Phenotypic and genotypic characterisation of Pasteurella multocida strains isolated from pigs in Hungary

A total of 146 Pasteurella multocida strains isolated from swine in Hungary in the last 20 years were examined. Biochemical characterisation and PCR-based techniques were used to determine species, subspecies, biovar, capsule type and presence of the toxA gene. Eighty-seven percent of the isolates belonged to P. multocida ssp. multocida, and 98% of these had biovar 3 or were trehalose- or lactose-fermenting or ornithine decarboxylase negative variants of that. Ten percent of the strains were P. multocida ssp. septica, and within this group 80% of the strains showed sorbitol-negative biovars (5, 6 and 7). The rest of the strains (20%) were lactose positive. Only 3% of the porcine isolates were P. multocida ssp. gallicida and 3 out of the 4 strains belonged to the dulcitol-fermenting biovar 8. Using a capsule-specific multiplex PCR, 60% of the strains belonged to capsule type D, 38% to capsule type A, and only 1 isolate had capsule type F. In contrast with data published in the literature, only 3% of capsule type D isolates carried the toxA gene, while this ratio was 41% for the type A strains. A remarkable regional distribution of toxA gene positive strains was observed. All but two isolates were found in swine herds located in the Transdanubian region, separated from other parts of Hungary by the river Danube.     

Genetic diversity of Pasteurella species isolated from European vespertilionid bats

Pasteurella are an important cause of fatal infections in free-ranging bats, but the genetic diversity of bat-derived strains is unclear. In the current study, 81 Pasteurella strains associated with pneumonia, severe organ necroses and systemic infection in free-ranging European vespertilionid bats were characterized by biochemical and molecular typing methods. Genetic relationships and subspecies status of Pasteurella multocida strains were determined by comparative 16S rDNA and rpoB gene sequence analysis. In addition, 30 representatives of the bat-derived P. multocida strains were selected based on phenotypic and genotypic tests to be compared by pulsed-field gel electrophoresis using SmaI. Most (85%) of the Pasteurella strains obtained from free-ranging bats in this study represented P. multocida ssp. septica. P. multocida ssp. multocida and Pasteurella species B were also identified in a small number of isolates. PFGE analysis correlated well with the sequencing results and revealed a high genetic diversity among bat-derived strains of P. multocida ssp. septica. Strains sharing identical or closely related SmaI fragment patterns were cultured from bats of different species, geographic origins, and years of isolation. The presence of numerous different P. multocida strains allows the assumption that Pasteurella infections in vespertilionid bats are not solely based on intra- but also on inter-species transmission. And indeed, our results present evidence of P. multocida infections in bats following cat predation.     

Occurrence of Pasteurella multocida and related species in village free ranging chickens and their animal contacts in Tanzania

Investigation was done to determine the presence of Pasteurella multocida and related species in free ranging chickens and ducks, dogs, cats and pigs in three climatic zones (cool, warm and hot) of rural Morogoro, Tanzania. A total of 153 isolates of P. multocida ssp. multocida and related species were obtained by direct culture on blood agar, selective medium and mouse inoculation. P. multocida ssp. multocida was isolated from 0.7% of chickens and 7% of ducks. In dogs and cats, P. multocida ssp. multocida was isolated from 1 and 68%, respectively. One isolate of Pasteurella gallinarum was isolated from a duck. Other species obtained were; P. multocida ssp. septica, Pasteurella stomatis and taxon 16 from dogs and cats, while Pasteurella dagmatis and Pasteurella canis were found in dogs only. Prevalence of P. multocida ssp. multocida was significantly higher (P<0.01) in ducks of the warm zone (22%) than in ducks of other zones (0%). No significant difference was observed between the prevalence of P. multocida ssp. multocida in chickens of the warm zone (2%) and chickens of the cool and hot zones (0%). Extended phenotypic characterization revealed phenotypic similarities between two isolates from chickens and the duck strains. Mouse inoculation appeared to be more sensitive in detecting P. multocida ssp. multocida than blood agar and selective medium. Direct culture on blood agar recovered most of the isolates from dogs. This study has demonstrated for the first time the presence of P. multocida and related species in the village free ranging chickens, ducks, dogs and cats in Tanzania. Other non-classified Pasteurella spp. were also observed in the study, but further characterization is required before the final classification can be made. This paper reports for the first time the isolation of unclassified Pasteurella from dogs and cats in Africa. The results implies that fowl cholera might be occurring in free ranging poultry, and dogs and cats kept in contact might serve as sources of P. multocida to chickens and ducks. Subsequent applications of molecular techniques to analyse the epidemiological relatedness of clones isolated from different host species is indicated.     

Characterization of Pasteurella species isolated from lungs of calves with pneumonia.

During routine bacteriological examination of pneumonic calf lungs it was experienced that many Pasteurella multocida-like isolates had a fermentation pattern different from what is generally accepted for P. multocida sensu stricto. Forty-one out of 50 strains selected for further investigation were phenotypically related and formed a group of indole-, mannitol-and sorbitol-negative P. multocida-like strains, which was tentatively designated taxon 13. Deoxyribonucleic acid/deoxyribonucleic acid hybridizations including both ornithine positive and ornithine negative strains of taxon 13 allowed the classification of the former as P. multocida biovar 6 and the latter as V factor independent strains of Haemophilus avium.     

Distribution of indole-producing urease-negative pasteurellas in animals.

Three hundred fifty-six animal isolates of indole-positive urease-negative cultures of Pasteurella, which would formerly have been classified as P. multocida, were examined with respect to their relationship to the recently described P. multocida subspecies (ssp.) multocida, septica, and gallicida and P. canis, P. stomatis/Taxon 16, and Pasteurella sp. B. Two hundred sixty-three (73.9%) of the cultures could be identified with one of these taxa, and 93 isolates (26.1%), representing 17 different biotypes, were unassignable. Pasteurella multocida ssp. multocida was the predominant taxon throughout and in most of the 25 animal species from which isolations were made. In dogs, P. canis was the most frequent. Different degrees of host predilection were observed also in P. multocida ssp. septica for cats, P. canis for sheep, and 2 of the unassignable biotypes for cattle and dogs, respectively. Overall, the respiratory tract was the most frequent source of isolates, but a propensity of P. multocida ssp. septica for localization in the central nervous system of cats was noted.     

Comparison of DNA:DNA homology and enzymatic activity between Pasteurella haemolytica and related species

A commercially available microbiological identification system and DNA:DNA hybridization were used to determine relationships between and within serovars 1-13 of Pasteurella haemolytica, and between P haemolytica and P multocida and 4 species of Actinobacillus. All serovars of P haemolytica that belonged to biovar A were related with mean DNA homology of 78%, whereas all serovars of P haemolytica that belonged to biovar T were related to each other with mean DNA homology of 90%. The DNA:DNA hybridization between strains of biovars A and T ranged from 3 to 13%, indicating little or no genetic relationship between the 2 biovars of P haemolytica. The DNA homology between all serovars of P haemolytica and other species of non-P haemolytica bacteria tested (P multocida and actinobacilli) was less than 14%, suggestive of essentially no genetic relationship of P haemolytica with the ATCC reference strains of the genus Pasteurella or the genus Actinobacillus. Enzymatic differences were observed between P haemolytica and the other non-P haemolytica bacteria tested; however, the microbiological identification system that uses enzymatic reactions could not distinguish among biovars of P haemolytica. Results of this research support other data that suggest that biovars A and T of P haemolytica should be classified as separate species, but do not support the inclusion of either biovar A or T within the genus Actinobacillus.     

Differentiation of Pasteurella multocida isolates from cases of atrophic rhinitis in pigs from Zimbabwe by RAPD and ribotyping

Atrophic rhinitis in pigs is rarely reported in Southern Africa. To determine the relationship between Pasteurella multocida clones from clinical cases of atrophic rhinitis, twenty-one strains were characterised by selected phenotypic and genotypic methods. Biochemical analysis classified 18 strains as P. multocida subspecies multocida, whilst the remainder were grouped into separate unassigned biotypes. Capsular groups A (16/21) and D (l/21) were found among the isolates by PCR. Four ribotype patterns were obtained following HpaII ribotyping, whilst random amplification of polymorphic DNA (RAPD) revealed three main clusters. However, subclusters were also noted for each RAPD cluster. Our results indicate that RAPD offers a better discrimination of strains than ribotyping and that none of the phenotypic characters were directly related to the genotypic clusters.     

Relationships among Pasteurellaceae isolated from free ranging chickens and their animal contacts as determined by quantitative phenotyping, ribotyping and REA-typing

One hundred and forty-three Pasteurella spp. strains and 10 unclassified strains obtained from free ranging poultry, dogs and cats were investigated by extended phenotypic characterization. One hundred and forty-nine of these strains were selected for further studies using ribotyping and REA-typing to evaluate the role of dogs and cats in Pasteurella multocida transmission. Seven and six type strains were included for comparison in phenotyping and genotyping, respectively. Eleven clusters and six unclustered strains were revealed by phenotyping. Ribotyping outlined 12 clusters and six unclustered strains. A correlation between clusters obtained by phenotyping and ribotyping was demonstrated which indicated that a genetic basis exists for clusters outlined by quantitative evaluation of phenotypic data. Similarities and differences in hosts, phenotype, ribotype, and zone of isolation were demonstrated among Pasteurella strains investigated. Isolates of P. multocida from ducks were shown to be clonal by both phenotyping and ribotyping. These strains were identical to one of the chickens strains. REA-typing, however, showed that the chicken strain was different underlining that exchange of clones of P. multocida between avian species rarely happens under village conditions. Management practise in the villages suggest the potential for exchange of P. multocida between poultry and animals kept in contact. The present findings, however, did not indicate that clones of P. multocida are widely exchanged between poultry and other animal species, even though close contact exists. In the present investigation exchange of clones of P. multocida was only demonstrated among animals belonging to the same species. Caution is drawn to the use of ribotyping as the sole method for epidemiological typing and tracing of P. multocida. The present results also underline the importance of proper phenotyping in the identification of P. multocida and related species.     

Atypical Pasteurella strains producing a toxin similar to the dermonecrotic toxin of Pasteurella multocida subspecies multocida

This paper is the first report of the production of a dermonecrotic toxin by pasteurella strains that do not belong to the species Pasteurella multocida subspecies multocida. Four strains, isolated from cattle with atrophic rhinitis, were characterised phenotypically. The strains were related to pasteurellaceae, but their taxonomic position remained unclear. The strains produced a toxin that caused a haemorrhagic dermonecrosis in guinea pigs and was lethal to mice. Both effects were neutralised by an antiserum against the purified dermonecrotic toxin of P multocida subspecies multocida. Western blot analysis of culture filtrates of the bovine strains revealed a protein, with the same molecular weight as dermonecrotic toxin, which reacted with both polyclonal and monoclonal antibodies against the toxin. In an immunodiffusion test, anti-dermonecrotic toxin serum did not discriminate between the toxin of the bovine strains and the toxin of P multocida subspecies multocida. It is concluded that these atypical pasteurella strains produce a toxin that is closely related to the dermonecrotic toxin of P multocida subspecies multocida.     

Characterization of toxin from different strains of Pasteurella multocida serotype A and D

Chromosomal DNA from 13 different selected Pasteurella multocida spp. multocida strains of serotypes A and D were isolated and compared. All 10 toxigenic strains were recognized by a DNA probe which included the toxA gene coding for the Pasteurella multocida toxin (PMT). None of the three nontoxigenic strains reacted with the DNA probe. Toxin from the 10 toxigenic strains were isolated and compared. All were found to possess the biological characteristics previously described for the PMT isolated from P. multocida ssp. multocida NCTC 12178, including molecular mass of approx. 143 kDa and reactivity with a series of monoclonal antibodies. Toxin prepared from different toxigenic strains could not be differentiated immunologically by tandem crossed immunoelectrophoresis, Toxin, which was affinity purified from four of the strains and subsequently inactivated by formaldehyde, was cross-protective when used for vaccination of mice before challenge with PMT. It is concluded that the toxin from toxigenic strains of P. multocida ssp. multocida must be very similar, if not identical.     

The molecular epidemiology of four outbreaks of porcine pasteurellosis

Biochemical profiles, restriction endonuclease analysis (REA) and ribotyping were used to investigate a total of 38 Pasteurella multocida isolates from four separate outbreaks of pasteurellosis in Australian piggeries. Six isolates were obtained from Outbreak 1, 16 from Outbreak 2 and eight each from outbreaks 3 and 4. Outbreaks 1 and 2 were cases of pneumonic pasteurellosis while outbreaks 3 and 4 involved systemic pasteurellosis. Biochemical characterisation established that a number of different types of P. multocida were present in outbreaks 1 and 3 while outbreaks 2 and 4 were associated with a single type of P. multocida. Outbreaks 1 and 3 yielded isolates of P. multocida that belonged to the subspecies multocida and gallicida, with the subspecies multocida isolates being identified as biovar 3 (6 in total) or 12 (1 in total) and the subspecies gallicida isolates (7 in total) being identified as biovar 8. All 24 isolates from outbreaks 2 and 4 belonged to the subspecies multocida and were all biovar 3. REA and ribotyping showed that, in outbreaks 1 and 3, there were three different types of P. multocida in each outbreak with no common strains between the outbreaks. The molecular methods showed that only a single strain of P. multocida was associated with outbreaks 2 and 4, although the outbreaks were associated with strains that differed in REA profiles but shared a ribotype profile. This study has shown that both, systemic and pneumonic pasteurellosis can be associated with either a single strain or multiple strains of P. multocida. The results also indicate that the molecular typing methods of REA and ribotyping are superior to biochemical characterisation for epidemiological investigation of porcine pasteurellosis.     

Molecular Variability among Strains of Pasteurella multocida Isolated from an Outbreak of Haemorrhagic Septicaemia in India

The applicability of conventional and molecular methods for rapid detection and differentiation of Pasteurella multocida serogroup B isolates involved in an outbreak of haemorrhagic septicaemia affecting Indian buffaloes, was studied. Five isolates were obtained and were subjected to phenotypic and genotypic characterization. None of the five isolates could be differentiated on the basis of cultural, biochemical, pathogenicity and antimicrobial sensitivity patterns. Polymerase chain reaction (PCR)-based techniques were found to be specific and sensitive for rapid detection and differentiation of isolates. Repetitive extragenic palindromic (REP-) PCR, enterobacterial repetitive intergenic consensus (ERIC-) PCR and single-primer PCR differentiated all the five isolates into different profiles. All the isolates involved in the outbreak were found to have a genetic profile different from standard P. multocida strain (P52). However, three isolates had similar profiles, whereas each of the remaining two had a different profile. The study indicates the involvement of multiple strains of P. multocida in a single outbreak of haemorrhagic septicaemia in buffaloes. The results also indicate that molecular methods of detection and typing are superior to conventional methods for rapid epidemiological investigations of haemorrhagic septicaemia.     

Taxonomy of pasteurella anatipestifer. 1. DNA base composition and DNA-DNA hybridization analysis

DNA was isolated from 15 strains of Pasteurella anatipestifer and from one strain each of Moraxella nonliquefaciens, M. bovis, Pasteurella multocida, P. haemolytica, P. gallinarum, P. pneumotropica, and P. ureae. The guanine-plus-cytosine contents of P. anatipestifer ranged from 32 to 35 mole %, whereas those of Moraxella and Pasteurella spp. were much higher, ranging from 40 to 45 mole %. DNA-DNA hybridization analysis revealed that homology of nine P. anatipestifer strains to strains ATCC 11845 and PA 15 was 52 to 100%, whereas homology of Moraxella and Pasteurella strains to these strains was only 3 to 17%. Similarly, homology of P. anatipestifer strains, Moraxella, and Pasteurella species other than P. multocida to P. multocida reference strain P-2192 was low. These results strongly suggest that P. anatipestifer is genetically unrelated to either Pasteurella or Moraxella.     

Lack of evidence for the occurrence of Pasteurella ureae in rodents

The taxonomy of five typical human isolates of Pasteurella ureae, one strain of Actinobacillus hominis, and three murine isolates which had been designated as Pasteurella ureae in published reports were re-examined. Their taxonomic relationships were investigated by both conventional phenotypic characterization and by DNA/DNA hybridization using the renaturation method. The human Pasteurella urea strains were highly homogeneous in their phenotypes and in their DNA reassociation. The strain of Actinobacillus hominis studied was genetically distinct from Pasteurella ureae, but was located, like Pasteurella ureae, in the Actinobacillus group. The remaining strains exhibited only low DNA relatedness with Pasteurella ureae and each other; this agreed with their phenotypic divergence. Two of the murine isolates were identified as indole-negative variant strains of Pasteurella pneumotropica sensu stricto (i.e., type Jawetz), or of the type Heyl of Pasteurella pneumotropica, respectively. The remaining murine isolate appears to represent a hitherto unrecognized species of Pasteurellaceae. So far, there is no evidence for the occurrence of Pasteurella ureae outside the human host.     

tRNA-intergenic spacer PCR for the identification of Pasteurella and Mannheimia spp

tRNA-intergenic spacer PCR (tDNA-PCR) was evaluated for its effectiveness in differentiating Pasteurella and Mannheimia (sub)species predominantly of ruminant origin. For this purpose, 38 reference strains and 13 field isolates belonging to both genera were investigated. tDNA-PCR enabled discrimination of all Pasteurella species tested (Pasteurella (P.) aerogenes, P. avium, P. canis, P. lymphangitidis, P. multocida, P. trehalosi). For the differentiation of the subspecies of P. multocida, an additional dulcitol reaction was required. Two of the five so far-defined Mannheimia species, M. granulomatis and M. varigena, had a distinct fingerprinting profile. The remaining three phylogenetically highly related species (M. haemolytica, M. glucosida, and M. ruminalis) clustered together. Nevertheless, M. ruminalis is non-haemolytic, and M. haemolytica and M. glucosida can be differentiated on the basis of two additional phenotypic characteristics (beta-glucosidase and aesculin hydrolysis). In conclusion, tDNA-PCR is a useful tool in differentiating organisms belonging to the genera Pasteurella and Mannheimia.     

Further investigations on Pasteurella multocida infections in feral birds injured by cats.

A total of 64 Pasteurella multocida strains (46 out of 11 different feral bird species, partly injured by cat bites, and 18 strains originating from clinically healthy cats) were biochemically differentiated. As a result, 67.4% of the strains from feral birds and 61.1% from the cats were classified as the subspecies multocida, whilst 21.7% and 27.8% were identified as the subspecies septica. The percentage of frequency for both the subspecies was of a comparable order of magnitude from the birds injured by cat bites and from cats (58.6% and 61.1%, 24.1% and 27.8% resp.), whereas the frequency from other feral birds differed considerably (82.4% and 17.6%). Maltose-positive strains were only demonstrable in birds with wounds inflicted by cats. To date, maltose-positive strains have only been obtained from one cat and one human being with an injury caused by a cat. The results of this investigation confirm the possibility of the direct transmission of Pasteurella multocida via cat bites. 19 strains from feral birds and 15 strains from cats were tested for their capability to produce toxins. The results of these tests were negative. The present paper also describes the pathologic-anatomical and histopathological lesions caused by the infection in feral birds.     

Development of a typing system for epidemiological studies of porcine toxin-producing Pasteurella multocida ssp. multocida in Denmark

The aim of the present study was to evaluate capsular-typing, plasmid-profiling, phage-typing and ribotyping for epidemiological studies of toxin-producing Pasteurella multocida ssp. multocida in Denmark. The evaluation of methods was based on 68 strains from nasal swabs and 14 strains from pneumonic lungs. Strains from lungs were all of capsular Type A, whereas strains from nasal swabs were of both capsular Types A and D. Only 9% of the strains contained plasmids, which could not be associated with antibiotic resistance. Phage-typing divided 61% of strains into 10 groups, while 39% were non-typable. CfoI ribotyping divided strains into four groups of which one type contained 94% of isolates. HindIII ribotyping divided strains into 18 types. A total of 18 strains from The Netherlands, UK and USA were subjected to HindIII ribotyping, resulting in 13 types of which six were identical to ribotypes of Danish strains. Phage-typing of isolates from an outbreak of atrophic rhinitis involving six herds in 1985 showed the existence of an epidemic strain. This type was recognised in the herd suspected of being the source of the infections and in four of the five infected herds. These findings were supported by HindIII ribotyping, as 85% of isolates from all herds were assigned to one ribotype. In conclusion, HindIII ribotyping seems to represent a useful tool for epidemiological studies of toxigenic P. multocida ssp. multocida.     

Prevalence and characteristics of Pasteurella multocida in commercial turkeys

The oropharyngeal regions of 680 meat turkeys and 55 breeder turkeys from nine outbreak farms, three history-outbreak farms, and 19 nonoutbreak farms in Ohio, Indiana, and Pennsylvania were cultured to determine the prevalence of Pasteurella multocida in turkeys. Pasteurella multocida was recovered from 32 out of 105 turkeys belonging to outbreak farms. Pasteurella multocida was not recovered from either history-outbreak or nonoutbreak farms. Characterization via capsular and somatic serotyping, biotyping, restriction endonuclease analysis, and antimicrobial susceptibility testing was performed on all recovered P. multocida isolates. Pasteurella multocida serotype A:1 and somatic serotype 1 with an un-typable capsular serogroup (UT:1) were the most common serogroups found. All isolates belonged to biotype P. multocida ssp. multocida. EcoRI, HpaII, and HindIII restriction enzyme digestions identified three, five, and five restriction fragment length polymorphism profiles, respectively. A majority of the isolates were susceptible to amikacin, ampicillin, ceftiofur, cephalothin, enrofloxacin, florfenicol, gentamicin, neomycin, novobiocin, oxacillin with 2% NaCl, sarafloxacin, tilmicosin, and trimethoprim with sulphadiazine and resistant to clindamicin, penicillin, tiamulin, and tylosin.     

Associations between biovar and virulence factor genes in Pasteurella multocida isolates from pigs in Spain

Two hundred and five isolates of Pasteurella multocida from pigs were phenotypically and genetically characterised by determining their biovar, capsular type, virulence-associated genes and pulsed-field gel electrophoresis (PFGE) profiles. All isolates were identified as P multocida subspecies multocida and most were assigned to biovar 3 (58 per cent) and biovar 2 (39.5 per cent). Biovar 1 represented 2.4 per cent of the isolates. According to the capsular type, the great majority of the isolates (79.0 per cent) belonged to capsular type A, 18.5 per cent belonged to capsular type D and 2.4 per cent were of capsular type F. All isolates harboured ompH, psl, oma87, ptfA, nanB, nanH, tonB, hgbA, sodA and sodC genes, while none of them possessed the transferrin-binding protein gene tbpA. The prevalence of toxA, pfhaA and hgbB genes was variable (7.8, 40.5 and 60.5 per cent of the isolates, respectively). After PFGE typing, isolates of biovar 2 and 3 were grouped in two different clusters (A and B) at a level of 45 per cent similarity. In addition, isolates of biovar 2 and 3 exhibited statistically significant differences (P<0.05) in the virulence-associated hgbB and pfhA genes (biovar 3 was hgbB(+) pfhA(-), while biovar 2 was hgbB- pfhA(+)).     

Molecular epidemiology of two fowl cholera outbreaks on a free-range chicken layer farm.

Two outbreaks of fowl cholera on a multiage free-range egg farm were investigated. The outbreaks occurred in 1994 and 2002. A total of 22 strains of Pasteurella multocida were available for study, 11 from the 1994 outbreak and 11 from the 2002 outbreak. Lesions typical of acute fowl cholera were seen in the 1994 outbreak, whereas both acute and chronic fowl cholera occurred in the 2002 outbreak. The isolates were examined in an extended phenotypic typing methodology, by a P. multocida-specific polymerase chain reaction (PCR), by the Heddleston somatic serotyping scheme, and by restriction endonuclease analysis (REA) typing using the enzyme HpaII. All 22 strains had the same phenotypic properties, all were confirmed as P. multocida by PCR, all were Heddleston serovar 4, and all had the same REA pattern. The results indicate that these 2 outbreaks were caused by the same clone of P. multocida--despite the 8-year time period between the outbreaks.