JOURNAL OF CLINICAL MICROBIOLOGY, Apr. 2000, p. 1488–1494 0095-1137/00/$04.00⫹0 Copyright © 2000, American Society for Microbiology. All Rights Reserved.
Vol. 38, No. 4
Identification of Campylobacter jejuni, C. coli, C. lari, C. upsaliensis, Arcobacter butzleri, and A. butzleri-Like Species Based on the glyA Gene SHAHNAZ TAHIHRA AL RASHID,1 IRENE DAKUNA,1 HELENA LOUIE,1 DAVID NG,1 PETER VANDAMME,2 WENDY JOHNSON,3 AND VOON LOONG CHAN1* Department of Medical Genetics and Microbiology, University of Toronto, Ontario, Canada M5S 1A81; Laboratory of Microbiology, University of Ghent, B-9000 Ghent, Belgium2; and Laboratory Centre for Disease Control, Winnipeg, Manitoba, Canada R3E 3R23 Received 9 October 1998/Returned for modification 20 January 1999/Accepted 26 November 1999
Currently, the detection and identification of Campylobacter and Arcobacter species remains arduous, largely due to cross-species phenotypic similarities and a relatively narrow spectrum of biochemical reactivity. We have developed a PCR-hybridization strategy, wherein degenerate primers are used to amplify glyA fragments from samples, which are then subjected to species-specific oligodeoxyribonucleotide probe hybridizations, to identify and distinguish between Campylobacter jejuni, C. coli, C. lari, C. upsaliensis, Arcobacter butzleri, and an A. butzleri-like species. Evaluation of this strategy with genomic DNA from different type strains suggests that this approach is both specific and sensitive and thus may be applicable in a diagnostic assay to identify and differentiate these highly related species. Campylobacter species are common human and animal pathogens (26). Reevaluation of their phylogenetic relationships has resulted in the current classification scheme of Campylobacter, Helicobacter, and Arcobacter species (18, 31, 32, 36). Since phenotypic tests have not been well standardized, variable results have been reported in several studies (14, 18, 19), which have rendered these tests unreliable as a sole method of identification. Alternatively, computerized schemes based on probability matrices, comprised of results from numerous phenotypic tests for a large number of strains (18, 20), and molecular-based approaches, including enzymology (7), serology (11), cellular fatty acid compositions (10), electrophoretic protein patterns (6), ribotyping (8, 27), DNA-DNA hybridization (15, 23, 38), PCR-DNA fingerprinting (9, 34, 35), and PCRrestriction fragment length polymorphism (RFLP) analysis (4) have been developed to improve species-level identification. Several oligodeoxyribonucleotide (oligo) probes (23, 38) and PCR assays (16, 28, 33, 37) have also been developed to identify Campylobacter species but were unable to differentiate Campylobacter jejuni from C. coli. A method using both PCR and RFLP was developed to differentiate Campylobacter, Arcobacter, Helicobacter, and Wolinella species (4). However, like the previous assays, this method was unable to distinguish C. jejuni from C. coli. Species-specific oligo probes for four thermophilic Campylobacter species, A. butzleri, and an A. butzleri-like strain were developed based on glyA (encoding serine hydroxymethyltransferase), a highly conserved gene (per GenBank, National Center for Biotechnology Information). A combined PCR-oligo hybridization strategy was then explored to test the sensitivity and specificity of these probes.
MATERIALS AND METHODS Bacterial strains, plasmids, and growth conditions. Bacterial strains used in this study are listed in Table 1. Campylobacter species were grown on either Columbia agar Base (Oxoid) supplemented with 5% defibrinated horse blood or Mueller-Hinton agar supplemented with 10% defibrinated sheep’s blood and then incubated at 37°C from 2 to 6 days, in 3-liter anaerobic jars under microaerophilic conditions created by Campylobacter Gas Generating Kits (Oxoid). For gene cloning experiments, the plasmid pBluescript II KS(⫹) (Stratagene) and Escherichia coli JM101 (24) were used. Competent cells were prepared by the rubidium chloride-calcium chloride protocol and transformed by standard procedures (24). Extraction of genomic DNA. Genomic DNA from Campylobacter spp. with the strain designations “R-” and “LMG” (Table 1) was extracted as described by Pitcher et al. (21). Genomic DNA from the remaining bacterial species was extracted as previously described (2) or by using DNAzol reagent (GIBCO BRL). PCR. Three degenerate oligo primers, S1 [5⬘-AA(C/T) AAA TA(C/A) GC(A/T) GAA GG(T/A) TAT-3⬘], S2 [5⬘-ATG CAT (C/T)AA (A/T)GG (A/ T)CC (A/T)CC TTG-3⬘] and S5 [5⬘-C(G/T)G C(G/A)A T(G/A)T G(G/A)G CAA TAT C(A/T)G C-3⬘] were designed based on three conserved regions within glyA. All PCRs were performed on a model 480 thermal cycler (PerkinElmer/Applied Biosystems). The 100-l S1-S2 PCRs were optimized with 1 mM MgCl2 for all species, except for A. nitrofigilis with 2 mM MgCl2. Each PCR contained 1 g of genomic DNA (except H. pylori, 0.5 g), 20 pmol of each primer, 20 mol of each deoxyribonucleotide triphosphate (dNTP), amplification buffer (50 mM KCl, 10 mM Tris-Cl; pH 8.3), and 2.5 U of Taq DNA polymerase (Promega and Boehringer Mannheim). The 50-l S1-S5 PCRs were optimized with 1.5 mM MgCl2, 0.4 g of genomic DNA, 50 pmol of each primer, 10 mol of each dNTP, amplification buffer, and 2.5 U of Taq DNA polymerase. All samples were overlaid with equal volumes of mineral oil, denatured at 95°C for 3 min, followed by 30 cycles of amplification, each with denaturation at 95°C for 1.5 min, primer annealing at 42°C (S1-S2) or 48°C (S1-S5) for 2 min, and chain extension at 72°C for 1 min. A final extension step at 72°C for 5 (S1-S2) or 10 (S1-S5) min was then performed. To test the sensitivity of the PCR-hybridization method, serial dilutions of C. jejuni ATCC 33560, C. upsaliensis ATCC 43954, A. butzleri ATCC 49616, and A. butzleri-like 13162 genomic DNA ranging from 1 fg to 1 g were used in the PCRs. Cloning and sequencing. PCR products of C. jejuni ATCC 33560, C. coli ATCC 33559, C. lari ATCC 35221, and C. upsaliensis ATCC 43954 were subcloned into pBluescript II KS(⫹) at the EcoRV site and sequenced by using either S1, S2, or M13 (⫺20) forward or reverse primer (Stratagene), with the Sequenase version 2.0 DNA Sequencing Kit (U.S. Biochemicals). PCR products of C. upsaliensis 14096, A. butzleri ATCC 49616, A. butzleri 13218, A. butzleri-like 13432, and A. butzleri-like 13207 were directly sequenced in 25-cycle reactions using either S1, S2, or an internal primer GlyA-In1 (5⬘-GAT AAA ATA TTA
* Corresponding author. Mailing address: Department of Medical Genetics and Microbiology, Medical Sciences Bldg., University of Toronto, Toronto, Ontario, Canada M5S 1A8. Phone: 416-978-6077. Fax: 416-978-6885. E-mail:
[email protected]. 1488
VOL. 38, 2000
IDENTIFICATION OF BACTERIAL SPECIES BASED ON glyA GENE
1489
TABLE 1. Bacterial strains used in this studya Bacterium
Strain(s)b
Campylobacter jejuni....................................†‡ATCC 33560*, †ATCC 43429, †ATCC 43430, †ATCC 43431, †ATCC 43432, †ATCC 43433, †CEPA-3C2, †COO6-852, †INN73832, †V482, †D5943, †D6033, †D19163 Campylobacter coli .......................................†ATCC 33559*, †LMG 75356, †LMG 85306, †LMG 98536, †LMG 98546, †LMG 98556, †LMG 98566, †LMG 98576, †LMG 98596, †LMG 158826 Campylobacter lari........................................†ATCC 35221*, †‡PC 6374, †LMG 88456, †LMG 88446, †LMG 79296, †LMG 98876, †LMG 98886, †LMG 98896, †LMG 99136, †LMG 99146, †LMG 91526, †LMG 92536, †LMG 112516, †R-3126, †R-7496 Campylobacter upsaliensisc ..........................†‡ATCC 43954*, †‡12030, †‡13064, †‡13950, †14013, †14080, †14506, †14510, †14526, †14529, †14530, †14532, †‡14967, †15172, ‡5424, ‡12034, ‡14096*, ‡17501, ‡17606 Arcobacter butzlerid ......................................‡ATCC 49616* or Ref. 13217*, †‡ Ref. 11556, ‡Ref. 13135, ‡Ref. 13218*, ‡Ref. 13443, ‡Ref. 13129, ‡Ref. 13075, ‡Ref. 12052, ‡Ref. 13220, ‡Ref. 11667 Arcobacter butzleri-liked ...............................‡Ref. 13162, ‡Ref. 13163, ‡Ref. 13128, ‡Ref. 13432*, ‡Ref. 13207*, ‡Ref. 13209, ‡Ref. 13114, ‡Ref. 13447, ‡Ref. 14064, ‡Ref. 14841 Campylobacter sputorum .............................†ATCC 33562 Arcobacter nitrofigilis....................................†‡ATCC 33309 Helicobacter cinaedi .....................................†‡ATCC 35683 Helicobacter pylori ........................................†‡Clinical isolate Helicobacter canisd .......................................‡ATCC 51401 or reference 16953, ‡Ref. 17656, ‡Ref. 16485 Escherichia coli.............................................†‡ATCC 9637, JM1011 Bifidobacterium adolescentis........................†ATCC 15703 Lactobacillus casei........................................No strain designation5 Pseudomonas aeruginosa .............................†‡ATCC 10145 Shigella sonnei ..............................................†‡ATCC 11803 a
All strains were from the American Type Culture Collection (ATCC), Rockville, Md., except as noted. Ref., reference. Superscript symbols and numbers: 1, supE thi ⌬(lac-proAB) F⬘ (traD36 proAB⫹ lacIq lacZ⌬M15) (24); 2, clinical isolate; 3, hippuricase-negative variants (29); 4, from J. L. Penner, University of Toronto, Ontario, Canada; 5, from A. Bognar, University of Toronto, Ontario, Canada; 6, BCCM-LMG Culture Collection, Ghent, Belgium; *, Strains used for the PCR, subcloned, and sequenced to generate the species-specific oligo probes; †, strains used for the PCR and Southern hybridizations to determine species specificity of the series 1 and 2 probes; and ‡, strains used for the PCR and Southern hybridizations to determine species specificity of the GlyAseries probes. c C. upsaliensis strains (3). d LCDC, Health Canada, Ottawa, Ontario, Canada. b
GGT ATG-3⬘), with the Ampli-Cycle Sequencing Kit (Perkin-Elmer/Applied Biosystems). DNA sequence alignment, probe designs, and syntheses. Nucleotide sequences were analyzed by using the Microgenie Sequence Analysis Program version 5 (Beckman Instruments), DNAsis (Helix Corporation), and CLUSTAL W (12, 13). Alignment of the glyA nucleotide sequences of C. jejuni ATCC 33560, C. coli ATCC 33559, C. lari ATCC 35221, C. upsaliensis ATCC 43954, C. upsaliensis 14096, A. butzleri ATCC 49616, A. butzleri 13218, A. butzleri-like 13432, and A. butzleri-like 13207 identified four regions which were used to design the speciesspecific oligo probes CJATC-1, CC-1, CL-1, CU-1 (series 1, 28-mer) (Dalton Chemical Laboratories), CJATC-2, CC-2, CL-2, CU-2 (series 2, 32-mer), GlyACU, GlyA-AB, GlyA-BL1, and GlyA-BL2 (31- to 35-mer) (ACGT Corp.). End labeling of the probes. Oligos (20 to 50 pmol) were radioactively labeled with [␥-32P]ATP (4,500 Ci/mmol; ICN Biomedicals Canada, Ltd.), kinase buffer (70 mM Tris-Cl [pH 7.6], 10 mM MgCl2, 5 mM dithiothreitol), and 20 U of T4 polynucleotide kinase (Pharmacia and New England Biolabs). Labeled probes were purified through an STE-equilibrated Sephadex G-50 spin column (24) or a Sephadex G-25 MicroSpin column (Pharmacia). Southern blot. Equal amounts of PCR products from all the species (Table 1) were electrophoresed in 1% agarose gels and transferred onto GeneScreen Plus nylon-based membranes (DuPont Canada, Inc.) or Hybond membrane (Amersham) by vacuum transfer using the LKB 2016 VacuGene Vacuum Blotting System (Pharmacia) or by capillary action (2). Southern hybridizations. Prior to hybridization, membranes were fixed by air drying them at room temperature and then soaked in 2⫻ SSC; (1⫻ SSC is 0.15 M NaCl plus 0.015 M sodium citrate); membranes were then prehybridized at 42, 45, or 37°C (series 1, series 2, or series GlyA- probes, respectively) for 30 min in 10 ml of solution (series 1 and 2 probes: 1% sodium dodecyl sulfate [SDS], 1 M NaCl, 10% dextran sulfate, and 5 mg of denatured salmon sperm DNA per ml; GlyA- series probes: same solution plus 50% formamide). Labeled probes were added with a specific activity of 3 ⫻ 105 cpm/ml (series 1 and 2) or 1 ⫻ 106 cpm/ml (GlyA- series), hybridized at 42, 45, and 37°C (series 1, series 2, and GlyA- probes, respectively) for 8 to 24 h, and finally washed twice as follows: series 1 and 2, with 0.2⫻ SSC for 10 min, at 60 and 50°C, respectively; GlyA-CU and GlyA-AB, with 0.2⫻ SSC plus 0.1% SDS at 62°C for 15 min; and GlyA-BL and GlyA-BL2, with 0.2⫻ SSC plus 0.1% SDS at 60°C for 15 min. Bands were visualized by autoradiography (X-Omat AR; Kodak Scientific Imaging Film) using exposure times from 40 min to 10 h at room temperature and from 10 to 22 h at ⫺70°C. Nucleotide sequence accession number. Sequences obtained in this study have been assigned GenBank accession numbers as follows: C. jejuni ATCC 33560
(AF136493), C. coli ATCC 33559 (AF136494), C. lari ATCC 35221 (AF136495), C. upsaliensis ATCC 43954 (AF136496), C. upsaliensis 14096 (AF136497), A. butzleri ATCC 49616 (AF136498), A. butzleri 13218 (AF136499), A. butzleri-like 13432 (AF136500), and A. butzleri-like 13207 (AF136501).
RESULTS PCR amplification using S1-S2 and S1-S5 primer pairs. PCR amplification of genomic DNA from C. jejuni, C. coli, C. lari, C. upsaliensis, C. sputorum, Helicobacter cinaedi, Helicobacter pylori, H. canis, A. nitrofigilis, A. butzleri, A. butzleri-like, E. coli, Pseudomonas aeruginosa, and Shigella sonnei strains in this study (Table 1) using S1-S2 and S1-S5 primer pairs generated the expected products of 640 and 460 bp, respectively (Fig. 1). However, the same primers did not generate PCR products from B. adolescentis and L. casei (data not shown). DNA sequences, alignment, and design of species-specific oligo probes. Two independent recombinant clones from each of the four thermophilic Campylobacter species were sequenced to ensure accuracy. The partial glyA sequences and the four regions chosen to design the species-specific probes are shown in Fig. 2. The melting temperatures (Tm [17]) of the probes are as follows: CJATC-1, 66.92°C; CC-1, 64.36°C; CL-1, 60.73°C; CU-1, 66.58°C; CJATC-2, 54.87°C; CC-2, 57.80°C; CL-2, 57.80°C; CU-2, 59.26°C; GlyA-AB, 64.36°C; GlyA-BL1, 63.08°C; GlyA-BL2, 65.86°C; and, GlyA-CU, 65.52°C. Southern blot and hybridizations. Each probe’s species specificity was tested against the bacteria listed in Table 1. Results of the series 1 and 2 probe hybridizations are shown in Fig. 3, and results of the GlyA- series probe hybridizations are shown in Fig. 4. Both CJATC-1 and CJATC-2 were positive for all 13 C. jejuni strains tested, with no cross-hybridizations to other species after 4 h of exposure. With longer exposure times
1490
AL RASHID ET AL.
J. CLIN. MICROBIOL.
FIG. 1. PCR products resolved in 1% (S1-S2) and 1.5% (S1-S5) agarose gels and used in the Southern hybridization experiments. S1-S2 and S1-S5 PCRs resulted in the expected product sizes of 640 and 460 bp, respectively. A total of 4 representative strains from each of the C. jejuni (of 13 strains), C. coli (of 10 strains), C. lari (of 15 strains), C. upsaliensis (of 19 strains), A. butzleri (of 10 strains), and A. butzleri-like species (of 10 strains), 3 strains of H. canis species, and single strains of C. sputorum, H. cinaedi, H. pylori, A. nitrofigilis, E. coli, P. aeruginosa, and S. sonnei are shown. For Southern hybridizations, equivalent amounts of products for all species were transferred onto membranes.
(22 h), however, weak cross-hybridizations were observed, i.e., CJATC-1 to C. coli strains and CJATC-2 to C. upsaliensis strains. Similarly, both CC-1 and CC-2 were positive for all 10 C. coli strains. After 22 h of exposure, CC-2 did not cross-hybridize to the other species’ PCR products, while CC-1 weakly crosshybridized to C. jejuni and C. upsaliensis strains. Both CL-1 and CL-2 were positive for 12 of the 15 C. lari strains (C. lari LMG 9253, LMG 11251, and R-749 were negative) after 2 h of exposure. However, after 22 h weak crosshybridizations were observed: CL-1 to A. nitrofigilis and CL-2 to C. upsaliensis strains. After 4 h of exposure, CU-1 was observed to be positive for all 14 C. upsaliensis strains, while CU-2 was positive for 12 of the 14 strains (C. upsaliensis 14080 and 14529 were not detected), and GlyA-CU was positive for all the 10 strains that were tested. The CU-1, CU-2, and GlyA-CU probes were also observed to cross-hybridize to A. butzleri 11556 (CU-1 and CU-2, results not shown; GlyA-CU, result shown in Fig. 4) after 4 h of exposure. Besides cross-hybridizing to A. butzleri 11556 strain, CU-1 was negative for other non-C. upsaliensis strains, as observed from exposure times of 4 to 22 h. Likewise, other than A. butzleri 11556, the CU-2 and GlyA-CU probes were negative for other non-C. upsaliensis strains after 4 h of exposure. However, after longer exposure times, weak cross-hybridizations were observed: CU-2 to C. jejuni strains and GlyA-CU to C. jejuni, C. coli, C. lari, and H. canis strains. GlyA-AB hybridized to all 10 A. butzleri strains after 4 h of exposure. However, GlyA-AB weakly cross-hybridized to A. butzleri-like strains after 20 h of exposure. GlyA-BL1 was positive for 7 (13162, 13163, 13128, 13432, 13207, 13209, and 13114) of the 10 A. butzleri-like strains, while GlyA-BL2 was positive for 8 (13162, 13128, 13432, 13207,
13209, 13114, 13447, and 14841) of the 10 strains. Both probes were negative for A. butzleri-like 14064, and neither GlyA-BL1 nor GlyA-BL2 cross-hybridized to the A. butzleri and C. upsaliensis strains tested. Sensitivity. PCRs with the S1-S2 primer pairs were performed on serially diluted C. jejuni ATCC 33560, C. upsaliensis ATCC 43954, A. butzleri ATCC 49616, and A. butzleri-like 13162 genomic DNA, and the lowest amount of genomic DNA required in order to yield sufficient PCR product to be detected by CJATC-1 was 0.4 pg (200 template copies), by GlyA-CU was 0.5 pg (230 copies), by GlyA-AB was 50 pg (23,000 copies), and by GlyA-BL1 and GlyA-BL2 was 500 pg (230,000 copies). DISCUSSION Three degenerate primers were designed, based on conserved glyA sequences, to amplify a glyA region encompassing the domain suggested to be part of the serine hydroxymethyltransferase active site and the domain implicated for binding its coenzyme, pyridoxal-5⬘-phosphate (1, 22, 25). Nucleotide sequence alignment of the amplified region from the Campylobacter and Arcobacter species revealed variable regions which were used to design three series of species-specific oligo probes to identify C. jejuni, C. coli, C. lari, C. upsaliensis, A. butzleri, and A. butzleri-like strains. The currently described combined strategy uses PCR amplification of the partial glyA gene with degenerate primer pairs (S1-S2 and S1-S5), followed by species-specific oligo probe hybridizations to identify and distinguish the various species. Based on the probes’ Tm values, four types of hybridization and washing conditions were optimized for species specificity. Each probe’s ability to detect different strains of the different
VOL. 38, 2000
IDENTIFICATION OF BACTERIAL SPECIES BASED ON glyA GENE
1491
FIG. 2. Multiple nucleotide sequence alignment of the partial glyA sequences. Alignment of the sequences from C. jejuni ATCC 33560, C. coli ATCC 33559, C. lari ATCC 35221, C. upsaliensis 43954, (C. ups) C. upsaliensis 14096, A. butzleri (A. b) ATCC 49616, A. butzleri 13218, A. butzleri-like 13432, and A. butzleri-like 13207. Each single-underlined sequence corresponds to each of the species-specific probes.
species was tested by using strains from human and animal sources. Strains were also obtained as unknown samples to objectively evaluate probe specificities. The CJATC-1, CJATC-2, CC-1, CC-2, CL-1, CL-2, GlyAAB, GlyA-BL1, and GlyA-BL2 probes were observed to be
species specific by using short exposure times of 4 h or less. However, with longer exposures of 20 to 22 h, weak crosshybridizations were observed for most probes. The cross-hybridizations observed were, at most, one-tenth the intensity observed from the species-specific hybridizations. However,
1492
AL RASHID ET AL.
J. CLIN. MICROBIOL.
FIG. 3. Autoradiographs of the Southern hybridizations testing the species specificity of the series 1 and 2 probes. Results of the series 1 and 2 probe hybridizations are shown for the specific detection of four representative strains for each species, as well as for species that were undetected, and for weak cross-hybridizations. Altogether, 13 strains of C. jejuni, 10 strains of C. coli, 15 strains of C. lari, 14 strains of C. upsaliensis, and 1 strain each of A. butzleri 11556, C. sputorum, H. cinaedi, H. pylori, A. nitrofigilis, E. coli, P. aeruginosa, and S. sonnei were tested in these series’ hybridizations (Table 1).
the CU-1, CU-2, and GlyA-CU probes were observed to crosshybridize as strongly to A. butzleri 11556 as to their target C. upsaliensis strains (Fig. 3 and 4). Interestingly, A. butzleri 11556 is the only strain originating from a water source in Thailand, while the other A. butzleri strains originate from human or animal sources in North America or Europe. Thus, it is possible that A. butzleri 11556 is an aberrant strain, and the target sequences are highly similar to those of the C. upsaliensisspecific probes, which resulted in the equal hybridization strength of the probes to this strain as to its species-specific targets. Since an unknown sample may contain any one of the species, use of all of the probes would resolve any discrepancies due to cross-hybridizations. For example, if an unknown sample was detected by both CJATC-1 and CC-2, since CJATC-1 cross-hybridizes weakly C. coli and CC-2 only detects C. coli, the sample would be determined to be C. coli. If the unknown sample is detected by CJATC-1 but not by CC-2, then the sample would be determined to be C. jejuni. Thus, cross-hybridizations would not be a factor for misidentification. Crosshybridization would also not affect the identification of either C. jejuni or C. lari strains, since C. jejuni probes do not crosshybridize to C. lari strains and vice versa. The cross-hybridization of the CU-1, CU-2, and GlyA-CU probes to A. butzleri
strains would not affect the identification of C. upsaliensis or A. butzleri strains, since GlyA-AB does not cross-hybridize with C. upsaliensis strains. The strength of the hybridizations do vary because there are minor nucleotide sequence variations between different strains. Most of the variations probably correspond to the third base of codons and thus, may not cause changes at the amino acid sequence level due to codon degeneracy. The probe may not detect strains if the target sequence has more substitutions, deletions, or insertions. For example, both CL-1 and CL-2 did not detect the C. lari strains LMG 9253, LMG 11251, and R-749. These C. lari strains have been shown to be aberrant C. lari strains by comparative whole-cell protein electrophoresis and by a PCR assay based on a novel putative GTPase gene (33). Similarly, the observation that GlyA-BL1 and GlyA-BL2 were unable to hybridize to A. butzleri-like 14064 may also be due to the aberrant nature of this isolate. However, this oligo probe identification system does permit a small degree of target sequence variation. For example, although the glyA sequences of C. jejuni ATCC 43431 (5) (GenBank accession no. X53816) and C. jejuni ATCC 33560 differ by two nucleotides at the CJATC-1 target sequence, CJATC-1 positively identified both C. jejuni strains. Furthermore, none of the probes hybridized to the other bacterial species such as C. sputorum, H.
VOL. 38, 2000
IDENTIFICATION OF BACTERIAL SPECIES BASED ON glyA GENE
1493
FIG. 4. Autoradiographs of the Southern hybridizations testing the species specificity of the GlyA- series probes. Results of the GlyA- series probe hybridizations are shown for specific detection of four representative strains for each species, as well as for species that were undetected, and for weak cross-hybridizations. Altogether, 10 strains of C. upsaliensis, 10 strains of A. butzleri, 10 strains of A. butzleri-like, 3 strains of H. canis, and 1 strain each of C. jejuni ATCC 33560, C. coli ATCC 33559, C. lari PC637, C. sputorum, H. cinaedi, H. pylori, A. nitrofigilis, E. coli, P. aeruginosa, and S. sonnei were tested in these series’ hybridizations (Table 1).
cinaedi, H. pylori, E. coli, P. aeruginosa, and S. sonnei. Thus, the advantages of having multiple probes for each species are to detect a wider range of strains, to improve the confidence of positive identification, and to reduce false-negative identification. The sensitivity of this PCR-oligo hybridization strategy varied with the different probes. As previously mentioned, since there may be target sequence variation between strains, sensitivity largely depends on the homology between the degenerate primers’ and probes’ target sequences and the strains’ sequences. Thus, oligo probes have high specificity but are not as sensitive as probes based on DNA fragment hybridizations. Furthermore, this PCR-hybridization strategy can differentiate between the more closely related thermophilic C. jejuni, C. coli, and C. lari species. Thus, the development of this strategy into a diagnostic kit would be more species specific than the currently available rapid Campylobacter identification kits, such as the AccuProbeAssay System (Gen-Probe, Inc.), which does not differentiate between these species. While the use of isotopic detection in this strategy has disadvantages, such as isotopic decay and radiation exposure, it is significantly more rapid than conventional diagnostic methods involving culturing of the various Campylobacter species, since the PCR can be used to directly amplify products from specimens such as stool samples (37). In addition, the hybridizations can be adapted to a nonradioactive detection system. Moreover, once the PCR product concentrations are equalized, dot blots could be used as an alternative to Southern blots. Another advantage of this strategy is that only two primer pairs are required to amplify the target fragment from a wide range of genera, including Campylobacter, Helicobacter, and Arcobacter. Thus, it will be an efficient diagnostic tool allowing many samples to be tested concurrently. In addition, this advantage provides a basis by which the PCR-oligo hybridization identification system can be easily expanded. Therefore, specific probes for the remaining Campylobacter, Helicobacter, and Arcobacter species and also the newly identified species can be developed, generated, and used by following this strategy. Another strategy that has been used successfully is a GeneChip
probe array strategy, which simultaneously identifies Mycobacterium species and the presence of specific rifampin resistance mutations in a test sample (30). With more species-specific target sequences identified, the glyA system described here has the potential to be further developed and expanded into the GeneChip probe array strategy. ACKNOWLEDGMENTS We thank E. Hani for valuable discussions and David Woodward from the Laboratory Centre for Disease Control (LCDC), Health Canada, for the A. butzleri, A. butzleri-like, and H. canis strains. P.V. is indebted to the Fund for Scientific Research–Flanders (Belgium) for a position as a postdoctoral research fellow. This work was supported by a grant from the Natural Sciences and Engineering Research Council of Canada and by a contract grant from the LCDC, Health Canada. REFERENCES 1. Angelaccio, S., S. Pascarella, E. Fattori, F. Bossa, W. Strong, and V. Schirch. 1992. Serine hydroxymethyltransferase: origin of substrate specificity. Biochemistry 31:155–162. 2. Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl. 1990. Current protocols in molecular biology. Greene Publishing Associates/Wiley-Interscience, New York, N.Y. 3. Bourke, B., P. M. Sherman, D. Woodward, H. Lior, and V. L. Chan. 1996. Pulsed-field gel electrophoresis indicates genotypic heterogeneity among Campylobacter upsaliensis strains. FEMS Microbiol. Lett. 143:57–61. 4. Cardarelli-Leite, P., K. Blom, C. M. Patton, M. A. Nicholson, A. G. Steigerwalt, S. B. Hunter, D. J. Brenner, T. J. Barrett, and B. Swaminathan. 1996. Rapid identification of Campylobacter species by restriction fragment length polymorphism analysis of a PCR-amplified fragment of the gene coding for 16S rRNA. J. Clin. Microbiol. 34:62–67. 5. Chan, V. L., and H. Bingham. 1991. Complete sequence of the Campylobacter jejuni glyA gene encoding serine hydroxymethyltransferase. Gene 101:51–58. 6. Costas, M., R. J. Owen, and P. J. M. Jackman. 1987. Classification of Campylobacter sputorum and allied campylobacters based on numerical analysis of electrophoretic protein patterns. Syst. Appl. Microbiol. 9:125–131. 7. Elharrif, Z., and F. Megraud. 1986. Characterization of thermophilic campylobacter. II. Enzymatic profiles. Curr. Microbiol. 13:317–322. 8. Fitzgerald, C., R. J. Owen, and J. Stanley. 1996. Comprehensive ribotyping scheme for heat-stable serotypes of Campylobacter jejuni. J. Clin. Microbiol. 34:265–269. 9. Giesendorf, B. A. J., H. Goossens, H. G. M. Niesters, A. van Belkum, A.
1494
10.
11.
12. 13. 14. 15.
16.
17. 18. 19. 20. 21. 22. 23. 24. 25.
AL RASHID ET AL.
Koeken, H. P. Endtz, H. Stegeman, and W. G. V. Quint. 1994. PCR-mediated DNA fingerprinting for epidemiological studies on Campylobacter spp. J. Med. Microbiol. 40:141–147. Goodwin, C. S., W. McConnell, R. K. McCulloch, C. McCulloch, R. Hill, M. A. Bronsdon, and G. Kasper. 1989. Cellular fatty acid composition of Campylobacter pylori from primates and ferrets compared with those of other campylobacters. J. Clin. Microbiol. 27:938–943. Hebert, G. A., D. G. Hollis, R. E. Weaver, A. G. Steigerwalt, R. M. McKinney, and D. J. Brenner. 1983. Serogroups of Campylobacter jejuni, Campylobacter coli, and Campylobacter fetus defined by direct immunofluorescence. J. Clin. Microbiol. 17:529–538. Higgins, D. G., A. J. Bleasby, and R. Suchs. 1991. CLUSTAL V: improved software for multiple sequence alignment. CABIOS 8:189–191. Higgins, D. G., and P. M. Sharp. 1989. Fast and sensitive multiple sequence alignments on a microcomputer. CABIOS 5:151–153. Lastovica, A. J., E. Le Roux, R. Warren, and H. Klump. 1994. Additional data on clinical isolates of Campylobacter mucosalis. J. Clin. Microbiol. 32:2338–2339. Li, C., D. A. Ferguson, Jr., T. Ha, D. S. Chi, and E. Thomas. 1993. A highly specific and sensitive DNA probe derived from chromosomal DNA of Helicobacter pylori is useful for typing H. pylori isolates. J. Clin. Microbiol. 31:2157–2162. Linton, D., A. J. Lawson, R. J. Owen, and J. Stanley. 1997. PCR detection, identification to species level, and fingerprinting of Campylobacter jejuni and Campylobacter coli direct from diarrheic samples. J. Clin. Microbiol. 35: 2568–2572. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. On, S. L. W. 1996. Identification methods for campylobacters, helicobacters, and related organisms. Clin. Microbiol. Rev. 9:405–422. On, S. L. W. 1994. Confirmation of human Campylobacter concisus isolates misidentified as Campylobacter mucosalis and suggestions for improved differentiation between the two species. J. Clin. Microbiol. 32:2305–2306. On, S. L. W., and B. Holmes. 1995. Classification and identification of campylobacters, helicobacters, and allied taxa by numerical analysis of phenotypic tests. Syst. Appl. Microbiol. 18:374–390. Pitcher, D. G., N. A. Saunders, and R. J. Owen. 1989. Rapid extraction of bacterial genomic DNA with guanidium thiocyanate. Lett. Appl. Microbiol. 8:151–156. Plamann, M., L. T. Stauffer, M. L. Urbanowski, and G. V. Stauffer. 1983. Complete nucleotide sequence of the E. coli glyA gene. Nucleic Acids Res. 11:2065–2075. Romaniuk, P. J., and T. J. Trust. 1989. Rapid identification of Campylobacter species using oligonucleotide probes to 16S ribosomal RNA. Mol. Cell. Probes 3:133–142. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Schirch, V., S. Hopkins, E. Villar, and S. Angelaccio. 1985. Serine hydroxymethyltransferase from Escherichia coli: purification and properties. J. Bacteriol. 163:1–7.
J. CLIN. MICROBIOL. 26. Skirrow, M. B. 1994. Diseases due to Campylobacter, Helicobacter, and related bacteria. J. Comp. Pathol. 111:113–149. 27. Smith, S. I., D. K. Olukoya, A. J. Fox, and A. O. Coker. 1998. Ribosomal RNA gene restriction fragment diversity amongst Penner serotypes of Campylobacter jejuni and C. coli. Z. Naturforsch. 53:65–68. 28. Stucki, U., J. Frey, J. Nicolet, and A. P. Burnens. 1995. Identification of Campylobacter jejuni on the basis of a species-specific gene that encodes a membrane protein. J. Clin. Microbiol. 33:855–859. 29. Totten, P. A., C. M. Patton, F. C. Tenover, T. J. Barrett, W. E. Stamm, A. G. Steigerwalt, J. Y. Lin, K. K. Holmes, and D. J. Brenner. 1987. Prevalence and characterization of hippurate-negative Campylobacter jejuni in King County, Washington. J. Clin. Microbiol. 25:1747–1752. 30. Troesch, A., H. Nguyen, C. G. Miyada, S. Desvarenne, T. R. Gingeras, P. M. Kaplan, P. Cros, and C. Mabilat. 1999. Mycobacterium species identification and rifampin resistance testing with high-density DNA probe arrays. J. Clin. Microbiol. 37:49–55. 31. Trust, T. J., S. M. Logan, C. E. Gustafson, P. J. Romaniuk, N. W. Kim, V. L. Chan, M. A. Ragan, P. Guerry, and R. R. Gutell. 1994. Phylogenetic and molecular characterization of a 23S rRNA gene positions the genus Campylobacter in the epsilon subdivision of the Proteobacteria and shows that the presence of transcribed spacers is common in Campylobacter spp. J. Bacteriol. 176:4597–4609. 32. Ursing, J. B., H. Lior, and R. J. Owen. 1994. Proposal of minimal standards for describing new species of the family Campylobacteraceae. Int. J. Syst. Bacteriol. 44:842–845. 33. Van Camp, G., H. Fierens, P. Vandamme, H. Goossens, A. Huyghebaert, and R. De Wachter. 1993. Identification of enteropathogenic Campylobacter species by oligonucleotide probes and polymerase chain reaction based on 16S rRNA genes. Syst. Appl. Microbiol. 16:30–36. 34. van Doorn, L.-J., B. A. J. Giesendorf, R. Bax, B. A. M. van der Zeijst, P. Vandamme, and W. G. V. Quint. 1997. Molecular discrimination between Campylobacter jejuni, Campylobacter coli, Campylobacter lari, and Campylobacter upsaliensis by polymerase chain reaction based on a novel putative GTPase gene. Mol. Cell. Probes 11:177–185. 35. Vandamme, P., B. A. J. Giesendorf, A. van Belkum, D. Pierard, S. Lauwers, K. Kersters, J.-P. Butzler, H. Goossens, and W. G. V. Quint. 1993. Discrimination of epidemic and sporadic isolates of Arcobacter butzleri by polymerase chain reaction-mediated DNA fingerprinting. J. Clin. Microbiol. 31: 3317–3319. 36. Vandamme, P., L.-J. van Doorn, S. T. Al Rashid, W. G. V. Quint, J. Van Der Plas, V. L. Chan, and S. L. W. On. 1997. Campylobacter hyoilei Alderton et al. 1995. and Campylobacter coli Ve´ron and Chatelaine 1973 are subjective synonyms. Int. J. Syst. Bacteriol. 47:1055–1060. 37. Waegel, A., and I. Nachamkin. 1996. Detection and molecular typing of Campylobacter jejuni in fecal samples by polymerase chain reaction. Mol. Cell. Probes 10:75–80. 38. Wesley, I. V., R. D. Wesley, M. Cardella, F. E. Dewhirst, and B. J. Paster. 1991. Oligodeoxyribonucleotide probes for Campylobacter fetus and Campylobacter hyointestinalis based on 16S rRNA sequences. J. Clin. Microbiol. 29:1812–1817.