Simultaneous in situ analysis of the structure and function of bacterial cells present within complex communities is a key for improving our understanding of microbial ecology. A protocol for the in situ identification of Listeria spp. using fluorescently tagged, rRNA-targeted oligonucleotide probes was developed. Ethanol fixation and enzymatic pretreatment with lysozyme and proteinase K were used to optimize whole cell hybridization of exponential phase and stationary phase Listeria spp. cells. In parallel, transcript probes carrying multiple digoxigenin molecules were combined with anti-digoxigenin Fab antibody fragments labeled with horseradish peroxidase to detect, via the catalytic deposition of fluorescein-tyramide, the iap-mRNA in single Listeria monocytogenes cells. The iap gene encodes the associated virulence factor p60. Application of the new signal amplification technique resulted in strong signals comparable in intensity to those obtained with fluorescently labeled rRNA-targeted oligonucleotide probes.
Fluorescent in situ hybridization (FISH) with rRNA-targeted oligonucleotide probes is a powerful tool to analyze in a cultivation-independent way the structure and dynamics of complex microbial communities (e.g. [1, 2]). Combination of FISH with confocal laser scanning microscopy also permits analysis of the spatial distribution of probe-target populations in microbial biofilms (e.g. [2, 3]) and tissue samples . While extremely useful for identification, rRNA-based techniques usually do not provide the researcher with much information about in situ functions of bacteria within their respective ecosystems. Molecular techniques for the in situ visualization of mRNA of key functional proteins in single microbial cells would provide the opportunity to narrow this gap and to link abundance and spatial distribution of bacteria with their in situ physiological activities. Due to the low number and instability of most mRNA molecules in prokaryotic cells only a few reports of successful in situ detection of bacterial mRNA have been reported to date [5–9].
Listeria spp. are facultatively anaerobic, non-spore-forming rod-shaped bacteria which have been assigned by ribosomal DNA sequence analysis to the Gram-positive bacteria with low DNA G+C content . The genus Listeria comprises six well characterized species, among which L. monocytogenes is of particular importance as a pathogen causing serious food-borne infections in animals and humans. Newborns, pregnant women and immunocompromised individuals are at increased risk for infection with virulent strains of L. monocytogenes. Such strains are able to display an intracellular life cycle within nonprofessional phagocytes as well as macrophages and monocytes. Several proteins have been recognized as essential for the virulence of L. monocytogenes. Among these, the 60-kDa extracellular p60 protein, encoded by the iap (invasion associated protein) gene has been identified to be not only required for efficient adhesion to certain mammalian cell types but also essential for cell metabolism of L. monocytogenes due to its murein hydrolase activity . It has been shown that all Listeria species secrete a p60 protein, albeit with genus-specific conserved N- and C-termini, and species-specific variable internal portions .
In the present study we report on a rRNA-based in situ identification protocol for the genus Listeria. In addition we have developed a technique for in situ detection of iap-mRNA within intact cells of L. monocytogenes. To our knowledge this is the first time that successful in situ identification of mRNA from a virulence factor in bacteria has been demonstrated. This technique might offer the opportunity to analyze gene expression both in L. monocytogenes present in mixed microbial communities and during infection in mammalian cells.
2Materials and methods
2.1Bacterial strains and media
Five L. monocytogenes strains (sv4a L99, sv4b SLCC 4013, sv1/2a EGD PrfA−, sv1/2a EGD ActA−, RIII) as well as L. seeligeri SLCC 3954, L. grayi, L. welshimeri SLCC 5828, L. innocua sv6a NCTC 11288, and L. ivanovii ATCC 19119 were obtained from the Institute of Medical Microbiology and Hygiene (University of Würzburg, Würzburg, Germany). Listeria strains and Enterococcus hirae DSM 20160T were cultured aerobically in brain heart infusion (BHI) broth at 37°C. Listeria cells were harvested at various growth phases (see Section 3) by centrifugation (5000×g for 5 min). Escherichia coli ATCC 11775T, Pseudomonas aeruginosa DSM 2659, and Aeromonas hydrophila ATCC 7966T were grown aerobically in Luria-Bertani medium and harvested at mid-exponential phase by centrifugation (5000×g for 5 min) to obtain cells with high rRNA contents.
The following rRNA-targeted oligonucleotide probes were used: (i) LIS-1255, complementary to a region of the 16S rRNA of the genera Listeria and Brochothrix; (ii) GAM, complementary to a region of the 23S rRNA conserved in the gamma subclass of Proteobacteria ; (iii) EUB, complementary to a conserved region of bacterial 16S rRNA molecules . In addition four oligonucleotides were designed and used for detection of iap-mRNA: (i) iap-30 (5′-ACCGCAATCCCAGCTGTA-3′, derived from primer UnilisA ); (ii) iap-630 (5′-CCCAAATAGTGTCACCGC-3′), (iii) iap-1159 (5′-AGTTGGTCCGTTACCACC-3′); and (iv) iap-1437 (5′-TACGCGACCGAAGCCA-3′, a truncated version of primer iap-1436; see below). Oligonucleotides were synthesized with a C6-TFA aminolinker at the 5′ end (Interaktiva, Ulm, Germany). Labeling with 5(6)-carboxyfluorescein-N-hydroxysuccinimide ester (CFLUOS; Boehringer, Mannheim, Germany), the monofunctional, hydrophilic sulfoindocyanine dyes Cy3 and Cy5 (Amersham, Buckinghamshire, UK) or horseradish peroxidase (HRP) was performed as described previously [17, 18].
2.3Cell fixation and enzymatic pretreatment
Gram-negative reference cells were fixed by addition of paraformaldehyde solution (final concentration 4%) and stored in a 1:1 mixture of phosphate-buffered saline (130 mM NaCl, 10 mM sodium phosphate buffer, pH 7.2) and 96% ethanol at −20°C . Listeria cells were fixed either by the procedure used for Gram-negative cells or by addition of ethanol to a final concentration of 50% (v/v) . E. hirae cells were fixed by addition of ethanol . Listeria cells were spotted on precleaned six-well glass slides, dried at 46°C and dehydrated in 50, 80, and 96% ethanol (3 min each). For partial enzymatic digestion of the cell wall Listeria cells were treated with and without lysozyme and proteinase K prior to hybridization. In the developed optimized protocol fixed immobilized Listeria cells were covered with 20 μl of lysozyme solution (10 mg ml−1 of lysozyme [Serva, Heidelberg, Germany; 164.500 U mg−1] in a buffer containing 100 mM Tris-HCl and 50 mM EDTA, pH 8.0) and incubated for 20 min at room temperature. Lysozyme reaction was stopped by rinsing the slide thoroughly with dH2O followed by an additional ethanol series. Subsequently, Listeria cells were subjected to an aqueous proteinase K solution (0.05 mg ml−1, 20 U mg−1, Boehringer) and incubated for 5 min at room temperature. Finally, proteinase K was removed by rinsing with dH2O and an additional ethanol series. For preparing artificial mixtures of Gram-negative reference cells and Listeria cells, Gram-negative reference cells were spotted on the well containing the fixed Listeria cells after the enzymatic cell wall permeabilization procedure.
2.4Whole cell hybridization with oligonucleotide probes
Hybridizations with CFLUOS, Cy3, Cy5 and HRP-labeled oligonucleotide probes were performed following the protocols of Manz et al.  and Amann et al. . Specific hybridization with probe Lis-1255 required the addition of formamide to a final concentration of 35% to the standard hybridization buffer. Hybridizations with the iap-mRNA-targeting oligonucleotide probes were performed by simultaneous application of the four probes (each labeled with Cy3) using a range of formamide concentrations (0%, 10%, 20%, 30%, 40%, 50%, 60%) in the hybridization buffer. The NaCl concentration in the washing buffer was adjusted according to the formamide concentration in the respective hybridization buffer . Specifically bound horseradish peroxidase-labeled oligonucleotide probes were detected by addition of the substrate fluorescein tyramide as described by Schönhuber et al. . For dual staining of L. monocytogenes cells with fluorescent oligonucleotide probes and DAPI, cells were stained after in situ hybridization with 1 μg ml−1 DAPI for 5 min at RT. Fluorescence was detected with a Zeiss LSM 510 scanning confocal laser microscope (Carl Zeiss, Oberkochen, Germany).
2.5DNA isolation, PCR amplification, cloning and sequencing
High molecular mass genomic DNA was extracted from L. monocytogenes sv1/2a EGD ActA− using a modification of the protocol given by Lema et al. . For cell lysis we replaced the lysozyme step by an ultrasonic treatment (150 s) after the addition of an equal volume of glass beads (0.1–0.11 mm diameter; Braun, Melsungen, Germany). PCR primers iap-1159F (5′-GGTGGTAACGGACCAACT-3′) and iap-1436R (5′-ATACGCGACCGAAGCCAA-3′; derived from the primer Lis1B ) were used for PCR with a thermal capillary cycler (Idaho Technology, Idaho Falls, ID). The primer pair bound between nucleotide position 1159 and 1453 of the iap gene (nucleotide numbering in L. monocytogenes EGD ). Reaction mixtures were prepared according to the instruction of the manufacturer in a total volume of 50 μl by using the 20 mM MgCl2 reaction buffer. Thermal cycling was carried out as follows: an initial denaturation at 94°C for 45 s followed by 35 cycles of denaturation at 94°C for 15 s, annealing at 58°C for 25 s and elongation at 72°C for 22 s. Cycling was completed by a final elongation step at 72°C for 60 s. The presence and size of the amplification product were determined by agarose gel electrophoresis (1.0%). The amplified 294 bp long product was excised from the gel and purified using the Gel DNA Extraction Kit (Boehringer). Purified products were ligated into the cloning vector pCR2.10 and transformed into competent E. coli INVαF′ cells with the Original TA cloning kit (Invitrogen Corp., San Diego, CA). Identity and orientation of the inserts were determined by cycle sequencing of purified plasmid preparations (QIAgen Spin Plasmid Kit; Qiagen, Hilden, Germany) with a Thermo sequenase cycle sequencing kit (Amersham) and an infrared automated DNA sequencer (LiCor, Inc., Lincoln, NE) under the conditions recommended by the manufacturers.
2.6In vitro transcription
Digoxigenin (DIG)-labeled antisense and sense polyribonucleotide probes were generated by in vitro transcription from pCR2.10 vectors containing the 294 bp long, cloned iap fragments in the respective orientations. 396-nucleotide sense and antisense transcripts were transcribed from the T7 promoter after linearization of the plasmid with SpeI. Transcription reactions were performed with an RNA transcription kit (Boehringer) with a UTP/DIG-11-UTP ratio of 6.5:3.5. Transcription products were analyzed by denaturing polyacrylamide gel electrophoresis and spectrophotometrically following standard protocols .
2.7Northern blot analysis
Total RNA was isolated from L. monocytogenes strains sv1/2a EGD ActA− and 7972 PrfA−, as well as from P. aeruginosa, as described by Oelmüller et al. . For Northern blotting, formaldehyde-agarose gel electrophoresis was performed as described previously . DIG-labeled RNA molecular mass markers (Boehringer) were also loaded onto the gel for subsequent use as standards for the size determination of hybridizing RNA bands. After electrophoresis, nucleic acids were transferred with a pressure blotter onto positively charged nylon membranes (Hybond™-N+, Ver. 2,) and subsequently crosslinked by exposure to UV light for 150 s. Hybridizations were performed with HRP-labeled oligonucleotides or with DIG-conjugated polyribonucleotide probes. The membranes were prehybridized for 1 h at 46°C (polyribonucleotide probes) or at 35°C (oligonucleotide probes) with 20 ml solution containing 10.8×SSC, 4.3% blocking reagent (Boehringer), 0.2%N-laurylsarcosine and 0.04% SDS. Hybridizations were performed at 46°C for at least 12 h using polyribonucleotide probes. Oligonucleotide hybridizations were performed at 35°C for 90 min. Hybridization solution contained 0.9 M NaCl, 0.02 M Tris-HCl (pH 8.0), 1.8% blocking reagent, 0.1%N-laurylsarcosine, 0.01% SDS, 22% formamide (EUB, iap-mRNA-targeted oligonucleotide probes) or 40% formamide (polyribonucleotide probes) or 62% formamide (Lis-1255), and 100 ng of HRP-labeled oligonucleotide probe or 250 ng DIG-labeled polyribonucleotide probe. Following hybridization, membranes were washed twice at 35°C (oligonucleotide probes) or 48°C (polyribonucleotide probes) for 10 min (oligonucleotide probes) or 15 min (polyribonucleotide probes) in 10 ml washing buffer containing 20 mM Tris-HCl (pH 7.2), 4.6 mM EDTA, and 0.01% SDS. NaCl concentration in the washing buffer was adjusted according to the formamide concentration in the respective hybridization buffer . DIG-conjugated polyribonucleotide probes were detected with HRP-labeled anti-DIG Fab antibody fragments following instructions recommended by the manufacturer (Boehringer). Specifically bound nucleic acid probes were visualized via HRP-induced chemiluminescence (ECL; Amersham). Emitted light was recorded with an ultrasensitive photon-sensitive camera (Hamamatsu C2400-47, Hamamatsu Photonics, Hersching, Germany).
2.8Whole cell hybridization with DIG-conjugated polyribonucleotide probes
Fixed and enzymatically pretreated cells were covered with 10 μl hybridization mixture (75 mM NaCl, 80% formamide, 20 mM Tris, 0.01% SDS, 250 ng of transcript probe) in an isotonically equilibrated humid chamber. Following an initial denaturation at 80°C for 30 min, hybridization was performed at 46°C for 5 h. Hybridization was terminated by immersing the slides briefly in distilled H2O. After air-drying, wells were covered with 21 μl detection solution (20 mM Tris, 0.7 M NaCl, 1% blocking reagent) containing 0.6 μg HRP-labeled anti-DIG antibody Fab fragments. Slides were incubated at 27°C for 1 h, washed for 10 min at the same temperature in a solution containing 0.7 M NaCl and 100 mM Tris, and finally rinsed with distilled water. Specifically bound HRP-labeled Fab fragments were detected by using the substrate fluorescein tyramide . Subsequently, fixed cells were hybridized with Cy5-labeled oligonucleotide probes (see above). Fluorescence was observed with a Zeiss LSM 510 scanning confocal laser microscope (Carl Zeiss).
3Results and discussion
3.1In situ hybridization of Listeria spp. with rRNA-targeted oligonucleotide probes
The difference in cell wall composition and structure between Gram-negative and Gram-positive bacteria hampers the application of in situ hybridization techniques established for Gram-negative bacteria to Gram-positive bacteria. Previous work has demonstrated that successful in situ hybridization of most Gram-positive bacteria requires changes in the fixation protocol and/or additional enzymatic pretreatments to make them permeable for reporter molecule-labeled nucleic acid probes (e.g. [19, 24, 25]). Therefore, we initially tested different fixation and enzymatic pretreatments to adapt existing in situ hybridization protocols to Listeria spp.
BHI-cultivated L. monocytogenes sv1/2a EGD ActA− cells were harvested at different times during the growth curve (Fig. 1A) and fixed immediately by addition of either paraformaldehyde or ethanol. Subsequent in situ hybridization with the 16S rRNA-targeted oligonucleotide probe Lis-1255 revealed that permeabilization of Listeria cells due to ethanol fixation was more effective than with paraformaldehyde fixation (Fig. 1B). Both fixatives allowed the specific visualization of the majority of exponential Listeria cells while stationary cells were not detectable. Failure to detect stationary Listeria cells could be due to low cellular ribosome contents or to limited cell wall permeability. We determined the effect of treating ethanol- and paraformaldehyde-fixed Listeria cells with different concentrations of lysozyme (0.1–50 mg ml−1) on probe permeability. These experiments demonstrated that BHI-cultivated stationary Listeria cells (i) still possess sufficient ribosomes for in situ detection (Fig. 1B) and (ii) are permeable for fluorescently labeled oligonucleotide probes after being treated with 10 mg ml−1 lysozyme for 20 min at RT (Fig. 2A). Increasing lysozyme concentration or exposure time resulted in partial or complete lysis of the cells. At a formamide concentration of 35% (v/v), probe Lis-1255 hybridized strongly with representative members of all six Listeria species and not with non-target organisms (Table 1). Fluorescent in situ hybridization is a promising tool for rapid identification and quantification of Listeria spp. in complex samples and could become a valuable complement to standard cultivation, PCR and immunological techniques. However, Listeria cells present in natural samples might differ in their cell wall composition from BHI-cultivated cells. Consequently, it is recommended to perform in situ monitoring of Listeria in samples which have been subjected to different cell permeabilization procedures.
Table 1. Listing of strains analyzed by whole cell hybridization with probe Lis-1255
aBW, Biozentrum Universität Würzburg, Würzburg, Germany; ATTC, American Type Culture Collection, Rockville, MD, USA; NCTC, National Collection of Type Cultures, London, UK; SLCC, Special Listeria Culture Collection, Institut für medizinische Mikrobiologie und Hygiene, Universität Würzburg, Würzburg, Germany; DSM, Deutsche Sammlung von Mikroorganismen und Zellkulturen, Braunschweig, Germany.
Listeria monocytogenes sv1/2a EGD ActA−
Listeria monocytogenes sv1/2a EGD PrfA−
Listeria monocytogenes RIII
Listeria monocytogenes sv4a L99
Listeria monocytogenes sv4b
Listeria innocua sv6a
3.2In situ detection of mRNA of the iap gene in whole fixed cells of L. monocytogenes
The iap-mRNA was chosen as target for the development of a nucleic acid probe-based method for monitoring specific gene expression in L. monocytogenes for two reasons: first, iap-mRNA has been reported to be regulated at a post-transcriptional level and is present in a relatively high copy number ; second, expression of iap, which encodes a murein hydrolase involved in septum separation, is essential for cell viability of L. monocytogenes. Northern analysis using direct photon camera inspection demonstrated that iap-mRNA concentrations in L. monocytogenes were similar in cells from exponential and stationary growth phases (data not shown). However, there is less iap-mRNA than 16S rRNA in the cells (Fig. 3). Low numbers of iap-mRNA required the simultaneous application of the four iap-mRNA-targeted oligonucleotide probes to obtain a detectable signal after Northern hybridization (Fig. 3I). The sensitivity of the iap-mRNA Northern hybridization was significantly increased by using multiply labeled antisense polyribonucleotide transcript probes (Fig. 3 II). Polyribonucleotide-conferred iap-mRNA-specific Northern hybridization signals were comparable in their intensity to those obtained with 16S rRNA-targeted oligonucleotide probes labeled with a single HRP molecule (Fig. 3 III, IV). Specificity of polyribonucleotide hybridization was demonstrated by hybridization of the Northern blot membranes with a multiply labeled sense probe which did not lead to a detectable signal (data not shown).
Detection of iap-mRNA in fixed and lysozyme-treated L. monocytogenes cells was unsuccessful using the four iap-mRNA-targeted Cy3-labeled oligonucleotide probes simultaneously for in situ hybridization (data not shown). Increase of the formamide concentration in the hybridization buffer in order to remove higher order structure in the target mRNA did not result in detectable signals. The most probable reason for this was the relatively low cellular copy number of the iap-mRNA (see Section 2.7) which could not even be sufficiently compensated for by the simultaneous use of four oligonucleotide probes labeled with the currently brightest fluorescent dye Cy3 . In order to further increase the sensitivity of our assay we attempted to use HRP-labeled derivatives of the four iap-mRNA-directed oligonucleotide probes in combination with the fluorogenic TSA assay. This approach can confer up to 20-fold more fluorescence to target cells than can oligonucleotides labeled with a single fluorescent dye molecule . It requires, however, that the target cells are permeable for the relatively large oligonucleotide-enzyme conjugate. Hybridization of ethanol-fixed and lysozyme-pretreated L. monocytogenes cells with a HRP-labeled derivative of oligonucleotide probe Lis-1255 revealed that less than 10% of the cells could be visualized via enzymatic transformation of fluorescein tyramide. To further enhance cell permeability we implemented an additional optimized proteinase K treatment (0.05 mg ml−1; 5 min, RT) in the hybridization protocol. This resulted in specific visualization of more than 90% of the L. monocytogenes cells with HRP-labeled probe Lis-1255 (data not shown). However, heterogeneity in signal intensities between different L. monocytogenes cells, indicating differences in cell wall composition, was observable. This effect was most pronounced for the isogenic mutants L. monocytogenes sv1/2a EGD ActA− and PrfA−, and less pronounced for L. monocytogenes sv4b which was therefore selected as a model strain for subsequent studies. Simultaneous hybridization of ethanol-fixed, lysozyme- and proteinase K-pretreated L. monocytogenes sv4b with HRP-labeled derivatives of the four iap-mRNA-targeted oligonucleotide probes, however, still did not lead to detectable signals using the fluorogenic TSA system (data not shown), indicating that the cellular concentration of iap-mRNA is below the detection limit of the HRP-oligonucleotide-TSA technique.
For further sensitivity enhancement we applied a multiply DIG-labeled iap-mRNA antisense polyribonucleotide probe. Such polyribonucleotide probes have been recently shown to be nicely suited for signal amplification in in situ hybridization techniques  and have therefore been applied for non-fluorogenic detection of bacterial mRNA [5, 7]. After detection of probe-conferred DIG via a HRP-labeled anti-DIG Fab antibody fragment and the fluorogenic TSA system, L. monocytogenes sv4b cells showed bright signals while Aeromonas hydrophila reference cells remained unstained (Fig. 2B). These results demonstrate that the developed fixation and pretreatment protocol renders L. monocytogenes sv4b permeable for the large, 240-kDa HRP-Fab molecules. The specificity of the signals was corroborated by control hybridizations with a sense polyribonucleotide probe (using identical conditions and reaction reagents) which did not confer detectable fluorescence to the Listeria cells (Fig. 2C).
In the present study we have developed a protocol for 16S rRNA-based in situ identification of bacteria of the genus Listeria. Moreover, we have demonstrated the usefulness of multiply labeled polyribonucleotide probes in combination with the fluorogenic TSA system for in situ monitoring of mRNA of a virulence factor in L. monocytogenes. Our results, however, demonstrate that successful application of the PolyDIG-transcript–HRP-Fab–TSA system requires target strain-specific optimization of the permeabilization procedure and is therefore recommended for well defined target species. The method should be especially powerful when combined with rRNA hybridization techniques and confocal microscopy to study activities of pathogenic bacteria during infection or of indicator species in complex environments.
The authors would like to thank Sibylle Schadhauser for expert technical assistance, Leo Eberl for help with the photon camera, and Thomas Fritsche for discussions and revision of the manuscript. The technical assistance of the students participating in the Großpraktikum II 1997 is acknowledged. This work was supported by a grant from the Bayerische Forschungsverbund Grundlagen Gentechnischer Verfahren (FORGEN) to K.H.S.