Fluorescence In Situ Hybridization Using Peptide Nucleic Acid Probes for Rapid Detection of Mycobacterium avium subsp. avium and Mycobacterium avium subsp. paratuberculosis in Potable-Water Biofilms

Abstract

Here, we present for the first time a high-affinity peptide nucleic acid (PNA) oligonucleotide sequence for detecting Mycobacterium avium bacteria, including the opportunistically pathogenic subspecies M. avium subsp. avium, M. avium subsp. paratuberculosis, and M. avium subsp. silvaticum, by the fluorescence in situ hybridization (FISH) method. There is evidence that M. avium subsp. avium especially is able to survive and grow in drinking-water biofilms and possibly transmit via drinking water. The designed PNA probe (MAV148) specificity was tested with several bacterial species, including other mycobacteria and mycolic acid-containing bacteria. From the range of bacterial strains tested, only M. avium subsp. avium and M. avium subsp. paratuberculosis strains were hybridized. The PNA FISH method was applied successfully to detect M. avium subsp. avium spiked in water samples and biofilm established within a Propella biofilm reactor fed with potable water from a distribution supply.

Mycobacterium avium subsp. avium is an environmental opportunistic human and animal pathogen (13). In immunocompetent patients, M. avium subsp. avium causes principally pulmonary and soft-tissue infections, while infections in immunosuppressed patients are disseminated (13). M. avium subsp. avium has been found in drinking-water distribution systems and can survive and grow in drinking-water biofilms (8, 14, 25, 39).

Another clinically important subspecies of M. avium is M. avium subsp. paratuberculosis. It has the same 16S rRNA sequence as M. avium subsp. avium (3, 20). M. avium subsp. paratuberculosis differs from M. avium subsp. avium by slower growth rate and a requirement for mycobactin in culture medium (3, 35). It also has multiple copies of the insertion sequence IS900 (15). The concern over the human health effects of M. avium subsp. paratuberculosis has increased lately. M. avium subsp. paratuberculosis is known to be the causative agent of Johne's disease, an inflammatory bowel disease in cattle and sheep (13). M. avium subsp. paratuberculosis may also be connected to the etiology of Crohn's disease, an inflammatory syndrome in the human gastrointestinal tract (5, 30). A conceivable transmission route for M. avium subsp. paratuberculosis to humans is via consumption of contaminated food or drinking water (5). In a recent study by Pickup et al. (30), one possible route for M. avium subsp. paratuberculosis transmission to humans was assumed to be the inhalation of aerosols carrying M. avium subsp. paratuberculosis. A third subspecies of M. avium is M. avium subsp. silvaticum, which may cause paratuberculosis in mammals and tuberculosis in birds but has never been isolated from the environment.

Typically mycobacteria are very-slow-growing bacteria. Cultures of intermediate- to slow-growing mycobacteria may take from 1 to 2 weeks to several months to grow. Traditionally, the detection of mycobacteria has been based on acid-fast staining, growth characteristics (growth rate, temperature, pigmentation, photoreactivity, and colony morphology), and biochemical tests. However, in clinical work, it is important to rapidly differentiate the causative mycobacterial species, e.g., M. avium subsp. avium from M. tuberculosis or other mycobacteria, and therefore, more rapid methods have been developed for the identification. High-performance liquid chromatography analyses of the mycobacterial mycolic acids (4) and gas liquid chromatography analyses of the fatty acids, alcohols, and cleavage products of mycolic acids (36, 38) are widely used today for the more rapid identification of mycobacteria. In recent years, molecular methods, including gene probes, random amplified polymorphic DNA, PCR, hsp65, fluorescence in situ hybridization (FISH), etc., have increasingly been developed for the most important species of mycobacteria (10, 22, 26, 27, 32, 34). Some of the tests are already commercially available.

FISH is based on the specific binding of nucleic acid probes to specific regions on rRNA (1). Traditionally, FISH methods are based on the use of conventional DNA oligonucleotide probes, containing around 20 bases. Peptide nucleic acid (PNA) molecules are DNA mimics, where the negatively charged sugar-phosphate backbone is replaced by an achiral, neutral polyamide backbone formed by repetitive units of N-(2-aminoethyl) glycine. PNA can hybridize to complementary nucleic acid targets obeying the Watson-Crick base-pairing rules (12). PNA molecules have unique hybridization characteristics, exhibiting rapid and stronger binding to complementary targets and a lack of electrostatic repulsion. Typically, the probes used for PNA FISH are shorter (optimum, 15 bases) than conventional DNA probes (29, 33). Stender et al. (32, 33) found earlier that PNA probes are able to penetrate the rigid cell wall of M. tuberculosis under relatively mild conditions, which preserve the cell morphology.

The PNA FISH method is also a promising technique for the in situ visualization of microorganisms in biofilms, especially because the hydrophobic nature of the PNA molecule allows a better diffusion through the biofilm matrix and thus improved discrimination of microbes in their naturally occurring three-dimensional structures (2).

We now report for the first time a rapid PNA FISH method that can identify clinically important M. avium subsp. avium and M. avium subsp. paratuberculosis bacteria. The method was applied to smears of pure cultures of the bacteria and to biofilm and water samples taken from a Propella biofilm reactor supplied with potable water from a distribution supply and spiked with M. avium subsp. avium.

MATERIALS AND METHODS

Bacterial species.

The specificity of the designed PNA probe was tested with several M. avium subsp. avium and M. avium subsp. paratuberculosis strains (Table 1). The probe was evaluated with several M. avium subsp. avium strains and other bacteria to test the specificity and sensitivity of the probe. All mycobacterial strains, with the exception of those of M. avium subsp. paratuberculosis, were grown on mycobacteria 7H11 agar (Difco); M. avium subsp. paratuberculosis was cultured on BBL Herrolds egg yolk agar with mycobactin J (BD UK Ltd., Oxford, United Kingdom). All mycobacteria were incubated at 30 to 37°C in air until there was enough biomass for testing (5 to 12 days); cultures of M. avium subsp. paratuberculosis were incubated for 4 to 8 weeks. Corynebacterium sp., Corynebacterium michiganense, and Rhodococcus bronchialis were cultured on R2A agar (Oxoid) and incubated at 37°C. Other bacteria used in this study were grown on mycobacteria 7H11 agar, while Dietzia maris, Nocardia cellulans, and Skermania piniformis were cultured on R2A agar and incubated at 30°C until there was enough biomass for the FISH analyses (5 to 12 days). Escherichia coli was cultured on nutrient agar (Oxoid) and incubated for 24 h at 37°C.

TABLE 1.

The sequence match results in databases and the results of the M. avium-specific PNA probe with tested bacteria

Organism	Accession no.a or sourceb	FISH outcome
M. avium subsp. avium	DSM 44156	+
M. avium subsp. avium	ATCC 15796	+
M. avium subsp. avium	E89	+
M. avium subsp. avium	E91	+
M. avium subsp. avium	E110	+
M. avium subsp. avium	E286	+
M. avium subsp. avium	E338	+
M. avium subsp. avium	E342	+
M. avium subsp. avium	E343	+
M. avium subsp. avium	E345	+
M. avium subsp. avium	E346	+
M. avium subsp. avium	E348-1	+
M. avium subsp. avium	E529	+
M. avium subsp. avium	E714	+
M. avium subsp. avium	E750	+
M. avium subsp. avium	E771	+
M. avium subsp. avium	E790	+
M. avium subsp. avium	BA11	+
M. avium subsp. avium	BA12	+
M. avium subsp. avium	BA30	+
M. avium subsp. paratuberculosis	ATCC 19698	+c
M. bohemicum	DSM 44277	−
M. chelonae	ATCC 14472	−
M. fortuitum	ATCC 6841T	−
M. gastri	ATCC 15754	−
M. gordonae	ATCC 14470T	−
M. intracellulare	ATCC 12950	−
M. kansasii	ATCC 12478T	−
M. lentiflavum	ATCC 51985	−
M. marinum	ATCC 927T	−c
M. mucogenicum	DSM 44625T	−c
M. nonchromogenicum	ATCC 19530T	−
M. palustre	DSM 44572	−
M. peregrinum	ATCC 14467T	−
M. scrofulaceum	ATCC 35792	−
M. shimoidei	ATCC 27962T	−
M. terrae	ATCC 15755T	−c
M. tusciae	DSM 44338	−
M. xenopi	ATCC 19250T	−
Corynebacterium sp.	*	−
Corynebacterium michiganense	*	−
Dietzia maris	DSM 43672	−d
Gordonia sputi	ATCC 33609	−
Nocardia cellulans	ATCC 12830	−
Skermania piniformis	DSM 43998	−d
Tsukamurella paurometabolum	ATCC 8368	−
Rhodococcus bronchialis	**	−
Williamsia murale	DSM 44343	−
Escherichia coli	NCTC 12900	−

^a“E” strains are environmetal strains isolated from brook waters (16, 17, 18, 37); “BA” strains are environmental strains isolated from bioaerosols (31).

^b*, isolate received from Kuopio University Hospital; **, isolate received from the Department of Microbiology, University of Helsinki.

^cStrong autofluorescence of the cells.

^dStrong autofluorescence of the cells, but the shape and size of the cells differed clearly from mycobacteria.

Design of PNA oligonucleotide probes.

Selection of the sequence of the probe was based on the 16S rRNA sequence of the M. avium subsp. avium type strain DSM 44156 (corresponding GenBank accession number AJ536037). The 16S rRNA databases at Michigan State University (Ribosomal Database Project II, http://rdp.cme.msu.edu/html/ [7]) and at the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/BLAST/) were searched for a specific sequence testing the probe specificity. Criteria for the sequence of PNA FISH probes were high specificity (at least 1-bp difference from the 16S rRNA sequences of bacteria other than M. avium), 15-bp length, and no self-complementary structures (33). The N terminus (5′) of the oligomer was labeled with 6-carboxyfluorescein (6-FAM; Eurogentec Ltd., Romsey, United Kingdom).

Hybridization in slide tests.

The hybridization protocol was similar to that of Stender et al. (32) with small modifications. A loopful of a colony grown on agar medium was suspended in 350 μl phosphate-buffered saline and mixed, and 25 μl of the bacterial suspension was placed on a microscope slide. In these studies, we used ordinary microscope slides which were cleaned with 98% ethanol (Menzel GmbH, Germany), polytetrafluoroethylene (Teflon)-coated specialized multispot microscope slides (C. A. H. Hendley Ltd., England), and Teflon-coated diagnostic microscope slides (Erie Scientific Company, Portsmouth, NH). The smear was air dried and gently flamed. Before hybridization, smears were immersed in 80% (vol/vol) ethanol for 15 min and allowed to air dry. Smears were covered with 25 μl of the FAM-labeled PNA probe in a hybridization solution and covered with coverslips. The hybridization solution contained 10% (wt/vol) dextran sulfate (Sigma Chemical Co., St. Louis, MO), 10 mM NaCl (Sigma), 30% (vol/vol) formamide (Sigma), 0.1% (wt/vol) sodium pyrophosphate (Sigma), 0.2% (wt/vol) polyvinylpyrrolidone (Sigma), 0.2% (wt/vol) Ficoll (Sigma), 5 mM disodium EDTA (Sigma), 0.1% (vol/vol) Triton X-100 (Sigma), 50 mM Tris-HCl (Sigma), and 200 nM PNA probe. Slides were hybridized for 90 min at 59°C. For the testing of the autofluorescence of the bacteria, the hybridization protocol for the reference slides was used, but the PNA probe was omitted from the buffer.

Following the hybridization, slides were washed by immersion in prewarmed (59°C) washing solution containing 5 mM Tris base (pH 10; Sigma), 15 mM NaCl, and 0.1% (vol/vol) Triton X-100 (Sigma). Washing was done in a 59°C oven for 30 min. After gentle rinsing with distilled water, slides were allowed to air dry.

Finally, the smears were mounted with 1 drop of nonfluorescence immersion oil (Fluka Chemika or Olympus, Japan) and covered with coverslips. The slides were stored in the dark for a maximum of 48 h before microscopy. Microscopy was conducted using a Zeiss Axioplan 2 epifluorescence microscope equipped with an HBO 100 W bulb (Zeiss AttoArc), Plan-Neofluar 40× oil immersion objective, and Roper Scientific RTE-charge-coupled device 1300Y camera or Olympus BX-51 TF epifluorescence microscope (Olympus Co. Ltd., Japan) equipped with an Olympus UPlanFl 100× oil immersion objective and CC-12 digital camera (Soft Imaging System, Germany). M. avium subsp. avium and M. avium subsp. paratuberculosis were detected on the basis of their green fluorescence, which was brighter than the autofluorescence of the bacteria in reference slides, and their cell morphology.

Biofilm and water analyses.

A Propella (Xenard, Mechanique de Precision, Seichamps, France) reactor (28) was spiked with the M. avium subsp. avium strain isolated from fresh stream water (strain E89). Before spiking, the Propella reactor with polyvinyl chloride (PVC) coupons inside the reactor was run with potable water from a distribution supply for 1 month. The M. avium subsp. avium strain was grown in Middlebrook 7H9 broth (Difco) at 35°C for 11 days. The suspension was centrifuged (15 min, 1,650 × g, Sorvall GLC-3), and the cells were washed with 0.9% NaCl and suspended in autoclaved potable water. A dilution of this in 50 ml of autoclaved potable water was feed to the Propella reactor. The concentration of M. avium subsp. avium bacteria in Propella was 10⁵/ml, as determined with culturing.

Water was circulated inside the Propella reactor, and the Reynolds number for water flow was over 15,000, i.e., water flow was turbulent. Water flow through the Propella reactor was 183 ml/min, the volume of the reactor being 2.3 liters (retention time, 12.6 h). A water sample (200 ml) was taken from the outlet of Propella 5 days after the spiking. The water was filtered through a membrane filter (Ultipor, NR047100, 0.2-μm pore size; Pall Corp.). Retained bacteria were detached from the filter by shaking with 6 ml of the original water filtrate in the presence of glass beads, as described by Kusnetsov et al. (21). An aliquot (25 μl) of the concentrated bacterial suspension was pipetted onto a Teflon-coated diagnostic microscope slide (Erie Scientific Company, Portsmouth, NH) and hybridized as described above. Hybridized bacteria were detected with Olympus BX-51 TF epifluorescence microscope (Olympus Co. Ltd., Japan) and counted using an ocular grid. The number of bacteria was analyzed also using a culturing method; the concentrated water sample was decontaminated for 15 min at room temperature with cetylpyridinium chloride (final concentration, 0.005% [wt/vol]). After 15 min of centrifugation (8,600 × g), 30 ml of sterile deionized water was mixed with the pellet, which was centrifuged again. The pellet was resuspended in 200 μl of sterile deionized water and spread plated on mycobacterial 7H11 agar supplemented with oleic acid-albumin-dextrose-catalase enrichment (Difco). Plates were incubated for up to 4 weeks at 30°C. Colonies were identified as M. avium by using the PNA FISH method described above. Heterotrophic bacteria in water were analyzed with a spread plating method on R2A agar (Difco). Plates were incubated for 7 days at 22°C before colony counting.

Propella biofilm coupons (PVC) were removed from the reactor 1 week after spiking the bacteria and kept moist until analyzed. Coupons were immersed in 80% (vol/vol) ethanol for 15 min, allowed to air dry, and covered with 100 μl of the PNA hybridization buffer (as described above). Hybridization was done in a moist petri dish in a 59°C oven. After hybridization, the coupon was immersed in prewarmed (59°C) washing buffer and washed for 30 min at 59°C in the oven. Finally, the coupon was rinsed with Milli-Q water. After air drying, the coupon was mounted with 1 drop of nonfluorescence immersion oil (Olympus, Japan) and covered with a coverslip. The coupon was observed by using an Olympus BX-51 TF epifluorescence microscope (Olympus Co. Ltd., Japan). For the culturing method, bacteria were detached from three Propella coupons by 2-min sonication in 25 ml of ultrapure water. The biofilm-water mixture was then analyzed in the same way as the outlet water samples described above. Biofilm results were normalized to the amount of surface area (cm²).

RESULTS AND DISCUSSION

Probe design and testing of specificity.

According to the PNA rules, the chosen PNA oligomer sequence was 5′-TGCGTCTTGAGGTCC-3′. This oligomer is complementary to M. avium subsp. avium (DSM 44156) 16S rRNA positions 148 to 162; therefore, the probe was designated MAV148. Because subspecies of M. avium (M. avium subsp. avium, M. avium subsp. paratuberculosis, and M. avium subsp. silvaticum) share the identical 16S rRNA sequence (3, 20), the probe is able to hybridize to all these strains. The Ribosomal Database Project II and National Center for Biotechnology Information BLAST searches showed that this oligomer differed by at least 1 bp from the 16S rRNA sequences of bacteria other than M. avium subsp. avium, M. avium subsp. paratuberculosis, and M. avium subsp. silvaticum, which is found to be sufficient for discrimination from other bacteria (12, 33).

The PNA probe was tested with smears of several Mycobacterium species. It was clear in this study that the MAV148 probe reacted with only with M. avium subsp. avium and M. avium subsp. paratuberculosis (Table 1). There was no cross-hybridization to rRNA of the other mycobacteria, mycolic acid-containing bacteria, or E. coli used in this study. However, the probe could not differentiate between M. avium subsp. avium and M. avium subsp. paratuberculosis.

Some autofluorescence was noted in smear tests with mycobacteria and also some other mycolic acid-containing bacteria (Table 1). The strongest autofluorescence was found with M. avium subsp. paratuberculosis. The autofluorescence was especially strong if the smear was made from a highly concentrated bacterial mixture, resulting in aggregates of cells with a notable autofluorescence. However, even when there was some autofluorescence with M. avium subsp. avium and other mycobacteria tested, the differentiation of hybridized bacteria from nonhybridized reference bacteria was really difficult only with M. avium subsp. paratuberculosis. When work is undertaken with mycobacteria, a negative reference control should always be included. The negative control is prepared in exactly the same way as the sample, but the PNA probe is omitted from the hybridization step. In this way, it is possible to differentiate autofluorescence of the cells from the cells hybridized with the MAV148 probe during microscopy.

In some previous studies, it was found that the autofluorescence of M. tuberculosis is a disadvantage of the PNA FISH method (26, 32). Further, Davenport et al. (9) found that there was generally autofluorescence in mycolic acid-containing bacteria. One solution for this problem could be fitting a fluorescein isothiocyanate (FITC)/Texas Red double-band filter to the epifluorescence microscope (26, 32) or bleaching the bacteria by potassium permanganate, as applied in the auramine-rhodamine staining of mycobacteria (6). Also, autofluorescence of bacteria may be reduced by strong light irradiation. Neumann and Gabel (24) found that the autofluorescence of tissues was reduced or eliminated by irradiation with light before treatment with fluorescence probes. We found the same result when the smears on glass slides were observed by microscopy. During the microscopy, the autofluorescence of bacteria was reduced markedly due to the bleaching effect of the strong excitation light of the epifluorescence microscope. However, reducing the autofluorescence of bacteria by using light was not systematically studied in this study but would be worthy of study in the future. We also found with the smear tests that using lower concentrations of bacteria in the smears could reduce the autofluorescence. With the cells separated, it was easier to differentiate autofluorescence from the fluorescence of hybridized cells.

Biofilms and water samples.

It proved feasible to identify and separate hybridized M. avium subsp. avium bacteria from other bacteria and background material in the outlet water taken from the Propella reactor (Fig. 1). In this case, the number of bacteria analyzed by the FISH method was 1.8 × 10⁷ bacteria/liter and the results were obtained in several hours. The colony count number of heterotrophic bacteria in water was 2.7 × 10⁷ CFU/liter. It is assumed that the majority of the PNA-hybridized cells were still viable since they exhibited a strong fluorescence signal, indicative of a high rRNA content. Using an agar culture method that took up to 4 weeks for the colonies to grow, the number of M. avium subsp. avium bacteria detected in that sample was only 0.7 × 10³ CFU/liter. The large difference in these numbers is partly explained by the cetylpyridinium chloride decontamination step used in the culturing method, which may also destroy some unknown but significant proportion of the M. avium bacteria. There are several studies published earlier showing that decontamination can kill a remarkably large fraction of the mycobacteria present (11, 23). Also, environmentally stressed strains probably are not able to recover or resuscitate on culture medium, which can lead to higher counts when using FISH than when using an agar culture method.

FIG. 1.

Epifluorescence photomicrograph of FISH-labeled M. avium subsp. avium bacteria in the outlet water concentrate of the Propella reactor supplied with potable water.

The PNA FISH method was tested with biofilms on flat PVC coupons taken from the Propella biofilm reactor. Strong autofluorescence of the biofilm and the PVC substratum itself hampered the detection, and especially the quantification, of hybridized bacteria in biofilms. However, M. avium subsp. avium bacteria were able to be visualized against the autofluorescence background. The number of bacteria detected with FISH was around 100,000/cm², when it was 83 ± 15 CFU/cm² when analyzed with the agar culture method from the detached biofilm-water mixture. The reasons for this approximately 3 log difference are probably the same as in water, i.e., decontamination and environmental stress during culturing. Use of the FISH method to detect specific species in biofilms will be facilitated if the substrata exhibit low autofluorescence. This is easily achieved for defined laboratory experiments where a simple choice of substratum can be selected but is made more difficult for the distribution supply pipes taken directly from the ground. Preliminary results indicate that concrete pipes are particularly problematic for autofluorescence (unpublished data). Although there have been numerous papers published on the use of nucleic acid-based probes to detect microorganisms in biofilms, few authors refer to problems with the autofluorescence of their samples. Nevertheless, when many of the published images are scrutinized, it becomes evident that autofluorescent material is present. Azevedo et al. (2) commented on this phenomenon in their study of Helicobacter pylori persistence in potable-water biofilms: they observed that the biofilm became more autofluorescent with age and concluded that the glycocalyx of the biofilm was accumulating the very low concentrations of polyaromatic hydrocarbons present in the drinking water supplying the biofilm reactor. Judicious selection of DNA- or PNA-based probes fluorescing more toward the red end of the spectrum can help overcome this problem.

In the present study, only oil immersion objectives were available for the microscopy observations, but this did not present particular problems using the flat coupons from the Propella reactor. Pipes are, of course, highly curved, making observation of biofilms on the inner surface physically more difficult for oil immersion lenses. Using objectives with a long working distance would make it easier to detect hybridized bacteria directly in intact biofilms (19). This would then facilitate locating specific bacteria in biofilms in real distribution supply pipe work and provide valuable information as to structure-function relationships and physiological niches. This would also enable more informed studies investigating the effects of different physicochemical environmental stresses such as flow velocity, water chemistry, and nutrient concentration, the type of disinfection, and concentration/contact time on the survival of M. avium in biofilms.