SGM Journals
DIAGNOSTICS, TYPING AND IDENTIFICATION

Identification of non-fermenting Gram-negative bacteria of clinical importance by an oligonucleotide array

Siou Cing Su, Mario Vaneechoutte, Lenie Dijkshoorn, Yu Fang Wei, Ya Lei Chen, Tsung Chain Chang

  • 1Department of Medical Laboratory Science and Biotechnology, College of Medicine, National Cheng Kung University, Tainan, Taiwan, ROC
  • 2Laboratory Bacteriology Research (LBR), Department of Clinical Chemistry, Microbiology and Immunology, Blok A, Ghent University Hospital, Ghent, Belgium
  • 3Department of Infectious Diseases, Leiden University Medical Center, Leiden, The Netherlands
  • 4Department of Biotechnology, National Kaohsiung Normal University, Kaohsiung, Taiwan, ROC
  • Correspondence
    Tsung Chain Chang
    tsungcha{at}mail.ncku.edu.tw

    • Journal of Medical Microbiology 2009; 58(5):596–605 · https://doi.org/10.1099/jmm.0.004606-0

    Download PDF View at publisher PubMed

    Abstract

    Many species of non-fermenting Gram-negative bacilli (non-fermenters) are important opportunistic and nosocomial pathogens. Identification of most species of non-fermenters by phenotypic characteristics can be difficult. In this study, an oligonucleotide array was developed to identify 38 species of clinically relevant non-fermenters. The method consisted of PCR-based amplification of 16S–23S rRNA gene intergenic spacer (ITS) regions using bacterial universal primers, followed by hybridization of the digoxigenin-labelled PCR products with oligonucleotide probes immobilized on a nylon membrane. A total of 398 strains, comprising 276 target strains (i.e. strains belonging to the 38 species to be identified) and 122 non-target strains (i.e. strains not included in the array), were analysed by the array. Four target strains (three reference strains and one clinical isolate) produced discrepant identification by array hybridization. Three of the four discordant strains were found to be correctly identified by the array, as confirmed by sequencing of the ITS and 16S rRNA genes, with the remaining one being an unidentified species. The sensitivity and specificity of the array for identification of non-fermenters were 100 and 96.7 %, respectively. In summary, the oligonucleotide array described here offers a very reliable method for identification of clinically relevant non-fermenters, with results being available within one working day.

    • The GenBank/EMBL/DDBJ accession numbers for the ITS sequences generated in this study are listed in Table 2⇑.

    • Details of the non-target strains tested are available with the online version of this paper.

    INTRODUCTION

    Non-fermenting Gram-negative bacilli (non-fermenters) are ubiquitous in the environment. Non-fermenters can cause a vast variety of infections (Dijkshoorn et al., 2007; LiPuma et al., 2007) and account for approximately 15 % of all Gram-negative bacilli cultured from clinical specimens (unpublished data of the National Cheng Kung University Hospital, Tainan, Taiwan). Pseudomonas aeruginosa is the most frequently isolated micro-organism, followed by Acinetobacter baumannii and Stenotrophomonas maltophilia (Blondel-Hill et al., 2007; LiPuma et al., 2007; Schreckenberger et al., 2007). Non-fermenters may differ in their pathogenic potential and transmissibility, and many are multidrug resistant (Schreckenberger et al., 2007). For this reason, accurate identification of non-fermenters to species level is important for appropriate patient management.

    In the diagnostic clinical microbiology laboratory, identification of non-fermenters relies mainly on phenotypic characteristics. A variety of commercial identification kits, such as API 20 NE (bioMérieux), VITEK 2 (bioMérieux) and Phoenix (Becton Dickinson), are being used for routine identification of these bacteria. Studies investigating the performance of these commercial identification systems have shown contradictory results (Funke & Funke-Kissling, 2004; Kiska et al., 1996). In a recent study using API 20 NE, 54 % of non-P. aeruginosa non-fermenters were assigned to species level, 7 % to genus level and 39 % of isolates could not be discriminated at any taxonomic level, whilst with VITEK 2 the respective numbers were 53, 1 and 46 % (Bosshard et al., 2006). Non-fermenters recovered from cystic fibrosis patients pose particular identification problems due to their phenotypic variation, atypical phenotypic characteristics and slow growth rates (Coenye et al., 2005; Ferroni et al., 2002).

    Molecular identification techniques are emerging as alternatives for phenotypic identification methods. Among these, 16S rRNA gene sequencing is widely used (Kolbert & Persing, 1999; Patel, 2001; Vaneechoutte & De Baere, 2007). However, by 16S rRNA gene sequence analysis, only 92 % of non-P. aeruginosa non-fermenters were assigned to species level, with the remaining 8 % being assigned to genus level (Bosshard et al., 2006). The intergenic spacer (ITS) region separating the 16S and 23S rRNA genes has been found to be a good candidate for bacterial identification (Chen et al., 2004; Gürtler & Stanisich, 1996; Tung et al., 2007). Recently, DNA array technology has been applied to identify a variety of micro-organisms with promising results (Fukushima et al., 2003; Park et al., 2005; Tung et al., 2006). This study aimed to develop an oligonucleotide array based on ITS sequences to identify 38 species of non-fermenters with clinical relevance.

    METHODS

    Bacterial strains and DNA.

    A collection of 276 target strains (123 reference strains and 153 clinical isolates), representing 38 species of non-fermenters, were analysed (Table 1⇓). Reference strains were obtained from: the American type Culture Collection (ATCC); the Belgian Co-ordinated Collections (BCCM/LMG); the Bioresources Collection and Research Center (BCRC); the Culture Collection, University of Göteborg (CCUG); the Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (DSMZ); and the Netherlands Culture Collection of Bacteria (NCCB). Clinical isolates were obtained from Ghent University Hospital (Belgium), Leiden University Medical Center (The Netherlands) and Kaohsiung Chang Gung Memorial Hospital (Taiwan). In order to assess the specificity of the array, a collection of 122 non-target strains belonging to 106 species other than the 38 target species were tested (see Supplementary Table S1, available in JMM online). All strains were cultured on sheep blood agar, incubated at 35 °C for 24–48 h and then used for DNA extraction by a boiling method (Millar et al., 2000). DNA extracts of Burkholderia pseudomallei were obtained from clinical isolates recovered from patients with septicaemic melioidosis admitted to the Kaohsiung Veterans General Hospital, Taiwan (Chen et al., 2006). These Burkholderia pseudomallei isolates were identified using biochemical test profiling (API 20 NE), by the presence of specific 16S RNA gene PCR amplicons (243 and 405 bp) and by the presence of a fliC gene PCR amplicon (267 bp) (Su et al., 2007).

    Table 1.

    Non-fermenting Gram-negative bacilli used for sensitivity testing of the array

    Design of oligonucleotide probes and array preparation.

    Forty-nine oligonucleotide probes (18–31mers) (Table 2⇓), based on ITS sequences, were designed for identification of the 38 target species (Table 1⇑). They included one positive control probe (designed from the 3′ end of the 16S rRNA gene) used to check for successful PCR and hybridization, three genus-specific probes used to identify Acinetobacter and Pseudomonas, one Acinetobacter calcoaceticus–Acinetobacter baumannii complex-specific probe and 44 species-specific probes (Table 2⇓). The Pseudomonas-specific probe (P3/P4) consisted of a mixture of probes P3 and P4 at an equimolar concentration. Probes used to identify the genera Pseudomonas and Acinetobacter and species in the A. calcoaceticus–A. baumannii complex were described in a recent study (Ko et al., 2008). Between 5 and 15 additional bases of thymine were added to the 3′ ends of some probes to increase the hybridization signal (Brown & Anthony, 2000) (Table 2⇓). A digoxigenin (DIG)-labelled irrelevant probe (5′-DIG-GCATATCAATAAGCGGAGGA-3) was spotted on the array and used as a position marker. The arrays (9×5 mm) were prepared with an automatic arrayer (model SR-A300; Ezspot) using a 400 μm-diameter solid pin as described previously (Tung et al., 2006). The layout of different probes on the array is shown in Fig. 1⇓.

    Figure image not available in archive
    Fig. 1.

    Layout of oligonucleotide probes on the array (9×5 mm). Probe PC (F10) is the positive control probe designed from a conserved region in the 16S rRNA gene. Probe NC (D6) is the negative control (tracking dye only). Probe M is a DIG-labelled oligonucleotide that was used as a position marker. Genus-specific and Acinetobacter calcoaceticus–Acinetobacter baumannii complex-specific probes are underlined. The sequences of the oligonucleotide probes are listed in Table 2.

    Table 2.

    Oligonucleotide probes used for identification of 38 species of non-fermenting Gram-negative bacilli

    Amplification of the ITS regions for array hybridization.

    The bacteria-specific universal primers 2F (5′-DIG-TTGTACACACCGCCCGTC-3′) and 10R (5′-DIG-TTCGCCTTTCCCTCACGGTA-3′) (Gürtler & Stanisich, 1996) were used to amplify the ITS regions, with each primer being labelled with a DIG molecule at its 5′ end. PCR was performed as described previously (Tung et al., 2006).

    Species determination by array hybridization.

    The procedures for pre-hybridization, hybridization (50 °C for 90 min) and colour development using anti-DIG antibodies have been described previously (Tung et al., 2006). The hybridized spots (400 μm in diameter) could be read by the naked eye. A strain was identified as one of the species listed in Table 1⇑ when both the positive control probe and the species-specific probe were hybridized (Table 2⇑). A strain was identified as belonging to the genus Acinetobacter or Pseudomonas when the genus-specific probe, Aci2 or P3/P4, respectively (Table 2⇑), was hybridized. In addition, a strain was identified as belonging to the A. calcoaceticus–A. baumannii complex when the complex-specific probe Acb2 was also hybridized (Ko et al., 2008). A strain was identified as Acinetobacter junii only when both of probes Ajun2-1 and Ajun5 were hybridized. In contrast, a strain was identified as Pseudomonas stutzeri if at least one of the four probes Pstu1-3, Pstu2-2R, Pstu2-4 and Pstu3-7R designed for the species was hybridized, and as Shewanella putrefaciens if at least one of the two probes Sput1-3 and Sput3-4 was hybridized.

    Discrepancy analysis.

    In cases where the array identification did not correspond with the original species name of the strain, the species identity of the strain was determined by sequencing of the ITS (Tung et al., 2007) and the 16S rRNA gene. Amplification of the ITS by PCR was carried out as described above. The method of Relman (1993) was followed to amplify the 16S rRNA gene. PCR products were sequenced (Tung et al., 2007) and the determined sequences were compared with reference sequences in GenBank using blast (). For ITS sequence comparison, final species identification was obtained only when the best-scoring ITS reference sequence had an identity of ≥97 % with the query sequence (Chen et al., 2004).

    RESULTS AND DISCUSSION

    Probe development

    For most species, a single probe was sufficient to obtain identification, but multiple probes were needed to identify A. junii, Acinetobacter lwoffii, P. stutzeri and Shewanella putrefaciens (Table 2⇑) due to intra- or interspecies sequence variation in the ITS regions. A. junii was identified only by simultaneous hybridization to two probes (Ajun2-1 and Ajun5), because Acinetobacter johnsonii LMG 1002 cross-hybridized with probe Ajun5 (Fig. 2⇓, chip 8), whereas ‘Acinetobacter venetianus’ CCUG 45561T (a non-target species) cross-reacted with probe Ajun2-1 (Fig. 2⇓, chip 46). Similarly, Acinetobacter gen. sp. 6 (Fig. 2⇓, chip 6) and Acinetobacter radioresistens (Fig. 2⇓, chip 11), in addition to their specific probes, cross-hybridized with probe A13TU used to identify Acinetobacter gen. sp. 13TU (a member of the A. calcoaceticus–A. baumannii complex), but neither one was misidentified as Acinetobacter gen. sp. 13TU as the complex-specific probe Acb2 was not hybridized.

    Figure image not available in archive
    Fig. 2.

    Hybridization results for 38 species of non-fermenters. Chip numbers are shown in parentheses and the corresponding probes hybridized on the arrays are indicated in Fig. 1. The cross-hybridization patterns of ‘Acinetobacter venetianus’ CCUG 45561T, Acinetobacter gen. sp. 14BJ CCUG 34435 and Burkholderia cenocepacia BCRC 17448T are also shown (chips 46–48).

    Different genomovars of P. stutzeri (Guasp et al., 2000) produced different hybridization patterns with the four probes Pstu1-3, Pstu2-2R, Pstu2-4 and Pstu3-7R designed for this species. For example, P. stutzeri ATCC 17588T (genomovar 1) (Fig. 2⇑, chip 36) and DSM 6082 (genomovar 5), in addition to the Pseudomonas-specific probe P3/P4, hybridized to probe Pstu1-3, whilst strain BCRC 14821 (genomovar 2) hybridized to another two probes (Pstu2-2R and Pstu2-4) (Fig. 2⇑, chip 37). A. lwoffii and Acinetobacter gen. sp. 9 are synonyms (Tjernberg & Ursing, 1989), and identification of the organism was made if at least one of the two probes (Alwo3 and Aun9-1) was hybridized (Fig. 2⇑, chips 9 and 10).

    Identification of reference strains by the array

    The hybridization patterns of the 38 species of non-fermenters on the arrays are shown alphabetically in Fig. 2⇑. Of 123 reference strains belonging to the 38 target species, 120 hybridized to their respective oligonucleotide probes and were correctly identified. Acinetobacter gen. sp. 9 LMG 1027, Pseudomonas fluorescens BCRC 10907 (=ATCC 13430) and Pseudomonas putida BCRC 14365 (=ATCC 31800) were identified as A. baumannii, P. putida and P. fluorescens, respectively (Table 3⇓). Discrepancy analysis by sequencing of the 16S rRNA genes revealed that all three identifications obtained by the array were correct. Identification of Acinetobacter gen. sp. 9 LMG 1027 as A. baumannii by the array was further confirmed by ITS sequencing. However, the identifications of P. fluorescens BCRC 10907 as P. putida and P. putida BCRC 14365 as P. fluorescens by the array could not be confirmed by ITS sequence comparison (sequence identities <97 %) (Table 3⇓), due to the lack of corresponding ITS sequences in public databases. P. putida and P. fluorescens are phenotypically similar; several tests (gelatin, trehalose, inositol and lecithinase) could be used to differentiate the two micro-organisms, with P. fluorescens being positive for the four tests and P. putida being negative (Blondel-Hill et al., 2007; Palleroni, 1984). In brief, all 123 reference strains, including the three strains that had been named incorrectly, were correctly identified by the array.

    Table 3.

    Analysis of discrepant strains by sequencing of the ITS and 16S rRNA genes

    Identification of clinical isolates by the array

    In this study, only clinical isolates identified by other molecular methods (Baele et al., 2000; Dijkshoorn et al., 1998; Nemec et al., 2001; Vaneechoutte et al., 1998) were used for sensitivity testing of the array, in order to avoid unnecessary discrepant identifications that should be reconfirmed by other molecular techniques. Of 153 target clinical isolates tested, only one (Moraxella caviae #003) was not identified by the array (Table 3⇑). However, the identity of this strain as M. caviae could not be confirmed by either ITS or 16S rRNA gene sequencing. If M. caviae #003 was excluded for sensitivity calculation and if reference strains (123 strains) and clinical isolates (152 strains) were taken together, the sensitivity of the array was 100 % (275/275).

    In addition, a collection of 128 clinical isolates, identified by API 20 NE as belonging to one of the 38 target species, were also analysed by the array (data not shown). Of these, 12 produced discrepant identifications between API 20 NE and the array. Eleven of the 12 discrepant isolates were correctly identified by array hybridization, as further confirmed by both ITS and 16S rRNA gene sequencing. One discrepant strain (5630A) was not identified by the array, and both ITS and 16S rRNA gene sequence analyses revealed that the strain was Bordetella hinzii, which was not a target species on the array (data not shown). The results again highlight the accuracy of the array and the inaccuracy of biochemical tests for identification of non-fermenters.

    Specificity of the array

    A collection of 122 non-target strains (106 species) were used for specificity testing of the array (see Supplementary Table S1). A total of four strains were misidentified. Burkholderia cenocepacia BCRC 17448T was misidentified as Burkholderia cepacia (Fig. 2⇑, chip 48), Chryseobacterium gleum BCRC 17270 was misidentified as Chryseobacterium indologenes, and Shewanella algae LMG 2267 and #017 (a clinical isolate) were misidentified as Shewanella putrefaciens by the array, resulting in a specificity of 96.7 % (118/122) for the array. Sequence analysis of the 16S rRNA gene revealed that the identities of the four cross-hybridization strains were correct (data not shown).

    In this study, most individual species were identified by a single oligonucleotide probe, but multiple probes were used to identify A. junii, A. lwoffii, P. stutzeri and Shewanella putrefaciens (Table 2⇑). The disadvantage of using multiple probes is the increased potential of cross-hybridization caused by other species that may have partial sequence similarity with one of the multiple probes. P. stutzeri has been reported to be involved in a variety of severe infections including bacteraemia, endocarditis and meningitis (Noble & Overman, 1994). Bacteria identified as P. stutzeri, on the basis of phenotypic tests, are recognized to be very heterogeneous and can be divided into nine genomovars based on DNA similarities (Guasp et al., 2000). Sequencing of the ITS region was proposed as a good alternative for genomovar differentiation of species of the P. stutzeri complex (Guasp et al., 2000). Due to the heterogeneity of the complex, it was difficult to find a consensus region that covered all genomovars of the micro-organism and consequently four probes were designed to identify the bacterium (Table 2⇑).

    The successful design of different probes was based on known sequences in the ITS regions and multiple sequence alignment played an important role in pinpointing the regions that could be used for probe synthesis. It should be noted that some non-fermenters (e.g. Acinetobacter haemolyticus and A. lwoffii) possess multiple ITS regions of different length and sequence (Chang et al., 2005). Oligonucleotide probes targeting any of these ITS regions can be used for identification.

    In conclusion, an oligonucleotide array was developed to identify 38 species of Gram-negative non-fermenters. With a sensitivity of 100 % and a specificity of 96.7 %, this array provides a rapid and relatively accurate method for species identification of clinically relevant non-fermenters. The results of array hybridization are available within one working day.

    Acknowledgments

    This project was partially supported by grants from the National Science Council (96-2320-B-006-024-MY3), Department of Economic Affairs (96-EC-17-A-10-S1-0013), Taiwan, and from the Center for Frontier Materials and Micro/Nano Science and Technology (D97-2700, D97-2720), National Cheng Kung University, Taiwan.

    References

    1. Baele, M., Baele, P., Vaneechoutte, M., Storms, V., Butaye, P., Devriese, L. A., Verschraegen, G., Gillis, M. & Haesebrouck, F. (2000). Application of tDNA-PCR for the identification of Enterococcus species. J Clin Microbiol 38, 4201–4207.
    2. Blondel-Hill, E., Henry, E. A. & Speert, D. P. (2007). Pseudomonas. In Manual of Clinical Microbiology, 9th edn, pp. 734–748. Edited by P. R. Murray, E. J. Baron, J. H. Jorgensen, M. L. Landry & M. A. Pfaller. Washington, DC: American Society for Microbiology.
    3. Bosshard, P. P., Zbinden, R., Abels, S., Böddinghaus, B., Altwegg, M. & Böttger, E. C. (2006). 16S rRNA gene sequencing versus the API 20 NE system and the VITEK 2 ID-GNB card for identification of nonfermenting Gram-negative bacteria in the clinical laboratory. J Clin Microbiol 44, 1359–1366.
    4. Brown, T. J. & Anthony, R. M. (2000). The addition of low numbers of 3′ thymine bases can be used to improve the hybridization signal of oligonucleotides for use within arrays on nylon supports. J Microbiol Methods 42, 203–207.
    5. Chang, H. C., Wei, Y. F., Dijkshoorn, L., Vaneechoutte, M., Tang, C. T. & Chang, T. C. (2005). Species-level identification of isolates of the Acinetobacter calcoaceticus–Acinetobacter baumannii complex by sequence analysis of the 16S–23S rRNA gene spacer region. J Clin Microbiol 43, 1632–1639.
    6. Chen, C. C., Teng, L. J. & Chang, T. C. (2004). Identification of clinically relevant viridans streptococci by sequence analysis of the 16S–23S rDNA spacer region. J Clin Microbiol 42, 2651–2657.
    7. Chen, Y. S., Hsiao, Y. S., Lin, H. H., Liu, Y. & Chen, Y. L. (2006). CpG-modified plasmid DNA encoding flagellin improves immunogenicity and provides protection against Burkholderia pseudomallei infection in BALB/c mice. Infect Immun 74, 1699–1705.
    8. Coenye, T., Spilker, T., Reik, R., Vandamme, P. & LiPuma, J. J. (2005). Use of PCR analyses to define the distribution of Ralstonia species recovered from patients with cystic fibrosis. J Clin Microbiol 43, 3463–3466.
    9. Dijkshoorn, L., van Harsselaar, B., Tjernberg, I., Bouvet, P. J. M. & Vaneechoutte, M. (1998). Evaluation of amplified ribosomal DNA restriction analysis for identification of Acinetobacter gen. sp. Syst Appl Microbiol 21, 33–39.
    10. Dijkshoorn, L., Nemec, A. & Seifert, H. (2007). An increasing threat in hospitals: multidrug-resistant Acinetobacter baumannii. Nat Rev Microbiol 5, 939–951.
    11. Ferroni, A., Sermet-Gaudelus, I., Abachin, E., Quesne, G., Lenoir, G., Berche, P. & Gaillard, J. L. (2002). Use of 16S rRNA gene sequencing for identification of nonfermenting Gram-negative bacilli recovered from patients attending a single cystic fibrosis center. J Clin Microbiol 40, 3793–3797.
    12. Fukushima, M., Kakinuma, K., Hayashi, H., Nagai, H., Ito, K. & Kawaguchi, R. (2003). Detection and identification of Mycobacterium species isolates by DNA microarray. J Clin Microbiol 41, 2605–2615.
    13. Funke, G. & Funke-Kissling, P. (2004). Evaluation of the new VITEK 2 card for identification of clinically relevant Gram-negative rods. J Clin Microbiol 42, 4067–4071.
    14. Guasp, C., Moore, E. R., Lalucat, J. & Bennasar, A. (2000). Utility of internally transcribed 16S–23S rDNA spacer regions for the definition of Pseudomonas stutzeri genomovars and other Pseudomonas species. Int J Syst Evol Microbiol 50, 1629–1639.
    15. Gürtler, V. & Stanisich, V. A. (1996). New approaches to typing and identification of bacteria using the 16S–23S rDNA spacer region. Microbiology 142, 3–16.
    16. Kiska, D. L., Kerr, A., Jones, M. C., Caracciolo, J. A., Eskridge, B., Jordan, M., Miller, S., Hughes, D., King, N. & Gilligan, P. H. (1996). Accuracy of four commercial systems for identification of Burkholderia cepacia and other Gram-negative nonfermenting bacilli recovered from patients with cystic fibrosis. J Clin Microbiol 34, 886–891.
    17. Ko, W.-C., Lee, N.-Y., Su, S. C., Dijkshoorn, L., Vaneechoutte, M., Wang, L.-R., Yan, J.-J. & Chang, T. C. (2008). Oligonucleotide-array based identification of species in the Acinetobacter calcoaceticus–A. baumannii complex isolated from blood cultures and antimicrobial susceptibility testing of the isolates. J Clin Microbiol 46, 2052–2059.
    18. Kolbert, C. P. & Persing, D. H. (1999). Ribosomal DNA sequencing as a tool for identification of bacterial pathogens. Curr Opin Microbiol 2, 299–305.
    19. LiPuma, J. J., Currie, B. J., Lum, G. D. & Vandamme, P. A. R. (2007). Burkholderia, Stenotrophomonas, Ralstonia, Cupriavidus, Pandoraea, Brevundimonas, Comamonas, Delftia, and Acidovorax. In Manual of Clinical Microbiology, 9th edn, pp. 749–769. Edited by P. R. Murray, E. J. Baron, J. H. Jorgensen, M. L. Landry & M. A. Pfaller. Washington, DC: American Society for Microbiology.
    20. Millar, B. C., Jiru, X., Moore, J. E. & Earle, J. A. (2000). A simple and sensitive method to extract bacterial, yeast and fungal DNA from blood culture material. J Microbiol Methods 42, 139–147.
    21. Nemec, A., De Baere, T., Tjernberg, I., Vaneechoutte, M., van der Reijden, T. J. K. & Dijkshoorn, L. (2001). Acinetobacter ursingii sp. nov. and Acinetobacter schindleri sp. nov., isolated from human clinical specimens. Int J Syst Evol Microbiol 51, 1891–1899.
    22. Noble, R. C. & Overman, S. B. (1994). Pseudomonas stutzeri infection: a review of hospital isolates and a review of the literature. Diagn Microbiol Infect Dis 19, 51–56.
    23. Palleroni, N. J. (1984). Pseudomonas. In Bergey's Manual of Systematic Bacteriology, vol. 1, pp. 141–199. Edited by N. R. Drieg & J. G. Holt. Baltimore: Williams & Wilkins.
    24. Park, H., Jang, H., Song, E., Chang, C. L., Lee, M., Jeong, S., Park, J., Kang, B. & Kim, C. (2005). Detection and genotyping of Mycobacterium species from clinical isolates and specimens by oligonucleotide array. J Clin Microbiol 43, 1782–1788.
    25. Patel, J. B. (2001). 16S rRNA gene sequencing for bacterial pathogen identification in the clinical laboratory. Mol Diagn 6, 313–321.
    26. Relman, D. A. (1993). Universal bacterial 16S rDNA amplification and sequencing. In Diagnostic Molecular Microbiology, pp. 489–495. Edited by D. H. Persing, F. C. Tenover, T. F. Smith & T. J. White. Washington, DC: American Society for Microbiology.
    27. Schreckenberger, P. C., Daneshvar, M. I., Weyant, R. S. & Hollis, D. G. (2007). Acinetobacter, Achromobacter, Chryseobacterium, Moraxella, and other nonfermentative Gram-negative rods. In Manual of Clinical Microbiology, 9th edn, pp. 770–802. Edited by P. R. Murray, E. J. Baron, J. H. Jorgensen, M. L. Landry & M. A. Pfaller. Washington, DC: American Society for Microbiology.
    28. Su, H. P., Yang, H. W., Chen, Y. L., Ferng, T. L., Chou, Y. L., Chung, T. C., Chen, C. H., Chiang, C. S., Kuan, M. M. & other authors (2007). Prevalence of melioidosis in the Er-Ren River Basin, Taiwan: implications for transmission. J Clin Microbiol 45, 2599–2603.
    29. Tjernberg, I. & Ursing, J. (1989). Clinical strains of Acinetobacter classified by DNA–DNA hybridization. APMIS 97, 595–605.
    30. Tung, S. K., Teng, L. J., Vaneechoutte, M., Chen, H. M. & Chang, T. C. (2006). Array-based identification of species of the genera Abiotrophia, Enterococcus, Granulicatella, and Streptococcus. J Clin Microbiol 44, 4414–4424.
    31. Tung, S. K., Teng, L. J., Vaneechoutte, M., Chen, H. M. & Chang, T. C. (2007). Identification of species of Abiotrophia, Enterococcus, Granulicatella and Streptococcus by sequence analysis of the ribosomal 16S–23S intergenic spacer region. J Med Microbiol 56, 504–513.
    32. Vaneechoutte, M. & De Baere, T. (2007). Taxonomy of the genus Acinetobacter based on 16S ribosomal RNA gene sequences. In Acinetobacter Molecular Biology, pp. 35–60. Edited by U. Gerischer. Norfolk, UK: Caister Academic Press.
    33. Vaneechoutte, M., Boerlin, P., Tichy, H.-V., Bannerman, E., Jäger, B. & Bille, J. (1998). Comparison of PCR-based DNA fingerprinting techniques for the identification of Listeria species and their use for atypical Listeria isolates. Int J Syst Bacteriol 48, 127–139.