCIP; see Materials and Methods). In addition to avoiding background autofluorescence, this approach also allows the use of tissue staining and light microscopy to visualize hybridized microbes and colocalize them with the areas of mammalian tissue pathology. A representative micrograph of an ileum tissue specimen from a cow with Johne's disease probed with the MAVP oligonucleotides is shown in Fig. 2D. Arrows indicate the reddish brown M. avium subsp. paratuberculosis bacilli.
MAC-specific probes for ISH detect MAC species in human clinical specimens. The MAC-specific probes used for ISH were applied to a variety of archival tissue specimens from patients with MAC infections in order to determine the applicabilities of these probes for the identification and diagnosis of mycobacterial infections. As shown in Fig. 3A to D, tissues were probed with the M. avium subsp. avium- and M. avium subsp. paratuberculosis-specific oligonucleotides MAVP187ssu and MAVP515lsu and visualized by either fluorescence microscopy (Fig. 3B and C) or light microscopy following treatment with antifluorescein-coupled AP (Fig. 3A and D). Figure 3A shows a tissue section from a lung resection from a patient infected with M. avium subsp. avium, as determined by culture and with the AccuProbe (GenProbe). The MAVP probes were able to detect many bacilli (Fig. 3A, arrows) throughout the tissue. The M. intracellulare-specific probes did not hybridize to the same resected lung tissue (data not shown). Tissue from a patient with tenosynovitis of the hand was negative by acid-fast staining but M. avium subsp. avium positive by culture after several weeks of growth; the AccuProbe (GenProbe) assay confirmed the presence of M. avium subsp. avium. The MAVP probes were clearly able to detect isolated bacilli (Fig. 3B) of the appropriate size and morphology. Figure 3C and D shows tissue sections from the duodenum and small bowel, respectively, of a pediatric CD patient. In these specimens, the MAVP probes hybridized to bacilli located predominantly within cells of the lamina propria.
Figure 3E shows a representative micrograph of lung tissue containing large granulomas that was probed with the M. intracellulare-specific oligonucleotides MIN351lsu and MIN1586lsu. The hybridization signal was visualized by the antifluorescein-AP colorimetric assay. The lung tissue was acid-fast positive, with one to two bacilli located in the granulomas in each tissue section; M. intracellulare was cultured from a section of this tissue, and its presence was confirmed with the AccuProbe (GenProbe). The M. intracellulare-specific FAM-labeled probes were able to detect more M. intracellulare bacilli (Fig. 3E, arrows) in the granulomas than were detected by acid-fast staining. In contrast, the MAVP probes did not hybridize to this lung tissue section (data not shown). The ISH methodology also worked well with sputum samples. Figure 3F shows a micrograph of an acid-fast positive sputum sample from a patient with a history of smoking, chronic cough, and an upper-zone cavity with a diagnosis of pulmonary MAC infection 2 years earlier. AccuProbe identification was inconclusive, and broad-range 16S rRNA sequence data detected a Corynebacterium sp. and small numbers of M. intracellulare cells (data not shown). ISH with the MAC-specific probe MAC2543-FAM revealed sparse, labeled bacilli among a background of DAPI-labeled bacteria of corynebacterial morphology. In addition, none of the MAC-specific probes hybridized to mediastinal lymph node tissue specimens from guinea pigs infected with M. tuberculosis (data not shown).
Comparison of rRNA ISH and IS900 ISH methodologies. Some recently published studies have used double-stranded DNA probes that target genomic DNA for the localization of mycobacteria in situ (34, 58). Because ribosomes, the targets of the ISH probes described in this study, are present in much greater cellular abundance than genomic DNA loci, a direct comparison of the DNA- and rRNA-based approaches was in order. To this end, MLN tissues from healthy cows and cows with Johne's disease were used as negative and positive controls, respectively, for ISH. Ziehl-Neelsen staining determined the absence or presence of acid-fast bacilli in each tissue sample (Fig. 4A and B, respectively). M. avium subsp. paratuberculosis infection of the positive control tissue was also demonstrated by IS900 and 16S rRNA PCRs (data not shown).
As expected, the MAC rRNA-specific probe MAVP187ssu-FAM stained only the positive control tissue, as indicated by a reddish brown precipitate (the INT-BCIP-stained AP product) with the morphology of clearly defined bacilli and a size and morphology consistent with M. avium subsp. paratuberculosis (Fig. 4D). No reddish brown signal was evident in the negative control tissue (Fig. 4C). In contrast, application of IS900 probes to bovine tissue sections resulted in the appearance of a variety of rod-like and amorphous forms in both the negative (Fig. 4E) and the positive (Fig. 4F) control MLN tissues. Under high-power oil-immersion magnification, however, these objects had a granular, amorphous appearance not necessarily suggestive of mycobacteria, whereas the cells stained with either Ziehl-Neelsen stain or rRNA-specific probes (Fig. 4B and D, respectively) had clear morphological boundaries, structures that resemble microorganisms.
To determine whether the false-positive results that were obtained with the IS900 probe were due to nonspecific precipitation or aggregation of the colorimetric dye, we synthesized a fluorescein-labeled IS900 probe (IS900-fluorescein) that allowed direct epifluorescent detection of cells. ISH of positive and negative control bovine MLN tissues with this IS900-fluorescein probe also resulted in significant nonspecific staining in both control tissue specimens (Fig. 4G and H). Indeed, application of the published IS900 ISH protocol (35) to organisms other than M. avium subsp. paratuberculosis that were grown in pure culture resulted in strong nonspecific signals (Fig. 5A). For example, Fig. 5A shows B. subtilis cells that were dually probed by the IS900 ISH procedure with IS900-fluorescein isothiocyanate (FITC) (green) and EUB338-Cy3 (red). Although the rRNA probe is uniformly distributed within the cells, consistent with the expected distribution of ribosomes, the IS900-FITC probe results in a punctate staining pattern that is localized to the outside of the cells. Similarly, Fig. 5B and C shows false-positive hybridization of the IS900-biotin probe to a Rhodococcus sp. and M. kansasii. Of the bacteria tested, only gram-negative microbes, such as E. coli and Pseudomonas aeruginosa, were resistant to the nonspecific hybridization with the IS900-biotin probe (data not shown). Figure 5D shows an example of an M. avium subsp. avium isolate probed with EUB338-FAM and visualized enzymatically with antifluorescein-AP Fab fragments and INT-BCIP. The reddish brown signal is distributed evenly throughout the cell, which had the expected morphology. Thus, both our permeabilization protocol and the use of small rRNA-based oligonucleotide probes rather than large DNA-based probes, such as IS900, are necessary for successful ISH of mycobacteria.
DISCUSSION
The members of MAC are increasingly recognized as important environmental pathogens and are implicated in several chronic, idiopathic diseases (16, 20, 24). CD, for example, is hypothesized to be a zoonotic infection that arises from exposure to dairy products contaminated with M. avium subsp. paratuberculosis, the causative agent of bovine Johne's disease (29). The public health implications of a link between human Crohn's disease and bovine Johne's disease demand careful consideration, given the increasing incidence of Johne's disease in cattle herds (6) and the presence of viable M. avium subsp. paratuberculosis isolates in commercial milk supplies (9, 46). The development of rapid and accurate assays for the detection of MAC organisms in tissue is urgently needed. In this paper we report on a rapid, highly reliable technique for the permeabilization and identification of MAC members in tissues and cultures by ISH. We describe rRNA-based oligonucleotide probes that specifically detect M. intracellulare (probes MIN351lsu and MIN1586lsu), the two M. avium subspecies (M. avium subsp. avium and M. avium subsp. paratuberculosis; probes MAVP187ssu and MAVP515lsu), and all members of MAC (probe MAC2543lsu). To assess the specificities and sensitivities of the MAC probes, we applied these probes to three types of samples: (i) axenic bacterial cultures (M. kansasii and M. tuberculosis, pathogens closely related to MAC), (ii) tissues from cows with Johne's disease infected with M. avium subsp. paratuberculosis, and (iii) human clinical samples with a variety of confirmed mycobacterial infections. By our methodology the ISH probes specifically labeled the mycobacterial species against which they were designed in both tissues and culture samples.
Because of their waxy, mycolic acid-laden cell walls, mycobacteria are normally recalcitrant to the typical permeabilization steps used for ISH with bacteria. When ISH was performed with mycobacteria, the results were reproducible only when the cells were permeabilized with a combination of xylene, lysozyme, and achromopeptidase treatments. Omission of any one of these steps either diminished or abolished the signals from the probes. Our permeabilization protocol also works well with other gram-positive bacteria, such as Corynebacterium sp., Rhodococcus sp., and B. subtilis; but in these instances, the xylene step is not critical.
Proteinase K is often used to permeabilize mycobacteria in culture and tissue in preparation for ISH (34, 58). In our experience, however, the use of proteinase K for mycobacterial ISH resulted either in inadequate permeabilization (i.e., the lack of a signal) or in significant tissue damage and nonspecific binding of the probes to tissue (i.e., excessive background signal). Overdigestion with proteinase K is therefore likely to increase the occurrence of false-positive results, as were observed by the IS900 ISH (discussed below). In contrast, lysozyme and achromopeptidase digest only bacterial cell walls and therefore do not affect the surrounding mammalian tissue.
The MAC-specific rRNA probes were visualized by two methods: (i) direct fluorescence and (ii) enzymatic amplification of a colorimetric substrate. One important benefit of the fluorescent method is its applicability to confocal laser scanning microscopy, which allows the high-resolution imaging of microbes in tissues. MAC cells were observed to be embedded in tissue sections of both bovine and human tissues, often within the cytoplasms of eukaryotic cells, thus eliminating the possibility of cross-contamination of specimens during the preparation of tissue sections. In our experience, it was absolutely essential to use high-resolution microscopy, either under a x1,000 oil-immersion or by confocal laser scanning microscopy, in order to confirm by the morphology the staining of microbial cells. In addition to direct visualization by fluorescence, the ISH probes also were indirectly visualized by bright-field microscopy following the reaction of a colorimetric dye (INT-BCIP) with antibody-conjugated AP. This method is not influenced by tissue autofluorescence, which can obscure the true signals of fluorescently labeled probes. Although autofluorescence typically is not a problem, tissues that are densely packed with red blood cells and/or mast cells can be prone to high autofluorescent backgrounds. Because this indirect method of visualization can be performed with tissue sections following direct fluorescent detection, the two approaches can be used to corroborate the results and thereby diminish the possibility of false-positive results.
The 16S and 23S rRNA sequences of M. avium subsp. paratuberculosis and M. avium subsp. avium are identical, and so rRNA-specific probes cannot distinguish between the two organisms. PCR amplification of the IS900 gene, which is present in the genome of M. avium subsp. paratuberculosis but not that of M. avium subsp. avium, is an effective means of typing the two subspecies. Although it is theoretically possible, the detection of M. avium subsp. paratuberculosis by IS900-based ISH presents several technical challenges. First, as was demonstrated with the rRNA-specific oligonucleotide probes, proper permeabilization is absolutely essential for gaining access to hybridization targets within a cell. Large hybridization probes, such as those produced by PCR amplification or nick translation of the IS900 gene, should require even more careful control of permeabilization conditions than is necessary for smaller probes. Second, M. avium subsp. paratuberculosis genomes are reported to carry only 1 to 14 copies of the IS900 gene, which necessitates an extremely sensitive assay for the detection of so few targets. Bacterial mRNAs are very difficult to detect by ISH, so IS900 mRNA is unlikely to be detectable in these assays. Nevertheless, IS900-based ISH has been suggested as a means of detecting M. avium subsp. paratuberculosis in clinical tissue specimens. Using IS900 ISH, both Sechi et al. (58) and Hulten et al. (34) have reported on the presence of M. avium subsp. paratuberculosis in specimens from patients with CD. In our experience, however, ISH with IS900-specific probes resulted in false-positive signals, especially if the colorimetric dye (INT-BCIP) was allowed to react with AP for prolonged periods (from 2 h to overnight). When the reaction times were limited to less than 2 h, no signal was seen for positive control tissues (data not shown).
It is crucial that the bacterial cell morphology be used as a benchmark for the validation of a positive signal with any kind of ISH assay because of the possibility of nonspecific signals (Fig. 4 and 5) (54). Often, this morphology can be visualized only by examination at a magnification of x1,000 under oil immersion because single, 1-μm rods (the typical size of M. avium subsp. paratuberculosis cells) are extremely difficult to visualize at x400 magnification, the magnification typically used in clinical analyses. When cells were visualized under oil immersion, the structures labeled with IS900-specific probes and INT-BCIP had large indistinct boundaries, whereas the cells labeled with rRNA-specific probes and INT-BCIP had clearly defined morphologies.
The false-positive results obtained with the IS900-specific probes were not due solely to the colorimetric labeling method. Under high magnification, fluorescently labeled IS900-specific probes were observed to bind to the outer cell walls of a variety of bacterial species. In contrast, rRNA-specific probes were observed to label the interiors of cells uniformly. Thus, the false-positive results obtained with the IS900-specific probes most likely arose from nonspecific binding of the DNA-specific probes to cells, coupled with nonspecific precipitation and aggregation of dye. Overall, the lack of specificity and apparently artifactual nature of the IS900-specific probes make their use questionable for the assay of mammalian tissues, such as tissue specimens from patients with CD, for the presence or absence of M. avium subsp. paratuberculosis bacteria.
Furthermore, we demonstrate the ability of the MAC rRNA-specific probes to detect and identify MAC organisms in a variety of animal and human tissue and clinical specimens, including sputum. Our probes offer researchers and clinicians an incisive method for identifying members of MAC in tissue specimens and culture. Future work with specimens from diseased and normal tissues will further define the sensitivities and specificities of the MAC-specific probes. In several cases, including M. intracellulare-infected lung tissue and tissue from a patient with M. avium subsp. avium tenosynovitis, the rRNA probes detected more bacilli than were evident by standard acid-fast staining procedures. Mycobacteria could not be detected by PCR in many of these specimens, possibly due to the low abundance of organisms or the presence of PCR inhibitors in the tissue specimen. On the other hand, ISH can detect single cells and, thus, may possibly provide greater sensitivity than traditional PCR. ISH probes hold the promise for the more rapid detection and differentiation of mycobacterial species than can be achieved by traditional clinical microbiological techniques.
ACKNOWLEDGMENTS
We thank Leonid Heifets, Sarah Wilson, Kayte Fulton, Randall Reves, Ginger Hildred, Michael Moynihan, Burton Golub, Randall Basaraba, and Mark Hernandez for providing strains, clinical specimens, and valuable feedback.
This work was supported by a grant from the National Institutes of Health (to N.R.P.).
REFERENCES
Amann, R. I., L. Krumholz, and D. A. Stahl. 1990. Fluorescent-oligonucleotide probing of whole cells for determinative, phylogenetic, and environmental studies in microbiology. J. Bacteriol. 172:762-770.
Behr, M. A., M. Semret, A. Poon, and E. Schurr. 2004. Crohn's disease, mycobacteria, and NOD2. Lancet Infect. Dis. 4:136-137.
Benson, C. A. 1994. Disease due to the Mycobacterium avium complex in patients with AIDS: epidemiology and clinical syndrome. Clin. Infect. Dis. 18(Suppl. 3):S218-S222.
Bull, T. J., J. Hermon-Taylor, I. Pavlik, F. El-Zaatari, and M. Tizard. 2000. Characterization of IS900 loci in Mycobacterium avium subsp. paratuberculosis and development of multiplex PCR typing. Microbiology 146(Pt 9):2185-2197.
Bull, T. J., E. J. McMinn, K. Sidi-Boumedine, A. Skull, D. Durkin, P. Neild, G. Rhodes, R. Pickup, and J. Hermon-Taylor. 2003. Detection and verification of Mycobacterium avium subsp. paratuberculosis in fresh ileocolonic mucosal biopsy specimens from individuals with and without Crohn's disease. J. Clin. Microbiol. 41:2915-2923.
Chacon, O., L. E. Bermudez, and R. G. Barletta. 2004. Johne's disease, inflammatory bowel disease, and Mycobacterium paratuberculosis. Annu. Rev. Microbiol. 58:329-363.
Chao, S. S., K. S. Loh, K. K. Tan, and S. M. Chong. 2002. Tuberculous and nontuberculous cervical lymphadenitis: a clinical review. Otolaryngol. Head Neck Surg. 126:176-179.
Chiodini, R. J. 1989. Crohn's disease and the mycobacterioses: a review and comparison of two disease entities. Clin. Microbiol. Rev. 2:90-117.
Chiodini, R. J., and J. Hermon-Taylor. 1993. The thermal resistance of Mycobacterium paratuberculosis in raw milk under conditions simulating pasteurization. J. Vet. Diagn. Investig. 5:629-631.
Chiodini, R. J., H. J. Van Kruiningen, W. R. Thayer, and J. A. Coutu. 1986. Spheroplastic phase of mycobacteria isolated from patients with Crohn's disease. J. Clin. Microbiol. 24:357-363.
Collins, M. T., G. Lisby, C. Moser, D. Chicks, S. Christensen, M. Reichelderfer, N. Hoiby, B. A. Harms, O. O. Thomsen, U. Skibsted, and V. Binder. 2000. Results of multiple diagnostic tests for Mycobacterium avium subsp. paratuberculosis in patients with inflammatory bowel disease and in controls. J. Clin. Microbiol. 38:4373-4381.
Cousins, D. V., R. Whittington, I. Marsh, A. Masters, R. J. Evans, and P. Kluver. 1999. Mycobacteria distinct from Mycobacterium avium subsp. paratuberculosis isolated from the faeces of ruminants possess IS900-like sequences detectable IS900 polymerase chain reaction: implications for diagnosis. Mol. Cell. Probes 13:431-442.
De Groote, M. A. 2004. Pulmonary infection in non-HIV infected individuals. In J. Bartram and A. Dufour (ed.), M. avium in water. U.S. Environmental Protection Agency/World Health Organization, London, United Kingdom.
De Groote, M. A. 2004. Skin, bone, and soft tissue infections. In J. Bartram and A. Dufour (ed.), M. avium in water. U.S. Environmental Protection Agency/World Health Organization, London, United Kingdom.
De Groote, M. A., D. N. Frank, E. Nachtway, A. L. St. Amand, H. Hearst, F. Garry, R. Basaraba, and N. R. Pace. Unpublished data.
De Groote, M. A., and M. D. Iseman. 2003. Environmental mycobacterial infections. In J. G. Bartlett and N. R. Blacklow (ed.), Gorbach textbook of infectious diseases. Lipincott Williams and Wilkins, Philadelphia, Pa.
DeLong, E. F., G. S. Wickham, and N. R. Pace. 1989. Phylogenetic stains: ribosomal RNA-based probes for the identification of single cells. Science 243:1360-1363.
Drake, W. P., Z. Pei, D. T. Pride, R. D. Collins, T. L. Cover, and M. J. Blaser. 2002. Molecular analysis of sarcoidosis tissues for Mycobacterium species DNA. Emerg. Infect. Dis. 8:1334-1341.
Drobniewski, F. A., P. G. More, and G. S. Harris. 2000. Differentiation of Mycobacterium tuberculosis complex and nontuberculous mycobacterial liquid cultures by using peptide nucleic acid-fluorescence in situ hybridization probes. J. Clin. Microbiol. 38:444-447.
Eccles, E., and J. Ptak. 1995. Mycobacterium avium complex infection in AIDS: clinical features, treatment, and prevention. J. Assoc. Nurses AIDS Care 6:37-47.
El-Zaatari, F. A., S. A. Naser, D. C. Markesich, D. C. Kalter, L. Engstand, and D. Y. Graham. 1996. Identification of Mycobacterium avium complex in sarcoidosis. J. Clin. Microbiol. 34:2240-2245.
El-Zaatari, F. A., M. S. Osato, and D. Y. Graham. 2001. Etiology of Crohn's disease: the role of Mycobacterium avium paratuberculosis. Trends Mol. Med. 7:247-252.
Englund, S. 2003. IS900/ERIC-PCR as a tool to distinguish Mycobacterium avium subsp. paratuberculosis from closely related mycobacteria. Vet. Microbiol. 96:277-287.
Falkinham, J. O., III. 1996. Epidemiology of infection by nontuberculous mycobacteria. Clin. Microbiol. Rev. 9:177-215.
Falkinham, J. O., III. 2003. Mycobacterial aerosols and respiratory disease. Emerg. Infect. Dis. 9:763-767.
Fitzpatrick, E. L., and F. E. LeJeune, Jr. 1996. Mycobacterial cervical lymphadenitis: a review. J. La. State Med. Soc. 148:451-454.
Giovannoni, S. J., E. F. DeLong, G. J. Olsen, and N. R. Pace. 1988. Phylogenetic group-specific oligodeoxynucleotide probes for identification of single microbial cells. J. Bacteriol. 170:720-726.
Greenstein, R. J. 2003. Is Crohn's disease caused by a mycobacterium Comparisons with leprosy, tuberculosis, and Johne's disease. Lancet Infect. Dis. 3:507-514.
Greenstein, R. J., and M. T. Collins. 2004. Emerging pathogens: is Mycobacterium avium subspecies paratuberculosis zoonotic Lancet 364:396-397.
Hance, A. J. 1998. The role of mycobacteria in the pathogenesis of sarcoidosis. Semin. Respir. Infect. 13:197-205.
Hermon-Taylor, J., N. Barnes, C. Clarke, and C. Finlayson. 1998. Mycobacterium paratuberculosis cervical lymphadenitis, followed five years later by terminal ileitis similar to Crohn's disease. BMJ 316:449-453.
Hermon-Taylor, J., and T. Bull. 2002. Crohn's disease caused by Mycobacterium avium subspecies paratuberculosis: a public health tragedy whose resolution is long overdue. J. Med. Microbiol. 51:3-6.
Horsburgh, C. R., Jr. 1999. The pathophysiology of disseminated Mycobacterium avium complex disease in AIDS. J. Infect. Dis. 179(Suppl. 3):S461-S465.
Hulten, K., H. M. El-Zimaity, T. J. Karttunen, A. Almashhrawi, M. R. Schwartz, D. Y. Graham, and F. A. El-Zaatari. 2001. Detection of Mycobacterium avium subspecies paratuberculosis in Crohn's diseased tissues by in situ hybridization. Am. J. Gastroenterol. 96:1529-1535.
Hulten, K., T. J. Karttunen, H. M. El-Zimaity, S. A. Naser, A. Almashhrawi, D. Y. Graham, and F. A. El-Zaatari. 2000. In situ hybridization method for studies of cell wall deficient M. paratuberculosis in tissue samples. Vet. Microbiol. 77:513-518.
Hulten, K., T. J. Karttunen, H. M. El-Zimaity, S. A. Naser, M. T. Collins, D. Y. Graham, and F. A. El-Zaatari. 2000. Identification of cell wall deficient forms of M. avium subsp. paratuberculosis in paraffin embedded tissues from animals with Johne's disease by in situ hybridization. J. Microbiol. Methods 42:185-195.
Iivanainen, E., J. Northrup, R. D. Arbeit, M. Ristola, M. L. Katila, and C. F. von Reyn. 1999. Isolation of mycobacteria from indoor swimming pools in Finland. APMIS 107:193-200.
Kanazawa, K., Y. Haga, O. Funakoshi, H. Nakajima, A. Munakata, and Y. Yoshida. 1999. Absence of Mycobacterium paratuberculosis DNA in intestinal tissues from Crohn's disease by nested polymerase chain reaction. J. Gastroenterol. 34:200-206.
Kilby, J. M., P. H. Gilligan, J. R. Yankaskas, W. E. Highsmith, Jr., L. J. Edwards, and M. R. Knowles. 1992. Nontuberculous mycobacteria in adult patients with cystic fibrosis. Chest 102:70-75.
Kon, O. M., and R. M. du Bois. 1997. Mycobacteria and sarcoidosis. Thorax 52(Suppl. 3):S47-S51.
Lambert, P. A. 2002. Cellular impermeability and uptake of biocides and antibiotics in gram-positive bacteria and mycobacteria. Symp. Ser. Soc. Appl. Microbiol. 2002:46S-54S.
Lane, D. J., B. Pace, G. J. Olsen, D. A. Stahl, M. L. Sogin, and N. R. Pace. 1985. Rapid determination of 16S ribosomal RNA sequences for phylogenetic analyses. Proc. Natl. Acad. Sci. USA 82:6955-6959.
Ludwig, W., O. Strunk, R. Westram, L. Richter, H. Meier, Yadhukumar, A. Buchner, T. Lai, S. Steppi, G. Jobb, W. Forster, I. Brettske, S. Gerber, A. W. Ginhart, O. Gross, S. Grumann, S. Hermann, R. Jost, A. Konig, T. Liss, R. Lussmann, M. May, B. Nonhoff, B. Reichel, R. Strehlow, A. Stamatakis, N. Stuckmann, A. Vilbig, M. Lenke, T. Ludwig, A. Bode, and K. H. Schleifer. 2004. ARB: a software environment for sequence data. Nucleic Acids Res. 32:1363-1371.
Mangiapan, G., and A. J. Hance. 1995. Mycobacteria and sarcoidosis: an overview and summary of recent molecular biological data. Sarcoidosis 12:20-37.
McFadden, J. J., and H. M. Fidler. 1996. Mycobacteria as possible causes of sarcoidosis and Crohn's disease. Soc. Appl. Bacteriol. Symp. Ser. 25:47S-52S.
Millar, D., J. Ford, J. Sanderson, S. Withey, M. Tizard, T. Doran, and J. Hermon-Taylor. 1996. IS900 PCR to detect Mycobacterium paratuberculosis in retail supplies of whole pasteurized cows' milk in England and Wales. Appl. Environ. Microbiol. 62:3446-3452.
Mirmirani, P., T. A. Maurer, B. Herndier, M. McGrath, M. D. Weinstein, and T. G. Berger. 1999. Sarcoidosis in a patient with AIDS: a manifestation of immune restoration syndrome. J. Am. Acad. Dermatol. 41:285-286.
Moss, M. T., E. P. Green, M. L. Tizard, Z. P. Malik, and J. Hermon-Taylor. 1991. Specific detection of Mycobacterium paratuberculosis by DNA hybridisation with a fragment of the insertion element IS900. Gut 32:395-398.
Murray, R. G. E., N. Doetsch, and C. F. Robinow. 1994. Determinative and cytological light microscopy, p. 32-33. In P. Gerhardt, R. G. E. Murray, W. A. Wood, and N. R. Krieg (ed.), General and molecular bacteriology. American Society for Microbiology, Washington, D.C.
Naser, S. A., I. Shafran, D. Schwartz, F. El-Zaatari, and J. Biggerstaff. 2002. In situ identification of mycobacteria in Crohn's disease patient tissue using confocal scanning laser microscopy. Mol. Cell. Probes 16:41-48.
Oliver, A., L. Maiz, R. Canton, H. Escobar, F. Baquero, and E. Gomez-Mampaso. 2001. Nontuberculous mycobacteria in patients with cystic fibrosis. Clin. Infect. Dis. 32:1298-1303.
Primm, T. P., C. A. Lucero, and J. O. Falkinham III. 2004. Health impacts of environmental mycobacteria. Clin. Microbiol. Rev. 17:98-106.
Richter, E., J. Wessling, N. Lugering, W. Domschke, and S. Rusch-Gerdes. 2002. Mycobacterium avium subsp. paratuberculosis infection in a patient with HIV, Germany. Emerg. Infect. Dis. 8:729-731.
Roholl, P. J., A. Herrewegh, and D. van Soolingen. 2002. Positive IS900 in situ hybridization signals as evidence for role of Mycobacterium avium subsp. paratuberculosis in etiology of Crohn's disease. J. Clin. Microbiol. 40:3112. (Author reply, 40:3112-3113.)
Rutherford, R. M., and J. J. Gilmartin. 2004. Mycobacteria in pathogenesis of sarcoidosis. Chest 125:354.
Ryan, P., M. W. Bennett, S. Aarons, G. Lee, J. K. Collins, G. C. O'Sullivan, J. O'Connell, and F. Shanahan. 2002. PCR detection of Mycobacterium paratuberculosis in Crohn's disease granulomas isolated by laser capture microdissection. Gut 51:665-670.
Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, vol. 3, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Sechi, L. A., M. Mura, E. Tanda, A. Lissia, A. Solinas, G. Fadda, and S. Zanetti. 2001. Identification of Mycobacterium avium subsp. paratuberculosis in biopsy specimens from patients with Crohn's disease identified by in situ hybridization. J. Clin. Microbiol. 39:4514-4517. (Erratum, 40:739, 2002.)
Sechi, L. A., M. Mura, E. Tanda, A. Lissia, G. Fadda, and S. Zanetti. 2004. Mycobacterium avium sub. paratuberculosis in tissue samples of Crohn's disease patients. New Microbiol. 27:75-77.
Sekar, R., A. Pernthaler, J. Pernthaler, F. Warnecke, T. Posch, and R. Amann. 2003. An improved protocol for quantification of freshwater actinobacteria by fluorescence in situ hybridization. Appl. Environ. Microbiol. 69:2928-2935.
Tortoli, E., M. T. Simonetti, and F. Lavinia. 1996. Evaluation of reformulated chemiluminescent DNA probe (AccuProbe) for culture identification of Mycobacterium kansasii. J. Clin. Microbiol. 34:2838-2840.
Van Kruiningen, H. J. 1999. Lack of support for a common etiology in Johne's disease of animals and Crohn's disease in humans. Inflamm. Bowel Dis. 5:183-191.
Vansnick, E., P. De Rijk, F. Vercammen, D. Geysen, L. Rigouts, and F. Portaels. 2004. Newly developed primers for the detection of Mycobacterium avium subspecies paratuberculosis. Vet. Microbiol. 100:197-204.
Wall, S., Z. M. Kunze, S. Saboor, I. Soufleri, P. Seechurn, R. Chiodini, and J. J. McFadden. 1993. Identification of spheroplast-like agents isolated from tissues of patients with Crohn's disease and control tissues by polymerase chain reaction. J. Clin. Microbiol. 31:1241-1245.
Watterson, S. A., and F. A. Drobniewski. 2000. Modern laboratory diagnosis of mycobacterial infections. J. Clin. Pathol. 53:727-732.
Woods, G. L., J. S. Bergmann, and N. Williams-Bouyer. 2001. Clinical evaluation of the GenProbe amplified Mycobacterium tuberculosis direct test for rapid detection of Mycobacterium tuberculosis in select nonrespiratory specimens. J. Clin. Microbiol. 39:747-749.(Allison L. St. Amand, Dan)
Department of Medicine, Division of Infectious Diseases, University of Colorado Health Sciences Center, Denver, Colorado
ABSTRACT
Members of the Mycobacterium avium complex (MAC) are important environmental pathogens that are implicated in several chronic, idiopathic diseases. Diagnosis of MAC-based diseases is compromised by the need to cultivate these fastidious and slowly growing organisms in order to identify which mycobacterial species are present. Detection is particularly difficult when MAC is intracellular or embedded within mammalian tissues. We report on the development of culture-independent, in situ hybridization (ISH) assays for the detection of MAC in culture, sputum, and tissue. This assay includes a highly reliable technique for the permeabilization of mycobacterial cells within culture and tissues. We describe a set of rRNA-based oligonucleotide probes that specifically detect either M. intracellulare, the two M. avium subspecies associated with human disease, or all members of MAC. The results call into question the validity of ISH results derived by the use of other gene loci, such as IS900.
INTRODUCTION
The Mycobacterium avium complex (MAC) consists of genetically similar, slowly growing bacteria that include M. avium subsp. avium, M. avium subsp. paratuberculosis, M. avium subsp. silvaticum, M. avium subsp. hominis, and M. intracellulare (16). MAC organisms are opportunistic pathogens that can be isolated from multiple environmental sources, including drinking water, soil, plants, dairy products, bioaerosols, and animals (25, 52). Although human exposure to MAC is ubiquitous, most individuals rarely develop infections. Immunocompromised individuals, such as those with AIDS or individuals who have had organ transplants, are at the greatest risk for MAC infection (3, 33). Before the emergence of AIDS, most MAC infections were pulmonary in nature and typically affected patients with preexisting lung diseases, such as emphysema or cystic fibrosis (13, 39, 51). MAC is also the most frequent cause of pediatric cervical lymphadenitis (7, 26). In recent years the emergence of cases of hot tub lung and lifeguard lung has been noted. In these cases disease occurs secondary to a hypersensitivity that develops due to aerosolized MAC bacteria in therapeutic pools, hot tubs, or indoor swimming pools (25, 37).
Although M. avium subsp. avium and M. intracellulare are the main causative agents of human MAC diseases (14), M. avium subsp. paratuberculosis has been suggested to be an emerging human pathogen (29). In addition to the well-known MAC infections, members of MAC are implicated as causes of human granulomatous diseases, such as sarcoidosis and Crohn's disease (CD) (45). Sarcoidosis is an idiopathic, multisystem disease that has many features in common with mycobacterial infections (30, 40, 55). Although acid-fast bacilli have not been detected in sarcoid granulomas, several studies (18, 21, 44, 47) have reported on the detection of mycobacteria, including M. avium subsp. paratuberculosis, in granulomatous tissue by molecular biology-based techniques. Two recent medical case reports demonstrate that M. avium subsp. paratuberculosis is able to infect humans. First, in 2002, a German AIDS patient developed a disseminated mycobacterial infection, and M. avium subsp. paratuberculosis was isolated from the patient (53). Second, Hermon-Taylor et al. (31) reported on a case of pediatric cervical lymphadenitis in which M. avium subsp. paratuberculosis was the primary infectious agent. It is perhaps noteworthy that this patient later developed CD.
M. avium subsp. paratuberculosis has long been suspected as a trigger of CD, a chronic human inflammatory bowel disease of unknown etiology (2, 8, 22, 28, 32). CD clinically resembles Johne's disease, which occurs in ruminants and nonhuman primates and which is characterized by severe gastroenteritis caused by M. avium subsp. paratuberculosis (6, 28, 62). Detection of M. avium subsp. paratuberculosis by either culture or molecular biology-based techniques in a subset of patients with CD has been reported (5, 11, 34, 50, 54, 56, 58). However, unambiguous identification of M. avium subsp. paratuberculosis as the etiologic agent of CD is hampered by the difficult and time-consuming methods necessary to detect the organism. M. avium subsp. paratuberculosis is extremely difficult to culture and requires prolonged incubation for weeks to months (65). At present, most molecular analyses of Johne's disease and CD rely on detection of the M. avium subsp. paratuberculosis-specific insertion element IS900 by nested and/or quantitative PCR (4, 12, 23, 63). However, as is the case with culture studies, the application of these molecular biology-based techniques to CD has led to contradictory results (10, 38, 48, 64). An alternative approach, in situ hybridization (ISH), is an attractive technique for the detection of M. avium subsp. paratuberculosis and other mycobacteria. Not only does ISH allow a more rapid diagnosis than culture methods, but also fewer bacilli can be detected by ISH than by current PCR methods. Furthermore, unlike PCR, ISH provides detailed, tissue-based morphological information.
One of the major obstacles to the use of ISH to identify mycobacteria is the tough, lipid-containing cell walls of these organisms, which renders cells impermeable to oligonucleotide probes (41). However, Hulten et al. (35, 36) developed an ISH methodology to identify M. avium subsp. paratuberculosis by microscopic detection of IS900. The IS900-based ISH (IS900 ISH) procedure uses a digoxigenin- or biotin-labeled double-stranded DNA probe and enzymatic amplification of the signal with alkaline phosphatase (AP)-conjugated antibodies. However, the IS900 ISH technique can detect only cell wall-deficient forms of M. avium subsp. paratuberculosis, which are hypothesized to play a role in colonization or persistence in CD (35, 64). By this method, both Hulten et al. (34) and Sechi et al. (58, 59) reported on the detection of M. avium subsp. paratuberculosis in tissues obtained from patients with CD. However, permeabilization issues due to the large size of the double-stranded DNA probe (241 bp) and the low copy number of IS900 targets (1 to 14 copies per M. avium subsp. paratuberculosis genome) render this technique and the results questionable. It seems unlikely that a few hybridization targets per cell, even including potential IS900 transcripts, would be able to sequester sufficient amounts of probe to visualize a single bacterial cell.
An alternative hybridization approach to IS900 ISH is the use of oligonucleotide probes that target rRNA (1, 27). rRNA is typically present in cells at high copy numbers (thousands of copies per cell) and thus provides a far greater number of potential targets than the IS900 genes. The abundance of rRNA in cells allows the ready visualization of individual organisms when fluorescent probes bind to their rRNA targets. Phylogenetic comparisons of rRNA gene sequences can be used to design oligonucleotide probes that can distinguish between very closely related organisms, such as species of mycobacteria (42). Several commercial products are approved for the molecular biology-based detection of mycobacteria, and most use 16S rRNA sequences as probes (e.g., AccuProbe rRNA hybridization [Gen-Probe, Inc., San Diego, Calif.]) or amplification primers (e.g., Amplified Mycobacterium Tuberculosis Direct Test [Gen-Probe, Inc.]) (19, 61, 65, 66). All of these commercial assays either require cultivation of the organism or are approved only for the assay of sputum samples. There is a clear need for rapid diagnostic methodologies that are applicable to many different tissue types, in addition to sputum. In this paper we report on the development of several different MAC rRNA-specific probes and techniques for the reproducible and efficient permeabilization of mycobacteria in tissue and culture. Using bovine Johne's disease as a model system, we compare our methods to previously published IS900-based protocols. Finally, we demonstrate the general applicability of rRNA-specific probes to the diagnosis of a variety of proven and suspected human cases of MAC disease.
MATERIALS AND METHODS
Culture preparations. Mycobacterial cultures were acquired from Leonid Heifets at The National Jewish Medical and Research Center, Denver, Colo. (M. avium subsp. paratuberculosis, M. intracellulare, M. kansasii, and M. tuberculosis) and Mark Hernandez at the University of Colorado, Boulder (M. avium subsp. avium, M. phlei, M. parafortuitum, Bacillus subtilis, Escherichia coli, Rhodococcus sp., and Corynebacterium sp.). All cultures were fixed in 10% formalin for 4 to 10 h, washed with phosphate-buffered saline (PBS; pH 8.0) (57), and stored in 70% ethanol-30% PBS at –20°C. Fixed cultures were applied to slides prepared with silane (Sigma-Aldrich). The slides were dipped in 100% xylene for 5 min and rehydrated in an ethanol series (100, 70, 30, and 0% ethanol in 10 mM Tris [pH 7.5]). The cultures were then treated with lysozyme (1 mg/ml; Sigma-Aldrich) and achromopeptidase (30 U/ml; Sigma-Aldrich) in 10 mM Tris (pH 7.5) at 37°C for 25 min. The cultures were washed for 5 min in 10 mM Tris (pH 7.5) and then dehydrated in an ethanol series (30, 70, and 100% ethanol in 10 mM Tris [pH 7.5]). The slides were allowed to air dry before hybridization.
Tissue preparation. Mesenteric lymph nodes and ileal tissues from diseased and healthy dairy cows were obtained from Randall Basaraba at Colorado State University, Fort Collins. Mediastinal lymph node sections from M. tuberculosis-infected guinea pigs were also obtained from Randall Basaraba. The presence of M. avium subsp. paratuberculosis in tissues from cows with Johne's disease was confirmed by Ziehl-Neelsen staining (49), IS900-based quantitative PCR (5), and M. avium subsp. paratuberculosis rRNA-specific quantitative PCR (15). M. avium subsp. paratuberculosis was not detected in negative control samples from healthy cattle by either acid-fast staining and microscopy or PCR. The National Jewish Medical and Research Center provided archived small-bowel sections from a pediatric CD patient and lung tissues from adult patients infected with M. avium and/or M. intracellulare. The Swedish Medical Center (Denver, Colo.) provided an archived tissue biopsy specimen from the hand of a patient with tenosynovitis secondary to M. avium infection. The Denver Health Medical Center (Denver, Colo.) provided acid-fast-positive sputum specimens from an elderly smoker with a pulmonary mycobacterial infection. All tissue samples were deidentified. This research was conducted under the University of Colorado at Boulder Human Research Committee exemption of protocol 0799.14 (Molecular analysis of microbes in human and animal diseased tissues) for analysis of deidentified human and animal tissues.
The tissue specimens were fixed in 10% formalin for 24 h, washed with PBS, and stored in 70% ethanol-30% PBS at –20°C. Tissue was embedded in paraffin wax, and 4-μm sections were placed on slides prepared with silane (Sigma-Aldrich). The tissue specimens were dewaxed in 100% xylene for 20 min and then rehydrated in an ethanol series (100, 70, 30, and 0% ethanol in 10 mM Tris [pH 7.5]). The sections were then treated with lysozyme (1 mg/ml; Sigma-Aldrich) and achromopeptidase (30 U/ml; Sigma-Aldrich) in 10 mM Tris (pH 7.5) at 37°C for 25 min. The tissue specimens were washed for 5 min in 10 mM Tris (pH 7.5) and dehydrated in an ethanol series (30, 70, and 100% ethanol in 10 mM Tris [pH 7.5]). All sections were allowed to air dry before hybridization.
16S and 23S rRNA-specific oligonucleotide probe hybridization. The probes used in this study were EUB338 (5'-CTG CTG CCT CCC GTA GGA GT-3') (1, 17, 27), MAVP187ssu (5'-TGC GTC TTG AGG TCC TAT CC-3'), MAVP515lsu (5'-TGT CCA TGC ATG CGG TTT-3'), MAC2543lsu (5'-ACG CCA CTA CAC CCC AAA-3'), MIN351ssu (5'-AGG TAG AGC TGA GAT GTA TCC T-3'), and MIN1586lsu (5'-CCC CGA AAC TCC ATG CCC-3'). All oligonucleotides were obtained from Integrated DNA Technologies, Inc., and were labeled at the 5' and 3' ends with either 6'-carboxyfluorescein (FAM) or Cy3. All probes described above except EUB338 were designed by using the probe design function of ARB software (43) and small- and large-subunit rRNA gene sequences obtained from the GenBank database. The hybridization buffer consisted of 900 mM NaCl, 20 mM Tris (pH 8.0), 0.01% sodium dodecyl sulfate, and 20% formamide. Probes were added to the hybridization buffer at a final concentration of 1 ng/μl, and the probe-hybridization buffer was applied to the specimens. Frame-Seal incubation chambers (MJ Research, Inc.) for slides were used to prevent evaporation. The hybridization slides were placed in an MJ Research, Inc., slide thermocycler and were heated at 94°C for 5 min. The slides were incubated at either 36°C (probes MIN351lsu, MIN1586lsu, and MAC2543lsu) or 40°C (probes MAVP187ssu, MAVP515lsu, and EUB338) for 6 to 18 h. Following hybridization, the slides were washed with 225 mM NaCl-5 mM EDTA-0.01% sodium dodecyl sulfate-20 mM Tris (pH 8.0) for 20 min at either 37 or 41°C. The slides were then dipped briefly in cold 20 mM Tris (pH 8.0) in order to remove excess salt and were finally mounted with Citifluor (Electron Microscopy Sciences) antifading reagent. The slides were examined under epifluorescence with a Nikon Eclipse E600 microscope or a Leica confocal laser scanning microscope.
Antifluorescein-AP visualization. Hybridization of the FAM-labeled oligonucleotide probes was carried out as described above. Hybridization was followed by incubation of the slides in blocking buffer (3% bovine serum albumin, 100 mM Tris [pH 7.5], 150 mM NaCl, 0.3% Triton X-100) at 4°C for 30 min. Fresh blocking buffer that contained 2 mU of antifluorescein-labeled AP (antifluorescein-AP) Fab fragments (Roche) per μl was then added to the sections. The slides were incubated at 4°C for 2 to 3 h. The slides were then washed for 15 min with buffer 1 (100 mM Tris [pH 7.5], 150 mM NaCl, 0.3% Triton X-100), followed by another wash with buffer 2 (100 mM Tris [pH 9.5], 150 mM NaCl, 50 mM MgCl2) for 15 min. The slides were placed in AP buffer (100 mM Tris [pH 9.5], 150 mM NaCl, 50 mM MgCl2, 0.2 mg of levamisol [Sigma-Aldrich] per ml, 0.1% Tween) for 10 min. 5-Bromo-4-chloro-3-indolylphosphate (BCIP)-2-(4-iodophenyl)-3-(4-nitrophenyl)-5-phenyltetrazolium chloride (INT) (Roche) was then added to fresh AP buffer (75 μl of INT-BCIP per 10 ml of AP buffer) and applied to the slides, and the slides were then incubated in the dark at 25°C for 1 to 2 h. The reaction was stopped by a brief wash in water. The sections were counterstained with methylene blue, mounted with Faramount (DakoCytomation), and observed under a bright-field microscope.
IS900-biotin and IS900-fluorescein hybridization. The IS900-biotin- and IS900-fluorescein-labeled probes were prepared as described by Sechi et al. (58), with the exception that fluorescein-12-dUTP (Roche) was incorporated into the IS900-fluorescein probe. Hybridization was performed as described by Hulten et al. (36), with the exception that biotin rather than digoxigenin was used to label the 241-bp double-stranded DNA probe. The tissue samples were deparaffinized in 100% xylene, rehydrated through a graded ethanol series (100, 70, and 30% ethanol in PBS), and washed with PBS (pH 7.4). The tissue specimens were then incubated with proteinase K (Sigma-Aldrich) in PBS at concentrations ranging from 0 to 1 mg/ml for 20 min at 37°C. Proteinase K was inactivated with 0.2% glycine in PBS. The sections were then dehydrated through a graded ethanol series and allowed to air dry. The sections were hybridized with 1 ng of labeled probe per μl in hybridization buffer (50% deionized formamide, 2x SSC [1x SSC is 0.15 M NaCl plus 0.015 M sodium citrate] [57], 10% dextran sulfate, 0.25 μg of yeast tRNA [Sigma-Aldrich] per μl, 0.5 μg of denatured salmon sperm DNA [Sigma-Aldrich] per μl, 1x Denhardt's solution [Sigma-Aldrich]). The probe was boiled for 10 min and cooled on ice for 10 min before application. The sections with the probe were heated for 10 min at 95°C and chilled on ice for 10 min. Hybridization was performed overnight at 37°C. The washing steps included washes with 2x SSC for 15 min at room temperature (RT), 1x SSC for 15 min at RT, 0.3x SSC for 15 min at 40°C, and 0.3x SSC for another 15 min at RT. The washes were followed by incubation of the slides in blocking buffer at room temperature for 30 min. Fresh blocking buffer that contained 2 mU of antibiotin-labeled AP (antibiotin-AP) Fab fragments (Roche) per μl was added to the slides, and the slides were incubated at room temperature for 2 to 3 h. The slides were then washed with buffer 1 for 15 min, followed by another wash with buffer 2 for 15 min. The slides were then placed in AP buffer for 10 min. INT-BCIP (75 μl per 10 ml of AP buffer) was added to fresh AP buffer, and the mixture was applied to the slides, which were incubated in the dark at RT for 4 to 12 h. The reaction was stopped by washing the slides in distilled water. The slides were mounted with Faramount (DakoCytomation) and observed under a bright-field microscope. IS900-fluorescein hybridizations were performed as described above for IS900-biotin, except that epifluorescence microscopy was used for direct visualization of the specimens.
Dot blot hybridization was used to confirm that the IS900-biotin and IS900-fluorescein probes were synthesized correctly. Briefly, M. avium subsp. paratuberculosis genomic DNA, E. coli genomic DNA, the IS900 PCR product, and the E. coli 16S rRNA PCR product were blotted onto nitrocellulose (Hybond+; Amersham Biosciences), according to the procedures of the manufacturer. The dot blot was hybridized with either probe by the procedures described above. The blots were visualized with either antibiotin-AP Fab fragments (Roche) or antifluorescein-AP Fab fragments (Roche). The AP and INT-BCIP colorimetric reaction mixture was incubated for 20 min at RT. The probes hybridized only to M. avium subsp. paratuberculosis genomic DNA and the IS900 PCR product (data not shown).
RESULTS
Development of ISH protocols and characterization of MAC rRNA ISH probes. A variety of potential MAC-specific ISH oligonucleotide probe sequences were chosen by using the Probe Design function of the ARB software package (43) and an alignment of small- and large-subunit rRNA sequences obtained from GenBank. M. avium subsp. paratuberculosis and M. avium subsp. avium rRNA sequences are identical, so it was impossible to design probes that would differentiate between the two subspecies. As a first test of specificity, fluorescently labeled probes were applied to pure cultures of cognate and noncognate species, including M. avium subsp. paratuberculosis, M. avium subsp. avium, M. intracellulare, M. tuberculosis, M. kansasii, M. parafortuitum, M. phlei, E. coli, B. subtilis, a Rhodococcus sp., and a Corynebacterium sp. A broad-range bacterial 16S probe, EUB338-FAM, was included in every hybridization experiment in order to test the effectiveness of the permeabilization and the hybridization protocols.
In initial experiments, none of the ISH probes, including EUB338, successfully hybridized to any of the mycobacterial cells. The complex, lipid-rich cell walls of mycobacteria are notoriously refractory to the lysis steps of standard ISH protocols (17). Consequently, we tested a number of techniques to develop a protocol that would enhance the permeabilization of mycobacterial cells. Enzymatic digestion of fixed cells with proteinase K (58, 59) failed to improve the fluorescent signal upon hybridization with EUB338 or the MAC-specific probes. Enzymatic digestion with lysozyme and achromopeptidase (60) resulted in a fluorescent signal for only a small fraction of the fixed cells from culture. We found, however, that a brief incubation of the fixed mycobacterial cells in 100% xylene, combined with subsequent lysozyme and achromopeptidase digestions, resulted in strong fluorescent hybridization signals. Treatment with all three reagents (xylene, lysozyme, and achromopeptidase) was absolutely required for successful ISH of mycobacterial cells. Although each oligonucleotide probe was tested individually (Table 1), we found that the combination of the species-specific probes for ISH resulted in a stronger fluorescent signal, even though the final probe concentration remained the same (data not shown).
By this enhanced ISH protocol, Cy3-conjugated probes MAVP187ssu and MAVP515lsu (Table 1) hybridized well to both M. avium subsp. avium and M. avium subsp. paratuberculosis (Fig. 1A and B, respectively) but not to M. intracellulare (Fig. 1D), M. tuberculosis (Fig. 1F), or other related and unrelated bacterial species (listed above; data not shown). Hybridization with probe EUB338-FAM clearly labeled the M. intracellulare and M. tuberculosis cells (Fig. 1C and E, respectively), indicating that cell wall permeabilization was effective. As expected, probe MAC2543lsu hybridized to all three members of the MAC complex, M. intracellulare, M. avium subsp. avium, and M. avium subsp. paratuberculosis, but failed to hybridize to the other species tested (Fig. 1G to L and Table 1). Finally, probes MIN351lsu and MIN1586lsu specifically hybridized to M. intracellulare cells but not to the cells of the other species tested, including other members of MAC, M. avium subsp. avium and M. avium subsp. paratuberculosis (Fig. 1M to R and Table 1).
Application of MAC-specific ISH probes to tissue samples. Tissues from dairy cows with well-characterized Johne's disease and healthy animals were used in order to establish the ability of the MAC-specific probes to stain the MAC cells embedded within tissues. The presence or absence of M. avium subsp. paratuberculosis in infected and uninfected mesenteric lymph nodes (MLNs) was first determined by Ziehl-Neelsen staining and quantitative PCR (data not shown) (5, 15). Both negative and positive control MLN samples were then probed with a combination of Cy3-labeled MAVP187ssu and MAVP515lsu fluorescent oligonucleotides. The negative control MLNs showed no sign of fluorescent bacilli when they were probed with the MAVP187ssu and MAVP515lsu oligonucleotides (Fig. 2A). In contrast, hybridization of the positive control MLNs resulted in an intense and abundant fluorescent signal (Fig. 2B) in an aggregate pattern similar to the acid-fast staining pattern (see Fig. 4B). The MIN351ssu and MIN1586lsu probes did not hybridize to either the bacillus-positive or the bacillus-negative MLN specimens (data not shown). The MAC2543lsu probe labeled the M. avium subsp. paratuberculosis-positive MLN specimen but not the M. avium subsp. paratuberculosis-negative MLN specimen (data not shown). Because the ISH protocol requires extensive digestion of the mycobacterial cell wall for probe accessibility, we were unable to find conditions that would allow dual staining with Ziehl-Neelsen and rRNA-specific probes.
Figure 2C demonstrates the ability of confocal laser scanning microscopy to detect clearly defined bacilli that hybridized with the MAVP187ssu-Cy3 and MAVP515lsu-Cy3 probes in M. avium subsp. paratuberculosis-infected bovine MLN tissue. The nuclei of lymphatic cells are stained with 4',6'-diamidino-2-phenylindole (DAPI) and appear blue. The confocal laser scanning microscope was able to resolve bacilli in different focal planes within the tissue, indicating that the probe had penetrated the entire tissue section. As expected for M. avium subsp. paratuberculosis, the labeled cells are rods approximately 1 μm in length. In general, tissue autofluorescence did not greatly affect the labeling or detection of M. avium subsp. paratuberculosis cells in most bovine or human tissues. However, high levels of autofluorescence were observed when erythrocytes, collagen, and/or elastic fibers were abundant, such as in lung tissue samples. We consequently developed a nonfluorescent approach to probe detection that uses AP conjugated to antifluorescein antibodies to enzymatically amplify a colorimetric dye (INT-BCIP; see Materials and Methods). In addition to avoiding background autofluorescence, this approach also allows the use of tissue staining and light microscopy to visualize hybridized microbes and colocalize them with the areas of mammalian tissue pathology. A representative micrograph of an ileum tissue specimen from a cow with Johne's disease probed with the MAVP oligonucleotides is shown in Fig. 2D. Arrows indicate the reddish brown M. avium subsp. paratuberculosis bacilli.
MAC-specific probes for ISH detect MAC species in human clinical specimens. The MAC-specific probes used for ISH were applied to a variety of archival tissue specimens from patients with MAC infections in order to determine the applicabilities of these probes for the identification and diagnosis of mycobacterial infections. As shown in Fig. 3A to D, tissues were probed with the M. avium subsp. avium- and M. avium subsp. paratuberculosis-specific oligonucleotides MAVP187ssu and MAVP515lsu and visualized by either fluorescence microscopy (Fig. 3B and C) or light microscopy following treatment with antifluorescein-coupled AP (Fig. 3A and D). Figure 3A shows a tissue section from a lung resection from a patient infected with M. avium subsp. avium, as determined by culture and with the AccuProbe (GenProbe). The MAVP probes were able to detect many bacilli (Fig. 3A, arrows) throughout the tissue. The M. intracellulare-specific probes did not hybridize to the same resected lung tissue (data not shown). Tissue from a patient with tenosynovitis of the hand was negative by acid-fast staining but M. avium subsp. avium positive by culture after several weeks of growth; the AccuProbe (GenProbe) assay confirmed the presence of M. avium subsp. avium. The MAVP probes were clearly able to detect isolated bacilli (Fig. 3B) of the appropriate size and morphology. Figure 3C and D shows tissue sections from the duodenum and small bowel, respectively, of a pediatric CD patient. In these specimens, the MAVP probes hybridized to bacilli located predominantly within cells of the lamina propria.
Figure 3E shows a representative micrograph of lung tissue containing large granulomas that was probed with the M. intracellulare-specific oligonucleotides MIN351lsu and MIN1586lsu. The hybridization signal was visualized by the antifluorescein-AP colorimetric assay. The lung tissue was acid-fast positive, with one to two bacilli located in the granulomas in each tissue section; M. intracellulare was cultured from a section of this tissue, and its presence was confirmed with the AccuProbe (GenProbe). The M. intracellulare-specific FAM-labeled probes were able to detect more M. intracellulare bacilli (Fig. 3E, arrows) in the granulomas than were detected by acid-fast staining. In contrast, the MAVP probes did not hybridize to this lung tissue section (data not shown). The ISH methodology also worked well with sputum samples. Figure 3F shows a micrograph of an acid-fast positive sputum sample from a patient with a history of smoking, chronic cough, and an upper-zone cavity with a diagnosis of pulmonary MAC infection 2 years earlier. AccuProbe identification was inconclusive, and broad-range 16S rRNA sequence data detected a Corynebacterium sp. and small numbers of M. intracellulare cells (data not shown). ISH with the MAC-specific probe MAC2543-FAM revealed sparse, labeled bacilli among a background of DAPI-labeled bacteria of corynebacterial morphology. In addition, none of the MAC-specific probes hybridized to mediastinal lymph node tissue specimens from guinea pigs infected with M. tuberculosis (data not shown).
Comparison of rRNA ISH and IS900 ISH methodologies. Some recently published studies have used double-stranded DNA probes that target genomic DNA for the localization of mycobacteria in situ (34, 58). Because ribosomes, the targets of the ISH probes described in this study, are present in much greater cellular abundance than genomic DNA loci, a direct comparison of the DNA- and rRNA-based approaches was in order. To this end, MLN tissues from healthy cows and cows with Johne's disease were used as negative and positive controls, respectively, for ISH. Ziehl-Neelsen staining determined the absence or presence of acid-fast bacilli in each tissue sample (Fig. 4A and B, respectively). M. avium subsp. paratuberculosis infection of the positive control tissue was also demonstrated by IS900 and 16S rRNA PCRs (data not shown).
As expected, the MAC rRNA-specific probe MAVP187ssu-FAM stained only the positive control tissue, as indicated by a reddish brown precipitate (the INT-BCIP-stained AP product) with the morphology of clearly defined bacilli and a size and morphology consistent with M. avium subsp. paratuberculosis (Fig. 4D). No reddish brown signal was evident in the negative control tissue (Fig. 4C). In contrast, application of IS900 probes to bovine tissue sections resulted in the appearance of a variety of rod-like and amorphous forms in both the negative (Fig. 4E) and the positive (Fig. 4F) control MLN tissues. Under high-power oil-immersion magnification, however, these objects had a granular, amorphous appearance not necessarily suggestive of mycobacteria, whereas the cells stained with either Ziehl-Neelsen stain or rRNA-specific probes (Fig. 4B and D, respectively) had clear morphological boundaries, structures that resemble microorganisms.
To determine whether the false-positive results that were obtained with the IS900 probe were due to nonspecific precipitation or aggregation of the colorimetric dye, we synthesized a fluorescein-labeled IS900 probe (IS900-fluorescein) that allowed direct epifluorescent detection of cells. ISH of positive and negative control bovine MLN tissues with this IS900-fluorescein probe also resulted in significant nonspecific staining in both control tissue specimens (Fig. 4G and H). Indeed, application of the published IS900 ISH protocol (35) to organisms other than M. avium subsp. paratuberculosis that were grown in pure culture resulted in strong nonspecific signals (Fig. 5A). For example, Fig. 5A shows B. subtilis cells that were dually probed by the IS900 ISH procedure with IS900-fluorescein isothiocyanate (FITC) (green) and EUB338-Cy3 (red). Although the rRNA probe is uniformly distributed within the cells, consistent with the expected distribution of ribosomes, the IS900-FITC probe results in a punctate staining pattern that is localized to the outside of the cells. Similarly, Fig. 5B and C shows false-positive hybridization of the IS900-biotin probe to a Rhodococcus sp. and M. kansasii. Of the bacteria tested, only gram-negative microbes, such as E. coli and Pseudomonas aeruginosa, were resistant to the nonspecific hybridization with the IS900-biotin probe (data not shown). Figure 5D shows an example of an M. avium subsp. avium isolate probed with EUB338-FAM and visualized enzymatically with antifluorescein-AP Fab fragments and INT-BCIP. The reddish brown signal is distributed evenly throughout the cell, which had the expected morphology. Thus, both our permeabilization protocol and the use of small rRNA-based oligonucleotide probes rather than large DNA-based probes, such as IS900, are necessary for successful ISH of mycobacteria.
DISCUSSION
The members of MAC are increasingly recognized as important environmental pathogens and are implicated in several chronic, idiopathic diseases (16, 20, 24). CD, for example, is hypothesized to be a zoonotic infection that arises from exposure to dairy products contaminated with M. avium subsp. paratuberculosis, the causative agent of bovine Johne's disease (29). The public health implications of a link between human Crohn's disease and bovine Johne's disease demand careful consideration, given the increasing incidence of Johne's disease in cattle herds (6) and the presence of viable M. avium subsp. paratuberculosis isolates in commercial milk supplies (9, 46). The development of rapid and accurate assays for the detection of MAC organisms in tissue is urgently needed. In this paper we report on a rapid, highly reliable technique for the permeabilization and identification of MAC members in tissues and cultures by ISH. We describe rRNA-based oligonucleotide probes that specifically detect M. intracellulare (probes MIN351lsu and MIN1586lsu), the two M. avium subspecies (M. avium subsp. avium and M. avium subsp. paratuberculosis; probes MAVP187ssu and MAVP515lsu), and all members of MAC (probe MAC2543lsu). To assess the specificities and sensitivities of the MAC probes, we applied these probes to three types of samples: (i) axenic bacterial cultures (M. kansasii and M. tuberculosis, pathogens closely related to MAC), (ii) tissues from cows with Johne's disease infected with M. avium subsp. paratuberculosis, and (iii) human clinical samples with a variety of confirmed mycobacterial infections. By our methodology the ISH probes specifically labeled the mycobacterial species against which they were designed in both tissues and culture samples.
Because of their waxy, mycolic acid-laden cell walls, mycobacteria are normally recalcitrant to the typical permeabilization steps used for ISH with bacteria. When ISH was performed with mycobacteria, the results were reproducible only when the cells were permeabilized with a combination of xylene, lysozyme, and achromopeptidase treatments. Omission of any one of these steps either diminished or abolished the signals from the probes. Our permeabilization protocol also works well with other gram-positive bacteria, such as Corynebacterium sp., Rhodococcus sp., and B. subtilis; but in these instances, the xylene step is not critical.
Proteinase K is often used to permeabilize mycobacteria in culture and tissue in preparation for ISH (34, 58). In our experience, however, the use of proteinase K for mycobacterial ISH resulted either in inadequate permeabilization (i.e., the lack of a signal) or in significant tissue damage and nonspecific binding of the probes to tissue (i.e., excessive background signal). Overdigestion with proteinase K is therefore likely to increase the occurrence of false-positive results, as were observed by the IS900 ISH (discussed below). In contrast, lysozyme and achromopeptidase digest only bacterial cell walls and therefore do not affect the surrounding mammalian tissue.
The MAC-specific rRNA probes were visualized by two methods: (i) direct fluorescence and (ii) enzymatic amplification of a colorimetric substrate. One important benefit of the fluorescent method is its applicability to confocal laser scanning microscopy, which allows the high-resolution imaging of microbes in tissues. MAC cells were observed to be embedded in tissue sections of both bovine and human tissues, often within the cytoplasms of eukaryotic cells, thus eliminating the possibility of cross-contamination of specimens during the preparation of tissue sections. In our experience, it was absolutely essential to use high-resolution microscopy, either under a x1,000 oil-immersion or by confocal laser scanning microscopy, in order to confirm by the morphology the staining of microbial cells. In addition to direct visualization by fluorescence, the ISH probes also were indirectly visualized by bright-field microscopy following the reaction of a colorimetric dye (INT-BCIP) with antibody-conjugated AP. This method is not influenced by tissue autofluorescence, which can obscure the true signals of fluorescently labeled probes. Although autofluorescence typically is not a problem, tissues that are densely packed with red blood cells and/or mast cells can be prone to high autofluorescent backgrounds. Because this indirect method of visualization can be performed with tissue sections following direct fluorescent detection, the two approaches can be used to corroborate the results and thereby diminish the possibility of false-positive results.
The 16S and 23S rRNA sequences of M. avium subsp. paratuberculosis and M. avium subsp. avium are identical, and so rRNA-specific probes cannot distinguish between the two organisms. PCR amplification of the IS900 gene, which is present in the genome of M. avium subsp. paratuberculosis but not that of M. avium subsp. avium, is an effective means of typing the two subspecies. Although it is theoretically possible, the detection of M. avium subsp. paratuberculosis by IS900-based ISH presents several technical challenges. First, as was demonstrated with the rRNA-specific oligonucleotide probes, proper permeabilization is absolutely essential for gaining access to hybridization targets within a cell. Large hybridization probes, such as those produced by PCR amplification or nick translation of the IS900 gene, should require even more careful control of permeabilization conditions than is necessary for smaller probes. Second, M. avium subsp. paratuberculosis genomes are reported to carry only 1 to 14 copies of the IS900 gene, which necessitates an extremely sensitive assay for the detection of so few targets. Bacterial mRNAs are very difficult to detect by ISH, so IS900 mRNA is unlikely to be detectable in these assays. Nevertheless, IS900-based ISH has been suggested as a means of detecting M. avium subsp. paratuberculosis in clinical tissue specimens. Using IS900 ISH, both Sechi et al. (58) and Hulten et al. (34) have reported on the presence of M. avium subsp. paratuberculosis in specimens from patients with CD. In our experience, however, ISH with IS900-specific probes resulted in false-positive signals, especially if the colorimetric dye (INT-BCIP) was allowed to react with AP for prolonged periods (from 2 h to overnight). When the reaction times were limited to less than 2 h, no signal was seen for positive control tissues (data not shown).
It is crucial that the bacterial cell morphology be used as a benchmark for the validation of a positive signal with any kind of ISH assay because of the possibility of nonspecific signals (Fig. 4 and 5) (54). Often, this morphology can be visualized only by examination at a magnification of x1,000 under oil immersion because single, 1-μm rods (the typical size of M. avium subsp. paratuberculosis cells) are extremely difficult to visualize at x400 magnification, the magnification typically used in clinical analyses. When cells were visualized under oil immersion, the structures labeled with IS900-specific probes and INT-BCIP had large indistinct boundaries, whereas the cells labeled with rRNA-specific probes and INT-BCIP had clearly defined morphologies.
The false-positive results obtained with the IS900-specific probes were not due solely to the colorimetric labeling method. Under high magnification, fluorescently labeled IS900-specific probes were observed to bind to the outer cell walls of a variety of bacterial species. In contrast, rRNA-specific probes were observed to label the interiors of cells uniformly. Thus, the false-positive results obtained with the IS900-specific probes most likely arose from nonspecific binding of the DNA-specific probes to cells, coupled with nonspecific precipitation and aggregation of dye. Overall, the lack of specificity and apparently artifactual nature of the IS900-specific probes make their use questionable for the assay of mammalian tissues, such as tissue specimens from patients with CD, for the presence or absence of M. avium subsp. paratuberculosis bacteria.
Furthermore, we demonstrate the ability of the MAC rRNA-specific probes to detect and identify MAC organisms in a variety of animal and human tissue and clinical specimens, including sputum. Our probes offer researchers and clinicians an incisive method for identifying members of MAC in tissue specimens and culture. Future work with specimens from diseased and normal tissues will further define the sensitivities and specificities of the MAC-specific probes. In several cases, including M. intracellulare-infected lung tissue and tissue from a patient with M. avium subsp. avium tenosynovitis, the rRNA probes detected more bacilli than were evident by standard acid-fast staining procedures. Mycobacteria could not be detected by PCR in many of these specimens, possibly due to the low abundance of organisms or the presence of PCR inhibitors in the tissue specimen. On the other hand, ISH can detect single cells and, thus, may possibly provide greater sensitivity than traditional PCR. ISH probes hold the promise for the more rapid detection and differentiation of mycobacterial species than can be achieved by traditional clinical microbiological techniques.
ACKNOWLEDGMENTS
We thank Leonid Heifets, Sarah Wilson, Kayte Fulton, Randall Reves, Ginger Hildred, Michael Moynihan, Burton Golub, Randall Basaraba, and Mark Hernandez for providing strains, clinical specimens, and valuable feedback.
This work was supported by a grant from the National Institutes of Health (to N.R.P.).
REFERENCES
Amann, R. I., L. Krumholz, and D. A. Stahl. 1990. Fluorescent-oligonucleotide probing of whole cells for determinative, phylogenetic, and environmental studies in microbiology. J. Bacteriol. 172:762-770.
Behr, M. A., M. Semret, A. Poon, and E. Schurr. 2004. Crohn's disease, mycobacteria, and NOD2. Lancet Infect. Dis. 4:136-137.
Benson, C. A. 1994. Disease due to the Mycobacterium avium complex in patients with AIDS: epidemiology and clinical syndrome. Clin. Infect. Dis. 18(Suppl. 3):S218-S222.
Bull, T. J., J. Hermon-Taylor, I. Pavlik, F. El-Zaatari, and M. Tizard. 2000. Characterization of IS900 loci in Mycobacterium avium subsp. paratuberculosis and development of multiplex PCR typing. Microbiology 146(Pt 9):2185-2197.
Bull, T. J., E. J. McMinn, K. Sidi-Boumedine, A. Skull, D. Durkin, P. Neild, G. Rhodes, R. Pickup, and J. Hermon-Taylor. 2003. Detection and verification of Mycobacterium avium subsp. paratuberculosis in fresh ileocolonic mucosal biopsy specimens from individuals with and without Crohn's disease. J. Clin. Microbiol. 41:2915-2923.
Chacon, O., L. E. Bermudez, and R. G. Barletta. 2004. Johne's disease, inflammatory bowel disease, and Mycobacterium paratuberculosis. Annu. Rev. Microbiol. 58:329-363.
Chao, S. S., K. S. Loh, K. K. Tan, and S. M. Chong. 2002. Tuberculous and nontuberculous cervical lymphadenitis: a clinical review. Otolaryngol. Head Neck Surg. 126:176-179.
Chiodini, R. J. 1989. Crohn's disease and the mycobacterioses: a review and comparison of two disease entities. Clin. Microbiol. Rev. 2:90-117.
Chiodini, R. J., and J. Hermon-Taylor. 1993. The thermal resistance of Mycobacterium paratuberculosis in raw milk under conditions simulating pasteurization. J. Vet. Diagn. Investig. 5:629-631.
Chiodini, R. J., H. J. Van Kruiningen, W. R. Thayer, and J. A. Coutu. 1986. Spheroplastic phase of mycobacteria isolated from patients with Crohn's disease. J. Clin. Microbiol. 24:357-363.
Collins, M. T., G. Lisby, C. Moser, D. Chicks, S. Christensen, M. Reichelderfer, N. Hoiby, B. A. Harms, O. O. Thomsen, U. Skibsted, and V. Binder. 2000. Results of multiple diagnostic tests for Mycobacterium avium subsp. paratuberculosis in patients with inflammatory bowel disease and in controls. J. Clin. Microbiol. 38:4373-4381.
Cousins, D. V., R. Whittington, I. Marsh, A. Masters, R. J. Evans, and P. Kluver. 1999. Mycobacteria distinct from Mycobacterium avium subsp. paratuberculosis isolated from the faeces of ruminants possess IS900-like sequences detectable IS900 polymerase chain reaction: implications for diagnosis. Mol. Cell. Probes 13:431-442.
De Groote, M. A. 2004. Pulmonary infection in non-HIV infected individuals. In J. Bartram and A. Dufour (ed.), M. avium in water. U.S. Environmental Protection Agency/World Health Organization, London, United Kingdom.
De Groote, M. A. 2004. Skin, bone, and soft tissue infections. In J. Bartram and A. Dufour (ed.), M. avium in water. U.S. Environmental Protection Agency/World Health Organization, London, United Kingdom.
De Groote, M. A., D. N. Frank, E. Nachtway, A. L. St. Amand, H. Hearst, F. Garry, R. Basaraba, and N. R. Pace. Unpublished data.
De Groote, M. A., and M. D. Iseman. 2003. Environmental mycobacterial infections. In J. G. Bartlett and N. R. Blacklow (ed.), Gorbach textbook of infectious diseases. Lipincott Williams and Wilkins, Philadelphia, Pa.
DeLong, E. F., G. S. Wickham, and N. R. Pace. 1989. Phylogenetic stains: ribosomal RNA-based probes for the identification of single cells. Science 243:1360-1363.
Drake, W. P., Z. Pei, D. T. Pride, R. D. Collins, T. L. Cover, and M. J. Blaser. 2002. Molecular analysis of sarcoidosis tissues for Mycobacterium species DNA. Emerg. Infect. Dis. 8:1334-1341.
Drobniewski, F. A., P. G. More, and G. S. Harris. 2000. Differentiation of Mycobacterium tuberculosis complex and nontuberculous mycobacterial liquid cultures by using peptide nucleic acid-fluorescence in situ hybridization probes. J. Clin. Microbiol. 38:444-447.
Eccles, E., and J. Ptak. 1995. Mycobacterium avium complex infection in AIDS: clinical features, treatment, and prevention. J. Assoc. Nurses AIDS Care 6:37-47.
El-Zaatari, F. A., S. A. Naser, D. C. Markesich, D. C. Kalter, L. Engstand, and D. Y. Graham. 1996. Identification of Mycobacterium avium complex in sarcoidosis. J. Clin. Microbiol. 34:2240-2245.
El-Zaatari, F. A., M. S. Osato, and D. Y. Graham. 2001. Etiology of Crohn's disease: the role of Mycobacterium avium paratuberculosis. Trends Mol. Med. 7:247-252.
Englund, S. 2003. IS900/ERIC-PCR as a tool to distinguish Mycobacterium avium subsp. paratuberculosis from closely related mycobacteria. Vet. Microbiol. 96:277-287.
Falkinham, J. O., III. 1996. Epidemiology of infection by nontuberculous mycobacteria. Clin. Microbiol. Rev. 9:177-215.
Falkinham, J. O., III. 2003. Mycobacterial aerosols and respiratory disease. Emerg. Infect. Dis. 9:763-767.
Fitzpatrick, E. L., and F. E. LeJeune, Jr. 1996. Mycobacterial cervical lymphadenitis: a review. J. La. State Med. Soc. 148:451-454.
Giovannoni, S. J., E. F. DeLong, G. J. Olsen, and N. R. Pace. 1988. Phylogenetic group-specific oligodeoxynucleotide probes for identification of single microbial cells. J. Bacteriol. 170:720-726.
Greenstein, R. J. 2003. Is Crohn's disease caused by a mycobacterium Comparisons with leprosy, tuberculosis, and Johne's disease. Lancet Infect. Dis. 3:507-514.
Greenstein, R. J., and M. T. Collins. 2004. Emerging pathogens: is Mycobacterium avium subspecies paratuberculosis zoonotic Lancet 364:396-397.
Hance, A. J. 1998. The role of mycobacteria in the pathogenesis of sarcoidosis. Semin. Respir. Infect. 13:197-205.
Hermon-Taylor, J., N. Barnes, C. Clarke, and C. Finlayson. 1998. Mycobacterium paratuberculosis cervical lymphadenitis, followed five years later by terminal ileitis similar to Crohn's disease. BMJ 316:449-453.
Hermon-Taylor, J., and T. Bull. 2002. Crohn's disease caused by Mycobacterium avium subspecies paratuberculosis: a public health tragedy whose resolution is long overdue. J. Med. Microbiol. 51:3-6.
Horsburgh, C. R., Jr. 1999. The pathophysiology of disseminated Mycobacterium avium complex disease in AIDS. J. Infect. Dis. 179(Suppl. 3):S461-S465.
Hulten, K., H. M. El-Zimaity, T. J. Karttunen, A. Almashhrawi, M. R. Schwartz, D. Y. Graham, and F. A. El-Zaatari. 2001. Detection of Mycobacterium avium subspecies paratuberculosis in Crohn's diseased tissues by in situ hybridization. Am. J. Gastroenterol. 96:1529-1535.
Hulten, K., T. J. Karttunen, H. M. El-Zimaity, S. A. Naser, A. Almashhrawi, D. Y. Graham, and F. A. El-Zaatari. 2000. In situ hybridization method for studies of cell wall deficient M. paratuberculosis in tissue samples. Vet. Microbiol. 77:513-518.
Hulten, K., T. J. Karttunen, H. M. El-Zimaity, S. A. Naser, M. T. Collins, D. Y. Graham, and F. A. El-Zaatari. 2000. Identification of cell wall deficient forms of M. avium subsp. paratuberculosis in paraffin embedded tissues from animals with Johne's disease by in situ hybridization. J. Microbiol. Methods 42:185-195.
Iivanainen, E., J. Northrup, R. D. Arbeit, M. Ristola, M. L. Katila, and C. F. von Reyn. 1999. Isolation of mycobacteria from indoor swimming pools in Finland. APMIS 107:193-200.
Kanazawa, K., Y. Haga, O. Funakoshi, H. Nakajima, A. Munakata, and Y. Yoshida. 1999. Absence of Mycobacterium paratuberculosis DNA in intestinal tissues from Crohn's disease by nested polymerase chain reaction. J. Gastroenterol. 34:200-206.
Kilby, J. M., P. H. Gilligan, J. R. Yankaskas, W. E. Highsmith, Jr., L. J. Edwards, and M. R. Knowles. 1992. Nontuberculous mycobacteria in adult patients with cystic fibrosis. Chest 102:70-75.
Kon, O. M., and R. M. du Bois. 1997. Mycobacteria and sarcoidosis. Thorax 52(Suppl. 3):S47-S51.
Lambert, P. A. 2002. Cellular impermeability and uptake of biocides and antibiotics in gram-positive bacteria and mycobacteria. Symp. Ser. Soc. Appl. Microbiol. 2002:46S-54S.
Lane, D. J., B. Pace, G. J. Olsen, D. A. Stahl, M. L. Sogin, and N. R. Pace. 1985. Rapid determination of 16S ribosomal RNA sequences for phylogenetic analyses. Proc. Natl. Acad. Sci. USA 82:6955-6959.
Ludwig, W., O. Strunk, R. Westram, L. Richter, H. Meier, Yadhukumar, A. Buchner, T. Lai, S. Steppi, G. Jobb, W. Forster, I. Brettske, S. Gerber, A. W. Ginhart, O. Gross, S. Grumann, S. Hermann, R. Jost, A. Konig, T. Liss, R. Lussmann, M. May, B. Nonhoff, B. Reichel, R. Strehlow, A. Stamatakis, N. Stuckmann, A. Vilbig, M. Lenke, T. Ludwig, A. Bode, and K. H. Schleifer. 2004. ARB: a software environment for sequence data. Nucleic Acids Res. 32:1363-1371.
Mangiapan, G., and A. J. Hance. 1995. Mycobacteria and sarcoidosis: an overview and summary of recent molecular biological data. Sarcoidosis 12:20-37.
McFadden, J. J., and H. M. Fidler. 1996. Mycobacteria as possible causes of sarcoidosis and Crohn's disease. Soc. Appl. Bacteriol. Symp. Ser. 25:47S-52S.
Millar, D., J. Ford, J. Sanderson, S. Withey, M. Tizard, T. Doran, and J. Hermon-Taylor. 1996. IS900 PCR to detect Mycobacterium paratuberculosis in retail supplies of whole pasteurized cows' milk in England and Wales. Appl. Environ. Microbiol. 62:3446-3452.
Mirmirani, P., T. A. Maurer, B. Herndier, M. McGrath, M. D. Weinstein, and T. G. Berger. 1999. Sarcoidosis in a patient with AIDS: a manifestation of immune restoration syndrome. J. Am. Acad. Dermatol. 41:285-286.
Moss, M. T., E. P. Green, M. L. Tizard, Z. P. Malik, and J. Hermon-Taylor. 1991. Specific detection of Mycobacterium paratuberculosis by DNA hybridisation with a fragment of the insertion element IS900. Gut 32:395-398.
Murray, R. G. E., N. Doetsch, and C. F. Robinow. 1994. Determinative and cytological light microscopy, p. 32-33. In P. Gerhardt, R. G. E. Murray, W. A. Wood, and N. R. Krieg (ed.), General and molecular bacteriology. American Society for Microbiology, Washington, D.C.
Naser, S. A., I. Shafran, D. Schwartz, F. El-Zaatari, and J. Biggerstaff. 2002. In situ identification of mycobacteria in Crohn's disease patient tissue using confocal scanning laser microscopy. Mol. Cell. Probes 16:41-48.
Oliver, A., L. Maiz, R. Canton, H. Escobar, F. Baquero, and E. Gomez-Mampaso. 2001. Nontuberculous mycobacteria in patients with cystic fibrosis. Clin. Infect. Dis. 32:1298-1303.
Primm, T. P., C. A. Lucero, and J. O. Falkinham III. 2004. Health impacts of environmental mycobacteria. Clin. Microbiol. Rev. 17:98-106.
Richter, E., J. Wessling, N. Lugering, W. Domschke, and S. Rusch-Gerdes. 2002. Mycobacterium avium subsp. paratuberculosis infection in a patient with HIV, Germany. Emerg. Infect. Dis. 8:729-731.
Roholl, P. J., A. Herrewegh, and D. van Soolingen. 2002. Positive IS900 in situ hybridization signals as evidence for role of Mycobacterium avium subsp. paratuberculosis in etiology of Crohn's disease. J. Clin. Microbiol. 40:3112. (Author reply, 40:3112-3113.)
Rutherford, R. M., and J. J. Gilmartin. 2004. Mycobacteria in pathogenesis of sarcoidosis. Chest 125:354.
Ryan, P., M. W. Bennett, S. Aarons, G. Lee, J. K. Collins, G. C. O'Sullivan, J. O'Connell, and F. Shanahan. 2002. PCR detection of Mycobacterium paratuberculosis in Crohn's disease granulomas isolated by laser capture microdissection. Gut 51:665-670.
Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, vol. 3, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Sechi, L. A., M. Mura, E. Tanda, A. Lissia, A. Solinas, G. Fadda, and S. Zanetti. 2001. Identification of Mycobacterium avium subsp. paratuberculosis in biopsy specimens from patients with Crohn's disease identified by in situ hybridization. J. Clin. Microbiol. 39:4514-4517. (Erratum, 40:739, 2002.)
Sechi, L. A., M. Mura, E. Tanda, A. Lissia, G. Fadda, and S. Zanetti. 2004. Mycobacterium avium sub. paratuberculosis in tissue samples of Crohn's disease patients. New Microbiol. 27:75-77.
Sekar, R., A. Pernthaler, J. Pernthaler, F. Warnecke, T. Posch, and R. Amann. 2003. An improved protocol for quantification of freshwater actinobacteria by fluorescence in situ hybridization. Appl. Environ. Microbiol. 69:2928-2935.
Tortoli, E., M. T. Simonetti, and F. Lavinia. 1996. Evaluation of reformulated chemiluminescent DNA probe (AccuProbe) for culture identification of Mycobacterium kansasii. J. Clin. Microbiol. 34:2838-2840.
Van Kruiningen, H. J. 1999. Lack of support for a common etiology in Johne's disease of animals and Crohn's disease in humans. Inflamm. Bowel Dis. 5:183-191.
Vansnick, E., P. De Rijk, F. Vercammen, D. Geysen, L. Rigouts, and F. Portaels. 2004. Newly developed primers for the detection of Mycobacterium avium subspecies paratuberculosis. Vet. Microbiol. 100:197-204.
Wall, S., Z. M. Kunze, S. Saboor, I. Soufleri, P. Seechurn, R. Chiodini, and J. J. McFadden. 1993. Identification of spheroplast-like agents isolated from tissues of patients with Crohn's disease and control tissues by polymerase chain reaction. J. Clin. Microbiol. 31:1241-1245.
Watterson, S. A., and F. A. Drobniewski. 2000. Modern laboratory diagnosis of mycobacterial infections. J. Clin. Pathol. 53:727-732.
Woods, G. L., J. S. Bergmann, and N. Williams-Bouyer. 2001. Clinical evaluation of the GenProbe amplified Mycobacterium tuberculosis direct test for rapid detection of Mycobacterium tuberculosis in select nonrespiratory specimens. J. Clin. Microbiol. 39:747-749.(Allison L. St. Amand, Dan)