Development of multi-color FISH method for analysis of seven Bifidobacterium species in human feces

Journal of Microbiological Methods 58 (2004) 413 – 421 www.elsevier.com/locate/jmicmeth

Development of multi-color FISH method for analysis of seven Bifidobacterium species in human feces Toshihiko Takada, Kazumasa Matsumoto, Koji Nomoto * Yakult Central Institute for Microbiological Research, 1796 Yaho, Kunitachi, Tokyo 186-8650, Japan Received 23 February 2004; received in revised form 14 May 2004; accepted 14 May 2004 Available online 26 June 2004

Abstract We have developed a multi-color fluorescence in situ hybridization (FISH) method which detects, by a single reaction, all seven species of Bifidobacterium (B. adolescentis, B. angulatum, B. bifidum, B. breve, B. catenulatum, B. dentium, and B. longum), the dominant bacteria in human feces. First, eight new types of oligonucleotide probe were designed, complementary with the 16S rRNA sequence specific to genus Bifidobacterium and each bifidobacterial species described above. Using whole cell hybridization, the fluorescent intensity was measured against the bacterial species targeted by each probe, to show that each probe is specific to the targeted bacteria and that the relative fluorescent intensity (RFI) as an indicator of probe accessibility is high at 61 – 117%. Then, bacterial species-specific probes were labeled with fluorochromes (FITC, TAMRA and Cy5) in seven different ways, singly or in combination. Using these probes, seven species of Bifidobacterium were differentially stained in mixed samples of cultured bacteria and feces from adult volunteers, proving the efficacy of this technique. D 2004 Elsevier B.V. All rights reserved. Keywords: Multi-color FISH; Bifidobacterium; Oligonucleotide; Human feces

1. Introduction Most bacteria constituting the intestinal bacterial flora are strictly anaerobic bacteria and are difficult to cultivate (Harmsen et al., 1999; Langendijk et al., 1995). Even for those cultivable bacteria, specific media and culture techniques are required for enumeration (Nelson and George, 1995), necessitating much time and effort to obtain results. Recently, with the development of phylogeny based on the 16S

* Corresponding author. Tel.: +81-42-577-8960; fax: +81-42577-3020. E-mail address: [email protected] (K. Nomoto). 0167-7012/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.mimet.2004.05.006

rRNA sequences (Woese, 1987), molecular biological techniques such as the quantitative PCR method (Saiki et al., 1988) and fluorescence in situ hybridization (FISH) method have been developed (Franks et al., 1998; Harmsen et al., 2002), and the identification and quantitation of microorganisms in a variety of environments have become rapid and easy, without requiring culture (Amann et al., 1995; Wang et al., 1996). With the FISH method, microorganisms are chemically fixed with paraformaldehyde and ethanol, hybridized with fluorescence-labeled probes and visualized under fluorescent microscopy (Amann et al., 1995; Delong et al., 1989). Fluorescent signals from these cells are captured with a charge-coupled device (CCD) camera, and these images are used to quanti-

414

T. Takada et al. / Journal of Microbiological Methods 58 (2004) 413–421

tate microorganisms in samples (Jansen et al., 1999). When the FISH method is applied for the analysis of intestinal bacterial flora in humans, more than seven types of bacterial group-specific probe are necessary to detect all the bacteria present, and at least 20 images must be taken from one sample in order to obtain accurate results (Harmsen et al., 2002). Therefore, an automated system from image acquisition to analysis is required to efficiently process a large number of samples (Jansen et al., 1999). With the FISH method, a variety of microorganism can be quantitatively identified by a single hybridization reaction using different combinations of fluorescent dyes with different fluorescent spectra (multi-color FISH) (Amann et al., 1996). This technique, in addition to the efficient detection of multiple microorganisms, appears to be a useful technique to visualize the bacterial composition and distribution. There are reports of the use of the multi-color FISH method for the analysis of microorganism in activate sludge (Amann et al., 1996) and the sea (Maruyama and Sunamura, 2000), but there is no report of its application in the analysis of microorganisms in human feces. To develop a multi-color FISH method for analysis of the bifidobacterial flora in human feces, three steps were taken: (i) preparation of new species-specific probes for bifidobacteria, (ii) evaluation of the specificity and probe accessibility, and (iii) examination of combinations of fluorochromes used to label the species-specific probes and application of the multi-color FISH method to mixture of bifidobacterial cultures and human feces.

2. Materials and methods 2.1. Design of oligonucleotide probes The 16S rRNA sequences of the genus Bifidobacterium and reference organisms were obtained from the DDBJ/GenBank/EMBL database. They were aligned with the Clustal W software (Thompson et al., 1994). The rRNA sequences specific to Genus Bifidobacterium, B. angulatum, B. bifidum, B. breve, B. catenulatum group (B. catenulatum and B. pseudocatenulatum), B. dentium and B. longum were selected from the variable region (V1, V2, V3 and V6 regions) (Table 1). The selected oligonucleotide target sites were tested for specificity against the 16S rRNA sequences available in the Ribosomal Database Project (RDP) by using the Probe Match analysis function of the RDP (Maidak et al., 1997). 2.2. Bacterial strains and growth conditions Table 2 shows the bifidobacterial strains used in this study. These bacterial strains were supplied by the American Type Culture Collection (ATCC), the National Institute of Biosciences and Human-Technology (FERM), the Deutsche Sammlung von Mikroorganismen und Zellkulturen (DSM), the Japan Collection of Microorganisms (JCM), National Collection of Food Bacteria (NCFB), and the Yakult Central Institute for Microbiological Research Tokyo (YIT). These bacterial strains were cultured to the late logarithmic phase in GAM broth (Nissui Seiyaku,

Table 1 Sequences and relative fluorescence intensities of all probes used in this study Name of probes

Sequence (5V– 3V)

Aimed bacteria

Target sitea (16S rRNA positions)

Relative fluorescence intensitiesb(% Bif153)

Bact338 Bif153 Bado434 Bang198 Bbif186 Bbre198 Bcat187 Bden82 Blon1004

GCTGCCTCCCGTAGGAGT ACCACCCGTTTCCAGGAG GCTCCCAGTCAAAAGCG AATCTTTCCCAGACCACC CCACAATCACATGCGATCATG AAAGGCTTTCCCAACACACC ACACCCCATGCGAGGAGT ACTCTCACCCGGAGGCGAA AGCCGTATCTCTACGACCGT

Bacteria Genus Bifidobacterium B. adolescentis B. angulatum B. bifidum B. breve B. catenulatum groupc B. dentium B. longum

338 – 355 153 – 170 434 – 450 198 – 216 186 – 206 198 – 218 187 – 204 82 – 103 1004 – 1024

85 100 67 82 75 72 61 78 117

a

Numbering corresponds to the structure model of E. coli 16S rRNA (Neefs et al., 1993). The relative fluorescent intensity (RFI) of each species-specific probe was expressed as percentages of the fluorescent intensity of Bif153. c The B. catenulatum group consists of B. catenulatum and B. pseudocatenulatum. b

T. Takada et al. / Journal of Microbiological Methods 58 (2004) 413–421

Tokyo, Japan) supplemented with 1% (wt/vol) lactose under anaerobic conditions at 37 jC.

415

(automatic fluorescent microscopy Leica DMRXA2, image acquisition software QFluoro [Wetzlar, Germany], and a cooled black and white CCD camera CoolSNAP HQ [Nippon Roper, Chiba, Japan]). The Leica DMRXA2 was equipped with a HCX PLAN APO objective (100  ; numerical aperture, 1.35), an HBO103W/2 mercury lamp, and four fluorescent

2.3. Image analysis equipment The observation and acquisition of fluorescent images were performed with a Leica Q550FW system

Table 2 List of bacterial strains and the results of whole cell hybridization using species- and group-specific probe Species

Strain

B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B. B.

ATCC 15703 NCFB 2229 NCFB 2230 NCFB 2231 ATCC 27535 ATCC 25527 ATCC 25910 ATCC 29521 ATCC 15696 ATCC 11863 Strain Yakult JCM 1211 ATCC 15700 ATCC 15698 Strain Yakult ATCC 27539 JCM 7130 ATCC 27686 ATCC 25911 ATCC 27916 ATCC 27534 JCM 8224 JCM 6291 ATCC 25864 ATCC 25912 ATCC 15697 DSM 10107 ATCC 15707 ATCC 15708 FERM P-6548 JCM 1218 JCM 8219 ATCC 27538 JCM 1200 JCM 1205 JCM 1214 JCM 8222 DSM 6531 DSM 20096 ATCC 27533 ATCC 25866

adolescentis adolescentis adolescentis adolescentis angulatum animalis asteroides bifidum bifidum bifidum bifidum boum breve breve breve catenulatum catenulatum choerinum coryneforme cuniculi dentium gallicum gallinarum globosum indicum infantis inopinatum longum longum longum magnum merycicum minimum pseudocatenulatum pseudolongum pullorum ruminantium saeculare subtile suis thermophirum a

Symbols:+, positive;

, negative.

Probesa Bif153

Bado434

+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +

+ + + +

Bang198

Bbif186

Bbre198

Bcat187

Bden82

Blon1004

+

+ + + + + + + + +

+ +

+ + +

+

+

416

T. Takada et al. / Journal of Microbiological Methods 58 (2004) 413–421

filters, A4 (excitation 360/40 nm, dichroic mirror 400 nm, emission 470/40 nm), L5 (excitation 480/40 nm, dichroic mirror 505 nm, emission 527/30 nm), N3 (excitation 546/12 nm, dichroic mirror 565 nm, emission 600/40 nm) or Y5 (excitation 620/60 nm, dichroic mirror 660 nm, emission 700/75 nm). Image analysis software Image-Pro Plus ver. 4.5 (Media Cybernetics, Silver Spring, MD) was used for the analysis of fluorescent images. 2.4. Probe specificity studies by whole cell hybridization method Cultures of bacteria were washed twice with PBS solution (Nissui Seiyaku), suspended in 3.7% formalin – PBS solution, and left on ice for 1 h. After two washes with PBS solution again, bacterial preparations were suspended in 50% ethanol – PBS solution and stored at 80 jC until use. Then, 10 Al fixed bacterial suspension was dropped on a MAS-coated slide glass (Matsunami, Osaka, Japan), air dried, and treated with 96% ethanol solution for 10 min. The oligonucleotide probes were synthesized as shown in Table 1, and labeled with 5-carboxytetramethyl-rhodamine-N-hydroxy-succinimide-ester (TAMRA) at the 5Vend. To the bacteria fixed on the slide glass, 100 Al hybridization solution (750 mM NaCl, 100 mM Tris – HCl, 5 mM EDTA, 0.01% BSA [wt/vol], 0.2% poly A [wt/vol] and 10% dextran sulfate [wt/vol]) containing 450 ng of the probe was laid and covered with a cover glass. Slide glasses were left overnight at 45 jC in a dark box humidified with SET solution (750 mM NaCl, 100 mM Tris – HCl, and 5 mM EDTA). After the hybridization reaction, the slide glasses were left for 20 min in wash solution (50 mM NaCl, 4 mM Tris – HCl, 0.02 mM EDTA) at 50 jC and washed once with distilled water. After drying in air, bacteria were embedded in VECTASHIELD (Vector Laboratories, Burlingame, CA). Slide glasses were examined using a Leica Q550FW system with a N3 filter, and the fluorescent intensity with a species-specific probe was judged as positive or negative with reference to that of Bif153, the probe specific to the genus Bifidobacterium. Fluorescent images were obtained from 10 fields for each target species, and the mean fluorescent intensity was obtained for each bacterial species.

2.5. Evaluation of probe accessibility The probe accessibility was determined in comparison to the fluorescent intensity of the standard probe after hybridization (Fuchs et al., 1998). In the present study, the Bif153 probe specific to the genus Bifidobacterium was used as the standard probe. The relative fluorescent intensity (RFI) of each speciesspecific probe was expressed as percentages of the fluorescent intensity of Bif153. 2.6. Combination of fluorochromes used for labeling probes for multi-color FISH The number of probes that can be used for multicolor FISH depends on the number of fluorochromes whose fluorescent spectra can be clearly differentiated. In the present study, FITC (Ex 494 nm, Em 518 nm), TAMRA (Ex 555 nm, Em 580 nm), and Cy5 (Ex 650 nm, Em 667 nm) were chosen based on analysis of the spectrum of each fluorochrome. TAMRA, FITC and Cy5 were assigned to red, green and blue, respectively, on the basis of the RGB color model. In addition to these three primary colors, the combination of two dyes, TAMRA and FITC, TAMRA and Cy5, and FITC and Cy5, made yellow, magenta and cyan, respectively, and the combination of three colors, TAMRA, FITC and Cy5, made white, thus forming a group of seven colors that were clearly differentiated visually. To select the fluorochrome used for labeling each species-specific probe, we decided to assign green, yellow, cyan and white, which were easily identified in multi-color images, to Bifidobacterium adolescentis, B. catenulatum group, B. bifidum and B. longum, respectively, species frequently detected in adult feces. 2.7. Effect of dual fluorochromes on probe accessibility Amongst the three fluorochromes used for labeling probes, Cy5 could be labeled only at the 5V end, while FITC and TAMRA could be labeled at either the 5V or 3V end. Therefore, using these three fluorochromes, probes could be labeled at both ends with different fluorochromes. When probes labeled with different fluorochromes are used at the same time, they compete for the target site, resulting in a decrease in the RFI value. However, a decrease in

T. Takada et al. / Journal of Microbiological Methods 58 (2004) 413–421

RFI values, we expected, could be avoided, due to lack of competition, when dual fluorochromes are used. Regarding the species-specific probes labeled with combinations of two fluorochromes [B. bifidum (Cy5 at 5V end and FITC at 3V end), B. catenulatum group (TAMRA at 5Vend and FITC at 3Vend) and B. dentium (Cy5 at 5V end and TAMRA at 3V end)], the probe accessibility of these probes was examined from the RFI value in comparison to the fluorescent intensity of the probes labeled singly with one of the fluorochromes. 2.8. Preparation of mixtures of seven species of bifidobacteria and fecal samples Mixed samples of cultures of seven strains of bifidobacteria, each belonging to different seven species, were prepared by mixing in equal amounts fixed bacteria from seven bifidobacterial species cultures. Fresh feces, obtained from three healthy adults, was diluted 5-fold with PBS solution filtered through a 0.2-Am filter, vigorously shaken with sterile glass beads (2 mm in diameter), and filtered through two layers of sterile gauze. The filtrate was mixed well with 3 volumes of 4% (wt/vol) paraformaldehyde– PBS solution (Amann et al., 1990) and left at 4 jC for 16 h. Then, preparations were stored at 80 jC until use. 2.9. Multi-color FISH method Ten microliters of a fixed sample, after dilution 100- to 400-fold with cold PBS solution, was spread evenly onto a 1-cm2 frame on an MAS-coated glass slide, dried and dehydrated in 96% [vol/vol] ethanol for 10 min. Then, 100 Al of the hybridization solution (750 mM NaCl, 100 mM Tris – HCl, 5 mM EDTA, 0.01% [wt/vol] BSA, 0.2% [wt/vol] poly A, 10% [wt/vol] dextran sulfate) containing 450 ng of each of the seven probes for multi-color fluorescent labeling plus Marina Blue (Ex 365 nm, Em 460 nm)-labeled Bif153 was dropped on the cell smears. After the smears covered with a coverslip, glass slide incubated overnight at 45 jC in a dark box humidified with SET solution (750 mM NaCl, 100 mM Tris – HCl, 5 mM EDTA). After the hybridization reaction was complete, the glass slides were incubated for 20 min in wash solution (50 mM NaCl, 4 mM

417

Tris – HCl, 0.02 mM EDTA) at 50 jC. After washing once more with distilled water and drying in air, the bacteria were embedded in VECTASHIELD (Vector Laboratories). Observation and image acquisition from glass slides were performed using a Q550FW system. L5, N3, Y5 and A4 fluorescent filter blocks were used for visualizing FITC, TAMRA, Cy5 and Marina Blue, respectively, in 10 fields for each sample. Image processings was performed on images with FITC, TAMRA, Cy5 and Marina Blue to remove the autofluorescence. After image processing, we analyzed whether the bacteria detected in each image with FITC, TAMRA and Cy5 were also labeled with Marina Blue. After pseudo-colorizing with green, red and blue for FITC, TAMRA and Cy5, respectively, and adjusting the contrast in each image, all colors were multi-layered in one image for multi-colored image analysis.

3. Results 3.1. Specificity and RFI value of probes The DNA sequence of Bif153, which is unique to the genus Bifidobacterium, has one mismatch with Gardnerella vaginalis and two mismatches with the DNA sequences for other bacteria outside of the genus Bifidobacterium. We concluded that the probes (Bang198, Bbre198, Bcat187, Bden82, and Blon1004) used for detection of B. angulatum, B. breve, B. catenulatum group, B. dentium and B. longum, respectively, were specific, since they had mismatches of more than two bases with the DNA sequences of bacteria outside the target species or group. The DNA sequence used for detection of B. adolescentis in the present study was the same as the one reported by Yamamoto et al. (1992) and was also complementary with that of B. ruminantium. However, B. ruminantium is a species that is isolated from the ruminant lumen (Biavati and Mattarelli, 1991) and not from human feces, and mismatches were identified in more than two bases with the DNA sequences of others besides these two species. We concluded that Bado434 could be used for the analysis of B. adolescentis in human feces. The DNA sequence of the Bbif186 probe, used for detection of B. bifidum, was

418

T. Takada et al. / Journal of Microbiological Methods 58 (2004) 413–421

complementary with that of B. cuniculii which is isolated from rabbits and not from humans (Scardovi et al., 1979), and had mismatches in more than 6 bases with the DNA sequences of other species than these two species, similarly allowing the use of Bbif186 as a specific probe for B. bifidum in human feces. The specificity and RFI value of the probes for detection of the genus Bifidobacterium and each bifidobacterial species were tested by the whole cell hybridization method (Table 1). The genus specific probe Bif153 hybridized to all bifidobacterial strains tested. All species-specific probes hybridized only with the target species, with the exception of Bado434 and Bbif186, which hybridized with B. adolescentis and B. ruminantium and with B. bifidum and B. cuniculii, respectively. The RFI values of the seven probes described above were in the range of 61 –117% of the fluorescent intensity of Bif153 (Table 1). Since the RFI value of Bact338, widely used as a probe consensus to domain bacteria (Amann et al., 1990), was 85% when similarly measured, these species-specific probes were judged to have high probe accessibility.

accessibility. Although the RFI value of the Cy5labeled Bbre198 was 27%, we judged this accessibility to be applicable. The RFI values of the Bden82 and Bbif186 probes, when labeled doubly with Cy5 and TAMRA (Bden82) or with Cy5 and FITC (Bbif186), were 78% and 49%, respectively, showing that these probes have sufficiently high probe accessibility. On the other hand, when the Bcat187 probes, labeled singly with TAMRA or FITC, were used in combination, as expected because of the competition between the probes, the RFI value of each probe was 40% or 27%, respectively, showing applicable probe accessibility to the target site. When three types of Blon1004 probe labeled singly with each of the three dyes (FITC, TAMRA and Cy5) were used in combination, although the RFI value of each probe decreased (38%, 24%, and 52%, respectively), we judged this as the total accessibility (114%) to be applicable.

3.2. Fluorescent labeling method for species-specific probes used for multi-color FISH

Fig. 1 shows the results of multi-color FISH, using mixed cultures of seven species of bifidobacteria. Black and white images of TAMRA, FITC and Cy5 were obtained and pseudo-colorized green, red and blue, respectively, to compose multi-colored images. The color of each species was green (G), yellow (Y), red (R), cyan (C), blue (B) and magenta (M) for B. adolescentis, B. catenulatum, B. angulatum, B. bifidum, B. breve and B. dentium, respectively, as expected. Although the color of B. longum was blue-ish white (W), but not true white as expected,

The effect of labeling with each fluorochrome on the probe accessibility was examined to determine the fluorescent labeling method for each speciesspecific probe used in the multi-color FISH method. As shown in Table 3, the RFI values of Bado434 and Bang198 probes, when labeled with FITC (Bado434) or TAMRA (Bang198), were 75% and 82%, respectively, showing that these probes have high probe

3.3. Multi-color FISH analysis of bifidobacterial species in mixed samples and human feces

Table 3 Relative fluorescence intensities of the fluorescently labeled probes used in multi-colored FISH method Name of probes

Target species

Fluorescently label

Relative fluorescence intensitya(% Bif153)

Bado434 Bang198 Bbre198 Bcat187 Bbif186 Bden82 Blon1004

B. B. B. B. B. B. B.

5VFITC 5VTAMRA 5VCy5 5VFITC + 5VTAMRA 5VCy5-3VFITC 5VCy5-3VTAMRA 5VCy5 + 5VFITC + 5VTAMRA

75

FITC

a b

adolescentis angulatum breve catenulatum groupb bifidum dentium longum

TAMRA

Cy5

82 27 27 33 38

40 55 24

49 78 52

The relative fluorescent intensity (RFI) of each species-specific probe was expressed as percentages of the fluorescent intensity of Bif153. The B. catenulatum group consists of B. catenulatum and B. pseudocatenulatum.

T. Takada et al. / Journal of Microbiological Methods 58 (2004) 413–421

Fig. 1. Epifluorescent image of mixed culture of seven different Bifidobacterium species by multi-color FISH. In situ hybridization was performed on a mixed culture of B. adolescentis ATCC 15703T, B. angulatum ATCC 27535T, B. bifidum ATCC 29521T, B breve ATCC 15700T, B. catenulatum ATCC 27539T, B. dentium ATCC 27535T and B. longum ATCC 15707T with FITC-labeled Bado434, TAMRA-labeled Bang198, FITC/Cy5-labeled Bbif186, Cy5-labeled Bbre198, FITC/TAMRA-labeled Bcat187, Cy5/TAMRA-labeled Bden82, and Cy5/FITC/TAMRA-labeled Blon1004, respectively. Black and white images of TAMRA, FITC and Cy5 were obtained and pseudo-colorized green, red and blue, respectively, to compose multi-colored images. In image, B. adolescentis, B. angulatum, B. bifidum, B. breve, B. catenulatum, B. dentium and B. longum are shown in G (green), R (red), C (cyan), B (blue), Y (yellow), M (magenta) and W (blue-ish white), respectively.

and was close to the color of B. bifidum (cyan), it was possible to differentiate these two bacteria visually, thus differentiating all seven bacterial species separately in the images. When multi-color FISH was applied to feces from three individuals, bacterial species were clearly idenFig. 2. Identification of Bifidobacterium species in fecal samples from three volunteers by multi-color FISH. In situ hybridization was performed on fecal samples from three healthy adults, simultaneously using Cy5-labeled Bbre198, FITC-labeled Bado434, TAMRA-labeled Bang198, FITC/Cy5-labeled Bbif186, FITC/ TAMRA-labeled Bcat187, Cy5/TAMRA-labeled Bden82, and Cy5/FITC/TAMRA-labeled Blon1004. Black and white images of TAMRA, FITC and Cy5 were obtained and pseudo-colorized green, red and blue, respectively, to compile multi-colored images. In image (A) on the left, B. adolescentis (G), B. longum (W) and B. bifidum (C) were detected. In image (B), B. adolescentis, B. catenulatum group (Y) and B. longum were detected. In image (C), B. catenulatum group was only detected.

419

420

T. Takada et al. / Journal of Microbiological Methods 58 (2004) 413–421

tified as in the bacterial cultures (Fig. 2). In sample A, the dominant species were B. adolescentis and B. longum, and B. bifidum was also identified. The dominant species was B. adolescentis, and the B. catenulatum group and B. longum were detected in sample B. In sample C, only the B. catenulatum group was detected. All bifidobacteria detected by each species-specific probe were also identified with the genus-specific probe (data not shown), thus concluding that all bifidobacteria were detected. The application of multi-color FISH in human feces clearly identified differences in the composition of fecal bifidobacterial species between individuals.

4. Discussion The objective of this study was to establish a multicolor FISH method to detect seven bifidobacterial species isolated from human feces. It had been reported that the main species of the genus Bifidobacterium in human feces were B. adolescentis, B. angulatum, B. bifidum, B. breve, B. catenulatum and B. longum (Scardovi, 1984). In the present study, seven bacterial species, these six species and B. dentium, were targeted with the multi-color FISH method. Yamamoto et al. (1992) reported the probes specific to each of B. adolescentis, B. bifidum, B. breve, and B. longum to use dot-blot hybridization method. However, we have found that the previously reported probes for B. bifidum, B. breve and B. longum, respectively, used for FISH, cross-reacted with a few other bifidobacterial species (data not shown). Therefore, we have prepared a series of new species-specific probes. Since the three-dimensional structure of the ribosome may hinder the access of oligonucleotides to their target sites, the probe accessibility could be different depending on the target site (Fuchs et al., 1998), requiring evaluation of not only specificity, but also probe accessibility. The results showed that each probe was bacterial species-specific (Table 2) and showed sufficiently high probe accessibility (Table 1) for the multi-color FISH method. The methods of labeling a variety of probes are critical for establishing a multi-color FISH method. It is essential to use fluorochromes with fluorescent spectra that can be clearly differentiated to create a

unique spectral signature for each species in images after multi-color staining. Three of fluorochromes, Cy5, FITC and TAMRA were selected as labeling dyes, since they satisfied this requirement and the oligonucleotide probes were easily labeled at the time of synthesis. To make seven combinations using these three dyes, two or three dyes were used together. If probes, labeled with different dyes, competed at the same target site, the RFI value could be reduced. Such competition could be avoided if different specific probes could be made against different sites of the gene from each species. However, it is difficult to identify variety of specific sequences from the limited region on the DNA sequence that codes for 16S rRNA. Therefore, we labelled each end of the probe with a different fluorescent dye with an idea that it would effectively prevent the decrease in the RFI value, without competition between probes. Although it was found that the probe accessibility was influenced differently by each method of labeling probe, overall accessibilities were evaluated to be applicable (Table 3). All seven species of bifidobacteria were differentiated by color in multi-color images when septetlabeling FISH method was used in bifidobacterial cultures (Fig. 1). When the RFI value was compared for probes for use in multi-labeling (Table 3), the RFI values of Blon1004 and Bbif186 with FITC and Cy5 were similar, while that of Blon1004 was relatively lower with TAMRA than with the other probes. This appears to have resulted from close spectral overlap between these target bacteria. On the other hand, when the multi-color FISH method was applied to human feces, target bacteria were clearly identified in multi-color images (Fig. 2), although the RFI values of Bcat187, Bbif186 and Blon1004 with each dye were relatively lower than those of the other speciesspecific probes. This result suggests that RNA was sufficiently retained for detection with the multi-color FISH method in fecal samples. Therefore, the probes, as designed, appeared to enable unique detection with the multi-color FISH method.

References Amann, R.I., Binder, B.J., Olson, R.J., Chisholm, S.W., Devereux, R., Stahl, D.A., 1990. Combination of 16S rRNA-targeted oligo-

T. Takada et al. / Journal of Microbiological Methods 58 (2004) 413–421 nucleotide 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.I., 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. Biavati, B., Mattarelli, P., 1991. Bifidobacterium ruminantium sp. nov. from the rumens of cattle. Int. J. Syst. Bacteriol. 41, 163 – 168. Delong, E.F., Wickham, G.S., Pace, N.R., 1989. Phylogenetic stains: ribosomal RNA-based probes for the identification of single microbial cells. Science 243, 1360 – 1363. Franks, A.H., Harmsen, H.J.M., Raangs, G.C., Jansen, G.J., Schut, F., Welling, G.W., 1998. Variations of bacterial populations in human feces measured by fluorescent in situ hybridization with group-specific 16S rRNA-targeted oligonucleotide probes. Appl. Environ. Microbiol. 64, 3336 – 3345. Fuchs, B.M., Wallner, G., Beisker, W., Schwippl, I., Ludwig, W., Amann, R., 1998. Flow cytometric analysis of the in situ accessibility of Escherichia coli 16S rRNA for fluorescently labeled oligonucleotide probes. Appl. Environ. Microbiol. 64, 4973 – 4982. Harmsen, H.J.M., Gibson, G.R., Elfferich, P., Raangs, G.C., Wildeboer-Veloo, A.C.M., Argaiz, A., Roberfroid, M.B., Welling, G.W., 1999. Comparison of viable cell counts and fluorescence in situ hybridization using specific rRNA-based probes for the quantification of human fecal bacteria. FEMS Microbiol. Lett. 183, 125 – 129. Harmsen, H.J., Raangs, G.C., He, T., Degener, J.E., Welling, G.W., 2002. Extensive set of 16S rRNA-based probes for detection of bacteria in human feces. Appl. Environ. Microbiol. 68, 2982 – 2990. Jansen, G.J., Wildeboer-Veloo, A.C.M., Tonk, R.H.J., Franks, A.H., Welling, G.W., 1999. Development and validation of an automated, microscopy-based method for enumeration of groups of intestinal bacteria. J. Microbiol. Methods 37, 215 – 221. Langendijk, P.S., Schut, F., Jansen, G.J., Raangs, G.C., Kamphuis, G.R., Wilkinson, M.H.F., Welling, G.W., 1995. Quantitative fluorescence in situ hybridization of Bifidobacterium spp. with

421

genus-specific 16S rRNA-targeted probes and its application in fecal samples. Appl. Environ. Microbiol. 61, 3069 – 3075. Maidak, B.L., Olsen, G.J., Larsen, N., Overbeek, R., McCaughey, M.J., Woese, C.R., 1997. The RDP (Ribosomal Database Project). Nucleic Acids Res. 25, 109 – 110. Maruyama, A., Sunamura, M., 2000. Simultaneous direct counting of total and specific microbial cells in seawater, using a deep-sea microbe as target. Appl. Environ. Microbiol. 66, 2211 – 2215. Neefs, J.M., van de Peer, Y., de Rijk, P., Chapelle, S., de Wachter, R., 1993. Compilation of small ribosomal subunit RNA structures. Nucleic Acids Res. 21, 3025 – 3049. Nelson, G.M., George, S.E., 1995. Comparison of media for selection and enumeration of mouse fecal flora populations. J. Microbiol. Methods 22, 293 – 300. Saiki, R.K., Gelfand, D.H., Stoffel, S., Scharf, S.J., Higuchi, R., Horn, G.T., Mullis, K.B., Ehrlich, H.A., 1988. Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239, 487 – 491. Scardovi, V., 1984. Genus Bifidobacterium Orla-Jensen, 1924. In: Krieg, N.R., Holt, J.G. (Eds.), Bergey’s Manual of Systematic Bacteriology, vol. 1. Williams & Wilkins, Baltimore, MD, pp. 1418 – 1434. Scardovi, V., Trovatelli, L.D., Biavati, B., Zani, G., 1979. Bifidobacterium cuniculi, B. choerinum, B. boum, and B. pseudocatenulatum: four new species and their deoxyribonucleic acid homology relationships. Int. J. Syst. Bacteriol. 29, 291 – 311. Thompson, J.D., Higgens, D.G., Gibson, T.J., 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, posituon specific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673 – 4680. Wang, R.F., Cao, W.W., Wand Cerniglia, C.E., 1996. PCR detection and quantitation of predominant anaerobic bacteria in human and animal fecal samples. Appl. Environ. Microbiol. 62, 1242 – 1247. Woese, C.R., 1987. Bacterial evolution. Microbiol. Rev. 51, 221 – 271. Yamamoto, T., Morotomi, M., Tanaka, R., 1992. Species-specific oligonucleotide probes for five Bifidobacterium species detected in human intestinal microflora. Appl. Environ. Microbiol. 58, 4076 – 4079.