Sequencing of the rpoB Gene and Flanking Spacers for Molecular Identification of Acinetobacter Species
http://www.100md.com
微生物临床杂志 2006年第3期
Unite des Rickettsies, CNRS UMR 6020, Faculte de Medecine, Universite de la Mediterrannee, 27 Boulevard Jean Moulin, 13385 Marseille Cedex 05, France
ABSTRACT
Acinetobacter species are defined on the basis of several phenotypic characters, results of DNA-DNA homology, and more recently, similarities or dissimilarities in 16S rRNA gene sequences. However, the 16S rRNA gene is not polymorphic enough to clearly distinguish all Acinetobacter species. We used an RNA polymerase -subunit gene (rpoB)-based identification scheme for the delineation of species within the genus Acinetobacter, and towards that end, we determined the complete rpoB gene and flanking spacer (rplL-rpoB and rpoB-rpoC) sequences of the 17 reference strains of Acinetobacter species and 7 unnamed genomospecies. By using complete gene sequences (4,089 bp), we clearly separated all species and grouped them into different clusters. A phylogenetic tree constructed using these sequences was supported by bootstrap values higher than those obtained with 16S rRNA or the gyrB or recA gene. Four pairs of primers enabled us to amplify and sequence two highly polymorphic partial sequences (350 and 450 bp) of the rpoB gene. These and flanking spacers were designed and tested for rapid identification of the 17 reference strains of Acinetobacter species and 7 unnamed genomospecies. Each of these four variable sequences enabled us to delineate most species. Sequences of at least two polymorphic sequences should be used to distinguish Acinetobacter grimontii, Acinetobacter junii, Acinetobacter baylyi, and genomic species 9 from one another. Finally, 21 clinical isolates of Acinetobacter baumannii were tested for intraspecies relationships and assigned correctly to the same species by comparing the partial sequences of the rpoB gene and its flanking spacers.
INTRODUCTION
Acinetobacter species are ubiquitous in the environment and have emerged as important nosocomial pathogens (3). A propensity to tolerate drying (29, 30) and resistance to many commonly used antibiotics (28) are key factors in enabling the organism to survive and spread in the nosocomial environment. Furthermore, Acinetobacter species may serve as reservoirs of antibiotic-resistant genes, particularly in hospital environments (22).
The genus Acinetobacter, originally proposed by Brisou and Prevot (6), comprises a collection of bacteria which show a great deal of phenotypic similarities. Currently, the genus Acinetobacter comprises many taxons based on hybridization groups and classified as genomovars and genomospecies. From an official taxonomic point of view, there are currently 17 Acinetobacter nomenspecies with standing in nomenclature. However, several other genomospecies have been proposed, mainly on the basis of DNA hybridization studies (7, 20, 21, 28).
Delineation of species within the genus Acinetobacter is often problematic (22). A number of recent studies have attempted to set up reliable identification schemes based on DNA homology, phenotypic characters, and comparison of cell envelope and outer membrane protein patterns, but such methods yield inconsistent results (1, 10, 11, 12, 23). The phenotypic characters of Acinetobacter species are susceptible to the conditions under which they are cultured (10). Members of this genus have a large amount of genes devoted to catabolic pathways (named by authors as an "archipelago of catabolic diversity") that lead to adaptation to most substrates (2). Thus, auxanograms have only limited usefulness. It is therefore no wonder that the recent edition of a widely consulted clinical microbiology manual has chosen to separate Acinetobacter species into only two groups, depending on their abilities to oxidize glucose and show hemolytic activities (24).
The currently used 16S rRNA gene sequencing determinations have failed to distinguish closely related genomic species of Acinetobacter due to its extremely low polymorphic nature and cannot replace the standard DNA-DNA hybridization method (11, 32). We experienced misidentification of Acinetobacter baumannii isolates as Acinetobacter calcoaceticus using 16S rRNA gene sequence comparison (16) that was later correctly identified using internal transcribed spacer and recombinase A (recA) gene sequence analyses (18). Gene sequences, including those of internal transcribed spacer (9), gyrase B (gyrB), and recA genes, have been evaluated to delineate species within the genus Acinetobacter (15, 31, 33). However, unavailability of the sequences of gyrB and recA genes for 10 recently described Acinetobacter species precludes the use of these gene sequences for comparing different species of Acinetobacter (7, 19, 20).
Several studies have demonstrated the usefulness of RNA polymerase -subunit (rpoB) gene sequences for the identification and taxonomic classification of various bacterial species, including Bartonella, spirochete, Staphylococcus, Mycobacterium, Legionella, Streptococcus, Enterobacteriaceae, Afipia and Bosea, Corynebacterium, Pasteurellaceae, Streptomyces and Kitasatospora, Mycoplasma, Salmonella, Bacillus, Paenibacillus, Ehrlichia, Anaplasma, Neorickettsia, and Wolbachia species (19). In one study, of the seven genes tested, the rpoB gene had the highest discriminatory power in delineating Bartonella species (17). To our knowledge, rpoB has not been evaluated for the delineation of Acinetobacter species. Moreover, the presence of only one copy of the rpoB gene in the genome of Acinetobacter sp. ADP1 was reported, which shows that the gene may be phylogenetically reliable (2).
In this study, we investigated the usefulness of rpoB gene sequencing for the differentiation and identification of 24 Acinetobacter species. Moreover, we reported herein four highly variable regions, including two polymorphic areas in the rpoB gene and two flanking spacers bordered by conserved sequences, with the objective of designing universal primers for amplification of a small but discriminative sequence for routine molecular identification of Acinetobacter species.
MATERIALS AND METHODS
Bacterial strains. A total of 24 Acinetobacter strains representing type strains of currently accepted species were obtained from the Collection de l'Institut Pasteur (CIP) and examined in the present study (Table 1). Additionally, 21 isolates of A. baumannii, including the louse strain of La Scola and Raoult (16, 18), were also studied (see Table S1 in the supplemental material). All strains were grown on Columbia agar plates with 5% sheep blood (Trypticase soy agar; bioMerieux, Marcy-l'Etoile, France). These culture plates were incubated for 48 h at 37°C under aerobic conditions.
Amplification and sequencing of the rpoB gene. The sequences of rplL, rpoB, and rpoC of Acinetobacter sp. ADP1 (GenBank accession number NC005966) and Pseudomonas syringae pv. tomato strain DC3000 (GenBank accession number NC 006347) had been aligned to design primers that were subsequently used to amplify the complete rpoB gene and flanking spacers. Additional primers were designed as required. This procedure as well as the procedures for DNA extraction, PCR amplification, and sequencing has been detailed elsewhere (14).
Analysis of complete rpoB sequences. The nucleotide sequences of the target rpoB gene fragments obtained were processed into sequence data using standard software (Sequence Analysis Software; Applied Biosystems). The partially overlapping sequences encompassing the complete rpoB gene were combined into a single consensus sequence using sequence assembler software (Applied Biosystems). Multiple rpoB gene sequence alignments among all 24 species were made using the Clustal W program (26) on the EMBL-EBI web server (http://www.ebi.ac.uk/clustalw/).
The phylogenetic relationship between complete sequences of the rpoB gene and two spacers was estimated by the neighbor-joining method (8). Phylogenetic trees were constructed, and their node reliabilities were estimated by performing bootstrap replicates. Bootstrap values were obtained from 1,000 trees generated randomly using SEQBOOT (PHYLIP software package). Proportional similarities of partial and complete rpoB sequences and spacers (rplL-rpoB and rpoB-rpoC) among species were computed using the MEGA 2.1 program.
Determination of discriminative partial sequences of the rpoB gene and flanking spacers. SVARAP (for Sequence VARiability Analysis Program; hypertext link "Telechargements") software was used (http://ifr48.free.fr/recherche/jeu_cadre/jeu_rickettsie.html) to find the portions of sequences with high variability bordering highly conserved regions. These short fragments showing high variability were amplified and sequenced in all strains, including all isolates of A. baumannii. The primers used and their targets were as listed in Table 2. PCRs were performed as described previously (14), with the exception of primer annealing (48°C for 30 s) and extension (72°C for 1 min).
Nucleotide sequence accession numbers. GenBank accession numbers for sequences of the rpoB gene and flanking spacers obtained in the present study are listed in Table 1.
RESULTS
Complete rpoB sequences of Acinetobacter species. The primers, designed as described above, amplified specific regions of rpoB in all the strains tested. In all species tested, the total size of the rpoB gene was 4,089 bp. Pairwise comparison of the percent similarities of DNA of rpoB genes of various strains tested varied between 83 and 99% (see Table S2 in the supplemental material). Only two pairs of species, Acinetobacter junii/Acinetobacter grimontii and Acinetobacter baylyi/genomic species 11, showed 99% DNA homology. The others showed less than 95% DNA homology when pairwise comparisons were made.
Identification of Acinetobacter species based on partial sequences. Using SVARAP software, we have identified two variable areas bordered by highly conserved regions: zone 1, between positions 2900 and 3250, and zone 2, between positions 3250 and 3700 (Fig. 1). The primers targeting these variable areas were designed. Using the primers, as indicated above, we amplified zones 1 and 2 of all reference strains and all isolates of A. baumannii. The sizes of the variable zone 1 and zone 2 were 350 and 450 bp, respectively. The percent similarities of zone 1 varied between 78.6 and 99.1%. In a pairwise comparison of DNA homology of the complete rpoB gene, A. baylyi/genomic species 11 and A. junii/A. grimontii showed proportional similarities of 98 and 99.1%, respectively. In the remaining strains tested, such homology was <96%. The percent similarity of zone 2 ranged between 75.8 and 99.6%. Again, in a pairwise comparison, the species pairs A. junii/A. grimontii and A. baylyi/genomic species 11 showed percent similarities of 98.8 and 99.6%, respectively. In contrast, in other species tested, such percent similarity in a pairwise comparison, as indicated above, was 96%. Overall, the percentage of sequence variability in the partial rpoB gene ranged between 0.4 and 24.2%. In contrast, in the complete rpoB gene sequence, the percentage of such variability ranged between 0.8 and 16.9%. The two partial sequences unambiguously established the identities of all 24 strains of Acinetobacter.
Intraspecies similarities in the partial rpoB gene sequence were observed among 21 clinical isolates of A. baumannii. In a pairwise comparison, the strain CIP 103655 showed between 94.9 and 95.7% similarity of zone 1 sequences with other strains tested. The other strains tested showed between 98.3 and 100% similarity of zone 1 sequences among one another. For zone 2 sequences, there was between 98.7 and 100% similarity when two strains other than strain CIP 103655 were compared. Compared with other strains, zone 2 sequences of strain CIP 103655 showed between 93.6 and 94.4% similarity with those of other strains. Two species were found to be the closest to A. baumannii. These included genospecies 3 (percent similarity, 95.1%) and A. calcoaceticus (percent similarity, 93.6%) for zone 1 and zone 2, respectively. Except for that of the CIP 103655 strain, the percent similarities in zone 1 and 2 sequences between A. baumannii and its closest species were lower than the intraspecies similarities. For CIP 103655, the most distant isolate of A. baumannii was 94.9% and 93.6% similar in zone 1 and zone 2, respectively.
Analysis of flanking spacers of the rpoB gene. The lengths of two spacers (rplL-rpoB and rpoB-rpoC spacers) varied from species to species. The sizes of these spacers were as listed in Table 1 and Table S1. The rplL-rpoB spacer (size, 301 to 310 bp) showed between 80.8 and 100% sequence similarity among various species studied herein. As was the case with partial and complete rpoB sequences, A. junii and A. grimontii showed 100% similarity in rplL-rpoB spacer sequences. Between A. baylyi/genomic species 11 and Acinetobacter lwoffii/genomic species 9, there was 98.4 to 99.7% similarity in rplL-rpoB spacer sequences.
The rpoB-rpoC spacer (size, 86 to 177 bp) sequences showed 70.2 to 99.5% similarity among various species studied herein. A. junii and A. grimontii showed 99.5% similarity in rpoB-rpoC spacer sequences. Among one another, the members (A. calcoaceticus, A. baumannii, and genomic species 3) of the so-called A. calcoaceticus/A. baumannii complex showed 98.5 to 99.0% similarity in rpoB-rpoC spacer sequences. In contrast, there were only 83.8 and 87.9% similarities in rpoB-rpoC spacer sequences between A. baylyi/genomic species 11 and A. lwoffii/genomic species 9, respectively.
The size of the rplL-rpoB spacer in strain CIP 103655 (size, 304 bp) was 1 base pair shorter than that found in other strains of A. baumannii. However, there was no difference among various A. baumannii strains tested in terms of the size (86 bp) of the rpoB-rpoC spacer. The strain CIP 103655 showed 96.1 to 96.4% similarity in the rplL-rpoB spacer sequence with other strains. The remaining strains of A. baumannii tested showed 99 to 100% similarity in the rplL-rpoB spacer sequence among one another. Compared in a pair, there was 97.7 to 98.8% similarity in rpoB-rpoC spacer sequences between strain CIP103655 and other A. baumannii strains. For other strains of A. baumannii, intrastrain similarities in rpoB-rpoC were 100%. The species found to be the closest to A. baumannii were genospecies 3 for rplL-rpoB (similarity, 95.9%) and A. calcoaceticus for rpoB-rpoC (similarity, 98.5%). Except for that of CIP 103655, these similarities were clearly lower than the intraspecies similarities observed among other strains of A. baumannii tested.
Phylogenetic analysis of Acinetobacter species. A phylogenetic tree was constructed from complete rpoB gene sequences using the neighbor-joining method. The phylogenetic tree so constructed was supported by high bootstrap values (Fig. 2). Compared to that obtained by analyzing 16S rRNA genes (7 of 22), a bootstrap value of 75% was obtained significantly more frequently when complete rpoB genes of the strains (17 of 22) had been analyzed (P < 0.01). All the species were clearly separated into different groups. The tree based on partial rpoB (concatenated zone 1 and zone 2) showed homogeneous grouping of A. baumannii (Fig. 3). Although the strain 103655 remained distinctly apart from other isolates of A. baumannii, it appeared in the same cluster. This clustering was supported by a bootstrap value of 85%.
DISCUSSION
The interspecies relationships in the genus Acinetobacter need further definition. This may be understandable, considering the fact that many genospecies are very similar phenotypically and, therefore, are not readily distinguishable (10). The increasing pathogenic importance of Acinetobacter has stimulated the development of reliable identification methods for these strains. Although considered the gold standard, the DNA-DNA hybridization method is time consuming and labor intensive (25). Sequencing of housekeeping genes in general and protein-encoding genes in particular and sequencing of 16S DNA have been used in the identification of species or strains and delineation of phylogenetic relationships among different strains and species within the genus Acinetobacter. Compared to 16S rRNA genes, housekeeping genes provide a higher degree of resolution as the latter genes evolve faster than the former (32). It may not therefore be surprising that various protein-encoding genes, including recA, groEL, hsp75, rpoB, rpoD, and gyrB, have been used for the classification of several unrelated bacteria at the intrageneric level (13, 14, 15, 31, 32, 34). Unfortunately, the gyrB and recA gene sequences are not available for 10 recently described species of Acinetobacter. This makes it impossible to compare their performances with those of rpoB gene sequences in delineating various species of Acinetobacter.
Several investigators have sequenced gyrB and recA genes of 14 genomic species of Acinetobacter and studied them to delineate phylogenetic relationships among different members of the genus Acinetobacter (5, 15, 31, 33). Furthermore, these investigators compared the performance of the gene sequence-based identification scheme, as indicated above, with that based on DNA-DNA hybridization (4, 27). In our study, we constructed phylogenetic trees incorporating 14 genospecies of Acinetobacter and found no congruency among four individual phylogenetic trees constructed using the sequences of 16S rRNA, rpoB, gyrB, and recA genes (data not shown). However, a bootstrap value of 75% was a statistically significant (P < 0.05) occurrence when a phylogenetic tree was based on rpoB (11 of 12) gene sequences compared to that based on either 16S rRNA (4 of 12), gyrB (5 of 12), or recA (6 of 12) gene sequences. This shows that the robustness of the phylogenetic tree was based on the rpoB gene sequence. A. lwoffii and Acinetobacter genomic species 9 were shown to be 100% identical when their partial gyrB gene sequences had been analyzed and compared. However, sequencing of gyrD and recA genes (15, 33) separated them. In the present study, these two species were clearly differentiated as they showed variability between 4.9 and 6.7% in their complete rpoB gene sequence and two partial sequences of the rpoB gene. The sequence of the rpoB gene failed to clearly distinguish A. baylyi from genospecies 9 on one hand and A. grimontii from A. baylyi on the other. Unfortunately, the sequences of gyrB and recA genes are not available to use for such comparisons.
For the routine identification of Acinetobacter, either of two partial sequences and spacers can be used as they have high discriminative power and short lengths. However, such an identification scheme may not be sufficient to distinguish A. grimontii from A. junii and A. baylyi from Acinetobacter genomic species 9 (Table 3). One way of addressing this problem is to incorporate another short variable sequence into the identification scheme. In order to identify an Acinetobacter isolate, zone 1 should be determined first as it is the shortest and allows clear identification of 20 of 24 species. If the sequences of zone 1 in A. grimontii and A. junii are similar, then the sequences of zone 2 should be studied to make a clear distinction between these two species (Table 3). On the other hand, if the sequences of zone 1 in A. baylyi and genomic species 9 are similar, then it will be useful to determine the sequence of the rpoB-rpoC spacer as it may make a clear distinction between these two species (Table 3).
The intraspecies variability of the short discriminative fragments observed within the A. baumannii species showed that, with exception of CIP 103655, all isolates have clearly higher degrees of similarities than that observed between A. baumannii and its closest species. However, the similarities observed between A. baumannii CIP 103655 and other isolates of the genus suggest that the identification of some isolates in this species could be ambiguous. All in all, an identification scheme based on analyzing partial rpoB gene sequences and flanking spacers may prove useful in the molecular identification of Acinetobacter. However, due to possible intraspecies dissimilarities and interspecies high similarities, such a diagnostic scheme should be undertaken for a large collection of Acinetobacter isolates. Moreover, the status of some isolates (e.g., A. baumannii CIP 103655) or some species (e.g., A. grimontii and A. baylyi) should be assessed by DNA-DNA hybridization studies and analyses of the sequences of such housekeeping genes as recA and gyrB. The molecular identification scheme proposed herein will provide better genetic definition of large collections of strains representing named species and will then serve as the basis for a large-scale phenotypic study to determine better biochemical tests to separate groups, especially the group of Acinetobacter species isolated in humans.
Supplemental material for this article may be found at http://jcm.asm.org/.
REFERENCES
Alexander, M., F. Ismail, P. J. H. Jackman, and W. C. Nobel. 1984. Fingerprinting Acinetobacter strains from clinical sources by numerical analysis of electrophoretic patterns. J. Med. Microbiol. 18:55-68.
Barbe, V., D. Vallenet, N. Fonknechten, A. Kreimeyer, S. Oztas, L. Labarre, S. Cruveiller, C. Robert, S. Duprat, P. Wincker, L. N. Ornston, J. Weissenbach, P. Marliere, G. N. Cohen, and C. Medigue. 2004. Unique features revealed by the genome sequence of Acinetobacter sp. ADP1, a versatile and naturally transformation competent bacterium. Nucleic Acids Res. 32:5766-5779.
Bergogne-Berezin, E., and K. J. Towner. 1996. Acinetobacter spp. as nosocomial pathogens: microbiological, clinical, and epidemiological features. Clin. Microbiol. Rev. 9:148-165.
Bouvet, P. J. M., and P. A. D. Grimont. 1986. Taxonomy of the genus Acinetobacter with the recognition of Acinetobacter baumannii sp. nov., Acinetobacter haemolyticus sp. nov., Acinetobacter johnsonii sp. nov., and Acinetobacter junii sp. nov. and emended descriptions of Acinetobacter calcoaceticus and Acinetobacter lwoffii. Int. J. Syst. Bacteriol. 36:228-240.
Bouvet, P. J. M., and P. A. D. Grimont. 1987. Identification and biotyping of clinical isolates of Acinetobacter. Ann. Inst. Pasteur Microbiol. 138:569-578.
Brisou, J., and A. R. Prevot. 1954. [Studies on bacterial taxonomy. X. The revision of species under Achromobacter group.] Ann. Inst. Pasteur (Paris) 86:722-728.
Carr, E. L., P. Kampfer, B. K. Patel, V. Gurtler, and R. J. Seviour. 2003. Seven novel species of Acinetobacter isolated from activated sludge. Int. J. Syst. Evol. Microbiol. 53:953-963.
Felsenstein, J. 1989. PHYLIP—phylogeny inference package (version 3.2). Cladistics 5:164-166.
Garcia, A., R. Montoya, H. Bello, G. Gonzalez, M. Dominguez, and R. Zemelman. 1996. Differentiation of Acinetobacter baumannii biotypes by amplification of 16S-23S rRNA intergenic spacer sequences. Microbios 88:159-167.
Gerner-Smidt, P., I. Tjernberg, and J. Ursing. 1991. Reliability of phenotypic tests for identification of Acinetobacter species. J. Clin. Microbiol. 29:277-282.
Ibrahim, A., P. Gerner-Smidt, and W. Liesack. 1997. Phylogenetic relationship of the twenty-one DNA groups of the genus Acinetobacter as revealed by 16S ribosomal DNA sequence analysis. Int. J. Syst. Bacteriol. 47:837-841.
Ino, T., and Y. Nishimura. 1989. Taxonomic studies of Acinetobacter species based on outer membrane protein patterns. J. Gen. Appl. Microbiol. 16:213-224.
Karlin, S., G. M. Weinstock, and V. Brendel. 1995. Bacterial classifications derived from RecA protein sequence comparisons. J. Bacteriol. 177:6881-6893.
Khamis, A., D. Raoult, and B. La Scola. 2004. rpoB gene sequencing for identification of Corynebacterium species. J. Clin. Microbiol. 42:3925-3931.
Krawczyk, B., K. Lewandowski, and J. Kur. 2002. Comparative studies of the Acinetobacter genus and the species identification method based on the recA sequences. Mol. Cell. Probes 16:1-11.
La Scola, B., P.-E. Fournier, P. Brouqui, and D. Raoult. 2001. Detection and culture of Bartonella quintana, Serratia marcescens, and Acinetobacter spp. from decontaminated human body lice. J. Clin. Microbiol. 39:1707-1709.
La Scola, B., Z. Zeaiter, A. Khamis, and D. Raoult. 2003. Gene-sequence-based criteria for species definition in bacteriology: the Bartonella paradigm. Trends Microbiol. 11:318-321.
La Scola, B., and D. Raoult. 2004. Acinetobacter baumannii in human body louse. Emerg. Infect. Dis. 10:1671-1673.
Nemec, A., L. Dijkshoorn, I. Cleenwerck, T. De Baere, D. Janssens, T. J. Van Der Reijden, P. Jezek, and M. Vaneechoutte. 2003. Acinetobacter parvus sp. nov., a small-colony-forming species isolated from human clinical specimens. Int. J. Syst. Evol. Microbiol. 53:1563-1567.
Nemec, A., T. De Baere, I. Tjernberg, M. Vaneechoutte, T. J. van der Reijden, and L. Dijkshoorn. 2001. Acinetobacter ursingii sp. nov. and Acinetobacter schindleri sp. nov., isolated from human clinical specimens. Int. J. Syst. Evol. Microbiol. 51:1891-1899.
Nishimura, Y., T. Ino, and H. Lizuka. 1988. Acinetobacter radioresistens sp. nov. isolated from cotton and soil. Int. J. Syst. Bacteriol. 38:209-211.
Nowak, A., and J. Kur. 1996. Differentiation of seventeen genospecies of Acinetobacter by multiplex polymerase chain reaction and restriction fragment length polymorphism analysis. Mol. Cell. Probes 10:405-411.
Rainey, F. A., E. Lang, and E. Stackebrandt. 1994. The phylogenetic structure of the genus Acinetobacter. FEMS Microbiol. Lett. 124:349-353.
Schreckenberger, P. C., M. I. Daneshvar, R. S. Weyant, and D. G. Hollis. 2003. Acinetobacter, Achromobacter, Chryseobacterium, Moraxella, and other nonfermentative gram-negative rods, p. 749-779. In P. R. Murray, E. J. Baron, J. H. Jorgensen, M. A. Pfaller, and R. H. Yolken (ed.), Manual of clinical microbiology, 8th ed. American Society for Microbiology, Washington, D.C.
Stackebrandt, E., and B. M. Goebel. 1994. Taxonomic note: a place for DNA-DNA reassociation and 16S rRNA sequence analysis in the present species definition in bacteriology. Int. J. Syst. Bacteriol. 44:846-849.
Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673-4680.
Tjernberg, I., and J. Ursing. 1989. Clinical strains of Acinetobacter classified by DNA-DNA hybridization. APMIS 97:595-605.
Van Dessel, H., L. Dijkshoorn, T. Van Der Reijden, N. Bakker, A. Paauw, P. Van Den Broek, J. Verhoef, and S. Brisse. 2004. Identification of a new geographically widespread multiresistant Acinetobacter baumannii clone from European hospitals. Res. Microbiol. 155:105-112.
Webster, C. A., M. Crow, H. Humphreys, and K. S. Towner. 1998. Surveillance of an adult intensive care unit for a long-term persistence of a multi-resistant strain of Acinetobacter baumannii. Eur. J. Clin. Microbiol. Infect. Dis. 17:171-176.
Wendt, C., B. Dietz, E. Dietz, and H. Ruden. 1997. Survival of Acinetobacter baumannii on dry surfaces. J. Clin. Microbiol. 35:1394-1397.
Yamamoto, S., and S. Harayama. 1996. Phylogenetic analysis of Acinetobacter strains based on the nucleotide sequences of gyrB genes and on the amino acid sequences of their products. Int. J. Syst. Bacteriol. 46:506-511.
Yamamoto, S., and S. Harayama. 1998. Phylogenetic relationships of Pseudomonas putida strains deduced from the nucleotide sequences of gyrB, rpoD and 16S rRNA genes. Int. J. Syst. Bacteriol. 48:813-819.
Yamamoto, S., P. J. Bouvet, and S. Harayama. 1999. Phylogenetic structures of the genus Acinetobacter based on gyrB sequences: comparison with the grouping by DNA-DNA hybridization. Int. J. Syst. Bacteriol. 49:87-95.
Zeaiter, Z., P. E. Fournier, H. Ogata, and D. Raoult. 2002. Phylogenetic classification of Bartonella species by comparing groEL sequences. Int. J. Syst. Evol. Microbiol. 52:165-171.(Bernard La Scola, Vijay A)
ABSTRACT
Acinetobacter species are defined on the basis of several phenotypic characters, results of DNA-DNA homology, and more recently, similarities or dissimilarities in 16S rRNA gene sequences. However, the 16S rRNA gene is not polymorphic enough to clearly distinguish all Acinetobacter species. We used an RNA polymerase -subunit gene (rpoB)-based identification scheme for the delineation of species within the genus Acinetobacter, and towards that end, we determined the complete rpoB gene and flanking spacer (rplL-rpoB and rpoB-rpoC) sequences of the 17 reference strains of Acinetobacter species and 7 unnamed genomospecies. By using complete gene sequences (4,089 bp), we clearly separated all species and grouped them into different clusters. A phylogenetic tree constructed using these sequences was supported by bootstrap values higher than those obtained with 16S rRNA or the gyrB or recA gene. Four pairs of primers enabled us to amplify and sequence two highly polymorphic partial sequences (350 and 450 bp) of the rpoB gene. These and flanking spacers were designed and tested for rapid identification of the 17 reference strains of Acinetobacter species and 7 unnamed genomospecies. Each of these four variable sequences enabled us to delineate most species. Sequences of at least two polymorphic sequences should be used to distinguish Acinetobacter grimontii, Acinetobacter junii, Acinetobacter baylyi, and genomic species 9 from one another. Finally, 21 clinical isolates of Acinetobacter baumannii were tested for intraspecies relationships and assigned correctly to the same species by comparing the partial sequences of the rpoB gene and its flanking spacers.
INTRODUCTION
Acinetobacter species are ubiquitous in the environment and have emerged as important nosocomial pathogens (3). A propensity to tolerate drying (29, 30) and resistance to many commonly used antibiotics (28) are key factors in enabling the organism to survive and spread in the nosocomial environment. Furthermore, Acinetobacter species may serve as reservoirs of antibiotic-resistant genes, particularly in hospital environments (22).
The genus Acinetobacter, originally proposed by Brisou and Prevot (6), comprises a collection of bacteria which show a great deal of phenotypic similarities. Currently, the genus Acinetobacter comprises many taxons based on hybridization groups and classified as genomovars and genomospecies. From an official taxonomic point of view, there are currently 17 Acinetobacter nomenspecies with standing in nomenclature. However, several other genomospecies have been proposed, mainly on the basis of DNA hybridization studies (7, 20, 21, 28).
Delineation of species within the genus Acinetobacter is often problematic (22). A number of recent studies have attempted to set up reliable identification schemes based on DNA homology, phenotypic characters, and comparison of cell envelope and outer membrane protein patterns, but such methods yield inconsistent results (1, 10, 11, 12, 23). The phenotypic characters of Acinetobacter species are susceptible to the conditions under which they are cultured (10). Members of this genus have a large amount of genes devoted to catabolic pathways (named by authors as an "archipelago of catabolic diversity") that lead to adaptation to most substrates (2). Thus, auxanograms have only limited usefulness. It is therefore no wonder that the recent edition of a widely consulted clinical microbiology manual has chosen to separate Acinetobacter species into only two groups, depending on their abilities to oxidize glucose and show hemolytic activities (24).
The currently used 16S rRNA gene sequencing determinations have failed to distinguish closely related genomic species of Acinetobacter due to its extremely low polymorphic nature and cannot replace the standard DNA-DNA hybridization method (11, 32). We experienced misidentification of Acinetobacter baumannii isolates as Acinetobacter calcoaceticus using 16S rRNA gene sequence comparison (16) that was later correctly identified using internal transcribed spacer and recombinase A (recA) gene sequence analyses (18). Gene sequences, including those of internal transcribed spacer (9), gyrase B (gyrB), and recA genes, have been evaluated to delineate species within the genus Acinetobacter (15, 31, 33). However, unavailability of the sequences of gyrB and recA genes for 10 recently described Acinetobacter species precludes the use of these gene sequences for comparing different species of Acinetobacter (7, 19, 20).
Several studies have demonstrated the usefulness of RNA polymerase -subunit (rpoB) gene sequences for the identification and taxonomic classification of various bacterial species, including Bartonella, spirochete, Staphylococcus, Mycobacterium, Legionella, Streptococcus, Enterobacteriaceae, Afipia and Bosea, Corynebacterium, Pasteurellaceae, Streptomyces and Kitasatospora, Mycoplasma, Salmonella, Bacillus, Paenibacillus, Ehrlichia, Anaplasma, Neorickettsia, and Wolbachia species (19). In one study, of the seven genes tested, the rpoB gene had the highest discriminatory power in delineating Bartonella species (17). To our knowledge, rpoB has not been evaluated for the delineation of Acinetobacter species. Moreover, the presence of only one copy of the rpoB gene in the genome of Acinetobacter sp. ADP1 was reported, which shows that the gene may be phylogenetically reliable (2).
In this study, we investigated the usefulness of rpoB gene sequencing for the differentiation and identification of 24 Acinetobacter species. Moreover, we reported herein four highly variable regions, including two polymorphic areas in the rpoB gene and two flanking spacers bordered by conserved sequences, with the objective of designing universal primers for amplification of a small but discriminative sequence for routine molecular identification of Acinetobacter species.
MATERIALS AND METHODS
Bacterial strains. A total of 24 Acinetobacter strains representing type strains of currently accepted species were obtained from the Collection de l'Institut Pasteur (CIP) and examined in the present study (Table 1). Additionally, 21 isolates of A. baumannii, including the louse strain of La Scola and Raoult (16, 18), were also studied (see Table S1 in the supplemental material). All strains were grown on Columbia agar plates with 5% sheep blood (Trypticase soy agar; bioMerieux, Marcy-l'Etoile, France). These culture plates were incubated for 48 h at 37°C under aerobic conditions.
Amplification and sequencing of the rpoB gene. The sequences of rplL, rpoB, and rpoC of Acinetobacter sp. ADP1 (GenBank accession number NC005966) and Pseudomonas syringae pv. tomato strain DC3000 (GenBank accession number NC 006347) had been aligned to design primers that were subsequently used to amplify the complete rpoB gene and flanking spacers. Additional primers were designed as required. This procedure as well as the procedures for DNA extraction, PCR amplification, and sequencing has been detailed elsewhere (14).
Analysis of complete rpoB sequences. The nucleotide sequences of the target rpoB gene fragments obtained were processed into sequence data using standard software (Sequence Analysis Software; Applied Biosystems). The partially overlapping sequences encompassing the complete rpoB gene were combined into a single consensus sequence using sequence assembler software (Applied Biosystems). Multiple rpoB gene sequence alignments among all 24 species were made using the Clustal W program (26) on the EMBL-EBI web server (http://www.ebi.ac.uk/clustalw/).
The phylogenetic relationship between complete sequences of the rpoB gene and two spacers was estimated by the neighbor-joining method (8). Phylogenetic trees were constructed, and their node reliabilities were estimated by performing bootstrap replicates. Bootstrap values were obtained from 1,000 trees generated randomly using SEQBOOT (PHYLIP software package). Proportional similarities of partial and complete rpoB sequences and spacers (rplL-rpoB and rpoB-rpoC) among species were computed using the MEGA 2.1 program.
Determination of discriminative partial sequences of the rpoB gene and flanking spacers. SVARAP (for Sequence VARiability Analysis Program; hypertext link "Telechargements") software was used (http://ifr48.free.fr/recherche/jeu_cadre/jeu_rickettsie.html) to find the portions of sequences with high variability bordering highly conserved regions. These short fragments showing high variability were amplified and sequenced in all strains, including all isolates of A. baumannii. The primers used and their targets were as listed in Table 2. PCRs were performed as described previously (14), with the exception of primer annealing (48°C for 30 s) and extension (72°C for 1 min).
Nucleotide sequence accession numbers. GenBank accession numbers for sequences of the rpoB gene and flanking spacers obtained in the present study are listed in Table 1.
RESULTS
Complete rpoB sequences of Acinetobacter species. The primers, designed as described above, amplified specific regions of rpoB in all the strains tested. In all species tested, the total size of the rpoB gene was 4,089 bp. Pairwise comparison of the percent similarities of DNA of rpoB genes of various strains tested varied between 83 and 99% (see Table S2 in the supplemental material). Only two pairs of species, Acinetobacter junii/Acinetobacter grimontii and Acinetobacter baylyi/genomic species 11, showed 99% DNA homology. The others showed less than 95% DNA homology when pairwise comparisons were made.
Identification of Acinetobacter species based on partial sequences. Using SVARAP software, we have identified two variable areas bordered by highly conserved regions: zone 1, between positions 2900 and 3250, and zone 2, between positions 3250 and 3700 (Fig. 1). The primers targeting these variable areas were designed. Using the primers, as indicated above, we amplified zones 1 and 2 of all reference strains and all isolates of A. baumannii. The sizes of the variable zone 1 and zone 2 were 350 and 450 bp, respectively. The percent similarities of zone 1 varied between 78.6 and 99.1%. In a pairwise comparison of DNA homology of the complete rpoB gene, A. baylyi/genomic species 11 and A. junii/A. grimontii showed proportional similarities of 98 and 99.1%, respectively. In the remaining strains tested, such homology was <96%. The percent similarity of zone 2 ranged between 75.8 and 99.6%. Again, in a pairwise comparison, the species pairs A. junii/A. grimontii and A. baylyi/genomic species 11 showed percent similarities of 98.8 and 99.6%, respectively. In contrast, in other species tested, such percent similarity in a pairwise comparison, as indicated above, was 96%. Overall, the percentage of sequence variability in the partial rpoB gene ranged between 0.4 and 24.2%. In contrast, in the complete rpoB gene sequence, the percentage of such variability ranged between 0.8 and 16.9%. The two partial sequences unambiguously established the identities of all 24 strains of Acinetobacter.
Intraspecies similarities in the partial rpoB gene sequence were observed among 21 clinical isolates of A. baumannii. In a pairwise comparison, the strain CIP 103655 showed between 94.9 and 95.7% similarity of zone 1 sequences with other strains tested. The other strains tested showed between 98.3 and 100% similarity of zone 1 sequences among one another. For zone 2 sequences, there was between 98.7 and 100% similarity when two strains other than strain CIP 103655 were compared. Compared with other strains, zone 2 sequences of strain CIP 103655 showed between 93.6 and 94.4% similarity with those of other strains. Two species were found to be the closest to A. baumannii. These included genospecies 3 (percent similarity, 95.1%) and A. calcoaceticus (percent similarity, 93.6%) for zone 1 and zone 2, respectively. Except for that of the CIP 103655 strain, the percent similarities in zone 1 and 2 sequences between A. baumannii and its closest species were lower than the intraspecies similarities. For CIP 103655, the most distant isolate of A. baumannii was 94.9% and 93.6% similar in zone 1 and zone 2, respectively.
Analysis of flanking spacers of the rpoB gene. The lengths of two spacers (rplL-rpoB and rpoB-rpoC spacers) varied from species to species. The sizes of these spacers were as listed in Table 1 and Table S1. The rplL-rpoB spacer (size, 301 to 310 bp) showed between 80.8 and 100% sequence similarity among various species studied herein. As was the case with partial and complete rpoB sequences, A. junii and A. grimontii showed 100% similarity in rplL-rpoB spacer sequences. Between A. baylyi/genomic species 11 and Acinetobacter lwoffii/genomic species 9, there was 98.4 to 99.7% similarity in rplL-rpoB spacer sequences.
The rpoB-rpoC spacer (size, 86 to 177 bp) sequences showed 70.2 to 99.5% similarity among various species studied herein. A. junii and A. grimontii showed 99.5% similarity in rpoB-rpoC spacer sequences. Among one another, the members (A. calcoaceticus, A. baumannii, and genomic species 3) of the so-called A. calcoaceticus/A. baumannii complex showed 98.5 to 99.0% similarity in rpoB-rpoC spacer sequences. In contrast, there were only 83.8 and 87.9% similarities in rpoB-rpoC spacer sequences between A. baylyi/genomic species 11 and A. lwoffii/genomic species 9, respectively.
The size of the rplL-rpoB spacer in strain CIP 103655 (size, 304 bp) was 1 base pair shorter than that found in other strains of A. baumannii. However, there was no difference among various A. baumannii strains tested in terms of the size (86 bp) of the rpoB-rpoC spacer. The strain CIP 103655 showed 96.1 to 96.4% similarity in the rplL-rpoB spacer sequence with other strains. The remaining strains of A. baumannii tested showed 99 to 100% similarity in the rplL-rpoB spacer sequence among one another. Compared in a pair, there was 97.7 to 98.8% similarity in rpoB-rpoC spacer sequences between strain CIP103655 and other A. baumannii strains. For other strains of A. baumannii, intrastrain similarities in rpoB-rpoC were 100%. The species found to be the closest to A. baumannii were genospecies 3 for rplL-rpoB (similarity, 95.9%) and A. calcoaceticus for rpoB-rpoC (similarity, 98.5%). Except for that of CIP 103655, these similarities were clearly lower than the intraspecies similarities observed among other strains of A. baumannii tested.
Phylogenetic analysis of Acinetobacter species. A phylogenetic tree was constructed from complete rpoB gene sequences using the neighbor-joining method. The phylogenetic tree so constructed was supported by high bootstrap values (Fig. 2). Compared to that obtained by analyzing 16S rRNA genes (7 of 22), a bootstrap value of 75% was obtained significantly more frequently when complete rpoB genes of the strains (17 of 22) had been analyzed (P < 0.01). All the species were clearly separated into different groups. The tree based on partial rpoB (concatenated zone 1 and zone 2) showed homogeneous grouping of A. baumannii (Fig. 3). Although the strain 103655 remained distinctly apart from other isolates of A. baumannii, it appeared in the same cluster. This clustering was supported by a bootstrap value of 85%.
DISCUSSION
The interspecies relationships in the genus Acinetobacter need further definition. This may be understandable, considering the fact that many genospecies are very similar phenotypically and, therefore, are not readily distinguishable (10). The increasing pathogenic importance of Acinetobacter has stimulated the development of reliable identification methods for these strains. Although considered the gold standard, the DNA-DNA hybridization method is time consuming and labor intensive (25). Sequencing of housekeeping genes in general and protein-encoding genes in particular and sequencing of 16S DNA have been used in the identification of species or strains and delineation of phylogenetic relationships among different strains and species within the genus Acinetobacter. Compared to 16S rRNA genes, housekeeping genes provide a higher degree of resolution as the latter genes evolve faster than the former (32). It may not therefore be surprising that various protein-encoding genes, including recA, groEL, hsp75, rpoB, rpoD, and gyrB, have been used for the classification of several unrelated bacteria at the intrageneric level (13, 14, 15, 31, 32, 34). Unfortunately, the gyrB and recA gene sequences are not available for 10 recently described species of Acinetobacter. This makes it impossible to compare their performances with those of rpoB gene sequences in delineating various species of Acinetobacter.
Several investigators have sequenced gyrB and recA genes of 14 genomic species of Acinetobacter and studied them to delineate phylogenetic relationships among different members of the genus Acinetobacter (5, 15, 31, 33). Furthermore, these investigators compared the performance of the gene sequence-based identification scheme, as indicated above, with that based on DNA-DNA hybridization (4, 27). In our study, we constructed phylogenetic trees incorporating 14 genospecies of Acinetobacter and found no congruency among four individual phylogenetic trees constructed using the sequences of 16S rRNA, rpoB, gyrB, and recA genes (data not shown). However, a bootstrap value of 75% was a statistically significant (P < 0.05) occurrence when a phylogenetic tree was based on rpoB (11 of 12) gene sequences compared to that based on either 16S rRNA (4 of 12), gyrB (5 of 12), or recA (6 of 12) gene sequences. This shows that the robustness of the phylogenetic tree was based on the rpoB gene sequence. A. lwoffii and Acinetobacter genomic species 9 were shown to be 100% identical when their partial gyrB gene sequences had been analyzed and compared. However, sequencing of gyrD and recA genes (15, 33) separated them. In the present study, these two species were clearly differentiated as they showed variability between 4.9 and 6.7% in their complete rpoB gene sequence and two partial sequences of the rpoB gene. The sequence of the rpoB gene failed to clearly distinguish A. baylyi from genospecies 9 on one hand and A. grimontii from A. baylyi on the other. Unfortunately, the sequences of gyrB and recA genes are not available to use for such comparisons.
For the routine identification of Acinetobacter, either of two partial sequences and spacers can be used as they have high discriminative power and short lengths. However, such an identification scheme may not be sufficient to distinguish A. grimontii from A. junii and A. baylyi from Acinetobacter genomic species 9 (Table 3). One way of addressing this problem is to incorporate another short variable sequence into the identification scheme. In order to identify an Acinetobacter isolate, zone 1 should be determined first as it is the shortest and allows clear identification of 20 of 24 species. If the sequences of zone 1 in A. grimontii and A. junii are similar, then the sequences of zone 2 should be studied to make a clear distinction between these two species (Table 3). On the other hand, if the sequences of zone 1 in A. baylyi and genomic species 9 are similar, then it will be useful to determine the sequence of the rpoB-rpoC spacer as it may make a clear distinction between these two species (Table 3).
The intraspecies variability of the short discriminative fragments observed within the A. baumannii species showed that, with exception of CIP 103655, all isolates have clearly higher degrees of similarities than that observed between A. baumannii and its closest species. However, the similarities observed between A. baumannii CIP 103655 and other isolates of the genus suggest that the identification of some isolates in this species could be ambiguous. All in all, an identification scheme based on analyzing partial rpoB gene sequences and flanking spacers may prove useful in the molecular identification of Acinetobacter. However, due to possible intraspecies dissimilarities and interspecies high similarities, such a diagnostic scheme should be undertaken for a large collection of Acinetobacter isolates. Moreover, the status of some isolates (e.g., A. baumannii CIP 103655) or some species (e.g., A. grimontii and A. baylyi) should be assessed by DNA-DNA hybridization studies and analyses of the sequences of such housekeeping genes as recA and gyrB. The molecular identification scheme proposed herein will provide better genetic definition of large collections of strains representing named species and will then serve as the basis for a large-scale phenotypic study to determine better biochemical tests to separate groups, especially the group of Acinetobacter species isolated in humans.
Supplemental material for this article may be found at http://jcm.asm.org/.
REFERENCES
Alexander, M., F. Ismail, P. J. H. Jackman, and W. C. Nobel. 1984. Fingerprinting Acinetobacter strains from clinical sources by numerical analysis of electrophoretic patterns. J. Med. Microbiol. 18:55-68.
Barbe, V., D. Vallenet, N. Fonknechten, A. Kreimeyer, S. Oztas, L. Labarre, S. Cruveiller, C. Robert, S. Duprat, P. Wincker, L. N. Ornston, J. Weissenbach, P. Marliere, G. N. Cohen, and C. Medigue. 2004. Unique features revealed by the genome sequence of Acinetobacter sp. ADP1, a versatile and naturally transformation competent bacterium. Nucleic Acids Res. 32:5766-5779.
Bergogne-Berezin, E., and K. J. Towner. 1996. Acinetobacter spp. as nosocomial pathogens: microbiological, clinical, and epidemiological features. Clin. Microbiol. Rev. 9:148-165.
Bouvet, P. J. M., and P. A. D. Grimont. 1986. Taxonomy of the genus Acinetobacter with the recognition of Acinetobacter baumannii sp. nov., Acinetobacter haemolyticus sp. nov., Acinetobacter johnsonii sp. nov., and Acinetobacter junii sp. nov. and emended descriptions of Acinetobacter calcoaceticus and Acinetobacter lwoffii. Int. J. Syst. Bacteriol. 36:228-240.
Bouvet, P. J. M., and P. A. D. Grimont. 1987. Identification and biotyping of clinical isolates of Acinetobacter. Ann. Inst. Pasteur Microbiol. 138:569-578.
Brisou, J., and A. R. Prevot. 1954. [Studies on bacterial taxonomy. X. The revision of species under Achromobacter group.] Ann. Inst. Pasteur (Paris) 86:722-728.
Carr, E. L., P. Kampfer, B. K. Patel, V. Gurtler, and R. J. Seviour. 2003. Seven novel species of Acinetobacter isolated from activated sludge. Int. J. Syst. Evol. Microbiol. 53:953-963.
Felsenstein, J. 1989. PHYLIP—phylogeny inference package (version 3.2). Cladistics 5:164-166.
Garcia, A., R. Montoya, H. Bello, G. Gonzalez, M. Dominguez, and R. Zemelman. 1996. Differentiation of Acinetobacter baumannii biotypes by amplification of 16S-23S rRNA intergenic spacer sequences. Microbios 88:159-167.
Gerner-Smidt, P., I. Tjernberg, and J. Ursing. 1991. Reliability of phenotypic tests for identification of Acinetobacter species. J. Clin. Microbiol. 29:277-282.
Ibrahim, A., P. Gerner-Smidt, and W. Liesack. 1997. Phylogenetic relationship of the twenty-one DNA groups of the genus Acinetobacter as revealed by 16S ribosomal DNA sequence analysis. Int. J. Syst. Bacteriol. 47:837-841.
Ino, T., and Y. Nishimura. 1989. Taxonomic studies of Acinetobacter species based on outer membrane protein patterns. J. Gen. Appl. Microbiol. 16:213-224.
Karlin, S., G. M. Weinstock, and V. Brendel. 1995. Bacterial classifications derived from RecA protein sequence comparisons. J. Bacteriol. 177:6881-6893.
Khamis, A., D. Raoult, and B. La Scola. 2004. rpoB gene sequencing for identification of Corynebacterium species. J. Clin. Microbiol. 42:3925-3931.
Krawczyk, B., K. Lewandowski, and J. Kur. 2002. Comparative studies of the Acinetobacter genus and the species identification method based on the recA sequences. Mol. Cell. Probes 16:1-11.
La Scola, B., P.-E. Fournier, P. Brouqui, and D. Raoult. 2001. Detection and culture of Bartonella quintana, Serratia marcescens, and Acinetobacter spp. from decontaminated human body lice. J. Clin. Microbiol. 39:1707-1709.
La Scola, B., Z. Zeaiter, A. Khamis, and D. Raoult. 2003. Gene-sequence-based criteria for species definition in bacteriology: the Bartonella paradigm. Trends Microbiol. 11:318-321.
La Scola, B., and D. Raoult. 2004. Acinetobacter baumannii in human body louse. Emerg. Infect. Dis. 10:1671-1673.
Nemec, A., L. Dijkshoorn, I. Cleenwerck, T. De Baere, D. Janssens, T. J. Van Der Reijden, P. Jezek, and M. Vaneechoutte. 2003. Acinetobacter parvus sp. nov., a small-colony-forming species isolated from human clinical specimens. Int. J. Syst. Evol. Microbiol. 53:1563-1567.
Nemec, A., T. De Baere, I. Tjernberg, M. Vaneechoutte, T. J. van der Reijden, and L. Dijkshoorn. 2001. Acinetobacter ursingii sp. nov. and Acinetobacter schindleri sp. nov., isolated from human clinical specimens. Int. J. Syst. Evol. Microbiol. 51:1891-1899.
Nishimura, Y., T. Ino, and H. Lizuka. 1988. Acinetobacter radioresistens sp. nov. isolated from cotton and soil. Int. J. Syst. Bacteriol. 38:209-211.
Nowak, A., and J. Kur. 1996. Differentiation of seventeen genospecies of Acinetobacter by multiplex polymerase chain reaction and restriction fragment length polymorphism analysis. Mol. Cell. Probes 10:405-411.
Rainey, F. A., E. Lang, and E. Stackebrandt. 1994. The phylogenetic structure of the genus Acinetobacter. FEMS Microbiol. Lett. 124:349-353.
Schreckenberger, P. C., M. I. Daneshvar, R. S. Weyant, and D. G. Hollis. 2003. Acinetobacter, Achromobacter, Chryseobacterium, Moraxella, and other nonfermentative gram-negative rods, p. 749-779. In P. R. Murray, E. J. Baron, J. H. Jorgensen, M. A. Pfaller, and R. H. Yolken (ed.), Manual of clinical microbiology, 8th ed. American Society for Microbiology, Washington, D.C.
Stackebrandt, E., and B. M. Goebel. 1994. Taxonomic note: a place for DNA-DNA reassociation and 16S rRNA sequence analysis in the present species definition in bacteriology. Int. J. Syst. Bacteriol. 44:846-849.
Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673-4680.
Tjernberg, I., and J. Ursing. 1989. Clinical strains of Acinetobacter classified by DNA-DNA hybridization. APMIS 97:595-605.
Van Dessel, H., L. Dijkshoorn, T. Van Der Reijden, N. Bakker, A. Paauw, P. Van Den Broek, J. Verhoef, and S. Brisse. 2004. Identification of a new geographically widespread multiresistant Acinetobacter baumannii clone from European hospitals. Res. Microbiol. 155:105-112.
Webster, C. A., M. Crow, H. Humphreys, and K. S. Towner. 1998. Surveillance of an adult intensive care unit for a long-term persistence of a multi-resistant strain of Acinetobacter baumannii. Eur. J. Clin. Microbiol. Infect. Dis. 17:171-176.
Wendt, C., B. Dietz, E. Dietz, and H. Ruden. 1997. Survival of Acinetobacter baumannii on dry surfaces. J. Clin. Microbiol. 35:1394-1397.
Yamamoto, S., and S. Harayama. 1996. Phylogenetic analysis of Acinetobacter strains based on the nucleotide sequences of gyrB genes and on the amino acid sequences of their products. Int. J. Syst. Bacteriol. 46:506-511.
Yamamoto, S., and S. Harayama. 1998. Phylogenetic relationships of Pseudomonas putida strains deduced from the nucleotide sequences of gyrB, rpoD and 16S rRNA genes. Int. J. Syst. Bacteriol. 48:813-819.
Yamamoto, S., P. J. Bouvet, and S. Harayama. 1999. Phylogenetic structures of the genus Acinetobacter based on gyrB sequences: comparison with the grouping by DNA-DNA hybridization. Int. J. Syst. Bacteriol. 49:87-95.
Zeaiter, Z., P. E. Fournier, H. Ogata, and D. Raoult. 2002. Phylogenetic classification of Bartonella species by comparing groEL sequences. Int. J. Syst. Evol. Microbiol. 52:165-171.(Bernard La Scola, Vijay A)