| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ena M. Korczak1
1 Institute of Veterinary Bacteriology, University of Bern, CH-3001 Bern, Switzerland
2 Department of Veterinary Pathobiology, The Royal Veterinary and Agricultural University, DK-1870 Frederiksberg C, Denmark
Correspondence
Peter Kuhnert
peter.kuhnert{at}vbi.unibe.ch
 |
ABSTRACT |
|---|
 TOP ABSTRACT MAIN TEXTREFERENCES |
|---|
and Supplementary Figs S1–S3. Phylogenetic trees based on infB, rpoB and recN genes and a table showing the phenotypic characters of the strains investigated are available as supplementary material in IJSEM Online.
|
MAIN TEXT |
|---|
|
TOPABSTRACTMAIN TEXTREFERENCES |
|---|
). To the knowledge of the authors very few cases have been reported since then. Isolates of A. capsulatus from the uterus of rabbits, previously misclassified as Actinobacillus equuli, were reported by Blackall et al. (1997), while organisms closely related to A. capsulatus have been reported from hamsters (Krause et al., 1989), free-ranging snowshoe-hares (Lepus americanus) (Zarnke & Schlater, 1988) and European brown hares (Bisgaard, 1993). Additional isolates were included in the studies of Bercovier et al. (1984) and Escande et al. (1984).
The genus Actinobacillus Brumpt 1910 consists of 18 species (Olsen & Moller, 2005). The work of Dewhirst et al. (1992) on the 16S rRNA gene (rrs) phylogeny showed that a core group clusters around the type species Actinobacillus lignieresii, whereas other actinobacilli, including A. capsulatus, cluster outside this group. This core group is often referred to as Actinobacillus sensu stricto or ‘true actinobacilli’ (Olsen & Moller, 2005), even though a defined emended description of the genus is lacking. Here the core group of Actinobacillus is referred to as including the type species A. lignieresii and the species Actinobacillus arthritidis, A. equuli, Actinobacillus hominis, Actinobacillus pleuropneumoniae, Actinobacillus suis and Actinobacillus ureae. Based on the work of Dewhirst et al. (1992), an attempt to exclude A. capsulatus from the genus Actinobacillus began, even though DNA–DNA reassociation studies indicated that it is correctly named (Escande et al., 1984; Mutters et al., 1989). Phylogenetic analysis of members of Actinobacillus based upon partial sequences of the infB gene (encoding translation initiation factor 2) left the position of A. capsulatus unresolved (Nørskov-Lauritsen et al., 2004). However, phylogenetic studies of the family Pasteurellaceae based on rpoB (encoding the 
-subunit of RNA polymerase) sequences grouped A. capsulatus within core Actinobacillus species (Korczak et al., 2004). Clustering of the type strain of A. capsulatus with core Actinobacillus species was also seen by using multilocus sequence analysis (Kuhnert & Korczak, 2006).
Comparison of rrs gene sequences of strains deposited as A. capsulatus indicates that the taxon is heterogeneous, possibly reflecting the inclusion of strains that were not isolated from rabbits, strains that had been misclassified or sequences that had been wrongly annotated, such as the horse isolate CCUG 19799T that was originally deposited as A. capsulatus but was classified recently as the type strain of A. equuli subsp. haemolyticus (Christensen et al., 2002). To investigate the reasons behind the diversity observed and to clarify the classification of A. capsulatus, a strain collection representing isolates from various animal species and several continents was reinvestigated using a polyphasic approach. A total of 23 strains that had been characterized previously as, or closely related to, A. capsulatus (Table 1) were analysed phenotypically and genotypically.
|
with reference to ‘Cowan and Steel's Manual’ (Barrow & Feltham, 1993).
Sequencing of rrs, rpoB and infB genes, as established phylogenetic markers for the family Pasteurellaceae, was used to investigate the relationship between strains based on previously described protocols by which about 1360, 520 and 1340 bp, respectively, of the rrs, rpoB and infB genes were determined (Kuhnert et al., 2002, 2004; Korczak et al., 2004). In addition, recN gene sequences were used as a representative target for determining whole-genome sequence similarity as described by Kuhnert & Korczak (2006), with 1340 bp of the gene being sequenced. Whole-genome similarity values were calculated based on the formula: SIgenome=–1.30+2.25(SIrecN) of Zeigler (2003), where SI is sequence identity. The recN sequences were also included in the phylogenetic analysis.
The results from the analysis of the 80 phenotypic characters are shown in Supplementary Table S1 available in IJSEM Online. Characters that differentiate the strains investigated are given in Table 2. Phenotypic analysis allowed separation of the 23 strains into five groups. The main group consisted of 17 strains that exhibited the same phenotypic features as the A. capsulatus type strain (CCUG 12396T) and were therefore classified as A. capsulatus. Two hamster isolates (CCUG 23125 and CCUG 26442) were closely related to A. capsulatus, but differed from it in CAMP, (+)-L-arabinose, (–)-D-sorbitol, 
-glucosidase and growth on MacConkey agar. One rabbit (R80) and the snowshoe-hare (J3-241) isolates were related to A. suis. Strain R80 represents a non-haemolytic and (+)-L-arabinose-negative isolate, whereas isolate J3-241 represents a (–)-D-mannitol-positive and -glucosidase-negative isolate of A. suis (Christensen & Bisgaard, 2004). The two isolates P224 and R46 were both negative for urease and therefore differed in a key character from the core group of Actinobacillus (Christensen & Bisgaard, 2004). Whereas isolate P224 was related to Mannheimia granulomatis, only differing in (+)-L-arabinose and -glucosidase, strain R46 could not be identified further phenotypically.
|
. Growth on MacConkey agar appears to be isolate-dependent and most strains will not grow on this medium. Also, oxidase and catalase reactions are isolate-dependent, with most strains being positive for both. Comparing characters used for identifying taxa of the core group of Actinobacillus described by Christensen & Bisgaard (2004) the following results were obtained. Catalase, oxidase, glycerol, (+)-L-arabinose and -glucosidase (PNPG) are isolate-dependent with most strains being positive. -Haemolysis, (–)-D-arabinose, (–)-L-fucose, inulin and -xylosidase (ONPX) are negative. Acid is produced from (+)-D-xylose, (+)-D-mannose, cellobiose, -glucosidase (NPG), lactose, -galactosidase (ONPG), (+)-D-melibiose, trehalose, raffinose and -galactosidase. These characters also allow separation of A. capsulatus from the other species within the genus.
A phylogenetic analysis was carried out with all strains. Trees were built in BioNumerics v. 4.5 (Applied Maths) using the Jukes–Cantor correction for matrix calculation and neighbour-joining for generating the tree. Previous investigations have shown that the rrs gene often represents a phylogeny that is different from that derived from protein-encoding housekeeping genes, especially in the case of A. capsulatus (Korczak et al., 2004, 2006; Kuhnert & Korczak, 2006). We therefore built two separate trees. Fig. 1 shows the classical rrs-derived tree and Fig. 2 a consensus tree derived from the matrices of individual trees of the infB, rpoB and recN gene sequences. The three individual trees are available as Supplementary Figs S1–S3, respectively, in IJSEM Online. Calculated using the Pearson correlation, the three trees showed a high degree of congruence between each other (>92 %), whereas the congruence of each of the three trees with the rrs tree was only around 70 %. In the rpoB-, infB- and recN-based trees, the phenotypic group of 17 A. capsulatus strains clustered within the core of Actinobacillus species, as was also seen in the consensus tree (Fig. 2). This cluster of 17 strains was very homogeneous, except with respect to strains CCUG 37798, 266/2693 and P796 in the rpoB gene-derived tree where these strains formed a cluster outside the main A. capsulatus branch. The same cluster of 17 A. capsulatus strains was found in the rrs gene-based tree; however, it was clearly outside the core Actinobacillus species cluster (Fig. 1). Strains CCUG 37798, 266/2693 and P796 also clustered separately in the rrs gene-derived tree. A remarkable observation was the sequence heterogeneity within the A. capsulatus cluster. The variability of the rrs sequences of A. capsulatus was as high as 3.3 %, which is normally considered as a sequence difference that allows separation of species (Stackebrandt et al., 2002). In contrast, the rpoB, infB and recN sequence heterogeneity within the cluster of 17 A. capsulatus strains was as low as 2.7 (including the three divergent isolates), 0.4 and 0.5 %, respectively. This is remarkable since the variability of rpoB and infB as well as recN is normally much higher than that of the rrs gene, which is more conserved. The reason for this anomaly in this case is not clear but it might indicate that the rrs gene in A. capsulatus is prone to relatively high mutation rates and/or horizontal gene transfer that is known to occur with rrs genes (Yap et al., 1999; Acinas et al., 2004). This might also explain why A. capsulatus clusters outside the core group of Actinobacillus species in the rrs gene-based phylogeny, whereas in all the other trees it evidently belongs to this cluster. Therefore, this example shows that the analysis of rrs gene sequences, as helpful as they are, in some cases might lead to false classification if taken as the gold standard without additional genetic and phenotypic characterization in a polyphasic approach.
|
|
The two isolates that resembled A. suis phenotypically (J3-241 and R80) clustered with A. suis, as seen in the consensus tree (Fig. 2). This clustering was seen repeatedly within the individual rpoB, infB and recN trees, but not within the rrs tree, in which these isolates are closer to A. hominis, and the type strain of A. suis clusters between the two A. equuli subspecies (Fig. 1). However, the resolution of the rrs gene is known not to allow discrimination between these closely related Actinobacillus species (Christensen & Bisgaard, 2004).
The phylogenetic analysis of strain P224, which was phenotypically related to M. granulomatis, placed this strain within that genus in all trees, but without clear species association.
Strain R46 appeared to be phylogenetically unrelated to the other isolates analysed as well as to any other member of the family Pasteurellaceae and cannot be clearly assigned to any specific cluster, either in the consensus tree or in any of the individual trees.
Analysis of representative housekeeping genes was shown recently to be useful for the prediction of whole-genome relatedness within the family Pasteurellaceae (Kuhnert & Korczak, 2006). The sequence of the recN gene was shown to be sufficient to obtain an approximation of genetic relatedness between species and genera. We therefore calculated the genome similarity values based on the recN gene for all strains analysed and cross-compared them with all available species of the family. Table 3 shows some of the values obtained from this comparison that are relevant for our study. All 17 strains that clustered phenotypically and phylogenetically as A. capsulatus had similarity values between each other of close to, or identical with, the maximal possible value of 0.95. The type strain of A. capsulatus has been shown previously to share high recN-based similarity values, mostly above 0.7, with core members of Actinobacillus (Kuhnert & Korczak, 2006). This was also observed for each of the other 16 A. capsulatus strains analysed (data not shown) and is further reflected by the uniformly high values all the A. capsulatus strains showed with A. suis (Table 3). The hamster isolates had values of 0.7 with the A. capsulatus strains, clearly outside the species border of about 0.85. However, they showed high similarity values with all species of the core group of Actinobacillus, ranging from 0.64 with A. lignieresii to 0.75 with A. equuli, definitely within the range of the genus, i.e. above 0.4 (data not shown). The two strains that were closely related to A. suis also showed very high similarity values with the type strain of A. suis, confirming the results of the phenotypic and phylogenetic analysis. Similarly, results obtained with strain P224 confirmed classification with Mannheimia as the highest values were obtained with species from the genus Mannheimia, clearly above the 0.4 threshold level (Table 3). The highest similarity value was obtained with Mannheimia varigena but was still clearly below the species level. Finally, strain R46, which could not be classified phenotypically or phylogenetically, showed the highest similarity value with [Haemophilus] parasuis (0.37), but without clear association with any established genus.
|
Emended description of Actinobacillus capsulatus Arseculeratne 1962, 38AL
Actinobacillus capsulatus (cap.su.la'tus. L. n. capsula a small chest, capsule; M.L. masc. adj. capsulatus encapsulated).
Cells are non-motile, Gram-negative, small coccoid or pleomorphic rods. Colonies on bovine blood agar are regular, circular, slightly raised, greyish and 1.0–1.5 mm in diameter after 24 h incubation at 37 °C. Haemolysis is normally not observed on calf blood agar but may be observed, with the type strain being non-haemolytic. Weakly CAMP positive using bovine blood. Catalase and oxidase reactions may vary but, in most cases, are positive, including the type strain. Fermentative with (+)-D-glucose in Hugh & Leifson's medium. Symbiotic growth (NAD requirement) is not observed. Porphyrin test is positive. No growth is observed on Simmons' citrate agar and acid is not produced from mucate. An alkaline reaction is not observed in malonate broth. H2S is not produced in TSI and growth is not observed in KCN. Methyl red and Voges–Proskauer tests at 37 °C are negative. Nitrate is reduced without gas formation. Urease and alanine aminopeptidase tests are positive, whereas tests for arginine dihydrolase, lysine- and ornithine decarboxylase, phenylalanine deaminase, indole, gelatinase, Tweens 20 and 80 and pigment formation are negative. Phosphatase is produced and growth may occur on MacConkey agar, although rarely, with the type strain not growing on this medium. Acid is not produced from meso-erythritol, adonitol, (+)-D-arabitol, xylitol, (–)-D-arabinose, (–)-L-xylose, dulcitol, myo-inositol, (+)-D-fucose, (–)-L-fucose, (+)-L-rhamnose, (–)-L-sorbose, (+)-D-melezitose, (+)-D-glycogen, inulin, (+)-D-turanose and -N-CH3-glucosamid. Acid is produced from (–)-D-ribose, (+)-D-xylose, (–)-D-mannitol, (–)-D-sorbitol, (–)-D-fructose, (+)-D-glucose (without gas), (+)-D-galactose, (+)-D-mannose, cellobiose, lactose, maltose, (+)-D-melibiose, sucrose, trehalose, raffinose, dextrin, arbutin and salicin. Most isolates, including the type strain, produce acid from glycerol, (+)-L-arabinose, aesculin, amygdalin and gentiobiose. -Glucosidase (NPG), -galactosidase (ONPG) and -galactosidase are produced whereas -fucosidase (ONPF), -mannosidase, -glucuronidase (PGUA) and -xylosidase (ONPX) are negative. Most isolates, including the type strain, produce -glucosidase (PNPG). The major polyamine compound is 1,3-diaminopropane (DAP), common to the majority of members of Pasteurellaceae with no characteristic minor compound (Busse et al., 1997). recN-based genome similarity with the type species A. lignieresii is 0.71, whereas using the genes recN, rpoA and thdF this value is 0.74 (Kuhnert & Korczak, 2006).
Isolated from various pathological lesions in rabbits, septicaemia in a kitten and conjunctiva of a hare. The type strain is CCUG 12396T (=Frederiksen 243T=ATCC 51571T=NCTC 11408T=CIP 103283T), which was isolated from a joint of a rabbit in Sri Lanka. The DNA G+C content of the type strain is 42.4 mol% (Kuhnert & Korczak, 2006) and the genome size is 1.9 GDa (Krause et al., 1989).
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTMAIN TEXTREFERENCES |
|---|
Arseculeratne, S. N. (1962). Actinobacillosis in joints of rabbits. J Comp Pathol 72, 33–39.[Medline]
Barrow, G. I. & Feltham, R. K. A. (editors) (1993). Cowan and Steel's Manual for the Identification of Medical Bacteria, 3rd edn. Cambridge: Cambridge University Press.
Bercovier, H., Escande, F. & Grimont, P. A. (1984). Biological characterization of Actinobacillus species and Pasteurella ureae. Ann Microbiol 135A, 203–218.
Bisgaard, M. (1993). Ecology and significance of Pasteurellaceae in animals. Zentralbl Bakteriol 279, 7–26.[Medline]
Bisgaard, M., Houghton, S. B., Mutters, R. & Stenzel, A. (1991). Reclassification of German, British and Dutch isolates of so-called Pasteurella multocida obtained from pneumonic calf lungs. Vet Microbiol 26, 115–124.[CrossRef][Medline]
Blackall, P. J., Bisgaard, M. & McKenzie, R. A. (1997). Characterisation of Australian isolates of Actinobacillus capsulatus, Actinobacillus equuli, Pasteurella caballi and Bisgaard Taxa 9 and 11. Aust Vet J 75, 52–55.[Medline]
Busse, H.-J., Bunka, S., Hensel, A. & Lubitz, W. (1997). Discrimination of members of the family Pasteurellaceae based on polyamine patterns. Int J Syst Bacteriol 47, 698–708.
Christensen, H. & Bisgaard, M. (2004). Revised definition of Actinobacillus sensu stricto isolated from animals. A review with special emphasis on diagnosis. Vet Microbiol 99, 13–30.[CrossRef][Medline]
Christensen, H., Bisgaard, M. & Olsen, J. E. (2002). Reclassification of equine isolates previously reported as Actinobacillus equuli, variants of A. equuli, Actinobacillus suis or Bisgaard taxon 11 and proposal of A. equuli subsp. equuli subsp. nov. and A. equuli subsp. haemolyticus subsp. nov. Int J Syst Evol Microbiol 52, 1569–1576.[Abstract]
Christie, R., Atkins, N. E. & Munch-Petersen, E. (1944). A note on a lytic phenomenon shown by group B streptococci. Aust J Exp Biol 22, 197–200.
Dewhirst, F. E., Paster, B. J., Olsen, I. & Fraser, G. J. (1992). Phylogeny of 54 representative strains of species in the family Pasteurellaceae as determined by comparison of 16S rRNA sequences. J Bacteriol 174, 2002–2013.
Escande, F., Grimont, F., Grimont, P. A. D. & Bercovier, H. (1984). Deoxyribonucleic acid relatedness among strains of Actinobacillus spp. and Pasteurella ureae. Int J Syst Bacteriol 34, 309–315.
Korczak, B., Christensen, H., Emler, S., Frey, J. & Kuhnert, P. (2004). Phylogeny of the family Pasteurellaceae based on rpoB sequences. Int J Syst Evol Microbiol 54, 1393–1399.
Korczak, B. M., Stieber, R., Emler, S., Burnens, A. P., Frey, J. & Kuhnert, P. (2006). Genetic relatedness within the genus Campylobacter inferred from rpoB sequences. Int J Syst Evol Microbiol 56, 937–945.
Krause, T., Kunstyr, I. & Mutters, R. (1989). Characterization of some previously unclassified Pasteurellaceae isolated from hamsters. J Appl Bacteriol 67, 171–175.[Medline]
Kuhnert, P. & Korczak, B. M. (2006). Prediction of whole-genome DNA–DNA similarity, determination of G+C content and phylogenetic analysis within the family Pasteurellaceae by multilocus sequence analysis (MLSA). Microbiology 152, 2537–2548.
Kuhnert, P., Frey, J., Lang, N. P. & Mayfield, L. (2002). Phylogenetic analysis of Prevotella nigrescens, Prevotella intermedia and Porphyromonas gingivalis clinical strains reveals a clear species clustering. Int J Syst Evol Microbiol 52, 1391–1395.[Abstract]
Kuhnert, P., Korczak, B., Falsen, E., Straub, R., Hoops, A., Boerlin, P., Frey, J. & Mutters, R. (2004). Nicoletella semolina gen. nov., sp. nov., a new member of Pasteurellaceae isolated from horses with airway disease. J Clin Microbiol 42, 5542–5548.
Mutters, R., Mannheim, W. & Bisgaard, M. (1989). Taxonomy of the group. In Pasteurella and Pasteurellosis, pp. 3–34. Edited by C. Adlam & J. M. Rutter. London: Academic Press.
Nørskov-Lauritsen, N., Christensen, H., Okkels, H., Kilian, M. & Bruun, B. (2004). Delineation of the genus Actinobacillus by comparison of partial infB sequences. Int J Syst Evol Microbiol 54, 635–644.
Olsen, I. & Moller, K. (2005). Genus II Actinobacillus Brumpt 1910, 849AL. In Bergey's Manual of Systematic Bacteriology, 2nd edn, vol. 2, part B, pp. 866–883. Edited by D. J. Brenner, N. R. Krieg, J. T. Staley & G. M. Garrity. New York: Springer.
Stackebrandt, E., Frederiksen, W., Garrity, G. M., Grimont, P. A. D., Kämpfer, P., Maiden, M. C. J., Nesme, X., Rosselló-Mora, R., Swings, J. & other authors (2002). Report of the ad hoc committee for the re-evaluation of the species definition in bacteriology. Int J Syst Evol Microbiol 52, 1043–1047.[Abstract]
Yap, W. H., Zhang, Z. & Wang, Y. (1999). Distinct types of rRNA operons exist in the genome of the actinomycete Thermomonospora chromogena and evidence for horizontal transfer of an entire rRNA operon. J Bacteriol 181, 5201–5209.
Zarnke, R. L. & Schlater, L. (1988). Actinobacillosis in free-ranging snowshoe hares (Lepus americanus) from Alaska. J Wildl Dis 24, 176–177.[Abstract]
Zeigler, D. R. (2003). Gene sequences useful for predicting relatedness of whole genomes in bacteria. Int J Syst Evol Microbiol 53, 1893–1900.
This article has been cited by other articles:
|
|
 |
|
  M. Bisgaard, B. M. Korczak, H.-J. Busse, P. Kuhnert, A. M. Bojesen, and H. Christensen Classification of the taxon 2 and taxon 3 complex of Bisgaard within Gallibacterium and description of Gallibacterium melopsittaci sp. nov., Gallibacterium trehalosifermentans sp. nov. and Gallibacterium salpingitidis sp. nov. Int J Syst Evol Microbiol, April 1, 2009; 59(4): 735 - 744. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Bisgaard, J. P. Christensen, A. M. Bojesen, and H. Christensen Avibacterium endocarditidis sp. nov., isolated from valvular endocarditis in chickens Int J Syst Evol Microbiol, August 1, 2007; 57(8): 1729 - 1734. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
| J MED MICROBIOL | ALL SGM JOURNALS | |
