| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Clinical Microbiology, Aalborg Hospital, Aalborg, Denmark
2 Department of Medical Microbiology and Immunology, University of Aarhus, Aarhus, Denmark
3 Department of Veterinary Microbiology, The Royal Veterinary and Agricultural University, Copenhagen, Denmark
4 Department of Clinical Biochemistry, Aalborg Hospital, Aalborg, Denmark
5 Department of Clinical Microbiology, State Serum Institute, Copenhagen, Denmark
Correspondence
Niels Nørskov-Lauritsen
norskov{at}microbiology.au.dk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
A matrix table of infB sequence similarity values is available as supplementary material in IJSEM Online.
Present address: Department of Clinical Microbiology, Hillerød Hospital, Hillerød, Denmark. 
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). In the most recent edition, the number of species was 18, of which only a few are considered to belong to Actinobacillus sensu stricto (Olsen & Møller, 2004).
DNA hybridization results from the genus Actinobacillus have been presented by several groups. Based on the spectrophotometric method, Mutters et al. (1989) suggested that the genus Actinobacillus should be restricted to A. lignieresii, Actinobacillus pleuropneumoniae, A. equuli, A. suis, A. capsulatus, Actinobacillus ureae, Actinobacillus hominis and Bisgaard taxa 5, 9 and 11. This delineation was presented in the form of a review, and crucial information, including the number of strains and the individual DNA binding values, was not given. In a study using the S1 nuclease method on multiple strains from several species, A. equuli, A. suis, A. ureae and A. capsulatus were sufficiently related to A. lignieresii to justify their inclusion in the genus (Escande et al., 1984). A single isolate received as A. equuli was unexpectedly found to differ from the other representatives of this species, but was located within the boundaries of the genus by DNA hybridization. This strain was later found to comply with Bisgaard taxon 9 and has been made the type strain of a novel species, Actinobacillus arthritidis (Christensen et al., 2002a). Bisgaard taxon 11 is phenotypically a haemolytic A. equuli, closely related to the type strain of A. equuli by DNA hybridization, and the taxon has been reclassified as A. equuli subsp. haemolyticus (Christensen et al., 2002b).
Increasing emphasis has been directed to the division of bacteria into groups based on comparison of 16S rRNA sequences – so-called phylogenetic classification (Krieg & Garrity, 2001). Based on a large number of near-full-length 16S rRNA sequences from different members of the family Pasteurellaceae, single representatives of the following species or taxa constitute a monophyletic group: A. lignieresii, A. pleuropneumoniae, A. equuli, A. suis, A. ureae, A. hominis, A. arthritidis, Bisgaard taxon 8 and strain CCUG 28015 (a duck isolate). The other Actinobacillus species are pending placement in other genera of the family Pasteurellaceae (Olsen et al., 2004).
Translation initiation factor 2 is an essential protein involved in the initiation of protein synthesis in prokaryotes. So far, the encoding gene, infB, has only been found as a single copy in the prokaryotic genome; it encompasses 2490 bp in Haemophilus influenzae (Fleischmann et al., 1995) and 2502 bp in Pasteurella multocida (May et al., 2001). A partial sequence of infB has been used successfully in a study of the genus Haemophilus (Hedegaard et al., 2001). In the present study, the investigation was extended to the genus Actinobacillus. The principal scope was species that have qualified as true members of the genus by DNA hybridization or 16S rRNA sequencing. The following were included: 11 representatives of the type species of the genus, A. lignieresii, four of which were equine isolates that appear to constitute a new genomospecies (genomospecies 1) of Actinobacillus that is distinct from A. lignieresii (Christensen et al., 2002a); representatives of species belonging to the genus by both DNA hybridization and rRNA sequence data (A. pleuropneumoniae, A. equuli, A. suis, A. ureae, A. hominis and A. arthritidis); representatives of the sorbitol-negative subgroup of Bisgaard taxon 9 (genomospecies 2), which belong to the genus according to molecular data (Christensen et al., 2002a); representatives of A. capsulatus and Bisgaard taxon 5, which are included in the genus by DNA hybridization data, but excluded on the basis of rRNA sequencing; and a few single isolates that are candidate species of the genus by 16S rRNA sequence comparison (CCUG 28015 and Bisgaard taxon 8) or can be used as outgroups (Actinobacillus muris and two biotypes of Pasteurella pneumotropica).
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
. DNA was released using the chelex method as described previously (Hedegaard et al., 2001). PCR products were purified on silica gel columns (QIAquick PCR purification kit; Qiagen).
|
). A total of 30 cycles was run, each consisting of a denaturation step at 94 °C for 30 s, an annealing step at 51 °C for 60 s and an elongation step at 72 °C for 90 s. Purified products were cycle-sequenced using the non-degenerate internal primer 1748R (5'-GTAGCAACCGGACCACGACCTTTAT-3') and the BigDye terminator DNA sequencing kit (Applied Biosystems); analysis was on an ABI 377 DNA sequencer. Primers are named according to nucleotide positions in the infB gene of Escherichia coli (Hedegaard et al., 1999). The corresponding positions in the P. multocida gene are nt 1018, 1580 and 1665 (May et al., 2001). With A. arthritidis strain CCUG 24862T and P. pneumotropica strain P309, a probable 3' mismatch prevented use of the internal primer, and sequencing was performed with the primers used for PCR amplification. A 426 bp fragment was obtained corresponding to amino acid positions 368–509 [with reference to P. multocida translation initiation factor 2 (May et al., 2001)]; sequences of all bacterial strains were compared. Sequences were checked by visual examination of the electrophoretograms, by translation into amino acid sequences and finally by re-examination of polymorphic positions within individual species or taxa. Ambiguities were resolved by repeating the PCR and sequencing reactions and by sequencing in both directions with the primers used for PCR amplification.
16S rRNA gene sequencing.
Fragments of the 16S rRNA genes were amplified by PCR using primers 5'-TATTACCGCGGCTGCTGGCA-3' and 5'-TCAGATTGAACGCTGGCGGC-3' derived from the rrnB operon of E. coli (Brosius et al., 1978). A total of 25 cycles was run, each consisting of a denaturation step at 94 °C for 60 s, an annealing step at 60 °C for 60 s and an elongation step at 72 °C for 2 min. Purified products were cycle-sequenced from both ends using the PCR primers and a ThermoSequenase dye terminator sequencing kit (Pharmacia Biotech); analysis was on an ABI 377 DNA sequencer. Fragments of 429–437 bp (corresponding to nt 62–496 of E. coli 16S rRNA; GenBank/EMBL accession no. J01695) were obtained from all bacterial strains and used for analysis.
An unresolved nucleotide in one sequencing direction was ignored if the complementary position was determined unambiguously. If the position was determined ambiguously in both sequencing directions, the sequencing reactions were repeated. If the ambiguity persisted after the second round of reactions, no further resolution attempt was carried out.
Sequence analysis.
Sequences were aligned using the program CLUSTAL W with a Windows interface (Thompson et al., 1997). Distances were modified by the Kimura two-parameter method (Kimura, 1980). Neighbour-joining, maximum-parsimony and maximum-likelihood analysis were performed by PHYLIP (Felsenstein, 1995). Bootstrap analyses were done by 100 resamplings of the datasets. The calculation of similarity between clusters (see matrix table available as supplementary material in IJSEM Online) or sequences (Table 2) was performed with even-weighted characters. When 16S rRNA was compared, gaps and ambiguous positions were ignored.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
; 163 polymorphic positions were available for phylogenetic analysis. Compared with the type strain of the type species of the genus (A. lignieresii P151T), 2613 nucleotide substitutions were present in the 58 other strains; when the protein sequences were compared, a total of 214 amino acid substitutions were detected.
A dendrogram constructed by the neighbour-joining method is shown in Fig. 1, with 52 strains clustering in nine sequence groups. Phylogenetic trees were also generated by the maximum-parsimony and maximum-likelihood methods (dendrograms not shown). These analyses generally supported the major topology of Fig. 1; some contradictory findings are mentioned in the text.
|
, were placed on a common branch by all three phylogenetic methods. The branch was supported by a moderate bootstrap value of 69 (of 100 resamplings) (Fig. 1) and relatively large differences were observed between individual clusters in the branch. The branch contained A. lignieresii, A. pleuropneumoniae and two unnamed taxa.
Fourteen strains of A. lignieresii and A. pleuropneumoniae, including the two type strains, were found in a common cluster with a sequence similarity in excess of 96 % (cluster 1). The cluster was further divided into four subgroups, with some distinction between A. lignieresii and A. pleuropneumoniae: one subgroup included only strains of A. lignieresii and two subgroups contained only A. pleuropneumoniae. In a fourth subgroup, a strain of A. pleuropneumoniae (HK356) carried an infB sequence that was almost identical to that of the type strain of A. lignieresii (Fig. 1).
Four equine isolates phenotypically identified as A. lignieresii constituted cluster 2. These were clearly distinct from the type strain and the other bovine isolates of A. lignieresii located in cluster 1. However, these isolates were related to strains in cluster 1 (similarity of 89–92 %). Equine isolates of A. lignieresii have recently been described as a unique genomospecies (genomospecies 1) based on DNA hybridization and 16S rRNA gene sequencing (Christensen et al., 2002a). Two other equine isolates, F141 and H4272/4, constituted a separate cluster on the A. lignieresii branch (cluster 3). These isolates comply phenotypically with the sorbitol-negative subgroup of Bisgaard taxon 9 (genomospecies 2) (Christensen et al., 2002a).
A. equuli branch
Six clusters (4–9) representing A. equuli, A. suis, A. ureae, A. arthritidis, A. hominis and A. capsulatus were placed on a common branch by the neighbour-joining method (Fig. 1). The branch was supported by a bootstrap value of 64. The A. capsulatus cluster (cluster 9) branched rather deeply and its location on the A. equuli branch was not confirmed by maximum-parsimony or maximum-likelihood analyses (dendrograms not shown).
Twelve bacterial strains representing A. equuli and its haemolytic variants constituted cluster 4. The latter phenotype has been designated Bisgaard taxon 11 or A. equuli subsp. haemolyticus (Christensen et al., 2002b). Although some division of the cluster into subgroups was apparent, subgrouping according to infB genotype did not correlate with haemolytic activity. A bootstrap value of only 20 was found for cluster 4, indicating difficulties in the distinction between A. equuli and A. suis (cluster 5). The three representatives of A. suis carried almost identical infB sequences, with only a single polymorphic position found in the studied fragment. A similar situation was found in the three representatives of A. ureae (cluster 6); however, these strains were more clearly separated from sequences in the A. equuli cluster. A similarity of 95–97 % was found between clusters 4 and 6, whereas the differences between clusters 4 and 5 were of the same magnitude as the differences found within cluster 4 (see supplementary material in IJSEM Online).
Two representatives of the newly described species A. arthritidis were positioned on the A. equuli branch by all three phylogenetic methods (cluster 7). This species was formed from the sorbitol-positive subgroup of Bisgaard taxon 9 (Christensen et al., 2002a).
Identical infB sequences were observed in seven strains of A. hominis (cluster 8) and in five strains of A. capsulatus (cluster 9). The relationship of A. hominis to clusters 4–7 was supported by a bootstrap value of 77 and the topology was also seen in dendrograms constructed by maximum-parsimony and maximum-likelihood analyses (not shown). The location of the A. capsulatus cluster was more questionable, as the two other phylogenetic methods placed this cluster on another branch together with the group consisting of P. pneumotropica (P421T), P309, CCUG 28015 and P1572; indeed, by maximum-likelihood analyses, the A. capsulatus cluster was even less related to the ‘true’ actinobacilli (clusters 1–8) than was the P. pneumotropica group. The simple index of similarity also reveals the uncertain position of the A. capsulatus cluster, as this cluster (cluster 9) differed more from the A. lignieresii cluster (cluster 1) than from the other groups of actinobacilli (see supplementary material in IJSEM Online).
Strains more remotely related to the genus Actinobacillus
Compared to the other strains studied, the type strain of P. pneumotropica and an additional strain (biotype Heyl, P309), a duck isolate (CCUG 28015), one representative of Bisgaard taxon 8 (P1572), two representatives of Bisgaard taxon 5 (CCUG 16493 and CCUG 24854) and the type strain of A. muris were more remotely related to the type species of the genus Actinobacillus using infB sequence analysis (Fig. 1).
Comparison of partial 16S rRNA gene sequences
A fragment from the 5' end of the 16S rRNA gene was sequenced from all 59 bacterial strains included in the present study. The fragment encompassed 437 nt in strains of A. lignieresii and A. pleuropneumoniae (cluster 1); 435 nt in clusters 2–9, CCUG 28015, P1572 and the two representatives of Bisgaard taxon 5; 431 nt in P. pneumotropica biotype Heyl; and 429 nt in the type strains of P. pneumotropica and A. muris (GenBank/EMBL accession nos AJ438630–AJ438688). Phylogenetic analysis based on near-full-length 16S rRNA sequences from 114 representatives of the family Pasteurellaceae has recently been published (Olsen et al., 2004), so the 16S rRNA sequencing in the present study was carried out to ascertain the identity of the isolates and to study intraspecies variation. In most cases, minor differences were found between the sequence of the rRNA gene fragment determined by us and the sequences available in the databases for the same strains. The majority of discrepancies concerned single bases that were either resolved in our sequence, but undetermined in GenBank/EMBL, or vice versa.
The subgrouping of cluster 1 observed by infB analysis was not supported by 16S rRNA gene sequencing. Near-full-length 16S rRNA sequences obtained with the type strains of A. lignieresii and A. pleuropneumoniae were almost identical and differed in only 4 nt positions (Fussing et al., 1998). Two of the polymorphic positions occurred in the fragment used in the present study and these two signature bases were repeatedly confirmed in the rRNA sequence of the other 12 strains of infB cluster 1, allowing a distinction of A. lignieresii from A. pleuropneumoniae. The variation within the two species was restricted to a single polymorphic position in each species (0·2 %) in the fragment of the 16S rRNA gene studied (GenBank/EMBL accession nos AJ438630–AJ438642 and AJ438688; alignment not shown).
With the five representatives of A. capsulatus, three distinct sequences were obtained (GenBank/EMBL accession nos AJ438675–AJ438679). The previously published sequences of the type strain and P796 were essentially confirmed, whereas the three remaining isolates carried mutually identical 16S rRNA gene fragment sequences with 98·6 % similarity to the type strain of the species.
Relatively large differences between the multiple rrn operons were indicated in some species of the genus Actinobacillus. With the type strain of A. hominis and an additional strain (P575, GenBank/EMBL accession nos AJ438666–AJ438667), the nucleotides GCAA were found in positions where the rrnB operon of E. coli carries the sequence TTCG (nt 208–211, numbering with reference to E. coli 16S rRNA). With the five other isolates of A. hominis, the sequencing in this area was ambiguous, as two distinct nucleotides at each position were discernible in both sequencing directions. The 16S rRNA sequence of the type strain, available as L06076, recognizes the ambiguity at position 208, but not at 209–211. With all seven representatives of A. hominis (GenBank/EMBL accession nos AJ438666–AJ438672), as well as the deposited sequence (L06076), positions 199 and 218 were ambiguously determined.
Representatives of A. equuli were divided into two groups, one group being almost identical to A. suis and the other showing more resemblance to A. hominis. These separate sequence groups have been previously represented as GenBank/EMBL accession numbers M75072 (the type strain of A. equuli) and M75067 (CCUG 19799). The latter strain has erroneously been referred to as A. capsulatus (Dewhirst et al., 1992), but it is a haemolytic variant of A. equuli, formerly designated Bisgaard taxon 11 (Berthoud et al., 2002; Christensen et al., 2002b). Haemolytic activity was not correlated with 16S rRNA sequence, as both groups contained haemolytic as well as non-haemolytic strains (dendrogram not shown). In the fragment of 16S rRNA studied, the difference between the two groups of A. equuli was determined by 12 polymorphic positions (nt 456–461 and 471–476, E. coli numbering). However, with the type strain of A. equuli (P157T, GenBank/EMBL accession no. AJ438649), electrophoretograms were obtained from both sequencing directions that were clearly an average of the two sequences at the 12 polymorphic positions. The sequencing was repeated with bacteria subcultured twice from a single colony to ascertain the purity of the strain and the same results were obtained.
Delineation of the genus Actinobacillus by different molecular methods
In Table 2, the results of partial infB sequencing are shown along with results obtained by 16S rRNA gene sequencing and DNA hybridization data. The type strains of different species (or reference strains from unnamed taxa) are compared with the type strain of the type species of the genus, A. lignieresii. The results from DNA hybridization experiments were compiled from two separate sources and are incomplete. Gaps and ambiguous positions are ignored in the calculation of rRNA similarity.
The results from DNA hybridization are not precise. Rather different results were obtained using the two methods with the type strain of A. suis or using the same method on highly related biovars of A. equuli (Table 2). The DNA relationship is high between the type strain of A. lignieresii and the reference strain of equine lignieresii (genomospecies 1, CCUG 22229), but decreases to a level of around 40 % in relation to members of the A. equuli branch. With 16S rRNA gene comparison, 97·0–98·3 % similarity was seen between the type strains of A. equuli, A. suis, A. ureae, A. arthritidis and A. hominis (all of the A. equuli branch) and A. lignieresii. A similar level was observed with Bisgaard taxon 8, whereas the duck isolate CCUG 28015 was as different from A. lignieresii as Bisgaard taxon 5 (95·1–95·7 %), which is unclustered by phylogenetic analysis (Olsen et al., 2004). A conspicuous result is seen with the type strain of A. capsulatus, which is about as different from A. lignieresii as A. muris or P. pneumotropica. By infB comparison, the similarity values of members on the A. equuli branch (omitting A. capsulatus) with the type species of the genus Actinobacillus are in the range 86·9–88·8 %, the similarity level of A. capsulatus and Bisgaard taxon 8 (which is of unresolved taxonomic position) is 84·3–84·5 %, whereas strains clearly located outside the genus Actinobacillus showed similarity to A. lignieresii of 81·7 % or less (Table 2).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Comparison of 16S rRNA gene sequences must often be carried out with sequences of unequal lengths. While substitutions can be handled easily, either as evenly weighted characters or by assignment of a value according to the type of substitution, it is not easy to score differences in length of sequences. Between the type strains of A. muris and A. lignieresii, a difference in length of nearly 2 % was found in the studied fragment of 16S rRNA. This difference is likely to be of phylogenetic significance, but, due to the inability of currently used algorithms to take this into account, the information is ignored. The allowance for gaps in the comparison of sequences most likely underestimates the phylogenetic distance between bacterial strains.
Another problem arises when differences are encountered between the multiple rrn operons in the same bacterial strain. PCR amplification results in an average product, which may be biased towards any of the individual sequences. The method used in the present study does not permit the certain determination of different sequences in separate operons; however, a clear indication of this phenomenon was observed with A. hominis and A. equuli. With the type strain of the latter species, 12 polymorphic positions were found within a stretch of 21 nt, which were also seen in the sequence deposited by Berthoud et al. (2002) (GenBank/EMBL accession no. AF381186). The equivocally determined nucleotides were not seen with all representatives of the various species. It is possible that repeated sequencing would have revealed ambiguities in a larger number of strains, but this was not attempted, as the sequences were accepted on defined criteria and because another sequencing method would be needed in order to identify the individual operons with certainty. The accepted strategy when comparing 16S rRNA sequences is to ignore positions with ambiguously determined nucleotides. However, even when accurate information is available, e.g. from whole-genome sequencing projects, it is difficult to make use of the information, as different phylogenies could be inferred from different operons. The taxonomic problems associated with rrn operon heterogeneity in the genus Veillonella have recently been addressed (Marchandin et al., 2003).
In general, greater intraspecies variation was observed with infB sequences than with 16S rRNA gene sequences, with A. capsulatus as the notable exception. Five representatives of this species all carried an identical infB sequence, whereas 21 polymorphic positions were encountered in the studied fragment of the 16S rRNA gene. Low similarity (97·9 %) has previously been noted between the near-full-length sequences deposited in databases for the type strain of A. capsulatus and A. capsulatus strain P796 (Olsen et al., 2004). Identical gene sequences in different bacterial strains could indicate intraspecies horizontal spread of a preferred sequence; however, in four positions in the 16S rRNA gene where the type strain differed from three other representatives of A. capsulatus, nucleotides in the fifth strain (P796) were ambiguously determined. It is therefore possible that the apparent differences between strains of A. capsulatus reflect, to some extent, differences in the multiple rRNA operons, and further studies on the homogeneity of this species are needed.
DNA hybridization determinations are reported in the literature with decreasing frequency, which underscores the necessity for other molecular methods to be used for validating conclusions based on 16S rRNA sequence comparisons. Phylogenetic analysis of DNA sequences from translated genes is an approach with theoretical advantages (Palys et al., 1997). Ideally, the method should be based on a number of universally distributed genes and multiple strains of each taxon should be included to avoid unrepresentative sequences, which may be the result of recombination events. One candidate gene is infB, encoding the translation initiation factor 2, which has worked well with the genus Haemophilus (Hedegaard et al., 2001).
Using infB sequence comparison, the genus Actinobacillus sensu stricto should include A. lignieresii, A. pleuropneumoniae, A. equuli, A. suis, A. ureae, A. arthritidis and A. hominis, plus the unnamed taxa of ‘equine lignieresii’ (genomospecies 1) and the sorbitol-negative subgroup of Bisgaard taxon 9 (genomospecies 2). These taxa were all related to the type species of the genus at a level above 85 % and constituted a monophyletic cluster by the neighbour-joining, maximum-likelihood and maximum-parsimony methods. With five representatives of A. capsulatus and a single strain of Bisgaard taxon 8, the infB sequences obtained were related to the type species of the genus at a level below 85 %, and these taxa were positioned with the P. pneumotropica cluster by the majority of the phylogenetic methods employed. These taxa may not belong to the genus Actinobacillus sensu stricto.
A. pleuropneumoniae is the cause of porcine necrotizing pleuropneumonia and was originally allocated to the genus Haemophilus as a consequence of its V factor (NAD) requirement. The description of V factor-independent strains led to an investigation of its generic position. It was found to be so closely related to A. lignieresii that individual speciation was not justified on the basis of DNA relatedness (Pohl et al., 1983). Because of the differences in host range and pathological lesions, A. lignieresii and A. pleuropneumoniae were upheld as separate species. The close relationship of these species is reflected in their infB sequences (Fig. 1). A single strain of A. pleuropneumoniae (HK356) may have acquired a fragment of the infB gene by recombination with a strain of A. lignieresii.
A. lignieresii is associated with characteristic granulomatous lesions in cattle and sheep (Christiansen, 1917; Olsen & Møller, 2004) and bacteria of indistinguishable phenotype have been isolated from horses and human horse bites (Dibb et al., 1981). It was recently shown that equine isolates of A. lignieresii represent a unique genomospecies (genomospecies 1; Christensen et al., 2002a). This finding was substantiated by infB sequence comparison (cluster 2 in Fig. 1). Two representatives of the sorbitol-negative subgroup of Bisgaard taxon 9 (genomospecies 2) were also located on the A. lignieresii branch (cluster 3). Clusters 1–3 were placed on a common branch by all three phylogenetic methods, yet it is apparent that relatively large differences separate clusters 2 and 3 from cluster 1. It is therefore possible that the topology of the A. lignieresii branch is not robust and that inclusion of other species or taxa in the comparison will lead to a shift in the position of clusters 2 and 3. Their affiliation to the genus Actinobacillus is, however, supported by both DNA relatedness and nucleotide sequence similarities (Table 2).
A relatively large number of strains of A. equuli were included in the study. Haemolytic variants of A. equuli have been designated ‘equine isolates of A. suis’ or Bisgaard taxon 11; recently, a reclassification as A. equuli subsp. haemolyticus was proposed (Christensen et al., 2002b). The subspeciation has not been substantiated by 16S rRNA (Berthoud et al., 2002) or infB sequences (Fig. 1). The differences in infB sequence between the 12 haemolytic and non-haemolytic strains of A. equuli (cluster 4) were comparable with the differences encountered between the 14 representatives of cluster 1. However, in contrast to the latter cluster, A. equuli was not well separated from the other species of its branch, particularly A. suis (Fig. 1). A similar result was seen in a study of near-full-length 16S rRNA sequences of eight strains of A. equuli together with representative taxa of the remaining part of the genus Actinobacillus. The phylogeny was poorly resolved and representatives of A. equuli were scattered in the dendrogram among other Actinobacillus species (Christensen et al., 2002b).
Four other species on the A. equuli branch, A. suis, A. ureae, A. arthritidis and A. hominis, were very homogeneous with respect to their infB sequences. These species were each represented by two to seven isolates. Because of the restricted sample size, it is not clear whether the apparent similarity of these species is real.
A. capsulatus presents a particular taxonomic problem. A thorough investigation of DNA relatedness on multiple strains has shown this species to be a true member of the genus Actinobacillus (Escande et al., 1984). In contrast, analysis based on 16 rRNA sequences of five strains showed that the species constituted a separate cluster only remotely related to Actinobacillus sensu stricto (Olsen et al., 2004). Some additional strains were included in this study and the peculiar sequence of the 16S rRNA gene was confirmed. In the studied fragment, the type strains of A. capsulatus, A. muris and P. pneumotropica were equally different from that of A. lignieresii (Table 2); however, the type strains of A. muris and P. pneumotropica carried a 16S rRNA sequence that was considerably shorter (429 nt) than the 435 nt observed with all species on the A. equuli branch, including A. capsulatus. With respect to infB sequence, A. capsulatus took up an intermediate position (Table 2) and, although the cluster was placed on the A. equuli branch by the neighbour-joining method (Fig. 1), this was not the case with the maximum-likelihood and maximum-parsimony methods (dendrograms not shown). Further investigations are needed in order to clarify the taxonomic position of A. capsulatus.
Analysis based on comparison of 16S rRNA sequences has clustered the duck isolate CCUG 28015 and the reference strain of Bisgaard taxon 8 with Actinobacillus sensu stricto (Olsen et al., 2004). This inclusion into the genus was not apparent from investigations of infB sequences (Fig. 1; Table 2). The taxonomic position of these candidate species awaits analysis of a greater number of representative strains. Bisgaard taxon 5, included in the genus by anecdotal evidence, was clearly located outside the genus Actinobacillus sensu stricto by infB sequence comparison (Fig. 1; Table 2).
A large number of provisional taxa and species of the family Pasteurellaceae have been described in recent years, and the family is likely to undergo rearrangement and reclassification in the future. If relatively high levels of similarity are insisted upon within individual genera, a large number of genera will be needed in order to encompass the diversity of the family. Using infB sequence comparison, an 85 % similarity demarcation line appeared appropriate for delineation of the genus Actinobacillus sensu stricto. Further data on DNA sequence similarity in the genus and in the family are needed before taxonomic conclusions are drawn.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Brosius, J., Palmer, M. L., Kennedy, P. J. & Noller, H. F. (1978). Complete nucleotide sequence of a 16S ribosomal RNA gene from Escherichia coli. Proc Natl Acad Sci U S A 75, 4801–4805.
Christensen, H., Bisgaard, M., Angen, Ø. & Olsen, J. E. (2002a). Final classification of Bisgaard taxon 9 as Actinobacillus arthritidis sp. nov. and recognition of a novel genomospecies for equine strains of Actinobacillus lignieresii. Int J Syst Evol Microbiol 52, 1239–1246.[Abstract]
Christensen, H., Bisgaard, M. & Olsen, J. E. (2002b). 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]
Christiansen, M. (1917). En ejendommelig pyaemisk lidelse hos faar. Maanedsskr Dyrlaeg 29, 449–458 (in Danish).
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.
Dibb, W. L., Digranes, A. & Tønjum, S. (1981). Actinobacillus lignieresii infection after a horse bite. Br Med J 283, 583–584.
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.
Felsenstein, J. (1995). PHYLIP (phylogenetic inference package), version 3.57c. Department of Genetics, University of Washington, Seattle, WA, USA.
Fleischmann, R. D., Adams, M. D., White, O. & 37 other authors (1995). Whole-genome random sequencing and assembly of Haemophilus influenzae Rd. Science 269, 496–512.
Friis-Møller, A., Christensen, J. J., Fussing, V., Hesselbjerg, A., Christiansen, J. & Bruun, B. (2001). Clinical significance and taxonomy of Actinobacillus hominis. J Clin Microbiol 39, 930–935.
Fussing, V., Paster, B. J., Dewhirst, F. E. & Poulsen, L. K. (1998). Differentiation of Actinobacillus pleuropneumoniae strains by sequence analysis of 16S rDNA and ribosomal intergenic regions, and development of a species specific oligonucleotide for in situ detection. Syst Appl Microbiol 21, 408–418.[Medline]
Hedegaard, J., Steffensen, S. A., Nørskov-Lauritsen, N., Mortensen, K. K. & Sperling-Petersen, H. U. (1999). Identification of Enterobacteriaceae by partial sequencing of the gene encoding translation initiation factor 2. Int J Syst Bacteriol 49, 1531–1538.
Hedegaard, J., Okkels, H., Bruun, B., Kilian, M., Mortensen, K. K. & Nørskov-Lauritsen, N. (2001). Phylogeny of the genus Haemophilus as determined by comparison of partial infB sequences. Microbiology 147, 2599–2609.
Kilian, M. (1976). A taxonomic study of the genus Haemophilus, with the proposal of a new species. J Gen Microbiol 93, 9–62.[Medline]
Kilian, M., Nicolet, J. & Biberstein, E. L. (1978). Biochemical and serological characterization of Haemophilus pleuropneumoniae (Matthews and Pattison 1961) Shope 1964 and proposal of a neotype strain. Int J Syst Bacteriol 28, 20–26.
Kimura, M. (1980). A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. J Mol Evol 16, 111–120.[CrossRef][Medline]
Krieg, N. R. & Garrity, G. M. (2001). On using the manual. In Bergey's Manual of Systematic Bacteriology, 2nd edn, vol. 1, pp. 15–19. Edited by D. R. Boone & R. W. Castenholz. New York: Springer.
Marchandin, H., Teyssier, C., Siméon de Buochberg, M., Jean-Pierre, H., Carriere, H. & Jumas-Bilak, E. (2003). Intra-chromosomal heterogeneity between the four 16S rRNA copies in the genus Veillonella: implications for phylogeny and taxonomy. Microbiology 149, 1493–1501.
May, B. J., Zhang, Q., Li, L. L., Paustian, M. L., Whittam, T. S. & Kapur, V. (2001). Complete genomic sequence of Pasteurella multocida, Pm70. Proc Natl Acad Sci U S A 98, 3460–3465.
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.
Olsen, I. & Møller, K. (2004). Genus Actinobacillus Brumpt 1910, 849AL. In Bergey's Manual of Systematic Bacteriology, 2nd edn, vol. 2. Edited by D. J. Brenner, N. R. Krieg, J. T. Staley & G. M. Garrity. New York: Springer (in press).
Olsen, I., Dewhirst, F. E., Paster, B. J. & Busse, H.-J. (2004). Family Pasteurellaceae Pohl 1981, 382VP. In Bergey's Manual of Systematic Bacteriology, 2nd edn, vol. 2. Edited by D. J. Brenner, N. R. Krieg, J. T. Staley & G. M. Garrity. New York: Springer (in press).
Palys, T., Nakamura, L. K. & Cohan, F. M. (1997). Discovery and classification of ecological diversity in the bacterial world: the role of DNA sequence data. Int J Syst Bacteriol 47, 1145–1156.
Phillips, J. E. (1984). Genus III. Actinobacillus Brumpt 1910, 849AL. In Bergey's Manual of Systematic Bacteriology, pp. 570–575. Edited by N. R. Krieg & J. G. Holt. Baltimore: Williams & Wilkins.
Pohl, S., Bertschinger, H. U., Frederiksen, W. & Mannheim, W. (1983). Transfer of Haemophilus pleuropneumoniae and the Pasteurella haemolytica-like organism causing porcine necrotic pleuropneumonia to the genus Actinobacillus (Actinobacillus pleuropneumoniae comb. nov.) on the basis of phenotypic and deoxyribonucleic acid relatedness. Int J Syst Bacteriol 33, 510–514.
Thompson, J. D., Gibson, T. J., Plewniak, F., Jeanmougin, F. & Higgins, D. G. (1997). The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 25, 4876–4882.
This article has been cited by other articles:
|
|
 |
|
  P. Kuhnert, B. M. Korczak, H. Christensen, and M. Bisgaard Emended description of Actinobacillus capsulatus Arseculeratne 1962, 38AL Int J Syst Evol Microbiol, March 1, 2007; 57(3): 625 - 632. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Christensen, P. Kuhnert, H.-J. Busse, W. C. Frederiksen, and M. Bisgaard Proposed minimal standards for the description of genera, species and subspecies of the Pasteurellaceae Int J Syst Evol Microbiol, January 1, 2007; 57(1): 166 - 178. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Kuhnert and B. M. Korczak 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, September 1, 2006; 152(Pt 9): 2537 - 2548. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Norskov-Lauritsen, B. Bruun, and M. Kilian Multilocus sequence phylogenetic study of the genus Haemophilus with description of Haemophilus pittmaniae sp. nov. Int J Syst Evol Microbiol, January 1, 2005; 55(1): 449 - 456. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Christensen, M. Bisgaard, O. Angen, W. Frederiksen, and J. E. Olsen Characterization of Sucrose-Negative Pasteurella multocida Variants, Including Isolates from Large-Cat Bite Wounds J. Clin. Microbiol., January 1, 2005; 43(1): 259 - 270. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Kuhnert, B. Korczak, E. Falsen, R. Straub, A. Hoops, P. Boerlin, J. Frey, and R. Mutters Nicoletella semolina gen. nov., sp. nov., a New Member of Pasteurellaceae Isolated from Horses with Airway Disease J. Clin. Microbiol., December 1, 2004; 42(12): 5542 - 5548. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Christensen, P. Kuhnert, J. E. Olsen, and M. Bisgaard Comparative phylogenies of the housekeeping genes atpD, infB and rpoB and the 16S rRNA gene within the Pasteurellaceae Int J Syst Evol Microbiol, September 1, 2004; 54(5): 1601 - 1609. [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 | |
