- Email: [email protected]
Multiplex FISH analysis of a six-species bacterial biofilm
Journal of Microbiological Methods 56 (2004) 37 – 47 www.elsevier.com/locate/jmicmeth
Multiplex FISH analysis of a six-species bacterial biofilm T. Thurnheer *, R. Gmu¨r, B. Guggenheim Section for Oral Microbiology and General Immunology, Institute for Oral Biology, University of Zu¨rich, Plattenstrasse 11, CH-8028 Zu¨rich, Switzerland Received 10 April 2003; received in revised form 11 July 2003; accepted 3 September 2003
Abstract Established procedures use different and seemingly incompatible experimental protocols for fluorescent in situ hybridization (FISH) with Gram-negative and Gram-positive bacteria. The aim of this study was to develop a procedure, based on FISH and confocal laser scanning microscopy (CLSM), for the analysis of the spatial organization of in vitro biofilms containing both Gram-negative and Gram-positive oral bacteria. Biofilms composed of the six oral species Actinomyces naeslundii, Candida albicans, Fusobacterium nucleatum, Streptococcus oralis, Streptococcus sobrinus, and Veillonella dispar were grown anaerobically for 64.5 h at 37 jC on hydroxyapatite disks preconditioned with saliva. Conditions for the simultaneous in situ hybridization of both Gram-negative and Gram-positive bacteria were sought by systematic variation of fixation and exposure to lysozyme. After fixation and permeabilization biofilms were labeled by FISH with 16S rRNA-targeted oligonucleotide probes ANA103 (for the detection of A. naeslundii), EUK116 (C. albicans), FUS664 (F. nucleatum), MIT447 and MIT588 (S. oralis), SOB174 (S. sobrinus), and VEI217 (V. dispar). Probes were used as 6-FAM, Cy3 or Cy5 conjugates, resulting in green, orange-red or deep-red fluorescence of target cells, respectively. Thus, with two independent triple-hybridizations with three probes carrying different fluorescence-tags, all six species could be visualized. Results show that the simultaneous investigation by FISH of complex biofilms composed of multiple bacterial species with differential Gram-staining properties is possible. In combination with the optical sectioning properties of CLSM the technique holds great promise for the analysis of spatial alterations in biofilm composition in response to environmental challenges. D 2003 Elsevier B.V. All rights reserved. Keywords: 16S rRNA; Biofilm; CLSM; Fluorescent in situ hybridization
1. Introduction Biofilms are formed after rapid attachment and growth of microorganisms on a broad range of surfaces in contact with natural fluids. Consisting of single cells, cell aggregates, and microcolonies embedded
* Corresponding author. Tel.: +41-1-634-3256; fax: +41-1-6344310. E-mail address: [email protected] (T. Thurnheer). 0167-7012/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.mimet.2003.09.003
in an exocellular polymeric matrix gel, biofilms are a most common phenomenon on Earth and have been the subject of intense experimental and theoretical scrutiny over the past decade (Costerton et al., 1995; Marsh and Bradshaw, 1995; Wimpenny et al., 2000; Hermanowicz, 2003). Biofilm architecture and the physiological status of the cells contained within these structures are of prime interest for clinical, industrial, and environmental microbiology. Dental plaques are naturally occurring complex biofilms involved in the development of caries and
38
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47
periodontal diseases. Knowledge of their composition, their members’ metabolism and know-how on the successful manipulation of mixed bacterial biofilms may lead to improved preventive measures against these common diseases. Investigations of plaque in vivo are difficult due to the high degree of variability between individuals and since the site-specific microbial composition is reflecting the influences of a multitude of uncontrollable endogeneous and exogeneous dietary factors. In addition, working with humans always raises ethical questions. Therefore, we have developed and applied a six-species in vitro biofilm model of supragingival plaque (Guggenheim et al., 2001a) for the examination of the spatial arrangement and associative behavior of microorganisms, the study of the efficacy of antimicrobial mouthrinses, or the investigation of mass transport of macromolecules (Guggenheim et al., 2001b; Shapiro et al., 2002; Thurnheer et al., 2003). For analyses of the spatial arrangement of bacteria in a multi-species biofilm techniques, which use specific bacterial cell markers and maintain the biofilm’s natural architecture are required. The aim of this study was to develop a fluorescent in situ hybridization (FISH) technique using probes to target-specific 16S rRNA sequences and to combine it with confocal laser scanning microscopy (CLSM) for the simultaneous analysis of the spatial distribution of both Gram-positive and Gram-negative bacteria in biofilms. FISH is a recognized tool for the specific and sensitive identification of target organisms within complex microbial communities (Amann et al., 1995). Visualization of FISH labeled cells in biofilms can be carried out by techniques, including fluorescence microscopy and CLSM (Lawrence et al., 1991; Davey and O’Toole, 2000; Wimpenny et al., 2000). With biofilms, CLSM is often to be preferred, since it allows a three-dimensional noninvasive visualization of cells and the computational reconstruction of mature biofilms without distortion of their structure. However, the crucial problem was to find experimental conditions that render cell walls of Gram-positive bacteria penetrable to the probes without the simultaneous loss of signal from Gram-negatives that are simultaneously labeled with fluorescent oligonucleotide probes. Moreover, they should allow several consecutive hybridization procedures without harming the biofilm structure.
2. Materials and methods 2.1. Preparation of biofilms Actinomyces naeslundii OMZ 745, Veillonella dispar ATCC 17748T (OMZ 493), Fusobacterium nucleatum KP-F2 (OMZ 596), Streptococcus sobrinus OMZ 176, Streptococcus oralis SK248 (OMZ 607), and Candida albicans OMZ 110 were used as inocula for biofilm formation (Guggenheim et al., 2001a; Shapiro et al., 2002). Biofilms were grown in 24well polystyrene cell culture plates on sintered hydroxyapatite disks as described (Thurnheer et al., 2003). In brief, hydroxyapatite disks (F 10.6 mm) that had been preincubated (pellicle-coated) in processed whole unstimulated saliva were covered with 1.6 ml of substrate composed of 70% saliva + 30% mFUM. mFUM corresponds to a well established tryptone-yeast based broth medium designated as FUM (Gmu¨r and Guggenheim, 1983) and modified by supplementing 67 mmol/l Sørensen’s buffer (final pH 7.2). The carbohydrate concentration in mFUM was 0.3% (w/v), and consisted of either glucose (0– 16.5 h of biofilm cultivation) or a 1:1 (w/w) mixture of glucose and sucrose (16.5 – 64.5 h of biofilm cultivation). Wells were inoculated with mixed cell suspensions (200 Al) prepared from equal volumes and densities of each species and incubated anaerobically at 37 jC. Substrate was replenished after dipping (see below) at 16.5 and 40.5 h by aspirating spent substrate and adding back fresh substrate composed of saliva + mFUM (1.6 ml). At 16.5, 20.5, 24.5, 40.5, 44.5, 48.5, and 64.5 h biofilms were washed by three consecutive dips in 2 ml of sterile physiological saline (1 min per dip, room temperature). After the 64.5-h dip-wash biofilms were fixed immediately with paraformaldehyde (see below). To quantitiate the cell numbers of the six species, identically grown biofilms were scraped from disks and cultivated on appropriate agar media as described (Guggenheim et al., 2001b). 2.2. Oligonucleotide probes Oligonucleotide DNA probes, labeled at the 5Vend with Cy3, Cy5, fluorescein (FITC), or 6-carboxyfluorescein (FAM), were purchased from Thermo Hybaid (Interactiva Division, Ulm, Ger-
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47
many). All employed probes are listed with their sequences and specificities in Table 1. Appropriate probe sequences for the specific detection of each biofilm strain were retrieved from the literature or designed using the Alignment Slice and Probe Match services of the Ribosomal Database Project II (Maidak et al., 1999). EUB338 served as a positive control in specificity and permeability tests (Amann et al., 1990). 2.3. Fixation and permeabilization of planktonic and biofilm bacteria Planktonically grown Gram-positive strains were fixed in 50% ethanol and permeabilized by exposure to lysozyme as described (Thurnheer et al., 2001). Strains of Gram-negative cells were fixed in 4% paraformaldehyde and needed no additional permeabilization (Manz et al., 1992). Biofilms were fixed for 60 min in 4% paraformaldehyde/PBS in 24-well polystyrene cell culture plates, permeabilized by treatment with lysozyme (Sigma, Buchs, Switzerland; 70V000 U/ml in Tris – HCl pH 7.5) for 7 to 10 min at 37 jC, and rinsed with physiological saline. After the fixation step, biofilms were stored
Table 1 Characteristics of 16S rRNA-directed oligonucleotide probes used for FISH Probe
Target
Sequence of probe (5V! 3V)
ANA103 A. naeslundii CGG TTA TCC AGA AGA AGG GG EUB338 All biofilm GCT GCC TCC bacteria CGT AGG AGT EUK516 C. albicans ACC AGA CTT GCC CTC C FUS664 F. nucleatum CTT GTA GTT CCG C(C/T)T ACC TC MIT447 S. oralis CAC (C/T)CG TTC TTC TCT TAC A MIT588 S. oralis ACA GCC TTT AAC TTC AGA CTT ATC TAA SOB174 S. sobrinus TTA ACT CCT CTT ATG CGG VEI217 V. dispar AAT CCC CTC CTT CAG TGA
Reference This study (Amann et al., 1990) (Amann et al., 1995) This study
B. Paster, personal communication This study
(Paster et al., 1998), modified (Paster et al., 1998)
39
at 4 jC in physiological saline for maximum three days. 2.4. FISH FISH was performed using a modification of the method described by Manz et al. (1992). Conditions for the application of the eight oligonucleotide probes were optimized first with cells from pure cultures of both target and non-target strains, then with biofilms by gradually increasing the formamide concentration in the hybridization buffer while keeping both ionic strength and hybridization temperature constant (Manz et al., 1992; Thurnheer et al., 2001). The thus defined optimum experimental conditions for each probe-label combination are summarized in Table 2. Specifically, biofilms were pre-incubated for 15 min at 46 jC in final hybridization buffer (0.9 M NaCl, 20 mM Tris –HCl (pH 7.5), 0.01% SDS) containing a formamide concentration between 10% and 50% according to Table 2. Then, biofilms were immersed for 180 min at 46 jC in 2 ml of the same solution with up to three different probes (5– 20 Ag/ml) requiring all the same formamide concentration. After hybridization the biofilms were immersed for 15 min at 48 jC in 50 ml of washing buffer and rinsed briefly in physiological saline. Washing buffer consisted of 20 mM Tris – HCl (pH 7.5), 5 mM EDTA, 0.01% SDS, and between 159 and 636 mM NaCl depending on the formamide concentration used during hybridization. If required by the formamide concentrations of the desired probe combinations, the FISH procedure was executed two to four times per biofilm, always starting with the probe needing the highest formamide concentration (Wagner et al., 1994). After the last FISH cycle excess saline was gently aspirated from the disks without touching the biofilms, and disks were embedded upside-down in 20 Al of Mowiol (Guggenheim et al., 2001b). Embedded disks were stored at room temperature in the dark for at least 6 h prior to microscopic examination. 2.5. Staining of the entire biofilm cell population and of exocellular polysaccharides (EPS) The entire biofilm cell population was stained by a 30-min incubation at room temperature with 0.3% Syto13 in physiological saline (Molecular Probes,
40
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47
Table 2 Relationship between fluorescence intensity and label, probe concentration and formamide concentration Probe
Label
ANA103
FITC FAM Cy3 Cy5 FITC FAM Cy3 FITC FAM Cy3 Cy5 FITC FAM Cy3 Cy5 FITC FAM Cy3 Cy5 Cy3 FITC FAM Cy3 Cy5 FITC Cy3 Cy5
EUB338
EUK516
FUS664
MIT447
MIT588 SOB174
VEI217
Probe concentration (Ag/ml of hybridization buffer)
Formamide concentration (%)
Fluorescence intensity (a.u.)a
or
20
10
1
or
20 20 5
10 10 10 – 50b
4 2 4
or
5 20
10 – 50b 30
4 4
or
5 20 10 20 5 20 20 20 5 10 5 20
40 30 20 20 20 – 30b 20 20 20 – 30b 20 20 30 10
4 4 4 4 4 4 2 4 1 – 4c 4 4 –
20 20 20 5 20
20 10 40 30 – 50b 40
2 3 4 4 4
a Fluorescence intensities are expressed in arbitrary units (a.u.), on a scale from (no fluorescence) to 4+ (very brilliant fluorescent intensity), scoring visually the fluorescence intensity detectable in CSLM. b These probes yielded brilliant fluorescence over a range of formamide concentrations without any effect on specificity. Accordingly, these probes were used at the formamide concentration required by the probe(s) with which they were combined during biofilm staining. c Unspecific binding.
Leiden, The Netherlands). EPS were stained by incubating biofilms for 1 h with Calcofluor (Sigma; 10 Ag/ml solution in 10 mM sodium phosphate, pH 7.5). 2.6. CLSM and image analysis Biofilms were examined using a DM IRB E inverted microscope (Leica Mikroskopie, Wetzlar, Germany) fitted with a UV laser (Coherent, Santa
Clara, CA, USA), a He – Ne laser (Uniphase VertriebsGmbH, Eching/Munich, Germany), an Ar laser (Coherent), and a TCS SP2 computer-operated confocal laser scanning system (Leica Lasertechnik, Heidelberg, Germany). Filters were set to 400– 490 nm for detection of Calcofluor, 500– 540 nm for FITC and FAM, 570 –630 nm for Cy3, and 660 – 710 for Cy5. The fluorescent signal received from Cy5 was assigned a light blue color for better differentiation from Cy3. Confocal images were obtained using 40 (numeric aperture 1.25) and 100 (numeric aperture 1.4) oil immersion objectives. Each biofilm was scanned at randomly selected positions not close to the edges of the disks. Z-series were generated by vertical optical sectioning at every position with the slices’ thickness set to 1.018 Am. Image acquisition was done in 8 line average mode and the data were processed on a Silicon Graphics 320k visual workstation (Mountain View, CA, USA) fitted with Windows NT v4.0. Scans were recombined and volumes were estimated using Imarisk 3.1 software (Bitplane, Zurich, Switzerland).
3. Results 3.1. Probe specificity and bacterial permeabilization Specificity tests under stringent FISH conditions using planktonically cultivated cells showed that all probes displayed the anticipated specificity (data not shown). Successful probe penetration through biofilm grown cells was evaluated with both EUB338 and the specific probes for each target strain. Cell permeabilization of paraformaldehyde-fixed biofilms by exposure to lysozyme proved to be absolutely necessary but the duration of enzyme incubation strongly affected fluorescence intensity (Fig. 1). Without lysozyme pretreatment the Gram-positive bacteria A. naeslundii, S. oralis, and S. sobrinus showed no fluorescence. On the other hand, too long lysozyme exposure time resulted in reduced fluorescence intensity of the Gram-negative biofilm members F. nucleatum and V. dispar. The optimal lysozyme incubation time for this mixed-species biofilm was found to be 9 min (Fig. 1). Ethanol fixation of biofilms proved to be no alternative, resulting in very weak hybridization signals from Gram-negative bacteria.
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47
41
Fig. 1. Effect of incubation time with lysozyme on fluorescence intensity of paraformaldehyde fixed cells of A. naeslundii, S. oralis, S. sobrinus, F. nucleatum, V. dispar, and C. albicans. Fluorescence intensities are expressed in arbitrary units (a.u.); low values indicate low intensity (see footnote to Table 2 for a more detailed description of a.u.).
Fluorescence intensity with a given probe depended on both probe and formamide concentration and was further related to the fluorescence label used. Table 2 describes for each probe-label combination the probe and formamide concentrations that resulted in optimum fluorescence intensity. All biofilm stainings described in the following were performed using the here listed probe and formamide concentrations. Of all probes, only EUB338 revealed excellent fluorescence signals regardless of its label. With one exception (see below), Cy3 conjugates yielded the brightest signals. They did so at smaller probe and higher formamide concentrations than the other dyes. ANA103 displayed brilliant fluorescence with Cy3 only. With EUK516 high fluorescence intensities were obtained with all labels, but FITC, FAM, and Cy5 required fourfold higher probe concentrations and less formamide than
Cy3. Similar results were obtained with VEI217 and FUS664, except that the latter probe demanded only a twofold increase of probe concentration when labeled with FITC. A problem was encountered with the Cy3labeled MIT447 probe, which reacted with other members of the consortium if used on paraformaldehyde-fixed cells. In contrast, hybridization with MIT447-Cy3 following ethanol fixation had yielded very specific reactions with planktonically grown cells from the mitis group of streptococci (Thurnheer et al., 2001). If MIT447 was labeled with FAM, the specificity with paraformaldehyde-fixed biofilm cells of S. oralis was restored. Whenever red labeling of S. oralis was required another probe (MIT588-Cy3) was used. Hybridization of S. sobrinus OMZ 176 in biofilms yielded only satisfactory fluorescence intensities, if SOB174 was labeled with Cy5. Cy3-conjugated
42
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47
SOB174 yielded significantly weaker fluorescence, and no signals at all were observed with FITC or FAM conjugates of SOB174. 3.2. Multiple staining of biofilms Two biofilms grown in parallel were stained for complementary strain triplets as S. oralis, F. nucleatum, and V. dispar were visualized with MIT447FAM, FUS664-Cy3, and VEI217-Cy5, C. albicans, A. naeslundii, and S. sobrinus with EUK516-FITC, ANA103-Cy3, and SOB174-Cy5, respectively (Fig. 2). The figure demonstrates representatively that it was possible to simultaneously stain by FISH Gram-positives, Gram-negatives, and a yeast, even in the presence of A. naeslundii, which is considered to be particularly tough to permeabilize. A. naeslundii and S. oralis were widely distributed in these biofilms, whereas V. dispar was restricted to isolated microcolonies composed of hundreds of cells. C. albicans was present sparsely in form of isolated single cells. Fig. 3 shows a single biofilm stained for all six species. Since we could distinguish simultaneously only three fluorescence colors, two morphologically distinct organisms were marked with the same color. Distinction between green Veillonellae and green Candidia cells was easy, whereas A. naeslundii and S. oralis, or F. nucleatum and S. sobrinus could not be distinguished, although a 100 objective and thus maximum magnification was used to make this micrograph. The upper part of the biofilm was almost completely stained, predominately by probes to A. naeslundii and S. oralis, whereas the lower regions showed unstained areas.
The biofilm shown in Fig. 4 was hybridized with probes ANA103-Cy3, EUK516-FITC, and SOB174Cy5 to identify A. naeslundii, C. albicans, and S. sobrinus. In addition, it was stained with Calcofluor to visualize EPS. The binding of Calcofluor in Fig. 4 suggestes retrospectively that the unstained regions observed in Fig. 3 may be explained by the presence of EPS. Again, the upper part of the biofilm consisted predominantly of A. naeslundii, whereas C. albicans was sparsely present and S. sobrinus was confined to isolated colonies of a few hundred cells. To verify these visual assessments, cells from nine biofilms of three independent triplicate experiments were scraped off and then cultured. In these experiments the different species accounted on average for 48.5 F 12.7% (SD) (A. naeslundii), 30.0 F 11.0% (S. oralis), 11.2 F 3.6% (S. sobrinus), 8.5 F 5.5% (V. dispar), 0.3 F 0.37% (F. nucleatum), and 0.02 F 0.01% (C. albicans) of the total CFU. Furthermore, the spatial abundance of each of the six taxa was estimated using data from all biofilms used in this study that had been labeled only with one green-, red- or bluelabeled probe to warrant the unequivocal assignment of the label to a biofilm strain. The total volume occupied by all cells of the biofilm was visualized by staining with Syto13. Based on data from 37 biofilms from >12 separate biofilm cultures, Syto13 stained cells accounted for 31.9 F 5.2% of the total area of image filled by the biofilm (defined as the area covered by cells, EPS and non-stained extracellular matrix). The data for the individual species were derived from 16 separate biofilms from >5 separate biofilm cultures with the exception of C. albicans where 10 analyses were evaluated. The relative contributions of the species were as follows: A. naeslundii 47.9 F 5.7%, S.
Fig. 2. CLSM images showing perpendicular xy-, xz-, and yz-planes through randomly selected points of 64.5-h biofilms. (A) The biofilm of (A) was stained first for V. dispar (blue) with VEI217-Cy5 at 40% formamide and then during a second FISH for both S. oralis (green) and F. nucleatum (red), using MIT447-FAM and FUS664-Cy3 at 25% formamide. (B) This biofilm was stained for C. albicans (green) with EUK516-FITC at 30% formamide followed by a 2nd FISH for both A. naeslundii (red) and S. sobrinus (blue) with ANA103-Cy3 and SOB174-Cy5 at 10% formamide. Characteristic, homogeneous colonies formed by V. dispar are visible at the top of the biofilm. The toughest organism to label was S. sobrinus (blue in image B) because the SOB174 probe yielded a comparatively weak signal. Fig. 3. CLSM images showing perpendicular xy-, xz-, and yz-planes through a randomly selected point of a 64.5-h biofilm. This biofilm was stained for all six biofilm strains but with only three colors at our disposition. Performing three consecutive FISHs, C. albicans (green), V. dispar (green) and S. oralis (red) were stained first using EUK516-FITC, VEI217-FITC and MIT588-Cy3, respectively, at 30% formamide. Next F. nucleatum (blue) was stained with FUS664-Cy5 at 20% formamide, and finally A. naeslundii (red) and S. sobrinus (blue) were stained using ANA103-Cy3 and SOB174Cy-5 at 10% formamide.
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47
Fig. 2.
Fig. 3.
43
44
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47
Fig. 4. CLSM images showing perpendicular xy-, xz-, and yz-planes through a randomly selected point of a 64.5-h multiply stained biofilm. C. albicans (green), A. naeslundii (red) and S. sobrinus (light blue) were stained first with EUK516-FITC at 30% formamide, followed by a second FISH with ANA103-Cy3 and SOB174-Cy-5 at 10% formamide. EPS (dark blue) was stained with Calcofluor after the completion of FISHs.
oralis 23.4 F 3.4%, S. sobrinus 10.7 F 4.8%, V. dispar 5.5 F 2.6%, F. nucleatum 7.3 F 2.1%, and C. albicans 1.0 F 0.9%, of the total cellular volume (Syto13 positive cells).
4. Discussion In the present study we developed a procedure for the simultaneous spatial analysis of the distribution of both Gram-positive and Gram-negative bacteria in a multi-species biofilm model. The technique combines FISH with oligonucleotide probes to specific 16S rRNA sequences and CLSM to assess fixed but otherwise intact biofilms. Key elements of the pro-
cedure are: adequate fixation, optimal permeabilization, and careful selection of the best combination between probe, fluorescence label, and formamide concentration. Although established FISH protocols use ethanol for the fixation of Gram-positive cells (Roller et al., 1994), we found no negative effect from paraformaldehyde fixation of our mixed-species biofilms. However, preliminary experiments showed that ethanol fixation of Gram-negatives followed by in situ hybridization resulted in reduced staining (data not shown). The need for permeabilizing the cell wall of Gram-positive organisms prior to in situ hybridization is well documented (Beimfohr et al., 1993; Moter and Go¨bel, 2000). The difficulty encountered with bio-
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47
films was to obtain optimal fluorescence signals from Gram-positives without losing fluorescence of Gramnegatives. This could only be achieved by keeping the permeabilization time strictly to 9 min. Fluorescence intensity in the biofilms was further influenced by the fluorescence label used. Manz et al. (1998) found that, in general, FITC labeled probes required lower formamide concentrations than Cy3 conjugated oligonucleotides. This is confirmed by the observations from this study. Cy3 is currently one of the brightest fluorochromes characterized by a large extinction coefficient. In comparison, FITC has less favorable features, in particular it is not photostable and its fluorescence intensity is pH sensitive. Maximum emission is reached only in the basic pH range (Brelje et al., 1993), which likely explains why we observed reduced fluorescence intensity with FITC-labeled probes in combination with a nearly neutral hybridization buffer. Multiple staining of biofilms with fluorescent compounds has been described before, using nucleic acid specific dyes (acridine orange, DAPI, stains from Molecular Probes’ Syto series) or other fluorochromes (Neu et al., 2002). Multiple staining of biofilms by FISH is very common; Okabe et al. (1999) stained Gram-negative filamentous bacteria in nitrifying biofilms using oligonucleotide probes with different labels. The same method was used by Wecke et al. (2000) to monitor the development of bacterial biofilms in human periodontal pockets. Triple hybridizations using FITC, Cy3, and Cy5 labeled oligonucleotide probes for the simultaneous in situ visualization of the genetic diversity of activated sludge organisms or in a trickling filter biofilm have been described by Amann et al. (1996) and Schmid et al. (2000), respectively. All these studies established that FISH with multiple probes, either simultaneously or in sequence, of biofilms is feasible, but the technique was applied always to either Gram-positive or Gramnegative organisms. Perry-O’Keefe et al. (2001) described a method using peptide nucleic acid instead of DNA probes for the simultaneous identification of Gram-positive and Gram-negative organisms. In their protocol 1 h of paraformaldehyde fixation was enough to permeabilize Gram-positive cells. This was not so with our biofilm model. Treatment with lysozyme for 9 min was an absolute necessity for successful probe penetration into the Gram-positive bacteria of the
45
consortium. However, there are limitations to the multiple staining since only three different fluorescent probe colors are available. To differentiate all six species, two separate biofilms had to be analyzed (Fig. 3). Additionally, EPS, such as predominantly a(1 ! 3) linked mutan (produced by S. sobrinus) and a(1 ! 6) linked dextrans (produced by S. oralis and S. sobrinus) were visualized by staining the biofilm with Calcofluor. Preliminary experiments with both mutan and dextrans revealed, that a-(1 ! 3)-and a-(1 ! 6) linked polysaccharides did react with Calcofluor (data not shown), although Calcofluor was originally described for staining of h-(1 ! 3)- and h-(1 ! 4) linked polysaccharides (Maeda and Ishida, 1967). It is very difficult to determine the labeling efficiency of a given probe in a mixed biofilm by FISH/ CLSM, since unstained cells remain undetectable and cannot be enumerated or estimated by volume. This problem must be addressed in future studies, possibly by combining FISH and immunofluorescence with species-specific fluorescence-labeled F(ab) fragments. At present, we can only compare the total cell volumes of the six species estimated by FISH/CLSM with the cell volume of the total cell population stained using Syto13. It is noteworthy that the mean volumes detected for the six biofilm strains amounted together to 95.8% of the volume covered by Syto13 stained cells. This suggests, albeit quite indirectly, that a large portion of the cells was in fact labeled by FISH. This view is supported by the observation that proportional relationship between four of the six species was similar when estimated by FISH/CLSM and by cultivation of scraped off biofilm cells. Only F. nucleatum and C. albicans showed 20 and 50 times lower proportions in culture analyses. A likely reason for this discrepancy is that the selective agar media for F. nucleatum and C. albicans were too suppressive. In summary, our results show that it is possible to permeabilize both Gram-negative and Gram-positive bacteria simultaneously and to perform several consecutive FISHs with multiple rRNA probes to identify simultaneously several members of the six-species biofilm. To our knowledge, this is the first time that a multispecies biofilm of supragingival plaque consisting of Gram-negative and Gram-positive organisms was analyzed using the FISH technique. In combination with the nondestructive sectioning properties of CLSM, this modified FISH technique enabled the
46
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47
visualization of the three-dimensional arrangement of the microbial population as well as the estimation of the quantitative distribution of the biofilm species. Staining of the bulk of the EPS with Calcofluor reassured us that black areas in the biofilm were not due to substantial losses of microorganisms during the staining procedure. The procedure facilitates analyses of the microstructure and of spatial alterations in biofilm composition that may occur in response to environmental challenges and should prove useful in addressing some fundamental questions about the nature of dental biofilms. Acknowledgements We thank T. Ba¨chi and M. Ho¨chli at the Central Electron Microscopy Laboratory of the University of Zu¨rich for the ability to use the CLSM, A. Meier and V. Osterwalder for excellent technical assistance. References Amann, R.I., Binder, B.J., Olson, R.J., Chisholm, S.W., Devereux, R., Stahl, D.A., 1990. Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microbiol. 56, 1919 – 1925. Amann, R.I., Ludwig, W., Schleifer, K.-H., 1995. Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol. Rev. 59, 143 – 169. Amann, R., Snaidr, J., Wagner, M., Ludwig, W., Schleifer, K.H., 1996. In situ visualization of high genetic diversity in a natural microbial community. J. Bacteriol. 178, 3496 – 3500. Beimfohr, C., Krause, A., Amann, R., Ludwig, W., Schleifer, K.-H., 1993. In situ identification of Lactococci, Enterococci and Streptococci. Syst. Appl. Microbiol. 16, 450 – 456. Brelje, T.C., Wessendorf, M.W., Sorenson, R.L., 1993. Multicolor laser scanning confocal immunofluorescence microscopy: practical application and limitations. Methods Cell Biol. 38, 97 – 181. Costerton, J.W., Lewandowski, Z., Caldwell, D.E., Korber, D.R., Lappin-Scott, H.M., 1995. Microbial biofilms. Annu. Rev. Microbiol. 49, 711 – 745. Davey, M.E., O’Toole, G.A., 2000. Microbial biofilms: from ecology to molecular genetics. Microbiol. Mol. Biol. Rev. 64, 847 – 867. Gmu¨r, R., Guggenheim, B., 1983. Antigenic heterogeneity of Bacteroides intermedius as recognized by monoclonal antibodies. Infect. Immun. 42, 459 – 470. Guggenheim, B., Giertsen, E., Schu¨pbach, P., Shapiro, S., 2001a. Validation of an in vitro biofilm model of supragingival plaque. J. Dent. Res. 80, 363 – 370.
Guggenheim, M., Shapiro, S., Gmu¨r, R., Guggenheim, B., 2001b. Spatial arrangements and associative behavior of species in an in vitro oral biofilm model. Appl. Environ. Microbiol. 67, 1343 – 1350. Hermanowicz, S.W., 2003. Biofilm structure: an interplay of models and experiments. In: Wuertz, S., Bishop, P.L., Wilderer, P.A. (Eds.), Biofilms in wastewater treatment—an interdisciplinary approach. International Water Association Publishing, London, United Kingdom, pp. 32 – 47. Lawrence, J.R., Korber, D.R., Hoyle, B.D., Costerton, J.W., Caldwell, D.E., 1991. Optical sectioning of microbial biofilms. J. Bacteriol. 173, 6558 – 6567. Maeda, H., Ishida, N., 1967. Specificity of binding of hexopyranosyl polysaccharides with fluorescent brightener. J. Biochem. 62, 276 – 278. Maidak, B.L., Cole, J.R., Parker Jr., C.T., Garrity, G.M., Larsen, N., Li, B., Lilburn, T.G., McCaughey, M.J., Olsen, G.J., Overbeek, R., Pramanik, S., Schmidt, T.M., Tiedje, J.M., Woese, C.R., 1999. A new version of the RDP (Ribosomal Database Project). Nucleic Acids Res. 27, 171 – 173. Manz, W., Amann, R., Ludwig, W., Wagner, M., Schleifer, K.-H., 1992. Phylogenetic oligodeoxynucleotide probes for the major subclasses of proteobacteria: problems and solutions. Syst. Appl. Microbiol. 15, 593 – 600. Manz, W., Eisenbrecher, M., Neu, T.R., Szewzyk, U., 1998. Abundance and spatial organization of Gram-negative sulfate-reducing bacteria in activated sludge investigated by in situ probing with specific 16S rRNA targeted oligonucleotides. FEMS Microbiol. Ecol. 25, 43 – 61. Marsh, P.D., Bradshaw, D.J., 1995. Dental plaque as a biofilm. J. Ind. Microbiol. 15, 169 – 175. Moter, A., Go¨bel, U., 2000. Fluorescence in situ hybridization (FISH) for direct visualization of microorganism. J. Microbiol. Methods 41, 85 – 112. Neu, T.R., Kuhlicke, U., Lawrence, J.R., 2002. Assessment of fluorochromes for two-photon laser scanning microscopy of biofilms. Appl. Environ. Microbiol. 68, 901 – 909. Okabe, S., Satoh, H., Watanabe, Y., 1999. In situ analysis of nitrifying biofilms as determined by in situ hybridization and the use of microelectrodes. Appl. Environ. Microbiol. 65, 3182 – 3191. Paster, B.J., Bartoszyk, I.M., Dewhirst, F.E., 1998. Identification of oral streptococci using PCR-based, reverse-capture, checkerboard hybridization. Methods Cell Sci. 20, 223 – 231. Perry-O’Keefe, H., Rigby, S., Oliveira, K., Sorensen, D., Slender, H., Coull, J., Hyldig-Nielsen, J.J., 2001. Identification of indicator microorganisms using a standardized PNA FISH method. J. Microbiol. Methods 47, 281 – 292. Roller, C., Wagner, M., Amann, R., Ludwig, W., Schleifer, K.-H., 1994. In situ probing of gram-positive bacteria with high DNA G + C content using 23S rRNA-targeted oligonucleotides. Microbiology 140, 2849 – 2858. Schmid, M., Twachtmann, U., Klein, M., Strous, M., Juretschko, S., Jetten, M., Metzger, J.W., Schleifer, K.H., Wagner, M., 2000. Molecular evidence for genus level diversity of bacteria capable of catalyzing anaerobic ammonium oxidation. Syst. Appl. Microbiol. 23, 93 – 106. Shapiro, S., Giertsen, E., Guggenheim, B., 2002. An in vitro oral
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47 biofilm model for comparing efficacies of antimicrobial mouthrinses. Caries Res. 36, 93 – 100. Thurnheer, T., Gmu¨r, R., Giertsen, E., Guggenheim, B., 2001. Automated fluorescent in situ hybridization for the specific detection and quantification of oral streptococci in dental plaque. J. Microbiol. Methods 44, 39 – 47. Thurnheer, T., Gmu¨r, R., Shapiro, S., Guggenheim, B., 2003. Mass transport of macromolecules within an in vitro model of supragingival plaque. Appl. Environ. Microbiol. 69, 1702 – 1709. Wagner, M., Amann, R., Ka¨mpfer, P., Assmus, B., Hartmann, A.,
47
Hutzler, P., Springer, N., Schleifer, K.-H., 1994. Identification and in situ detection of Gram-negative filamentous bacteria in activated sludge. Syst. Appl. Microbiol. 17, 405 – 417. Wecke, J., Kersten, T., Madela, K., Moter, A., Go¨bel, U.B., Friedmann, A., Bernimoulin, J.P., 2000. A novel technique for monitoring the development of bacterial biofilms in human periodontal pockets. FEMS Microbiol. Lett. 191, 95 – 101. Wimpenny, J., Manz, W., Szewzyk, U., 2000. Heterogeneity in biofilms. FEMS Microbiol. Rev. 24, 661 – 671.
Multiplex FISH analysis of a six-species bacterial biofilm T. Thurnheer *, R. Gmu¨r, B. Guggenheim Section for Oral Microbiology and General Immunology, Institute for Oral Biology, University of Zu¨rich, Plattenstrasse 11, CH-8028 Zu¨rich, Switzerland Received 10 April 2003; received in revised form 11 July 2003; accepted 3 September 2003
Abstract Established procedures use different and seemingly incompatible experimental protocols for fluorescent in situ hybridization (FISH) with Gram-negative and Gram-positive bacteria. The aim of this study was to develop a procedure, based on FISH and confocal laser scanning microscopy (CLSM), for the analysis of the spatial organization of in vitro biofilms containing both Gram-negative and Gram-positive oral bacteria. Biofilms composed of the six oral species Actinomyces naeslundii, Candida albicans, Fusobacterium nucleatum, Streptococcus oralis, Streptococcus sobrinus, and Veillonella dispar were grown anaerobically for 64.5 h at 37 jC on hydroxyapatite disks preconditioned with saliva. Conditions for the simultaneous in situ hybridization of both Gram-negative and Gram-positive bacteria were sought by systematic variation of fixation and exposure to lysozyme. After fixation and permeabilization biofilms were labeled by FISH with 16S rRNA-targeted oligonucleotide probes ANA103 (for the detection of A. naeslundii), EUK116 (C. albicans), FUS664 (F. nucleatum), MIT447 and MIT588 (S. oralis), SOB174 (S. sobrinus), and VEI217 (V. dispar). Probes were used as 6-FAM, Cy3 or Cy5 conjugates, resulting in green, orange-red or deep-red fluorescence of target cells, respectively. Thus, with two independent triple-hybridizations with three probes carrying different fluorescence-tags, all six species could be visualized. Results show that the simultaneous investigation by FISH of complex biofilms composed of multiple bacterial species with differential Gram-staining properties is possible. In combination with the optical sectioning properties of CLSM the technique holds great promise for the analysis of spatial alterations in biofilm composition in response to environmental challenges. D 2003 Elsevier B.V. All rights reserved. Keywords: 16S rRNA; Biofilm; CLSM; Fluorescent in situ hybridization
1. Introduction Biofilms are formed after rapid attachment and growth of microorganisms on a broad range of surfaces in contact with natural fluids. Consisting of single cells, cell aggregates, and microcolonies embedded
* Corresponding author. Tel.: +41-1-634-3256; fax: +41-1-6344310. E-mail address: [email protected] (T. Thurnheer). 0167-7012/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.mimet.2003.09.003
in an exocellular polymeric matrix gel, biofilms are a most common phenomenon on Earth and have been the subject of intense experimental and theoretical scrutiny over the past decade (Costerton et al., 1995; Marsh and Bradshaw, 1995; Wimpenny et al., 2000; Hermanowicz, 2003). Biofilm architecture and the physiological status of the cells contained within these structures are of prime interest for clinical, industrial, and environmental microbiology. Dental plaques are naturally occurring complex biofilms involved in the development of caries and
38
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47
periodontal diseases. Knowledge of their composition, their members’ metabolism and know-how on the successful manipulation of mixed bacterial biofilms may lead to improved preventive measures against these common diseases. Investigations of plaque in vivo are difficult due to the high degree of variability between individuals and since the site-specific microbial composition is reflecting the influences of a multitude of uncontrollable endogeneous and exogeneous dietary factors. In addition, working with humans always raises ethical questions. Therefore, we have developed and applied a six-species in vitro biofilm model of supragingival plaque (Guggenheim et al., 2001a) for the examination of the spatial arrangement and associative behavior of microorganisms, the study of the efficacy of antimicrobial mouthrinses, or the investigation of mass transport of macromolecules (Guggenheim et al., 2001b; Shapiro et al., 2002; Thurnheer et al., 2003). For analyses of the spatial arrangement of bacteria in a multi-species biofilm techniques, which use specific bacterial cell markers and maintain the biofilm’s natural architecture are required. The aim of this study was to develop a fluorescent in situ hybridization (FISH) technique using probes to target-specific 16S rRNA sequences and to combine it with confocal laser scanning microscopy (CLSM) for the simultaneous analysis of the spatial distribution of both Gram-positive and Gram-negative bacteria in biofilms. FISH is a recognized tool for the specific and sensitive identification of target organisms within complex microbial communities (Amann et al., 1995). Visualization of FISH labeled cells in biofilms can be carried out by techniques, including fluorescence microscopy and CLSM (Lawrence et al., 1991; Davey and O’Toole, 2000; Wimpenny et al., 2000). With biofilms, CLSM is often to be preferred, since it allows a three-dimensional noninvasive visualization of cells and the computational reconstruction of mature biofilms without distortion of their structure. However, the crucial problem was to find experimental conditions that render cell walls of Gram-positive bacteria penetrable to the probes without the simultaneous loss of signal from Gram-negatives that are simultaneously labeled with fluorescent oligonucleotide probes. Moreover, they should allow several consecutive hybridization procedures without harming the biofilm structure.
2. Materials and methods 2.1. Preparation of biofilms Actinomyces naeslundii OMZ 745, Veillonella dispar ATCC 17748T (OMZ 493), Fusobacterium nucleatum KP-F2 (OMZ 596), Streptococcus sobrinus OMZ 176, Streptococcus oralis SK248 (OMZ 607), and Candida albicans OMZ 110 were used as inocula for biofilm formation (Guggenheim et al., 2001a; Shapiro et al., 2002). Biofilms were grown in 24well polystyrene cell culture plates on sintered hydroxyapatite disks as described (Thurnheer et al., 2003). In brief, hydroxyapatite disks (F 10.6 mm) that had been preincubated (pellicle-coated) in processed whole unstimulated saliva were covered with 1.6 ml of substrate composed of 70% saliva + 30% mFUM. mFUM corresponds to a well established tryptone-yeast based broth medium designated as FUM (Gmu¨r and Guggenheim, 1983) and modified by supplementing 67 mmol/l Sørensen’s buffer (final pH 7.2). The carbohydrate concentration in mFUM was 0.3% (w/v), and consisted of either glucose (0– 16.5 h of biofilm cultivation) or a 1:1 (w/w) mixture of glucose and sucrose (16.5 – 64.5 h of biofilm cultivation). Wells were inoculated with mixed cell suspensions (200 Al) prepared from equal volumes and densities of each species and incubated anaerobically at 37 jC. Substrate was replenished after dipping (see below) at 16.5 and 40.5 h by aspirating spent substrate and adding back fresh substrate composed of saliva + mFUM (1.6 ml). At 16.5, 20.5, 24.5, 40.5, 44.5, 48.5, and 64.5 h biofilms were washed by three consecutive dips in 2 ml of sterile physiological saline (1 min per dip, room temperature). After the 64.5-h dip-wash biofilms were fixed immediately with paraformaldehyde (see below). To quantitiate the cell numbers of the six species, identically grown biofilms were scraped from disks and cultivated on appropriate agar media as described (Guggenheim et al., 2001b). 2.2. Oligonucleotide probes Oligonucleotide DNA probes, labeled at the 5Vend with Cy3, Cy5, fluorescein (FITC), or 6-carboxyfluorescein (FAM), were purchased from Thermo Hybaid (Interactiva Division, Ulm, Ger-
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47
many). All employed probes are listed with their sequences and specificities in Table 1. Appropriate probe sequences for the specific detection of each biofilm strain were retrieved from the literature or designed using the Alignment Slice and Probe Match services of the Ribosomal Database Project II (Maidak et al., 1999). EUB338 served as a positive control in specificity and permeability tests (Amann et al., 1990). 2.3. Fixation and permeabilization of planktonic and biofilm bacteria Planktonically grown Gram-positive strains were fixed in 50% ethanol and permeabilized by exposure to lysozyme as described (Thurnheer et al., 2001). Strains of Gram-negative cells were fixed in 4% paraformaldehyde and needed no additional permeabilization (Manz et al., 1992). Biofilms were fixed for 60 min in 4% paraformaldehyde/PBS in 24-well polystyrene cell culture plates, permeabilized by treatment with lysozyme (Sigma, Buchs, Switzerland; 70V000 U/ml in Tris – HCl pH 7.5) for 7 to 10 min at 37 jC, and rinsed with physiological saline. After the fixation step, biofilms were stored
Table 1 Characteristics of 16S rRNA-directed oligonucleotide probes used for FISH Probe
Target
Sequence of probe (5V! 3V)
ANA103 A. naeslundii CGG TTA TCC AGA AGA AGG GG EUB338 All biofilm GCT GCC TCC bacteria CGT AGG AGT EUK516 C. albicans ACC AGA CTT GCC CTC C FUS664 F. nucleatum CTT GTA GTT CCG C(C/T)T ACC TC MIT447 S. oralis CAC (C/T)CG TTC TTC TCT TAC A MIT588 S. oralis ACA GCC TTT AAC TTC AGA CTT ATC TAA SOB174 S. sobrinus TTA ACT CCT CTT ATG CGG VEI217 V. dispar AAT CCC CTC CTT CAG TGA
Reference This study (Amann et al., 1990) (Amann et al., 1995) This study
B. Paster, personal communication This study
(Paster et al., 1998), modified (Paster et al., 1998)
39
at 4 jC in physiological saline for maximum three days. 2.4. FISH FISH was performed using a modification of the method described by Manz et al. (1992). Conditions for the application of the eight oligonucleotide probes were optimized first with cells from pure cultures of both target and non-target strains, then with biofilms by gradually increasing the formamide concentration in the hybridization buffer while keeping both ionic strength and hybridization temperature constant (Manz et al., 1992; Thurnheer et al., 2001). The thus defined optimum experimental conditions for each probe-label combination are summarized in Table 2. Specifically, biofilms were pre-incubated for 15 min at 46 jC in final hybridization buffer (0.9 M NaCl, 20 mM Tris –HCl (pH 7.5), 0.01% SDS) containing a formamide concentration between 10% and 50% according to Table 2. Then, biofilms were immersed for 180 min at 46 jC in 2 ml of the same solution with up to three different probes (5– 20 Ag/ml) requiring all the same formamide concentration. After hybridization the biofilms were immersed for 15 min at 48 jC in 50 ml of washing buffer and rinsed briefly in physiological saline. Washing buffer consisted of 20 mM Tris – HCl (pH 7.5), 5 mM EDTA, 0.01% SDS, and between 159 and 636 mM NaCl depending on the formamide concentration used during hybridization. If required by the formamide concentrations of the desired probe combinations, the FISH procedure was executed two to four times per biofilm, always starting with the probe needing the highest formamide concentration (Wagner et al., 1994). After the last FISH cycle excess saline was gently aspirated from the disks without touching the biofilms, and disks were embedded upside-down in 20 Al of Mowiol (Guggenheim et al., 2001b). Embedded disks were stored at room temperature in the dark for at least 6 h prior to microscopic examination. 2.5. Staining of the entire biofilm cell population and of exocellular polysaccharides (EPS) The entire biofilm cell population was stained by a 30-min incubation at room temperature with 0.3% Syto13 in physiological saline (Molecular Probes,
40
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47
Table 2 Relationship between fluorescence intensity and label, probe concentration and formamide concentration Probe
Label
ANA103
FITC FAM Cy3 Cy5 FITC FAM Cy3 FITC FAM Cy3 Cy5 FITC FAM Cy3 Cy5 FITC FAM Cy3 Cy5 Cy3 FITC FAM Cy3 Cy5 FITC Cy3 Cy5
EUB338
EUK516
FUS664
MIT447
MIT588 SOB174
VEI217
Probe concentration (Ag/ml of hybridization buffer)
Formamide concentration (%)
Fluorescence intensity (a.u.)a
or
20
10
1
or
20 20 5
10 10 10 – 50b
4 2 4
or
5 20
10 – 50b 30
4 4
or
5 20 10 20 5 20 20 20 5 10 5 20
40 30 20 20 20 – 30b 20 20 20 – 30b 20 20 30 10
4 4 4 4 4 4 2 4 1 – 4c 4 4 –
20 20 20 5 20
20 10 40 30 – 50b 40
2 3 4 4 4
a Fluorescence intensities are expressed in arbitrary units (a.u.), on a scale from (no fluorescence) to 4+ (very brilliant fluorescent intensity), scoring visually the fluorescence intensity detectable in CSLM. b These probes yielded brilliant fluorescence over a range of formamide concentrations without any effect on specificity. Accordingly, these probes were used at the formamide concentration required by the probe(s) with which they were combined during biofilm staining. c Unspecific binding.
Leiden, The Netherlands). EPS were stained by incubating biofilms for 1 h with Calcofluor (Sigma; 10 Ag/ml solution in 10 mM sodium phosphate, pH 7.5). 2.6. CLSM and image analysis Biofilms were examined using a DM IRB E inverted microscope (Leica Mikroskopie, Wetzlar, Germany) fitted with a UV laser (Coherent, Santa
Clara, CA, USA), a He – Ne laser (Uniphase VertriebsGmbH, Eching/Munich, Germany), an Ar laser (Coherent), and a TCS SP2 computer-operated confocal laser scanning system (Leica Lasertechnik, Heidelberg, Germany). Filters were set to 400– 490 nm for detection of Calcofluor, 500– 540 nm for FITC and FAM, 570 –630 nm for Cy3, and 660 – 710 for Cy5. The fluorescent signal received from Cy5 was assigned a light blue color for better differentiation from Cy3. Confocal images were obtained using 40 (numeric aperture 1.25) and 100 (numeric aperture 1.4) oil immersion objectives. Each biofilm was scanned at randomly selected positions not close to the edges of the disks. Z-series were generated by vertical optical sectioning at every position with the slices’ thickness set to 1.018 Am. Image acquisition was done in 8 line average mode and the data were processed on a Silicon Graphics 320k visual workstation (Mountain View, CA, USA) fitted with Windows NT v4.0. Scans were recombined and volumes were estimated using Imarisk 3.1 software (Bitplane, Zurich, Switzerland).
3. Results 3.1. Probe specificity and bacterial permeabilization Specificity tests under stringent FISH conditions using planktonically cultivated cells showed that all probes displayed the anticipated specificity (data not shown). Successful probe penetration through biofilm grown cells was evaluated with both EUB338 and the specific probes for each target strain. Cell permeabilization of paraformaldehyde-fixed biofilms by exposure to lysozyme proved to be absolutely necessary but the duration of enzyme incubation strongly affected fluorescence intensity (Fig. 1). Without lysozyme pretreatment the Gram-positive bacteria A. naeslundii, S. oralis, and S. sobrinus showed no fluorescence. On the other hand, too long lysozyme exposure time resulted in reduced fluorescence intensity of the Gram-negative biofilm members F. nucleatum and V. dispar. The optimal lysozyme incubation time for this mixed-species biofilm was found to be 9 min (Fig. 1). Ethanol fixation of biofilms proved to be no alternative, resulting in very weak hybridization signals from Gram-negative bacteria.
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47
41
Fig. 1. Effect of incubation time with lysozyme on fluorescence intensity of paraformaldehyde fixed cells of A. naeslundii, S. oralis, S. sobrinus, F. nucleatum, V. dispar, and C. albicans. Fluorescence intensities are expressed in arbitrary units (a.u.); low values indicate low intensity (see footnote to Table 2 for a more detailed description of a.u.).
Fluorescence intensity with a given probe depended on both probe and formamide concentration and was further related to the fluorescence label used. Table 2 describes for each probe-label combination the probe and formamide concentrations that resulted in optimum fluorescence intensity. All biofilm stainings described in the following were performed using the here listed probe and formamide concentrations. Of all probes, only EUB338 revealed excellent fluorescence signals regardless of its label. With one exception (see below), Cy3 conjugates yielded the brightest signals. They did so at smaller probe and higher formamide concentrations than the other dyes. ANA103 displayed brilliant fluorescence with Cy3 only. With EUK516 high fluorescence intensities were obtained with all labels, but FITC, FAM, and Cy5 required fourfold higher probe concentrations and less formamide than
Cy3. Similar results were obtained with VEI217 and FUS664, except that the latter probe demanded only a twofold increase of probe concentration when labeled with FITC. A problem was encountered with the Cy3labeled MIT447 probe, which reacted with other members of the consortium if used on paraformaldehyde-fixed cells. In contrast, hybridization with MIT447-Cy3 following ethanol fixation had yielded very specific reactions with planktonically grown cells from the mitis group of streptococci (Thurnheer et al., 2001). If MIT447 was labeled with FAM, the specificity with paraformaldehyde-fixed biofilm cells of S. oralis was restored. Whenever red labeling of S. oralis was required another probe (MIT588-Cy3) was used. Hybridization of S. sobrinus OMZ 176 in biofilms yielded only satisfactory fluorescence intensities, if SOB174 was labeled with Cy5. Cy3-conjugated
42
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47
SOB174 yielded significantly weaker fluorescence, and no signals at all were observed with FITC or FAM conjugates of SOB174. 3.2. Multiple staining of biofilms Two biofilms grown in parallel were stained for complementary strain triplets as S. oralis, F. nucleatum, and V. dispar were visualized with MIT447FAM, FUS664-Cy3, and VEI217-Cy5, C. albicans, A. naeslundii, and S. sobrinus with EUK516-FITC, ANA103-Cy3, and SOB174-Cy5, respectively (Fig. 2). The figure demonstrates representatively that it was possible to simultaneously stain by FISH Gram-positives, Gram-negatives, and a yeast, even in the presence of A. naeslundii, which is considered to be particularly tough to permeabilize. A. naeslundii and S. oralis were widely distributed in these biofilms, whereas V. dispar was restricted to isolated microcolonies composed of hundreds of cells. C. albicans was present sparsely in form of isolated single cells. Fig. 3 shows a single biofilm stained for all six species. Since we could distinguish simultaneously only three fluorescence colors, two morphologically distinct organisms were marked with the same color. Distinction between green Veillonellae and green Candidia cells was easy, whereas A. naeslundii and S. oralis, or F. nucleatum and S. sobrinus could not be distinguished, although a 100 objective and thus maximum magnification was used to make this micrograph. The upper part of the biofilm was almost completely stained, predominately by probes to A. naeslundii and S. oralis, whereas the lower regions showed unstained areas.
The biofilm shown in Fig. 4 was hybridized with probes ANA103-Cy3, EUK516-FITC, and SOB174Cy5 to identify A. naeslundii, C. albicans, and S. sobrinus. In addition, it was stained with Calcofluor to visualize EPS. The binding of Calcofluor in Fig. 4 suggestes retrospectively that the unstained regions observed in Fig. 3 may be explained by the presence of EPS. Again, the upper part of the biofilm consisted predominantly of A. naeslundii, whereas C. albicans was sparsely present and S. sobrinus was confined to isolated colonies of a few hundred cells. To verify these visual assessments, cells from nine biofilms of three independent triplicate experiments were scraped off and then cultured. In these experiments the different species accounted on average for 48.5 F 12.7% (SD) (A. naeslundii), 30.0 F 11.0% (S. oralis), 11.2 F 3.6% (S. sobrinus), 8.5 F 5.5% (V. dispar), 0.3 F 0.37% (F. nucleatum), and 0.02 F 0.01% (C. albicans) of the total CFU. Furthermore, the spatial abundance of each of the six taxa was estimated using data from all biofilms used in this study that had been labeled only with one green-, red- or bluelabeled probe to warrant the unequivocal assignment of the label to a biofilm strain. The total volume occupied by all cells of the biofilm was visualized by staining with Syto13. Based on data from 37 biofilms from >12 separate biofilm cultures, Syto13 stained cells accounted for 31.9 F 5.2% of the total area of image filled by the biofilm (defined as the area covered by cells, EPS and non-stained extracellular matrix). The data for the individual species were derived from 16 separate biofilms from >5 separate biofilm cultures with the exception of C. albicans where 10 analyses were evaluated. The relative contributions of the species were as follows: A. naeslundii 47.9 F 5.7%, S.
Fig. 2. CLSM images showing perpendicular xy-, xz-, and yz-planes through randomly selected points of 64.5-h biofilms. (A) The biofilm of (A) was stained first for V. dispar (blue) with VEI217-Cy5 at 40% formamide and then during a second FISH for both S. oralis (green) and F. nucleatum (red), using MIT447-FAM and FUS664-Cy3 at 25% formamide. (B) This biofilm was stained for C. albicans (green) with EUK516-FITC at 30% formamide followed by a 2nd FISH for both A. naeslundii (red) and S. sobrinus (blue) with ANA103-Cy3 and SOB174-Cy5 at 10% formamide. Characteristic, homogeneous colonies formed by V. dispar are visible at the top of the biofilm. The toughest organism to label was S. sobrinus (blue in image B) because the SOB174 probe yielded a comparatively weak signal. Fig. 3. CLSM images showing perpendicular xy-, xz-, and yz-planes through a randomly selected point of a 64.5-h biofilm. This biofilm was stained for all six biofilm strains but with only three colors at our disposition. Performing three consecutive FISHs, C. albicans (green), V. dispar (green) and S. oralis (red) were stained first using EUK516-FITC, VEI217-FITC and MIT588-Cy3, respectively, at 30% formamide. Next F. nucleatum (blue) was stained with FUS664-Cy5 at 20% formamide, and finally A. naeslundii (red) and S. sobrinus (blue) were stained using ANA103-Cy3 and SOB174Cy-5 at 10% formamide.
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47
Fig. 2.
Fig. 3.
43
44
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47
Fig. 4. CLSM images showing perpendicular xy-, xz-, and yz-planes through a randomly selected point of a 64.5-h multiply stained biofilm. C. albicans (green), A. naeslundii (red) and S. sobrinus (light blue) were stained first with EUK516-FITC at 30% formamide, followed by a second FISH with ANA103-Cy3 and SOB174-Cy-5 at 10% formamide. EPS (dark blue) was stained with Calcofluor after the completion of FISHs.
oralis 23.4 F 3.4%, S. sobrinus 10.7 F 4.8%, V. dispar 5.5 F 2.6%, F. nucleatum 7.3 F 2.1%, and C. albicans 1.0 F 0.9%, of the total cellular volume (Syto13 positive cells).
4. Discussion In the present study we developed a procedure for the simultaneous spatial analysis of the distribution of both Gram-positive and Gram-negative bacteria in a multi-species biofilm model. The technique combines FISH with oligonucleotide probes to specific 16S rRNA sequences and CLSM to assess fixed but otherwise intact biofilms. Key elements of the pro-
cedure are: adequate fixation, optimal permeabilization, and careful selection of the best combination between probe, fluorescence label, and formamide concentration. Although established FISH protocols use ethanol for the fixation of Gram-positive cells (Roller et al., 1994), we found no negative effect from paraformaldehyde fixation of our mixed-species biofilms. However, preliminary experiments showed that ethanol fixation of Gram-negatives followed by in situ hybridization resulted in reduced staining (data not shown). The need for permeabilizing the cell wall of Gram-positive organisms prior to in situ hybridization is well documented (Beimfohr et al., 1993; Moter and Go¨bel, 2000). The difficulty encountered with bio-
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47
films was to obtain optimal fluorescence signals from Gram-positives without losing fluorescence of Gramnegatives. This could only be achieved by keeping the permeabilization time strictly to 9 min. Fluorescence intensity in the biofilms was further influenced by the fluorescence label used. Manz et al. (1998) found that, in general, FITC labeled probes required lower formamide concentrations than Cy3 conjugated oligonucleotides. This is confirmed by the observations from this study. Cy3 is currently one of the brightest fluorochromes characterized by a large extinction coefficient. In comparison, FITC has less favorable features, in particular it is not photostable and its fluorescence intensity is pH sensitive. Maximum emission is reached only in the basic pH range (Brelje et al., 1993), which likely explains why we observed reduced fluorescence intensity with FITC-labeled probes in combination with a nearly neutral hybridization buffer. Multiple staining of biofilms with fluorescent compounds has been described before, using nucleic acid specific dyes (acridine orange, DAPI, stains from Molecular Probes’ Syto series) or other fluorochromes (Neu et al., 2002). Multiple staining of biofilms by FISH is very common; Okabe et al. (1999) stained Gram-negative filamentous bacteria in nitrifying biofilms using oligonucleotide probes with different labels. The same method was used by Wecke et al. (2000) to monitor the development of bacterial biofilms in human periodontal pockets. Triple hybridizations using FITC, Cy3, and Cy5 labeled oligonucleotide probes for the simultaneous in situ visualization of the genetic diversity of activated sludge organisms or in a trickling filter biofilm have been described by Amann et al. (1996) and Schmid et al. (2000), respectively. All these studies established that FISH with multiple probes, either simultaneously or in sequence, of biofilms is feasible, but the technique was applied always to either Gram-positive or Gramnegative organisms. Perry-O’Keefe et al. (2001) described a method using peptide nucleic acid instead of DNA probes for the simultaneous identification of Gram-positive and Gram-negative organisms. In their protocol 1 h of paraformaldehyde fixation was enough to permeabilize Gram-positive cells. This was not so with our biofilm model. Treatment with lysozyme for 9 min was an absolute necessity for successful probe penetration into the Gram-positive bacteria of the
45
consortium. However, there are limitations to the multiple staining since only three different fluorescent probe colors are available. To differentiate all six species, two separate biofilms had to be analyzed (Fig. 3). Additionally, EPS, such as predominantly a(1 ! 3) linked mutan (produced by S. sobrinus) and a(1 ! 6) linked dextrans (produced by S. oralis and S. sobrinus) were visualized by staining the biofilm with Calcofluor. Preliminary experiments with both mutan and dextrans revealed, that a-(1 ! 3)-and a-(1 ! 6) linked polysaccharides did react with Calcofluor (data not shown), although Calcofluor was originally described for staining of h-(1 ! 3)- and h-(1 ! 4) linked polysaccharides (Maeda and Ishida, 1967). It is very difficult to determine the labeling efficiency of a given probe in a mixed biofilm by FISH/ CLSM, since unstained cells remain undetectable and cannot be enumerated or estimated by volume. This problem must be addressed in future studies, possibly by combining FISH and immunofluorescence with species-specific fluorescence-labeled F(ab) fragments. At present, we can only compare the total cell volumes of the six species estimated by FISH/CLSM with the cell volume of the total cell population stained using Syto13. It is noteworthy that the mean volumes detected for the six biofilm strains amounted together to 95.8% of the volume covered by Syto13 stained cells. This suggests, albeit quite indirectly, that a large portion of the cells was in fact labeled by FISH. This view is supported by the observation that proportional relationship between four of the six species was similar when estimated by FISH/CLSM and by cultivation of scraped off biofilm cells. Only F. nucleatum and C. albicans showed 20 and 50 times lower proportions in culture analyses. A likely reason for this discrepancy is that the selective agar media for F. nucleatum and C. albicans were too suppressive. In summary, our results show that it is possible to permeabilize both Gram-negative and Gram-positive bacteria simultaneously and to perform several consecutive FISHs with multiple rRNA probes to identify simultaneously several members of the six-species biofilm. To our knowledge, this is the first time that a multispecies biofilm of supragingival plaque consisting of Gram-negative and Gram-positive organisms was analyzed using the FISH technique. In combination with the nondestructive sectioning properties of CLSM, this modified FISH technique enabled the
46
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47
visualization of the three-dimensional arrangement of the microbial population as well as the estimation of the quantitative distribution of the biofilm species. Staining of the bulk of the EPS with Calcofluor reassured us that black areas in the biofilm were not due to substantial losses of microorganisms during the staining procedure. The procedure facilitates analyses of the microstructure and of spatial alterations in biofilm composition that may occur in response to environmental challenges and should prove useful in addressing some fundamental questions about the nature of dental biofilms. Acknowledgements We thank T. Ba¨chi and M. Ho¨chli at the Central Electron Microscopy Laboratory of the University of Zu¨rich for the ability to use the CLSM, A. Meier and V. Osterwalder for excellent technical assistance. References Amann, R.I., Binder, B.J., Olson, R.J., Chisholm, S.W., Devereux, R., Stahl, D.A., 1990. Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microbiol. 56, 1919 – 1925. Amann, R.I., Ludwig, W., Schleifer, K.-H., 1995. Phylogenetic identification and in situ detection of individual microbial cells without cultivation. Microbiol. Rev. 59, 143 – 169. Amann, R., Snaidr, J., Wagner, M., Ludwig, W., Schleifer, K.H., 1996. In situ visualization of high genetic diversity in a natural microbial community. J. Bacteriol. 178, 3496 – 3500. Beimfohr, C., Krause, A., Amann, R., Ludwig, W., Schleifer, K.-H., 1993. In situ identification of Lactococci, Enterococci and Streptococci. Syst. Appl. Microbiol. 16, 450 – 456. Brelje, T.C., Wessendorf, M.W., Sorenson, R.L., 1993. Multicolor laser scanning confocal immunofluorescence microscopy: practical application and limitations. Methods Cell Biol. 38, 97 – 181. Costerton, J.W., Lewandowski, Z., Caldwell, D.E., Korber, D.R., Lappin-Scott, H.M., 1995. Microbial biofilms. Annu. Rev. Microbiol. 49, 711 – 745. Davey, M.E., O’Toole, G.A., 2000. Microbial biofilms: from ecology to molecular genetics. Microbiol. Mol. Biol. Rev. 64, 847 – 867. Gmu¨r, R., Guggenheim, B., 1983. Antigenic heterogeneity of Bacteroides intermedius as recognized by monoclonal antibodies. Infect. Immun. 42, 459 – 470. Guggenheim, B., Giertsen, E., Schu¨pbach, P., Shapiro, S., 2001a. Validation of an in vitro biofilm model of supragingival plaque. J. Dent. Res. 80, 363 – 370.
Guggenheim, M., Shapiro, S., Gmu¨r, R., Guggenheim, B., 2001b. Spatial arrangements and associative behavior of species in an in vitro oral biofilm model. Appl. Environ. Microbiol. 67, 1343 – 1350. Hermanowicz, S.W., 2003. Biofilm structure: an interplay of models and experiments. In: Wuertz, S., Bishop, P.L., Wilderer, P.A. (Eds.), Biofilms in wastewater treatment—an interdisciplinary approach. International Water Association Publishing, London, United Kingdom, pp. 32 – 47. Lawrence, J.R., Korber, D.R., Hoyle, B.D., Costerton, J.W., Caldwell, D.E., 1991. Optical sectioning of microbial biofilms. J. Bacteriol. 173, 6558 – 6567. Maeda, H., Ishida, N., 1967. Specificity of binding of hexopyranosyl polysaccharides with fluorescent brightener. J. Biochem. 62, 276 – 278. Maidak, B.L., Cole, J.R., Parker Jr., C.T., Garrity, G.M., Larsen, N., Li, B., Lilburn, T.G., McCaughey, M.J., Olsen, G.J., Overbeek, R., Pramanik, S., Schmidt, T.M., Tiedje, J.M., Woese, C.R., 1999. A new version of the RDP (Ribosomal Database Project). Nucleic Acids Res. 27, 171 – 173. Manz, W., Amann, R., Ludwig, W., Wagner, M., Schleifer, K.-H., 1992. Phylogenetic oligodeoxynucleotide probes for the major subclasses of proteobacteria: problems and solutions. Syst. Appl. Microbiol. 15, 593 – 600. Manz, W., Eisenbrecher, M., Neu, T.R., Szewzyk, U., 1998. Abundance and spatial organization of Gram-negative sulfate-reducing bacteria in activated sludge investigated by in situ probing with specific 16S rRNA targeted oligonucleotides. FEMS Microbiol. Ecol. 25, 43 – 61. Marsh, P.D., Bradshaw, D.J., 1995. Dental plaque as a biofilm. J. Ind. Microbiol. 15, 169 – 175. Moter, A., Go¨bel, U., 2000. Fluorescence in situ hybridization (FISH) for direct visualization of microorganism. J. Microbiol. Methods 41, 85 – 112. Neu, T.R., Kuhlicke, U., Lawrence, J.R., 2002. Assessment of fluorochromes for two-photon laser scanning microscopy of biofilms. Appl. Environ. Microbiol. 68, 901 – 909. Okabe, S., Satoh, H., Watanabe, Y., 1999. In situ analysis of nitrifying biofilms as determined by in situ hybridization and the use of microelectrodes. Appl. Environ. Microbiol. 65, 3182 – 3191. Paster, B.J., Bartoszyk, I.M., Dewhirst, F.E., 1998. Identification of oral streptococci using PCR-based, reverse-capture, checkerboard hybridization. Methods Cell Sci. 20, 223 – 231. Perry-O’Keefe, H., Rigby, S., Oliveira, K., Sorensen, D., Slender, H., Coull, J., Hyldig-Nielsen, J.J., 2001. Identification of indicator microorganisms using a standardized PNA FISH method. J. Microbiol. Methods 47, 281 – 292. Roller, C., Wagner, M., Amann, R., Ludwig, W., Schleifer, K.-H., 1994. In situ probing of gram-positive bacteria with high DNA G + C content using 23S rRNA-targeted oligonucleotides. Microbiology 140, 2849 – 2858. Schmid, M., Twachtmann, U., Klein, M., Strous, M., Juretschko, S., Jetten, M., Metzger, J.W., Schleifer, K.H., Wagner, M., 2000. Molecular evidence for genus level diversity of bacteria capable of catalyzing anaerobic ammonium oxidation. Syst. Appl. Microbiol. 23, 93 – 106. Shapiro, S., Giertsen, E., Guggenheim, B., 2002. An in vitro oral
T. Thurnheer et al. / Journal of Microbiological Methods 56 (2004) 37–47 biofilm model for comparing efficacies of antimicrobial mouthrinses. Caries Res. 36, 93 – 100. Thurnheer, T., Gmu¨r, R., Giertsen, E., Guggenheim, B., 2001. Automated fluorescent in situ hybridization for the specific detection and quantification of oral streptococci in dental plaque. J. Microbiol. Methods 44, 39 – 47. Thurnheer, T., Gmu¨r, R., Shapiro, S., Guggenheim, B., 2003. Mass transport of macromolecules within an in vitro model of supragingival plaque. Appl. Environ. Microbiol. 69, 1702 – 1709. Wagner, M., Amann, R., Ka¨mpfer, P., Assmus, B., Hartmann, A.,
47
Hutzler, P., Springer, N., Schleifer, K.-H., 1994. Identification and in situ detection of Gram-negative filamentous bacteria in activated sludge. Syst. Appl. Microbiol. 17, 405 – 417. Wecke, J., Kersten, T., Madela, K., Moter, A., Go¨bel, U.B., Friedmann, A., Bernimoulin, J.P., 2000. A novel technique for monitoring the development of bacterial biofilms in human periodontal pockets. FEMS Microbiol. Lett. 191, 95 – 101. Wimpenny, J., Manz, W., Szewzyk, U., 2000. Heterogeneity in biofilms. FEMS Microbiol. Rev. 24, 661 – 671.