Pasteurella multocida and Bordetella bronchiseptica infections in rabbits.

The natural history of infection with Pasteurella multocida and Bordetella bronchiseptica in domestic rabbits was studied prospectively at a commercial rabbitry. At weaning, about 25% of rabbits had nasal infections with P. multocida and 75% had infections with B. bronchiseptica. Infection of weanling rabbits paralleled nasal infections of their dams. The proportion of rabbits with both infections increased with age. At 2 to 4 months old, about 50% of rabbits with P. multocida or P. multocida and B. bronchiseptica infections had upper respiratory disease (URD), whereas rabbits with B. bronchiseptica infection had no disease. In rabbits about 10 months old, 75% with P. multocida or P. multocida and B. bronchiseptica infections had URD, whereas virtually none with B. bronchiseptica infection had disease. Disease of the nares, paranasal sinuses, middle ears, and lungs was associated with P. multocida and not B. bronchiseptica infection. In adult rabbits with nasal P. multocida infection, with or without signs of URD, about 80% had concurrent infection of the paranasal sinuses and middle ears and 20% had infection of the bronchi and lungs. In rabbits without nasal P. multocida infection, 20 to 35% had P. multocida infection of the paranasal sinuses and middle ears. Weanling rabbits with and without P. multocida infection had similar immunoglobulin G (IgG) levels. In rabbits observed prospectively, the only antibody differences between those transiently and persistently infected with P. multocida were a diminished IgA response in nasal lavages and an earlier IgM response in sera of transiently infected rabbits. IgG levels increased with the duration of infection. There was no relationship between immunoglobulin levels and freedom from P. multocida infection.     

Adherence of Pasteurella multocida or Bordetella bronchiseptica to the swine nasal epithelial cell in vitro.

The interaction of Bordetella bronchiseptica or Pasteurella multocida with swine nasal epithelial cells was studied in vitro. The mean number of B. bronchiseptica organisms adhered per cell was about three times as high as that of P. multocida (P less than 0.01), and the adherence was specifically inhibited by the homologous antiserum prepared with the whole-cell antigen of each bacterium. The poor affinity of P. multocida to the swine nasal mucosa as compared with that of B. bronchiseptica was also demonstrated in the cultured fragments of the nasal mucosa. When observed with a scanning electron microscope, B. bronchiseptica organisms colonized the fragments, whereas few P. multocida organisms adhered. Morphologically, the P. multocida-infected fragments had an essentially normal structure, whereas marked degeneration and marked desquamation of the epithelial cells and severe inflammatory reactions were observed in many areas of the B. bronchiseptica-infected fragments. These morphological observations were consistent with those for the nasal mucosa of P. multocida- or B. bronchiseptica-infected neonatal pigs (T. Nakai, K. Kume, H. Yoshikawa, T. Oyamada, and T. Yoshikawa, Jpn. J. Vet. Sci. 48:693-701, 1986; T. Oyamada, T. Yoshikawa, H. Yoshikawa, M. Shimizu, T. Nakai, and K. Kume, Jpn. J. Vet. Sci. 48:377-387, 1986). Cultured swine nasal fragments, however, were equally injured when they were incubated in a medium containing purified dermonecrotic toxin (DNT) preparations of B. bronchiseptica or P. multocida. Therefore, these DNT preparations can induce morphological damage closely resembling that induced in vivo. Hence, colonization of B. bronchiseptica and production of its DNT on the swine nasal mucosa appear to result in the production of mucosal damage. On the other hand, P. multocida seems to lack the ability to colonize normal swine nasal mucosa, thus resulting in no production or the slight production of DNT to such an extent as to produce mucosal damage. The present data support our previous hypothesis (Nakai et al.; Oyamada et al.) that B. bronchiseptica induces swine atrophic rhinitis, whereas P. multocida does not.     

Characterization of Pasteurella multocida from nasal cavities of piglets from farms with or without atrophic rhinitis.

A total of 137 strains of Pasteurella multocida isolated from the nasal tracts of pigs with and without clinical atrophic rhinitis (AR) were studied for their biochemical, antigenic, and toxigenic characteristics. There were no major biochemical differences among the P. multocida isolates. Capsular antigen types A and D were both present in the nasal cavities of the pigs with or without clinical AR. However, the prevalence of type D was higher on farms with pigs with AR. Types A and D with different somatic antigens could both be present in the same pig. There was no correlation between somatic types and/or capsular types with the clinical AR status of the pigs on the farm. Toxigenic isolates were found only in pigs which had a problem of clinical AR, and a great majority of these isolates belonged to type D. Since there was a high level of heterogeneity of the strains in the P. multocida population on a farm, several strains should be characterized before the diagnosis of AR could be excluded on the basis of the absence of isolation of rhinopathogenic P. multocida strains.     

Adherence of Bordetella bronchiseptica and Pasteurella multocida to swine nasal ciliated epithelial cells in vitro

A bacterial adherence assay using swine nasal turbinate fragments was established. Turbinate fragments were incubated with Bordetella bronchiseptica or Pasteurella multocida type D at different concentrations or for different incubation times at 37 degrees C on a shaker at 120 rev/min. B. bronchiseptica phase I strains exhibited strong adherence to swine nasal ciliated epithelial cells. The number of adherent bacteria per cell increased when the bacterial concentration or incubation time increased (0, 15, 30, and 60 min); however, the number of adherent bacteria decreased after 3 or 6 hours' incubation due to the loss of cilia from cells. The optimal bacterial concentration and incubation time were 1 x 10(9) organisms/ml and one hour respectively, which resulted in 7.48 +/- 0.66 (Mean +/- SEM; B. bronchiseptica strain 03) and 9.31 +/- 0.54 (B. bronchiseptica strain 013) adherent bacteria per cell. In contrast to B. bronchiseptica phase I strains, rough phase strains of B. bronchiseptica and all P. multocida strains tested showed no adherence to swine nasal ciliated epithelial cells. All B. bronchiseptica phase I strains could agglutinate calf RBC but rough phase strains could not. Furthermore, pretreatment of B. bronchiseptica phase I organisms with 1 mg/ml or 2 mg/ml of trypsin significantly inhibited the adherence of B. bronchiseptica to ciliated epithelial cells; however, trypsin (2 mg/ml) treatment of bacteria did not decrease their ability to agglutinate calf RBC. From these results we conclude that, in addition to hemagglutinin, other proteinaceous components exist on the surface of virulent B. bronchiseptica that are sensitive to 2 mg/ml trypsin; these are suggested to be the adhesins for the adherence of B. bronchiseptica to swine nasal ciliated epithelial cells.     

Two-Color Hybridization Assay for Simultaneous Detection of Bordetella bronchiseptica and Toxigenic Pasteurella multocida from Swine

Bordetella bronchiseptica and toxigenic Pasteurella multocida are the etiologic agents of swine atrophic rhinitis. Methods currently used for their identification are time-consuming and suffer from a lack of sensitivity. We describe a colony lift-hybridization assay for detection of B. bronchiseptica and toxigenic P. multocida that can be performed with a single colony lift derived from a primary isolation plate without the need for pure subcultures of suspect bacteria. Membranes are hybridized simultaneously to probes derived from the B. bronchiseptica alcA gene and the P. multocida toxA gene. A multicolor development procedure permits sequential detection of bound probes. The assay was tested with 84 primary isolation plates generated from nasal swabs from swine with clinical signs of atrophic rhinitis. Comparison of the results from the colony lift-hybridization assay with those from conventional testing, based on a combination of colony morphology, biochemical reactions, mouse lethality, and enzyme-linked immunosorbent assay, indicated that the colony lift assay has superior sensitivity and comparable specificity. This technique has wide application for diagnostic and experimental studies.     

Evaluation of transport media for Pasteurella multocida isolates from rabbit nasal specimens.

A suitable medium for the transport of Pasteurella multocida in nasal specimens from rabbits was investigated by using pure cultures of the organism and nasal swabs from infected rabbits. First, the ability of eight transport media to preserve the viabilities of P. multocida strains isolated from rabbits was studied. Cary-Blair medium and Leibovitz medium no. 15 (L-15) were found to be superior to the other six media tested, enabling survival of the organism for more than 14 days at room temperature. Second, the survival of P. multocida in nasal specimens was evaluated on both Cary-Blair medium and L-15. The recovery rate of the organism from these two media was more than 80 to 90% during 4 days of storage and decreased gradually with increasing preservation time. There were no significant differences (P > 0.05) in recovery rates of the organism between Cary-Blair medium and L-15. On the basis of these results, we recommend the use of Cary-Blair medium for the transport of P. multocida in rabbit nasal specimens because of the ease of transport of nasal swabs by mail.     

Evidence for sialyl glycoconjugates as receptors for Bordetella bronchiseptica on swine nasal mucosa.

The nature of the receptors for Bordetella bronchiseptica was investigated by using the in vitro adherence assay system. The results indicated that sialyl glycoconjugates acted as receptors on swine nasal mucosa. These results were obtained by two independent approaches: inhibition of epithelial cell adherence with sialic acid-containing compounds but not with compounds lacking sialic acid residues and loss of adherence after treatment of epithelial cells with periodate or neuraminidase. B. bronchiseptica seems to have strong affinity for mucin. This may help the bacterium to colonize the mucosal surfaces of the swine nasal cavity.     

Stimulation of bone resorption by inflamed nasal mucosa, dermonecrotic toxin-containing conditioned medium from Pasteurella multocida, and purified dermonecrotic toxin from P. multocida.

The effects of inflamed nasal mucosa from pigs with atrophic rhinitis (AR), cell extract from Bordetella bronchiseptica, conditioned medium from Pasteurella multocida, and purified dermonecrotic toxin (DNT) from P. multocida on mouse fetal long bones in organ culture were studied. Inflamed nasal "AR mucosa" stimulated the release of 45Ca from prelabeled cultures, while histologically the formation of calcified matrix was impaired as well. B. bronchiseptica cell extract only transiently increased 45Ca release, but also impaired the formation of matrix. 45Ca release was also stimulated by DNT-containing conditioned medium from P. multocida and by purified DNT. The effect of DNT was biphasic: low doses (1 to 25 ng/ml) slightly stimulated bone resorption, higher doses were inhibitory. The stimulatory action of DNT on 45Ca release was accompanied by an increase in numbers of preosteoclasts and osteoclasts. The significance of these findings for the pathogenesis of AR is discussed.     

Comparison of four charcoal media for the isolation of Bordetella pertussis.

Charcoal-horse blood agar with 40 micrograms of cephalexin per ml, charcoal-horse blood agar with 3 micrograms of lincomycin per ml, charcoal agar with 3 micrograms of lincomycin per ml, and Legionella (buffered charcoal-yeast extract) agar with 3 micrograms of lincomycin per ml were compared for isolation of Bordetella pertussis. Charcoal-horse blood agar with 40 micrograms of cephalexin per ml gave the best results, with a B. pertussis recovery rate of 100%. Growth was most rapid and the mean number of colonies was highest on this agar, and growth of pharyngeal flora was completely suppressed.     

Comparison of modified Bordet-Gengou and modified Regan-Lowe media for the isolation of Bordetella pertussis and Bordetella parapertussis.

Culture and fluorescent-antibody methods for detection of Bordetella species were evaluated by two state public health laboratories. Field-inoculated plates of Regan-Lowe agar medium were most useful if incubation was initiated on the day of collection. Regan-Lowe and Bordet-Gengou media were comparable for subculturing nasopharyngeal specimens that were transported and enriched in half-strength Regan-Lowe agar. Maximum sensitivity was achieved when the media were used in parallel. Fluorescent-antibody-stained smears of nasopharyngeal specimens were more sensitive for detection of Bordetella pertussis than for detection of Bordetella parapertussis. The fluorescent-antibody method, however, was too insensitive for use without culture.     

Characteristics and biotypes of Pasteurella multocida isolated from humans.

Fifty-two isolates of Pasteurella (48 strains of Pasteurella multocida and 4 strains of atypical Pasteurella) were identified by conventional and commercial test systems. All strains fermented glucose, sucrose, and fructose in purple broth base (Difco Laboratories) with bromocresol purple as indicator, although the atypical Pasteurella produced fermentation reactions that were barely perceptible. Eleven different biotypes were identified by fermentation reactions in maltose, mannitol, xylose, sorbitol, and trehalose media. There was a correlation of biotypes to cat bites, with 61% of cat bite isolates falling into biotype A and B. A correlation of biotype and dog bite isolates was not seen. The choice of medium used for fermentation tests was critical as evidenced by the inability of the organisms to grow in a second commercially purchased preparation of purple broth base. The reliability of commercial test systems in identifying Pasteurella was 81% for Oxi/Ferm (Roche Diagnostics, Div. Hoffmann-La Roche, Inc., Nutley, N.J.), 68% for API (Analytab Products, Plainview, N.Y.), and 11% for Minitek (BBL Microbiology Systems, Cockeysville, MD.).     

Specificity of DNA probes for the detection of toxigenic Pasteurella multocida subsp. multocida strains.

Five DNA probes directed against different regions of the gene that encodes the dermonecrotic toxin of Pasteurella multocida subsp. multocida were examined for their ability to identify toxigenic P. multocida subsp. multocida strains. The specificities of the probes were studied with 96 strains of P. multocida subsp. multocida and 22 strains of 11 other bacterial species. Results of colony hybridization assays using these probes indicated that two of the five probes have potential diagnostic value.     

Bacteriological variation among Bordetella bronchiseptica isolates from dogs and other species.

Bacteriological properties of 50 isolates of Bordetella bronchiseptica were compared. Phase variation, which involved colonial morphology and its associated characters of hemagglutination, hemolysis, acriflavine agglutination; crystal violet staining, flagellation, and fimbriation, occurred among these isolates. Organisms representing the three observed morphotypes did not have different growth rates, nor were any differences in their bacteriological characteristics observed after repeated subculture on agar. There were also variations in antimicrobial drug susceptibility, especially to sulfonamide-trimethoprim, and in nitrate reduction. The relationships among these variable parameters were not apparent. None of the observed variations could be attributed to differences in the species of origin.     

Evaluation of Bordetella bronchiseptica vaccines in specific-pathogen-free piglets with bacterial cell surface antigens in enzyme-linked immunosorbent assay.

The progenies of specific-pathogen-free sows which had been immunized with Bordetella bronchiseptica vaccines of various origin before parturition were challenged intranasally with B. bronchiseptica within 5 days of birth. Sera of piglets were taken weekly and investigated by enzyme-linked immunosorbent assay against a mixture of B. bronchiseptica cell surface antigens containing curled fibers and fimbriae, lipopolysaccharide, and a mixture of proteins mostly derived from the outer membrane. The serological response to this antigenic mixture was paradoxical; the highest titers were obtained with the least effective vaccines. Antibodies which did relate to protection were oriented against the outer-membrane-derived proteins, one of which, of 68,000 molecular weight, appeared to be particularly important for two reasons. First, its concentration within the antigenic mixture was dependent upon cultural conditions; of all the proteins present in virulent strains, it was the first to disappear upon modulation. Second, it was absent from a strain which was unable to induce atrophic rhinitis in specific-pathogen-free piglets. Although all vaccines tested had some beneficial effect on the various clinical manifestations of the disease, only two vaccines were effective (P less than 0.001) in the prevention of nasal pathological changes. These two vaccines also stimulated the highest titers against the 68,000-molecular-weight protein. A mouse protection test utilizing a lethal intraperitoneal challenge failed to monitor the efficacy of vaccines for protection against atrophic rhinitis.     

Isolation of Bordetella bronchiseptica from Blood and a Pancreatic Abscess

Bordetella bronchiseptica is a respiratory pathogen rarely encountered in human hosts. We describe a case of bacteremia and pancreatic abscess caused by this organism. To our knowledge, this is the first reported case of B. bronchiseptica causing intra-abdominal infection in the form of an abscess.     

Enrichment medium for the isolation of Bordetella.

The development of a specimen collection and transport medium outfit for the rapid laboratory diagnosis of whoping cough is described. The transport medium consisted of a semisolid agar containing charcoal, cephalexin, and defibrinated horse blood. It was also found to be an excellent enrichment medium for the selective isolation of Bordetella pertussis and B. parapertussis from scantily populated specimens. The investigation of 3,237 specimens that yielded 1,419 positive isolates of Bordetella, including 86 B. parapertussis, during a 20-month period is presented. A total of 3,076 specimens were processed in the laboratory by using the enrichment medium in addition to the routine procedure. Of these specimens, 757 were submitted in our medium, from which 137 (18%) were positive. Of the 567 specimens received in Amies transport medium, 290 (51%) positive cultures were obtained by the enrichment method only and not by primary culture.     

Bovine erythrocyte-agglutinin as a possible adhesin of Bordetella bronchiseptica responsible for binding to porcine nasal epithelium

The bovine erythrocyte-agglutinin (BEA) of Bordetella bronchiseptica, located on the cell surface in non-fimbrial form, has been identified as a possible adhesion responsible for binding to porcine nasal epithelium in studies with BEA-negative (BEA-) mutants and in competitive studies with bovine erythrocytes.     

Use of single-enzyme amplified fragment length polymorphism for typing Pasteurella multocida subsp. multocida isolates from pigs

Single-enzyme amplified fragment length polymorphism (SE-AFLP) analyses were used to differentiate 97 isolates of porcine Pasteurella multocida subsp. multocida. The strains, isolated from animals with pneumonia, rhinitis, and septicemia, were classified as capsular types A, D, and F. SE-AFLP showed a discriminatory index of 0.87 and identified 18 different profiles.     

Conditions for transformation of Pasteurella multocida by electroporation.

Conditions for electroporation of plasmid DNA into Pasteurella multocida were determined for use in developing a cloning system to study virulence factors of P. multocida. The highest efficiency of transformation (1.25 x 10(7) cfu/micrograms DNA) was obtained when 7.6 x 10(10) cells of P. multocida strain R473 were electroporated at 12.5 kV/cm (10 ms, 5 ng of pVM109). Transformation efficiencies of cells prepared at mid-log-phase were approximately 0.5 log10 lower than early, late, or stationary phases. Neither pBR322 nor pUC-19 were able to transform strain R473 under these conditions, even when DNA concentrations were increased to 1 microgram. When pBR322 was ligated with a Pasteurella plasmid, pLAR-1, the hybrid was able to transform strain R473 at an efficiency between 4.5 x 10(2) and 8 x 10(4) cfu/micrograms DNA. Six strains of P. multocida including serotypes A, B, D, and E were transformed successfully.