Evaluation of RIDOM, MicroSeq, and GenBank services in the molecular identification of Nocardia species

• Published in: International journal of medical microbiology Vol. 293; no. 5; pp. 359 - 370
• Main Authors: Mellmann, Alexander, Cloud, Joann L., Andrees, Sebastian, Blackwood, Kym, Carroll, Karen C., Kabani, Amin, Roth, Andreas, Harmsen, Dag
• Format: Journal Article
• Language: English
• Published: Jena Elsevier GmbH 01.11.2003; Elsevier; Elsevier Science Ltd
• Subjects: 16S rDNA; Bacteriology; Biological and medical sciences; Databases, Nucleic Acid; DNA, Bacterial - chemistry; DNA, Bacterial - genetics; DNA, Ribosomal - chemistry; DNA, Ribosomal - genetics; Fundamental and applied biological sciences. Psychology; Humans; Microbiology; Miscellaneous; molecular identification; Nocardia; Nocardia - classification; Nocardia - genetics; Phylogeny; Polymerase Chain Reaction; RIDOM; RNA, Ribosomal, 16S - chemistry; RNA, Ribosomal, 16S - genetics; Sequence Analysis, DNA; Nocardia; 16S rDNA; molecular identification; RIDOM; Actinomycetales; Microbiology; Bacteria; Nocardiaceae; Actinomycetes; Identification
• ISSN: 1438-4221; 1618-0607
• DOI: 10.1078/1438-4221-00271

The molecular identification of Nocardia species, when compared to phenotypic identification, has two primary advantages: rapid turn-around time and improved accuracy. The information content in the 5′-end of the 16S ribosomal RNA gene is sufficient for identification of most bacterial species. An evaluation was performed to demonstrate the quality of results provided by two specialized databases (RIDOM and MicroSeq 500 versions 1.1 and 1.4.3, library version 500-0125, respectively) and the more general GenBank database. In addition, these results were compared with phenotypic identifications. Partial 5′-16S rDNA sequences from 64 culture collection strains (DSM, CIP, JCM, and ATCC) were derived, in duplicate, independently in two laboratories. Furthermore, the sequences and the conventional identification results of 91 clinical Nocardia isolates were determined. With the exception of N. soli and N. cummidelens all Nocardia type strains were distinguishable using 5′-16S rDNA sequencing. Assuming a normal distribution for the pairwise distances of all unique Nocardia sequences and choosing a reporting criterion of ≥99.12% similarity for a “distinct species”, a statistical error probability of 1.0% can be calculated. When the various databases were searched with the clinical isolate sequences RIDOM gave a perfect match in 71.4% of cases whereas MicroSeq yielded a perfect match in only 26.4%. The GenBank service gave a 100% similarity in 59.3% but in 70.4% of these cases the results obtained were not exclusive for a single Nocardia species. Conventional methods gave a correct identification in 59 cases, although most recent taxonomic changes were not taken into account. The RIDOM service (http://www.ridom-rdna.de/) is in the process of making available a comprehensive and high-quality database for bacterial identification purposes and provides excellent results for the majority of Nocardia isolates.