| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Veterinary Microbiology, The Royal Veterinary and Agricultural University, Stigbøjlen 4, 1870 Frederiksberg C, Denmark
2 Institute of Veterinary Bacteriology, University of Bern, Laenggass-Strasse 122, Bern, Switzerland
3 Institut für Medizinische Mikrobiologie und Krakenhaushygiene, Philipps-Universität, Marburg, Germany
4 University of Zimbabwe, Harare, Zimbabwe
Correspondence
Henrik Christensen
henrik.christensen{at}vetmi.kvl.dk
 |
ABSTRACT |
|---|
 TOP ABSTRACT MAIN TEXTREFERENCES |
|---|
The GenBank/EMBL/DDBJ accession numbers of 16S rRNA gene sequences described in this report are AY465357–AY465374 and AY431030–AY431033, as listed in Table 1
and Fig. 2.
Supplementary tables showing the DNA reassociation data are available in IJSEM Online.
Present address: Institute for Animal Health, Compton, Newbury, Berks, UK. 
|
MAIN TEXT |
|---|
|
TOPABSTRACTMAIN TEXTREFERENCES |
|---|
) and subsequently included in the Approved Lists (Skerman et al., 1980). The species has been excluded from Pasteurella sensu stricto on the basis of genotypic studies, which is the reason for enclosing the genus name in brackets (Mutters et al., 1989). Presently, [P.] aerogenes seems rather heterogeneous. Twenty-three biovars were reported by Bisgaard (1993) and five DNA clusters outlined by Böhme (1993). Four additional biovars have subsequently been outlined. The species was initially isolated from the intestines, kidney and nose of necropsy specimens from piglets and young swine in the USA and found associated with diarrhoea and other bacterial infections (McAllister & Carter, 1974). Only a single isolate of [P.] aerogenes from a pig fetus was considered as a primary pathogen in the first report (McAllister & Carter, 1974). Later on, Fodor et al. (1991) also reported the bacterium isolated from a pig fetus in Hungary. The origins of pig isolates from Belgium were in accordance with the initial description; however, the bacteria were also reported from vaginal exudates of sows (Hommez & Devriese, 1976). In the 1990s, it was concluded that most isolates of [P.] aerogenes were obtained from pathological lesions and that the ecology of the species was poorly known (Bisgaard, 1993). More recently [P.] aerogenes has been identified from human infections related to pig bites (Lindberg et al., 1998).
rRNA–DNA hybridizations have shown that [P.] aerogenes was most closely related to Pasteurella multocida (De Ley et al., 1990). This relationship was not confirmed by 16S rRNA phylogenetic analysis, since 16S rRNA phylogenetic analysis of the family Pasteurellaceae showed that [P.] aerogenes, [Pasteurella] mairii and [Actinobacillus] seminis formed a monophyletic unit defined as the Seminis cluster (cluster 14) (Olsen et al., 2004).
[P.] mairii (Bisgaard taxon 19) was proposed as a separate species by Sneath & Stevens (1990). The species was named on basis of nine strains isolated from pigs and one strain isolated from a prairie marmot (Sneath & Stevens, 1985). Previous studies found [P.] aerogenes and [P.] mairii to be related by DNA–DNA hybridization at 46 % DNA reassociation (Mutters et al., 1985). [P.] mairii has also been suggested to be excluded from Pasteurella sensu stricto (Mutters et al., 2004).
[Actinobacillus] rossii was established by Sneath & Stevens (1990) on the basis of an unnamed bacterial taxon represented by 10 strains isolated from pigs (Sneath & Stevens, 1985), including four strains originally described by Ross et al. (1972). [A.] rossii could only be separated unequivocally from [P.] aerogenes by acid production from sucrose according to their descriptions (Sneath & Stevens, 1985, 1990).
Misidentification of [A.] rossii and [P.] aerogenes occurs frequently when reference is made to 16S rRNA gene sequence comparison. The 16S rRNA gene comparison of Olsen et al. (2004) found strains MCCM 01550 and MCCM 01551 (listed as P.sp.2 and 1989-5477, respectively, in Table 1) to be related to the type strain of [A.] rossii in accordance with the present study despite their phenotypic identification as [P.] aerogenes. 16S rRNA phylogenetic analysis also showed that the type strain of [A.] rossii was closely related to [Actinobacillus] porcinus (Christensen et al., 2003). Further studies of virulence and pathogenicity are dependent on an improved classification, including a correct identification of [A.] rossii.
|
). Unfortunately, previous descriptions of [P.] aerogenes do not include CAMP-test data, and bovine blood was used rather than ovine blood. Another representative of the RTX toxin family was found in the type strain of [A.] rossii (Schaller et al., 2000). Both genes showed similarity to the apxII and apxIII genes of Actinobacillus pleuropneumoniae. The RTX toxins of these organisms might represent a major virulence factor; however, further studies of their impact on pathogenicity and phylogenetic relationships remain to be performed. Such studies of virulence and pathogenicity have to await improved classification and identification of these organisms. The aim of the present study consequently was to improve the classification of [P.] aerogenes, [P.] mairii and [A.] rossii by comparing phenotypic results with genotypic characterization mainly obtained by 16S rRNA gene sequence analysis and DNA–DNA hybridizations.
Bacterial strains and phenotypic characterization
Eighty-nine strains including reference strains were selected for the study based upon previous reports on phenotypic (Bisgaard, 1993) and DNA–DNA hybridization (Ursing, 1981; Böhme, 1993) data (Table 1). Further phenotypic characterization was performed according to Bisgaard et al. (1991) including separation into biovars based on 17 characters variable within bacteria previously recognized as [P.] aerogenes (Table 2). Only some of the biotyped strains were included in the present genotypic investigation. The type strain of [A.] porcinus was also phenotyped to allow comparison with [P.] aerogenes, [P.] mairii, [A.] rossii and [A.] seminis. [A.] porcinus was cultivated on chocolate agar.
|
) were characterized in order to select representative strains for 16S rRNA gene sequence comparison. For additional comparison, the type strains of the species [P.] aerogenes, [P.] mairii, [A.] rossii and [A.] seminis were included. Cells were cultivated in brain heart infusion broth (Difco) and DNA was isolated as described by Christensen et al. (1993). DNA was dried and redissolved in water of Millipore quality and the concentration was determined by spectrophotometry (GeneQuant; Pharmacia Biotech). PCR amplification was performed by the Ready-To-Go RAPD analysis kit (Amersham Biosciences) as described previously (Dziva et al., 2001). Primer three (5'-GTAGACCCCGT) was selected for the study to give few and distinct bands on the basis of screening with the primers included in the kit against the type strain of [A.] rossii and a strain of [P.] aerogenes. The amplified fragments were resolved on agarose gels and stained in ethidium bromide for visualization.
Four groups, each with an identical profile, were found in addition to separate profiles for the type strains of [P.] aerogenes and [A.] seminis (Fig. 1). The distribution of strains within the four RAPD types is shown in Table 1. Between four (types IV–VI) and nine (type III) fragments were observed. No fragments were in common between the profiles II ([P.] mairii) and IV–VI ([A.] rossii), underlining the genotypic diversity of these taxa. Eleven strains of [P.] mairii including the type strain included under two different designations formed group II, while 14 strains representing [A.] rossii formed groups IV (seven strains including the type strain), V (four strains) and VI (three strains) (Table 1; Fig. 1). Up to four fragments with sizes of approx. 350, 600, 650, 700, 750, 800, 850, 900, 1000 and 1500 bp were shared between the profiles. Types IV–VI with strains of [A.] rossii shared fragments of 350, 600, 700 and 800 bp. None of the four type strains of [P.] aerogenes, [P.] mairii, [A.] rossii and [A.] seminis shared a profile. Three additional strains of [A.] seminis (CCUG 18730, CCUG 18731 and CCUG 23440) not included in the present study shared a profile and differed from the type strain in two fragments (data not shown).
|
–3). DNA–DNA reassociation data are available as two supplementary tables in IJSEM Online.
|
) was performed as outlined previously (Angen et al., 1998; Kuhnert et al., 2000; Christensen et al., 2002). These analyses were performed in different laboratories by common cycle-sequencing protocols. Strain BA436.5 was found difficult to analyse because of inter-operon differences.
Searches for 16S rRNA gene sequences were performed by BLAST (Altschul et al., 1997) at GenBank. Pairwise sequence comparisons were performed by BESTFIT (Wisconsin Sequence Analysis Package, GCG) and multiple alignments made by PILEUP (GCG). Maximum-likelihood analysis was performed by fastDNAmL including bootstrap analysis (Felsenstein, 1981; Olsen et al., 1994) on a Linux-compatible server. Maximum-parsimony analysis was performed by PHYLIP (Felsenstein, 1995).
Phylogenetic analysis based on 16S rRNA gene sequence comparison showed 17 strains to be closely related to the type strain of [P.] aerogenes, with high support for their common node by bootstrap analysis and parsimony analysis (Fig. 2). These 17 strains represented organisms identified phenotypically as [P.] aerogenes biovars 3 (n=1), 4 (5), 5 (1), 9 (1), 10 (1), 11 (1), 19 (1), 25 (1), 26 (1) and 27 (1) in addition to [P.] mairii (2) and an unclassified strain, and showed 99·6 % or higher 16S rRNA gene sequence similarity to the type strain of [P.] aerogenes (biovar 1). Strains JF2419 and JF2420 shared sequences with the previously deposited strains JF2011, JF2101, JF2154 and JF2039 under the accession numbers AF139577, AF139585, AF139587 and AF139582 (these strains had identical sequences). Strains P591 and P592 shared sequences with strain JF2072, previously deposited under accession no. AF139584. Strain C5527 shared a sequence with strain JF2034, previously deposited under accession no. AF139581.
|
; Olsen et al., 2004).
Fifteen strains were related to the type strain of [A.] rossii and formed a monophyletic group (Fig. 2). These strains represented organisms classified phenotypically as biovars 2, 6–8, 12–15, 21, 23, 24 and 26A of [P.] aerogenes, each represented by a single isolate. Members of this group showed 97·8 % or higher 16S rRNA gene sequence similarity to the type strain of [A.] rossii and the group was then slightly more diverse than was observed for the strains related to [P.] aerogenes. The species [A.] rossii and [A.] porcinus were related (Porcinus) (Christensen et al., 2003) (Fig. 2). Strains representing the two groups [P.] aerogenes and [A.] rossii showed 93·8–95·2 % 16S rRNA gene sequence similarity. Between [P.] aerogenes and [P.] mairii strains, 95·3–97·2 % similarity was found, while strains of [A.] rossii and [P.] mairii showed 93·9–95·5 % similarity.
Strains representing biovars 16, 17, 18 and 20 of [P.] aerogenes were found to be unrelated to the [P.] aerogenes–[P.] mairii–[A.] rossii complex by 16S rRNA gene sequence comparison. The non-haemolytic but CAMP-positive strain B96/12 of biovar 16 was related to the type strain of Actinobacillus equuli subsp. equuli (accession no. M75072) with 99·9 % similarity. Strains JF2423 and BA436.5, respectively from biovars 17 and 18, showed 98·9 % 16S rRNA gene sequence similarity to taxon 10 (strain CCUG 15572, accession no. AF024528), a taxon closely related to [Actinobacillus] succinogenes (Christensen et al., 2003; Olsen et al., 2004). Strain BA597.9 of biovar 20 was found to be variable in haemolysis, but showed a positive CAMP reaction and was 98–100 % similar in the 16S rRNA sequence to strains of Mannheimia varigena (accession nos AF053893, AF053899, U57076).
DNA–DNA hybridization
DNA–DNA hybridizations between 10 strains in nine pairs were performed according to the spectrophotometric method as described previously (Mutters et al., 1985). Further data on DNA–DNA hybridization were reported by Böhme (1993) (spectrophotometric method) and Ursing (1981) (hydroxyapatite method). DNA reassociation values are shown in two supplementary tables in IJSEM Online, including published data. These data are illustrated as a single linkage dendrogram in Fig. 3. Strains of biovars 1, 9, 10, 11 and 19, including the type strain of [P.] aerogenes, linked at 70 % DNA reassociation, while members of biovars 2, 6, 7, 8, 12, 15 and 21 linked at 81 % reassociation. Between the [P.] aerogenes group and type strain of [P.] mairii, a maximum of 47 % DNA reassociation was observed, indicating that these taxa might be related at the genus level based on DNA–DNA hybridizations (Fig. 3).
|
). The relationship with biovars 16, 18 and 20 and the relatedness between these biovars was at most 37 %, confirming the misidentification of these strains, since 16S rRNA gene sequence analysis showed that strains of these biovars were closely related to A. equuli subsp. equuli, taxon 10 of Bisgaard and M. varigena.
Emendation of the descriptions of [P.] aerogenes, [P.] mairii and [A.] rossii
Strains characterized by 16S rRNA gene sequence comparison, DNA–DNA hybridization as well as strains grouped by the RAPD analysis were all characterized phenotypically. Common characters for all strains of [P.] aerogenes, [P.] mairii and [A.] rossii investigated (Table 1) were Gram-negative staining and lack of motility at 22 and 37 °C. The Hugh & Leifson fermentation test with (+)-D-glucose was positive. Symbiotic growth (NAD requirement) was negative. The porphyrin test was positive. No growth was observed on Simmons' citrate agar and acid was not formed from mucate; an alkaline reaction was not seen in malonate broth. The H2S/TSI and KCN tests were negative. The Voges–Proskauer test at 37 °C was also negative. Nitrate was reduced without gas formation. Urease and alanine aminopeptidase tests were positive. Tests for arginine dehydrolase, lysine decarboxylase and phenylalanine deaminase were negative. The three species were gelatinase-negative and did not hydrolyse Tween 20 or 80. Pigment was not formed. Acid was not produced from meso-erythritol, adonitol, xylitol, (–)-L-xylose, dulcitol, (+)-D-fucose, (–)-L-sorbose, cellobiose, trehalose, (+)-D-melezitose, (+)-D-glycogen, inulin, aesculin, amygdalin, arbutin, gentiobiose, salicin, (+)-D-turanose and 
-N-CH3-glucosamid. Acid was produced from (–)-D-ribose, (–)-D-fructose, (–)-L-fucose and (+)-D-glucose. The ONGP test (-galactosidase tested with o-nitrophenyl D-galactopyranoside) was positive. Tests for 
-fucosidase and -mannosidase were negative.
All except a single isolate of [A.] rossii produced acid from (+)-D-galactose, while only a single isolate was PNPG positive (-glucosidase activity tested by 4-nitrophenyl -D-glucopyranoside as substrate). Using the limits that positive or negative reactions refer to 90 % of strains, these tests were excluded from Table 3, in which characters that separate the three taxa are stated. From this table, it is noted that at least four characters separate [P.] aerogenes from [P.] mairii and [P.] mairii from [A.] rossii, while not even a single character allows a clear-cut separation of [P.] aerogenes from [A.] rossii. However, a combination of ornithine decarboxylase, production of indole and acid from (–)-D-mannitol, (–)-D-sorbitol, (+)-D-melibiose and raffinose separates these taxa.
The present study showed that phenotypic characterization was insufficient for classification and identification of porcine members of the [P.] aerogenes–[P.] mairii–[A.] rossii complex. Comparison of 16S rRNA gene sequences showed that [P.] aerogenes should be restricted to only 11 biovars, whereas 12 other biovars showed high 16S rRNA gene sequence similarity to the type strain of [A.] rossii. Finally, four biovars were found to be unrelated to [P.] aerogenes and [A.] rossii by 16S rRNA gene sequence comparison, but related to other taxa of the Pasteurellaceae including Actinobacillus sensu stricto, Mannheimia and taxon 10 of Bisgaard. Both A. equuli subsp. equuli and M. varigena have been isolated from pigs. [P.] aerogenes biovar 16 was recorded as producing gas from (+)-D-glucose. However, reinvestigations showed that gas production could not be reproduced and misidentification is explained as the presence of gas in the Durham tube before inoculation. Biovar 18 was (–)-D-mannitol- and phosphatase-negative and (–)-L-fucose-positive, characters not previously reported for A. equuli. Only differences in phosphatase and formation of gas from glucose separate taxon 10 from A. equuli sensu stricto. For the same reasons, biovar 18 might be regarded as a (–)-D-mannitol-negative taxon 10 (M. Bisgaard, unpublished results). Biovar 17 differs from taxon 10 (CCUG 15572) only in (–)-D-mannitol and from biovar 18 in (+)-L-rhamnose. Similar phenotypic characters were, however, observed for biovar 17 and taxon 6 of Bisgaard. DNA–DNA hybridizations are needed to resolve the taxonomic positions of these taxa. Several phenotypic characters [-haemolysis, methyl red, urease, (+)-D-mannose, PNPG and ONPF] separate biovar 20 from M. varigena biovar 1. ONPF is testing for -fucosidase activity and was performed with 2-nitrophenyl -L-fucopyranoside.
Strains JF2006, JF1319, JF2032 and JF2118 of [P.] aerogenes with the RTX toxin (pax gene) (Kuhnert et al., 2000) could not be separated from the other strains based on 16S rRNA gene sequence comparison.
Polyamine profiles have been reported for two strains of [P.] aerogenes; however, strain BK1778/4 of biovar 19 differed significantly from the type strain. In the type strain, spermidine occurred in the highest proportion followed by putrescine, but in the other strain, BK1778/4, 1,3-diaminopropane was dominant, making up around 90 % of all polyamines (Busse et al., 1997). The results might be explained if the strains had been misidentified; however, strain BK1778/4 was included in the present study and found to give a DNA–DNA reassociation value of 87 % with the type strain of [P.] aerogenes. Further investigations are needed to solve this problem. In the type strain of [A.] rossii and in strain BK2661.4 of biovar 21, 1,3-diaminopropane dominated, with 94·7 and 91·0 % of polyamines, respectively. This similarity was in accordance with a 16S rRNA gene sequence similarity of 99·1 % between these two strains. For the type strain of [P.] mairii, 1,3-diaminopropane and putrescine constituted 47·7 and 43 % of the total polyamine content, respectively, diverging significantly from the strains of [P.] aerogenes and [A.] rossii characterized.
Human isolates obtained from infections related to pig bites seem to be related to biovar 17 of [P.] aerogenes (Lester et al., 1993; Lindberg et al., 1998), although their phenotype was also reportedly related to that of taxon 6 of Bisgaard (Bisgaard et al., 1983), a taxon otherwise isolated only from guinea pigs. In addition, taxon 6 of Bisgaard could not be separated phenotypically from biovar 17 in the present study. Finally, the SP group strains MCCM 00489 and MCCM 02119, which are phenotypically related to Bisgaard taxon 6 (M. Bisgaard, unpublished results), showed polyamine patterns related to [A.] rossii, with 1,3-diaminopropane dominance of around 92 % (Busse et al., 1997). The identification of a strain of biovar 17 of [P.] aerogenes by 16S rRNA gene sequence comparison as a misidentified strain of taxon 10 of Bisgaard could indicate that the human isolates might also belong to this taxon, since 98·9 % similarity was reported between strain JF2423 (=P.a.3) of the present study and taxon 6 of Bisgaard. Exactly the same 16S rRNA gene sequence similarity was found between taxa 6 and 10 of Bisgaard.
Ribotyping (RFLP with EcoRI probed by 16S–23S Escherichia coli rRNA) of 20 of the strains in the present study was able to separate the strains into groups with similarity to [P.] aerogenes or [A.] rossii as well as separating the divergent biovars 16, 17, 18 and 20 (M. Bisgaard, unpublished results). Previously, ribotyping with the application of EcoRI was reported by Lester et al. (1993). Five of the strains investigated in the present study (P591 of biovar 10, P592 and P593 of biovar 11, P595 of biovar 6 and the type strain P634T of [P.] aerogenes) belonged to two clusters related at 57 % similarity only. Strains of biovar 10 and 11 were related to the type strain of [P.] aerogenes at 92 % similarity, while the strain of biovar 6 (shown to represent [A.] rossii in the present study) clustered at 57 % (Lester et al., 1993). In the present study, strains of biovars 10 and 11 showed 99·8 % 16S rRNA gene sequence similarity to the type strain of [P.] aerogenes, whereas biovar 6 was related to the type strain of [P.] aerogenes with 95·1 % similarity only. It can be concluded that ribotyping with EcoRI is a reliable tool for separation of isolates related to either [P.] aerogenes or [A.] rossii.
Based on the present emended description of the three species [P.] aerogenes, [P.] mairii and [A.] rossii, it would be tempting to re-evaluate published phenotypic data by Ross et al. (1972), Sneath & Stevens (1985), McAllister & Carter (1974) and Fodor et al. (1991) on these species. However, such a comparison is difficult for several reasons. First of all, methods might differ slightly. As examples, an important character such as -haemolysis depends on the type of blood (bovine or ovine) and medium used, including pH (M. Bisgaard, unpublished observation). In addition, the outcome of the phosphatase test seems to depend on the type of test used (see Table 3). Another problem is that the original data do not include results for single strains, which is a precondition for clear-cut comparisons.
Ward et al. (1998) reported on equine isolates of [P.] mairii; however, the strains differed in trehalose, lactose and salicin according to the present description of [P.] mairii. For the same reasons, identification of isolates of [P.] mairii from horses should be questioned unless genetic data confirm their classification.
The accuracy of the present characterization was evaluated by the inclusion of three different designations of the type strain of [P.] mairii. Both phenotypic and RAPD characterization of strains P637T, Mair5143/70T and NCTC 10699T showed identical results.
The three taxa [P.] aerogenes, [P.] mairii and [A.] rossii can probably not be separated on the basis of DNA G+C content or genome size. The DNA G+C content of the type strain of [P.] aerogenes was 41·8 mol% (Mutters et al., 1985), with a genome size of 1·6–2·0 GDa (Mutters et al., 2004). The DNA G+C content of the type strain of [P.] mairii was 43·4 mol% and the genome size 1·9 GDa (Mutters et al., 1985; Piechulla et al., 1985), and the DNA G+C content of the type strain of [A.] rossii was 41·9 mol% with a genome size of 1·8 GDa (Piechulla et al., 1985). For two other strains related to [A.] rossii by 16S rRNA gene sequence comparison, slightly lower DNA G+C contents but larger genome sizes were found in the present study. The DNA G+C content of strain P.24 of [A.] rossii ([P.] aerogenes biovar 7) was 40·5 mol%, with a genome size of 2·5 GDa, and the DNA G+C content of strain BA657/3 of [A.] rossii ([P.] aerogenes biovar 12) was 39 mol% and the genome size was 2·2 GDa.
Based on the unique phylogenetic position of [P.] aerogenes, [P.] mairii and [A.] seminis, making up the Seminis group inferred from 16S rRNA gene sequence comparison (Christensen et al., 2003; Olsen et al., 2004), this group might deserve genus rank. With the emended description of [P.] aerogenes in the present study, this species showed 95·5–97·2 % 16S rRNA gene sequence similarity to the emended [P.] mairii, while this taxon showed 97·6–97·7 % similarity to the type strain of [A.] seminis. Unfortunately data are missing to document fully the relationships between [P.] aerogenes, [P.] mairii and [A.] seminis, since [A.] seminis has not been extensively characterized phenotypically and DNA–DNA hybridizations have not been made between this taxon and [P.] aerogenes and [P.] mairii. With the exception of a negative reaction in urease and ONPG and differences in colonial morphology, [A.] seminis shares the common phenotypic properties for [P.] aerogenes, [P.] mairii and [A.] rossii given in the present paper (M. Bisgaard, unpublished data). Characters that separate all four taxa are stated in Table 3. In addition, RAPD profiling seems to be a valuable tool to confirm the separation of the three species and their separation from [A.] seminis (see Fig. 1). However, until further information is provided, it is only possible to change the circumscription of the three species [P.] aerogenes, [P.] mairii and [A.] rossii to improve the identification of these taxa and not to name a new genus. The major changes compared to the original descriptions (McAllister & Carter, 1974; Kilian & Frederiksen, 1981; Sneath & Stevens, 1990) have been the extended number of strains characterized and different kinds of phenotypic properties evaluated. Few characters have changed compared with the original description; however, new characters useful for differentiation of the organisms have been added to the descriptions. Twelve characters (negative Gram stain, lack of motility at 22 and 37 °C, lack of growth in Simmons' citrate agar and acid formation from glucose, nitrate reduction, positive urease, negative lysine decarboxylase and gelatinase reactions and lack of acid production from dulcitol and salicin) were conserved between all three species and were in accordance with the original descriptions of McAllister & Carter (1974) and Sneath & Stevens (1990). A further 14, 28 and 31 reactions were in accordance with the original descriptions for each of the three species [P.] aerogenes, [P.] mairii and [A.] rossii, respectively. However, 10, 18 and 15 reactions diverged from the original descriptions. This might be related to different numbers of strains investigated changing the positive and negative signs in relation to the tolerated limits of 10 % variation allowed for a reaction. Different methods or variation in methods might also account for the variation and, finally, phenotypic characterization might have diverged from the 16S rRNA-based identification in previous studies, as already discussed. For 44, 22 and 22 characters, respectively, new information was provided for the three species [P.] aerogenes, [P.] mairii and [A.] rossii.
16S rRNA gene sequence-based phylogenetic analysis has indicated that [A.] rossii and [A.] porcinus form a genus-like group (Christensen et al., 2003; Olsen et al., 2004). Comparison of phenotypic characteristics of [A.] rossii with the type strain of [A.] porcinus showed divergence in seven characters. Reactions of three divergent characters [phosphatase, glycerol and (+)-L-arabinose] are shown in Table 3. In addition, the type strain of [A.] porcinus was NAD-requiring, urease-negative, without ability to form acid from (–)-D-ribose and ONPF-positive compared with [A.] rossii and also [P.] aerogenes and [P.] mairii, being without NAD requirement, urease-positive, with ability to form acid from (–)-D-ribose and ONPF-negative. Since only the type strain of [A.] porcinus has been characterized in detail, it is not currently possible to consider whether the Rossii 16S rRNA group can be named as a new genus; however, the distinctiveness of [A.] rossii and [A.] porcinus at the species level has been verified.
The type strain of [A.] porcinus has previously been characterized by Møller et al. (1996) and Kielstein et al. (2001) and agreement was found for 30 of 31 characters; for (–)-D-fructose, we determined a positive reaction compared with the negative result determined by Kielstein et al. (2001).
Emended description of [Pasteurella] aerogenes McAllister and Carter 1974
[Pasteurella] aerogenes (ae.ro.ge'nes. Gr. subst. masc. aeros air; Gr. subst. neut. genos generating; N.L. subst. aerogenes air-generating).
Cellular and colonial morphologies are as reported by McAllister & Carter (1974). Cells on blood agar are 0·5–1·0 µm wide and 1·1–2·0 µm long with filaments seen, especially in older cultures. After 24 h incubation on bovine blood agar, colonies are circular, smooth, convex, regular and greyish, 0·5–1·0 mm in diameter. Haemolysis is not observed on bovine blood agar. Cells are Gram-negatively stained and do not show motility at 22 or 37 °C. Catalase reaction is positive. Hugh & Leifson fermentation test with (+)-D-glucose is positive. Symbiotic growth (NAD requirement) is negative. Porphyrin test is positive. No growth is observed on Simmons' citrate agar and acid is not formed from mucate; alkaline reaction is not seen in malonate broth. H2S/TSI and KCN tests are negative. Voges–Proskauer test at 37 °C is also negative. Nitrate is reduced without gas formation. Urease and alanine aminopeptidase tests are positive. Tests for arginine dehydrolase, lysine decarboxylase and phenylalanine deaminase are negative and so are tests for indole and phosphatase. The species is gelatinase-negative and does not hydrolyse Tween 20 or 80. Growth on MacConkey is positive. Pigment is not formed. Acid is not produced from meso-erythritol, adonitol, (+)-D-arabitol, xylitol, (–)-L-xylose, dulcitol, (–)-D-sorbitol, (+)-D-fucose, (+)-L-rhamnose, (–)-L-sorbose, cellobiose, (+)-D-melibiose, trehalose, (+)-D-melezitose, (+)-D-glycogen, inulin, aesculin, amygdalin, arbutin, gentiobiose, salicin, (+)-D-turanose or -N-CH3-glucosamid. Acid is produced from glycerol, (–)-D-arabinose, (–)-D-ribose, meso-inositol, (–)-D-fructose, (–)-L-fucose, (+)-D-galactose, (+)-D-glucose, (+)-D-mannose, lactose, maltose, sucrose and dextrin. The ONGP test is positive. Tests for -fucosidase, -mannosidase and -galactosidase and PNPG are negative. The oxidase, methyl red, ornithine decarboxylase, NPG (-glucosidase test determined by 4-nitrophenyl -D-glucopyranoside), PGUA (-glucuronidase reaction determined by 4-nitrophenyl -D-glucopyranosiduronic acid) and onpx (-xylosidase test performed with 2-nitrophenyl -D-xylopyranoside) reactions are isolate-dependent in addition to production of acid from (+)-L-arabinose, (+)-D-xylose, (–)-D-mannitol, raffinose, hydrolysis of aesculin and gas formation from glucose. It is recommended to use supplementary genotyping methods in the identification of this species, e.g. 16S rRNA gene sequencing or ribotyping. The bacteria are mostly reported from pigs and infrequently from rabbit, dog and humans bitten by pigs. The bacteria have been isolated from tonsils, lung, bronchia, intestines, placenta, stomach of aborted pigs, joint, liver and lymph nodes and associated with sepsis, pneumonia and diarrhoea.
The type strain is ATCC 27883T (=CCUG 9995T=CIP 80.14T=DSM 10153T), isolated from pig intestine in USA. The DNA G+C content of the type strain is 41·8 mol% (Mutters et al., 1985) and the genome size is 1·6–2·0 GDa (Mutters et al., 2004).
Emended description of [Pasteurella] mairii Sneath and Stevens 1990
[Pasteurella] mairii (mair'i.i. N.L. gen. n. mairii of Mair).
Cellular and colony morphologies are mostly as previously reported by Sneath & Stevens (1990). Cells are small, non-motile bacilli or coccobacilli, mostly shorter than 1 µm and not more than 0·5 µm wide, occurring singly or as filaments. Colonies on bovine blood agar are regular, circular and convex, greyish, semi-transparent and approx. 1–1·5 mm in diameter after 24 h incubation at 37 °C. Haemolysis is normally not seen but may be observed, mainly under the colonies. Cells are Gram-negative and do not show motility at 22 or 37 °C. Catalase reaction is positive. Hugh & Leifson fermentation test with (+)-D-glucose is positive. Symbiotic growth (NAD requirement) is negative. Porphyrin test is positive. No growth is observed on Simmons' citrate agar and acid is not formed from mucate; an alkaline reaction is not seen in malonate broth. H2S/TSI and KCN tests are negative. Methyl red and Voges–Proskauer tests at 37 °C are also negative. Nitrate is reduced without gas formation. Urease and alanine aminopeptidase test are positive. Tests for arginine dehydrolase, lysine decarboxylase and phenylalanine deaminase are negative. Ornithine decarboxylase and phosphatase tests are positive. Indole and gelatinase tests are negative and members of the species do not hydrolyse Tween 20 or 80. Growth on MacConkey is negative and pigment is not formed. Acid is not produced from meso-erythritol, adonitol, xylitol, (–)-D-arabinose, (–)-L-xylose, dulcitol, (+)-D-fucose, (+)-L-rhamnose, (–)-L-sorbose, cellobiose, lactose, maltose, (+)-D-melibiose, trehalose, (+)-D-melezitose, raffinose, dextrin, (+)-D-glycogen, inulin, aesculin, amygdalin, arbutin, gentiobiose, salicin, (+)-D-turanose or -N-CH3-glucosamid. Acid is produced from (+)-L-arabinose, (+)-D-xylose, (–)-D-ribose, meso-inositol, (–)-D-mannitol, (–)-D-sorbitol, (–)-D-fructose, (–)-L-fucose, (+)-D-galactose, (+)-D-glucose (without gas), (+)-D-mannose and sucrose. ONGP, NPG and PGUA tests are positive. Tests for -fucosidase, -mannosidase, -galactosidase and PNPG are negative. Aesculin is hydrolysed. Oxidase and ONPX reactions are isolate-dependent and so is acid formation from glycerol and (+)-D-arabitol. It is recommended to use supplementary genotyping methods in the identification of this species, e.g. 16S rRNA gene sequencing and RAPD DNA fingerprinting. The bacteria have been isolated from the reproductive tract and from aborted pigs.
The type strain is NCTC 10699T (=CCUG 27189 T), isolated from a pig fetus in the UK. The DNA G+C content of the type strain is 43·4 mol% and the genome size is 1·9 GDa (Mutters et al., 1985; Piechulla et al., 1985).
Emended description [Actinobacillus] rossii Sneath and Stevens 1990
[Actinobacillus] rossii (ross'i.i. N.L. gen. n. rossii of Ross).
Cellular and colony morphologies are mostly as reported previously by Sneath & Stevens (1990). Cells are mostly small bacilli, rarely coccobacilli, less than 2 µm. Colonies on bovine agar are circular, regular, convex and greyish and semi-transparent and about 1–2 mm in diameter after 24 h incubation at 37 °C. -Haemolysis is not commonly observed. Cells are Gram-negatively stained and do not show motility at 22 or 37 °C. Hugh & Leifson fermentation test with (+)-D-glucose is positive. Symbiotic growth (NAD requirement) is negative. Porphyrin test is positive. No growth is observed on Simmons' citrate agar and acid is not formed from mucate; an alkaline reaction is not seen in malonate broth. H2S/TSI and KCN tests are negative. Voges–Proskauer test at 37 °C is also negative. Nitrate is reduced without gas formation. Urease and alanine aminopeptidase test are positive. Tests for arginine dehydrolase, lysine decarboxylase, ornithine decarboxylase and phenylalanine deaminase are negative. Phosphatase test is negative. The species is gelatinase-negative and does not hydrolyse Tween 20 or 80. Pigment is not formed. Acid is not produced from meso-erythritol, adonitol, xylitol, (–)-L-xylose, dulcitol, (+)-D-fucose, (–)-L-sorbose, cellobiose, trehalose, (+)-D-melezitose, (+)-D-glycogen, inulin, aesculin, amygdalin, arbutin, gentiobiose, salicin, (+)-D-turanose or -N-CH3-glucosamid. Acid is produced from glycerol, (+)-L-arabinose, (–)-D-ribose, (–)-D-mannitol, (–)-D-fructose, (–)-L-fucose, (+)-D-galactose, (+)-D-glucose and lactose. ONGP test is positive. Tests for -fucosidase, -glucosidase, PNPG and -mannosidase are negative. Catalase, oxidase, methyl red, indole, growth on MacConkey agar, NPG, -galactosidase, PGUA and ONPX reactions are isolate-dependent, as are acid formation from (+)-L-arabitol, (–)-D-arabinose, (+)-D-xylose, meso-inositol, (–)-sorbitol, (+)-mannose, (+)-L-rhamnose, maltose, (+)-D-melibiose, sucrose, raffinose and dextrin and gas formation from glucose. It is recommended to use supplementary genotyping methods in the identification of this species, e.g. 16S rRNA gene sequencing, RAPD and ribotyping. The bacteria have been isolated from vaginal exudate, abortion, tonsils, lungs, liver, intestines, bone and brain of pigs.
The type strain is ATCC 27072T (=CCUG 12395T=NCTC 10801T=CIP 102634T), isolated from vaginal exudates of a pig in the USA. The DNA G+C content is 39–41·9 mol% and the genome size is 2·2–2·5 GDa.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTMAIN TEXTREFERENCES |
|---|
Angen, Ø., Ahrens, P. & Tegtmeier, C. (1998). Development of a PCR test for identification of Haemophilus somnus in pure and mixed cultures. Vet Microbiol 63, 39–48.[CrossRef][Medline]
Bisgaard, M. (1993). Ecology and significance of Pasteurellaceae in animals. Zentbl Bakteriol 279, 7–26.
Bisgaard, M., Mutters, R. & Mannheim, W. (1983). Characterization of some previously unreported taxa isolated from guinea pigs (Cavia procellus) and provisionally classed with the "HPA-group". INSEM 114, 227–244.
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]
Böhme, A. (1993). Zelluläre Kohlenhydrate und Genomverwandtschaften bei Pasteurella aerogenes und ähnlichen Isolaten. Marburg: Fachbereich Biologie der Philipps-Universität (in German).
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, J. P., Olsen, J. E. & Bisgaard, M. (1993). Ribotypes of Salmonella enterica serovar Gallinarum biovars gallinarum and pullorum. Avian Pathol 22, 725–738.[Medline]
Christensen, H., Bisgaard, M., Angen, Ø. & Olsen, J. E. (2002). Final classification of Bisgaard taxon 9 in the novel species Actinobacillus arthritidis sp. nov., and proposal of genomospecies 1 for equine A. lignieresii. Int J Syst Evol Microbiol 52, 1239–1246.[Abstract]
Christensen, H., Foster, G., Christensen, J. P., Olsen, J. E. & Bisgaard, M. (2003). Characterization of bacteria associated with birds including phylogenetic analysis by 16S rRNA sequence comparison and the description of two new taxa of Pasteurellaceae. J Appl Microbiol 95, 354–363.[CrossRef][Medline]
De Ley, J., Mannheim, W., Mutters, R. & 7 other authors (1990). Inter- and intrafamilial similarities of rRNA cistrons of the Pasteurellaceae. Int J Syst Bacteriol 40, 126–137.
Dziva, F., Christensen, H., Olsen, J. E. & Mohan, K. (2001). Random amplification of polymorphic DNA and phenotypic typing of Zimbabwean isolates of Pasteurella multocida. Vet Microbiol 82, 361–372.[CrossRef][Medline]
Felsenstein, J. (1981). Evolutionary trees from DNA sequences: a maximum likelihood approach. J Mol Evol 17, 368–376.[CrossRef][Medline]
Felsenstein, J. (1995). PHYLIP (phylogeny inference package), version 3.57c. Distributed by the author. Department of Genetics, University of Washington, Seattle, USA.
Fodor, L., Hajtos, I. & Glavits, R. (1991). Abortion of a sow caused by Pasteurella aerogenes. Acta Vet Hung 39, 13–19.[Medline]
Hommez, J. & Devriese, L. A. (1976). Pasteurella aerogenes isolations from swine. Zentbl Veterinaermed B 23, 265–268.
Kielstein, P., Wuthe, H., Angen, Ø., Mutters, R. & Ahrens, P. (2001). Phenotypic and genetic characterization of NAD-dependent Pasteurellaceae from the respiratory tract of pigs and their possible pathogenetic importance. Vet Microbiol 81, 243–255.[CrossRef][Medline]
Kilian, M. & Frederiksen, W. (1981). Identification tables for the Haemophilus-Pasteurella-Actinobacillus group. In Haemophilus, Pasteurella and Actinobacillus, pp. 281–290. Edited by M. Kilian, W. Frederiksen & E. L. Biberstein. London: Academic Press.
Kuhnert, P., Heyberger-Meyer, B., Nicolet, J. & Frey, J. (2000). Characterization of PaxA and its operon: a cohemolytic RTX toxin determinant from pathogenic Pasteurella aerogenes. Infect Immun 68, 6–12.
Lester, A., Gerner-Smidt, P., Gahrn-Hansen, B., Søgaard, P., Schmidt, J. & Frederiksen, W. (1993). Phenotypical characters and ribotyping of Pasteurella aerogenes from different sources. Zentbl Bakteriol 279, 75–82.
Lindberg, J., Frederiksen, W., Gahrn-Hansen, B. & Bruun, B. (1998). Problems of identification in clinical microbiology exemplified by pig bite wound infections. Zentbl Bakteriol 288, 491–499.
McAllister, H. A. & Carter, G. R. (1974). An aerogenic Pasteurella-like organism recovered from swine. Am J Vet Res 35, 917–922.[Medline]
Møller, K., Fussing, V., Grimont, P. A. D., Paster, B. J., Dewhirst, F. E. & Kilian, M. (1996). Actinobacillus minor sp. nov., Actinobacillus porcinus sp. nov., and Actinobacillus indolicus sp. nov., three new V factor-dependent species from the respiratory tract of pigs. Int J Syst Bacteriol 46, 951–956.
Mutters, R., Ihm, P., Pohl, S., Frederiksen, W. & Mannheim, W. (1985). Reclassification of the genus Pasteurella Trevisan 1887 on the basis of deoxyribonucleic acid homology, with proposals for the new species Pasteurella dagmatis, Pasteurella canis, Pasteurella stomatis, Pasteurella anatis, and Pasteurella langaa. Int J Syst Bacteriol 35, 309–322.
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.
Mutters, R., Christensen, H. & Bisgaard, M. (2004). Genus Pasteurella Trevisan 1887, 94AL Nom. cons. Opin. 13, Jud. Comm. 1954. In Bergey's Manual of Systematic Bacteriology, 2nd edn, vol. 2. Edited by G. R. Garrity. New York: Springer (in press).
Olsen, G. J., Matsuda, H., Hagstrom, R. & Overbeek, R. (1994). FastDNAml: a tool for construction of phylogenetic trees of DNA sequences using maximum likelihood. Comput Appl Biosci 10, 41–48.
Olsen, I., Dewhirst, F. E., Paster, B. J. & Busse, H.-J. (2004). Family Pasteurellaceae. In Bergey's Manual of Systematic Bacteriology, 2nd edn, vol. 2. Edited by G. R. Garrity. New York: Springer (in press).
Piechulla, K., Bisgaard, M., Gerlach, H. & Mannheim, W. (1985). Taxonomy of some recently described avian Pasteurella/Actinobacillus-like organisms as indicated by deoxyribonucleic acid relatedness. Avian Pathol 14, 281–311.
Ross, R. F., Hall, J. E., Orning, A. P. & Dale, S. E. (1972). Characterization of an Actinobacillus isolated from the sow vagina. Int J Syst Bacteriol 22, 39–46.
Schaller, A., Kuhnert, P., Puente-Redondo, V. A., de la Nicolet, J. & Frey, J. (2000). Apx toxins in Pasteurellaceae species from animals. Vet Microbiol 74, 365–376.[CrossRef][Medline]
Skerman, V. B. D., McGowan, V. & Sneath, P. H. A. (1980). Approved lists of bacterial names. Int J Syst Bacteriol 30, 225–420.
Sneath, P. H. A. & Stevens, M. (1985). A numerical taxonomic study of Actinobacillus, Pasteurella and Yersinia. J Gen Microbiol 131, 2711–2738.[Medline]
Sneath, P. H. A. & Stevens, M. (1990). Actinobacillus rossii sp. nov., Actinobacillus seminis sp. nov., nom. rev., Pasteurella bettii sp. nov., Pasteurella lymphangitidis sp. nov., Pasteurella mairi sp. nov., and Pasteurella trehalosi sp. nov. Int J Syst Bacteriol 40, 148–153.
Ursing, J. (1981). Deoxyribonucleic acid hybridisation studies of gas producing pasteurellae. In Haemophilus, Pasteurella and Actinobacillus, pp. 255–263. Edited by M. Kilian, W. Frederiksen & E. L. Biberstein. London: Academic Press.
Ward, C. L., Wood, J. L. N., Houghton, S. B., Mumford, J. A. & Chanter, N. (1998). Actinobacillus and Pasteurella species isolated from horses with lower airway disease. Vet Rec 143, 277–279.
This article has been cited by other articles:
|
|
 |
|
  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]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||
