Identification of immunogenic, surface-exposed outer membrane proteins of Pasteurella haemolytica serotype 1

Pasteurella haemolytica serotype 1 (S1) is the bacterium most frequently recovered from the lungs of cattle that have succumbed to shipping fever pneumonia. P. haemolytica outer membrane proteins (OMPs) are important immunogens in the development of resistance to pneumonic pasteurellosis. The purpose of this study was to identify the repertoire of immunogenic, surface-exposed P. haemolytica (S1) OMPs, that could be important in the development of protective immunity. We determined surface exposure of OMPs by (1) their susceptibility to protease treatment and (2) their ability to adsorb out antibodies from bovine immune sera. For a comprehensive identification of immunogenic, surface-exposed OMPs, we used bovine antisera from calves that were resistant to experimental P. haemolytica challenge after (1) natural exposure to P. haemolytica, (2) vaccination with live P. haemolytica, or (3) vaccination with P. haemolytica OMPs. We identified 21 immunogenic, surface-exposed P. haemolytica OMPs. Most were recognized by all three immune sera. However, some were recognized by one or two of the three antisera. Our analyses identified surface-exposed, immunogenic proteins that were not identified in previous studies.     

Investigation of outer membrane proteins of Pasteurella. 2: Iron-regulated outer membrane proteins of Pasteurella multocida and Pasteurella haemolytica

Pasteurella multocida and Pasteurella haemolytica produce specific proteins in the outer membrane under iron-depleted conditions. Pasteurella multocida serovar A expresses these proteins of molecular masses of 76 and 96 kDa as determined by electrophoresis. The analogous serovar D produces a further iron-regulated protein of 85 kDa. The Pasteurella haemolytica strains of serovar A1, A6 and T contain iron-regulated outer membrane proteins of molecular masses of 71, 77 and 100 kDa. These proteins possess binding positions for iron ions. Both Pasteurella multocida and Pasteurella haemolytica strains utilize iron from porcine and bovine transferrin, but not from haemin and haemoglobin.     

Comparison of serological techniques to measure antibody to Pasteurella haemolytica A1.

Analysis of 45 sera was performed employing five techniques which are currently in use in three laboratories to measure anti-Pasteurella haemolytica antibodies. The enzyme linked immunosorbent assay, passive hemagglutination, complement fixation and direct and indirect bacterial agglutination assays were employed and a relationship between tests in the measurement of anti-P. haemolytica antibodies was demonstrated. Regression analysis together with prediction and confidence intervals were tabulated also. The conclusion drawn from statistical analysis was that all five tests are similar in their ability to detect immune responses (antibody and antigen(s) interactions) to Pasteurella haemolytica.     

Mannheimia haemolytica serotype 1 and Pasteurella trehalosi serotype 10 culture supernatants contain fibrinogen-binding proteins

Fibrinogen-binding proteins were found in the culture supernatants of Mannheimia haemolytica serotype 1 (ATCC 43270) and Pasteurella trehalosi serotype 10 (ECO-100). Sheep fibrinogen was biotinylated and shown to bind to proteins in the culture supernatants by modified western blot. Fibrinogen-binding proteins in the culture supernatant may be important virulence factors leading to the characteristic fibrinous pneumonia caused by these organisms and may be critical antigenic targets for immune prophylaxis.     

Comparison of antigens of Pasteurella haemolytica serotype 1 grown in vitro and in vivo.

To determine whether antigenic differences exist in Pasteurella haemolytica serotype 1 grown in different culture conditions, the bacteria was grown on solid enriched medium, in broth culture, and in tissue chambers subcutaneously implanted in the flanks of calves. The organisms obtained by each culture method were comparable with respect to encapsulation and lipopolysaccharide content. In the bacteria grown in vivo, several unique high molecular-mass (greater than 150 kDa) protein antigens were found by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and protein immunoblotting. Bacteria grown in vitro had higher concentrations of a 49- and a 26-kDa protein than the organisms grown in vivo. The concentration of several major proteins (30, 42, 55, 71, and 100 kDa) were similar among the organisms grown by the three cultural conditions. Although the high molecular-mass antigens were unique for the chamber-grown bacteria, they were recognized by serum from a calf that had been vaccinated with formalin-killed, solid medium-grown P haemolytica and were resistant to challenge exposure with the live organism. This recognition of antigens by serum from the P haemolytica-resistant calf that had been vaccinated with solid-medium-grown bacterium, indicates that the high molecular-mass antigens from chamber-grown P haemolytica may be precursors of or share antigenic determinants with other P haemolytica proteins and may not be important for consideration in vaccine formulation.     

Isolation of immunogenic outer membrane proteins from Mannheimia haemolytica serotype 1 by use of selective extraction and immunoaffinity chromatography

OBJECTIVE: To use antibodies produced by calves in response to infection with Mannheimia haemolytica in immunoaffinity chromatography for the identification and subsequent isolation of the dominant immunogenic antigens from bacteria grown in iron-deficient media. SAMPLE POPULATION: Serum from 10 calves actively infected with M haemolytica. PROCEDURE: An outer membrane protein fraction was obtained from sonicated salt-extracted M haemolytica cells by extraction with N-lauroyl sarcosinate. The immunoglobulin fraction of serum from calves actively infected with M haemolytica was used to prepare an immunoaffinity column. The immunoaffinity column was used to isolate the dominant immunogenic proteins from the outer membrane protein fraction. The resultant immunogenic protein fraction was subjected to ELISA and immunoblot methods as well as carbohydrate quantification. Sequencing of the N-terminal was performed on the most prominent protein. RESULTS: 5 immunogenic proteins with molecular weights of 42, 30, 24, 20, and 15 kd were isolated. The immunogenic protein fraction was found to contain 51% carbohydrate. The immunoaffinity column capacity was 1 microg of immunogenic protein/mL of gel. The N-terminal sequence of the 42-kd protein was Tyr-Gln-Thr-Tyr-Gln-Ser-X-Leu-Gln, where X could not be identified. CONCLUSIONS AND CLINICAL RELEVANCE: lmmunogenic proteins were isolated by use of immunoaffinity chromatography. A substantial amount of carbohydrates was co-purified in the process. Additional experiments are needed to determine whether the carbohydrates would hinder or enhance development of vaccine preparations. This method could potentially allow a more rapid production of antigens for use in vaccines.     

Expression of iron-regulated outer membrane proteins by porcine strains of Pasteurella multocida.

The outer membrane protein (OMP) profiles of two strains of capsular type A Pasteurella multocida isolated from the lungs of pigs with enzootic pneumonia were studied. Sarkosyl extracted OMPs from P. multocida grown under iron-restricted and iron-replete conditions were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blot analysis. Results showed that the iron-regulated outer membrane proteins (IROMPs) with molecular masses of 74 kDa, 94 kDa, 99 kDa and 109 kDa were expressed by strain A52, while 74 kDa, 82 kDa, 94 kDa and 99 kDa IROMPs were expressed by strain B80. Swine immune sera, obtained from pigs which were first immunized with a polyvalent P. multocida type A and type D bacterin and subsequently challenged with type A strain of P. multocida, contained antibodies against the IROMPs. These antibodies cross-reacted with the IROMPs expressed by avian strain P1059 of P. multocida. Convalescent-phase serum obtained from turkeys which survived fowl cholera, also cross-reacted with the IROMPs from porcine strains of P. multocida. These results suggested that IROMPs from porcine and avian strains of P. multocida may share common epitopes that were recognized by swine immune serum as well as turkey convalescent-phase serum.     

Serum IgG and IgM antibody response in cattle to antigens of Pasteurella haemolytica

The serum IgG and IgM antibody responses of 48 cattle vaccinated with live Pasteurella haemolytica (LIVE), formalin-killed P. haemolytica in Freund's incomplete adjuvant (FIA), or formalin-killed P. haemolytica in aluminum hydroxide adjuvant (ALH) to a variety of P. haemolytica antigens were evaluated. Enzyme-linked immunosorbent assays (ELISAs) were used to determine the sequential and day 21 IgG and IgM antibody responses to whole P. haemolytica (WB), a capsular carbohydrate-protein subunit (CPS) extracted from the organism, P. haemolytica capsular carbohydrate (CC), and P. haemolytica leukotoxin (LT). LIVE and FIA vaccinates developed generally higher IgG and IgM responses to all antigens compared to ALH vaccinates. LIVE vaccinates developed IgG responses to LT which were significantly higher (P less than 0.05) than all other vaccinates. In contrast, FIA vaccinates developed significantly higher IgG responses to CPS than all other vaccinates. On the basis of the ELISA results, similar or cross reacting antigenic sites were present in preparations containing surface antigens (WB, CPS and CC), but not LT. Disease resistance, as determined by experimental lesions induced in the 48 calves by transthoracic challenge with P. haemolytica, was significantly greater in the LIVE and FIA vaccinates compared with ALH vaccinates. No significant difference in resistance was detected between LIVE and FIA vaccinates. Lesions in ALH vaccinates were not significantly different than those in phosphate-buffered saline (PBS) controls. Increased IgG responses to all antigens were significantly associated with resistance to experimental disease; however, IgG responses to CPS were most highly correlated with resistance. The only IgM response which was significantly correlated with resistance was the response to CPS. These studies indicate that serum IgG antibody responses to various surface antigens of P. haemolytica, as well as LT, can enhance resistance to experimental pneumonic pasteurellosis. Serum IgM responses, however, do not appear to play a major role in resistance to experimental disease.     

Antibody titers to Pasteurella haemolytica A1 in Ontario beef cattle.

Indirect bacterial agglutination titers to Pasteurella haemolytica A1 were determined in serum, thoracic, pericardial, or peritoneal fluid from cattle necropsied as part of the Bruce County Beef Project in 1979-80 and 1980-81. Antibody titers were also assayed in serum from 84 calves on entry to feedlots in the fall of 1979. Titers on entry were low compared to antibody levels at necropsy. Cattle which died with pneumonia, in particular those dying of fibrinous pneumonia (shipping fever), had lower levels of antibody to P. haemolytica than did those dying of other causes.     

Multiple drug resistance in Pasteurella multocida and Pasteurella haemolytica from cattle and swine.

The results of antimicrobial susceptibility testing on 262 strains of Pasteurella multocida and 141 strains of Pasteurella haemolytica isolated from cattle and swine from 1971 to 1974 were analyzed for patterns of resistance to streptomycin, penicillin, tetracycline, and chloramphenicol, using a modified Kirby-Bauer procedure. Resistance was recorded for 80.5% of the isolants of P multocida and 92.2% of those of P haemolytica. Resistance to streptomycin was most frequent, followed by resistance to penicillin and tetracycline. Most cultures of P multocida and P haemolytica were susceptible to chloramphenicol. There were 9 patterns of resistance with the aforementioned antibiotics. The combinations, streptomycin and penicillin and streptomycin and tetracycline, each accounted for approximately 10% of the resistance patterns of P multocida. Approximately half of the 14 isolants of P haemolytica were resistant to the combination of streptomycin, penicillin, and tetracycline. These observations underscore the need for antimicrobial susceptibility testing of clinical isolants of P multocida and P haemolytica.     

Systemic and pulmonary antibody responses of calves to Pasteurella haemolytica after intrapulmonary inoculation.

Systemic and pulmonary antibody responses of calves to Pasteurella haemolytica were evaluated by measuring immunoglobulin production in blood for 9 days and in pulmonary lavage fluid for 7 days after intrapulmonary inoculation. Clinical signs, pulmonary lesions, pulmonary and systemic inflammatory response, and amount of antigen in lavage fluid were used to evaluate the response of calves to challenge with P haemolytica. The pulmonary response consisted of production of IgG, IgE, and IgM antibodies to P haemolytica antigens and a 17- to 68-fold increase of cells in lavage fluid 8 hours after inoculation, with a gradual decrease toward normal. Antibodies of the IgM isotype to P haemolytica were demonstrated as early as 8 hours through 7 days after inoculation in 3 of 3 calves. Of the anti-P haemolytica isotypes, IgM was found in the highest concentration. In all of the inoculated calves, IgE was found 1 to 2 days after inoculation, and IgG was found in 2 of 3 inoculated calves from day 1 through 7 after inoculation. Detection of IgG correlated with smaller pulmonary lesions. Immunoglobulin A was not detected in lavage fluid. Serum was evaluated for IgG and IgM antibody response to P haemolytica. Specific IgM was detectable 5 days after inoculation, and IgG was detectable 7 days after inoculation. Pasteurella haemolytica antigens were not detected in serum or plasma. A transient increase in neutrophil count was found 8 hours after inoculation, with return to baseline values by 24 hours after inoculation.(ABSTRACT TRUNCATED AT 250 WORDS)     

Characterization of outer membrane proteins participating in iron transport in Pasteurella multocida serotype A3

Iron-regulated outer membrane proteins (IROMPs) of P. multocida serotype A3, which function as receptors for complexes containing iron ions, are induced by iron deficiency in the bacterial growth environment. Analysis of an electrophoresis image of proteins isolated from bacteria grown on medium supplemented with 2,2'-dipyridyl revealed expression of 16 new proteins that were not noted in the case of the bacteria grown in standard conditions, with molecular weights from 30 to 160 kDa. Induction of IROMP expression occurred within 30 minutes after restricted iron conditions were established. In immunoblotting, distinct reactions were noted for proteins of molecular weight ranges of 25-49 kDa, 61-95 kDa, and 108-214 kDa. Proteins of the molecular weight of 68, 75 and 86 kDa were analysed using mass spectrometry and matched with the highest probability to proteins in the NCBI data base. Several dozen different proteins with similar amino acid sequences were matched to each sample.     

Two outer membrane proteins are bovine lactoferrin-binding proteins in Mannheimia haemolytica A1

Mannheimia haemolytica is a Gram negative bacterium that is part of the bovine respiratory disease, which causes important economic losses in the livestock industry. In the present work, the interaction between M. haemolytica A1 and bovine lactoferrin (BLf) was studied. This iron-chelating glycoprotein is part of the mammalian innate-immune system and is present in milk and mucosal secretions; Lf is also contained in neutrophils secondary granules, which release this glycoprotein at infection sites. It was evidenced that M. haemolytica was not able to use iron-charged BLf (BholoLf) as a sole iron source; nevertheless, iron-lacked BLf (BapoLf) showed a bactericidal effect against M. haemolytica with MIC of 4.88 ± 1.88 and 7.31 ± 1.62 μM for M. haemolytica strain F (field isolate) and M. haemolytica strain R (reference strain), respectively. Through overlay assays and 2-D electrophoresis, two OMP of 32.9 and 34.2 kDa with estimated IP of 8.18 and 9.35, respectively, were observed to bind both BapoLf and BholoLf; these OMP were identified by Maldi-Tof as OmpA (heat-modifiable OMP) and a membrane protein (porin). These M. haemolytica BLf binding proteins could be interacting in vivo with both forms of BLf depending on the iron state of the bovine.

Electronic supplementary material

The online version of this article (doi:10.1186/s13567-016-0378-1) contains supplementary material, which is available to authorized users.     

Serum antibody response to carbohydrate antigens of Pasteurella haemolytica serotype 1: relation to experimentally induced bovine pneumonic pasteurellosis

The antibody responses to the capsular carbohydrate (CC) purified from Pasteurella haemolytica serotype 1 were determined by an ELISA, using 135 sera from 6 calves vaccinated with phosphate-buffered saline solution, formalin-killed P haemolytica bacterins, live P haemolytica, or an extract of P haemolytica referred to as carbohydrate-protein subunit (CPS). Calves vaccinated with live P haemolytica, bacterins, or CPS developed serum antibodies to CC. Bacterins containing Freund incomplete adjuvant or Freund complete adjuvant induced higher antibody responses than did bacterins containing aluminum hydroxide. In 4 of 6 experiments, high antibody responses to CC were significantly (P less than 0.05) correlated with resistance to transthoracic challenge exposure with P haemolytica. When calves were challenge exposed with a dose of P haemolytica that was 4.5 times greater than the standard challenge exposure dose or when calves that had been vaccinated with CPS were challenge exposed, antibody responses did not significantly (P greater than 0.05) correlate with resistance to challenge exposure. The amount of serum antibodies to CPS increased significantly (P less than 0.05) when calves were vaccinated with live or killed P haemolytica or with CPS, compared with that in calves given saline solution. In 5 of 6 experiments, correlation between high antibody responses and resistance to challenge exposure was significant (P less than 0.05). The correlation between those variables was not significant (P less than 0.07) for CPS-vaccinated calves. In the ELISA, treatment of CPS with sodium m-periodate, to oxidize periodate-sensitive carbohydrate epitopes, failed to markedly alter the antibody response to CPS.(ABSTRACT TRUNCATED AT 250 WORDS)     

Efficacy of a streptomycin-dependent, live Pasteurella haemolytica vaccine against challenge exposure to Pasteurella haemolytica in cattle

A streptomycin-dependent, live Pasteurella haemolytica vaccine was given in 1 or 2 doses to 2 groups of weaned calves; 2 other groups of calves were not vaccinated. All calves in the vaccinated groups and calves in 1 of the nonvaccinated groups were stressed by transport, intratracheally inoculated with bovine herpesvirus type-1 (Cooper strain), and then intratracheally inoculated with P haemolytica type A1. The 4th group of calves (nonvaccinated controls) was not stressed and were not intratracheally inoculated with virus or bacteria. Mean daily weight gains, total clinical sign scores, lung lesion scores, plasma fibrinogen concentrations, and antibody titers against P haemolytica were determined at various intervals. Calves that had been vaccinated twice had greater mean daily weight gains and lower total clinical sign scores and lung lesion scores than did nonvaccinated, challenge-exposed calves, but the difference was not significant (P greater than 0.05). Calves vaccinated once had the greatest mean daily weight gains, the lowest total clinical sign scores, and the lowest lung lesion scores when compared with the other 2 challenge-exposed groups of calves. Mean daily weight gains and total clinical sign scores of calves vaccinated once were significantly different (P less than 0.05) than those of calves vaccinated twice. Nonvaccinated, nonchallenge-exposed control calves did not develop clinical signs of disease, did not develop lung lesions, and had consistently positive daily weight gains, and had scores in these areas that were significantly different (P less than 0.05) from those of all challenge-exposed groups of calves. Increases in plasma fibrinogen concentrations corresponded to infection with P haemolytica.(ABSTRACT TRUNCATED AT 250 WORDS)     

Pasteurella haemolytica cytotoxin neutralizing activity in sera from Ontario beef cattle.

A random sample of sera obtained from cattle necropsied as part of the Bruce County Beef Project in 1980-81 was assayed for the ability to neutralize the cytotoxin of Pasteurella haemolytica A1. Cattle dying of fibrinous pneumonia had significantly lower neutralizing activity in serum than cattle which died for reasons other than pneumonia. Activity in pneumonic cattle was also lower than the mean of twelve samples randomly chosen from sera of cattle bled on entry to feedlots in the fall of 1979. A role for the toxin neutralizing response in resistance to pneumonic pasteurellosis is proposed.     

Serum neutralization of cytotoxin from Pasteurella haemolytica, serotype 1 and resistance to experimental bovine pneumonic pasteurellosis

Sera from several groups of experimental calves were tested for cytotoxin neutralizing capacity. The relationship between this capability and resistance of the animals to an experimental challenge was examined. All undiluted bovine sera tested, other than fetal bovine serum, neutralized cytotoxin. Preadsorption of selected sera with formalin-killed P. haemolytica did not reduce their neutralizing capacity. Crude IgG fractions extracted from bovine sera retained neutralizing capacity as well. Cytotoxin neutralization titers were determined by serial dilution of sera from cattle which were previously unexposed, naturally exposed, or exposed by vaccination to the organism. Both live and killed vaccines were used. Prior exposure to live organisms resulted in the production of antibodies to both cell surface antigens and cytotoxin, whereas exposure to the killed vaccine resulted in the production of antibodies primarily to cell surface antigens. Resistance to experimental challenge with the organism correlated directly with serum cytotoxin neutralizing titers.     

Identification of Pasteurella multocida CHAPS-soluble outer membrane proteins

Fowl cholera continues to be of concern to the poultry industry, especially for turkey growers. This disease costs the turkey industry millions of dollars annually. In order to develop improved live attenuated vaccines or subunit vaccines, the outer-membrane proteins of Pasteurella multocida were examined with the use of proteomics. Of the 11 proteins total present in an outer-membrane subfraction of P. multocida, four additional proteins were identified, completing the composition of the detergent-soluble cross-protective protein fraction. These additional four proteins include protective bacterial surface antigen, OMA87 (Accession no. 15603857); heme-hemopexin receptor, HemR (Accession no. 15602441); lactate permease, LctP (Accession no. 15603717); and heptosyl transferase F, RfaF (Accession no. 15603709). Both the Oma87 and the HemR proteins would be of interest for subunit and modified live vaccine studies, respectively, because of their purported roles as virulence factors for P. multocida.     

Induction of acute bronchopneumonia in mice by intrabronchial inoculation of Pasteurella haemolytica serotype 1.

Dose dependent pulmonary lesions of acute bronchopneumonia were induced in male, outbred Swiss Webster mice by intrabronchial inoculation of Pasteurella haemolytica. Five exponential dilutions ranging from 5 x 10(4) to 5 x 10(8) colony forming units per mL (CFU/mL) of Pasteurella haemolytica serotype 1 were inoculated into five groups of mice. Mice were killed by cervical dislocation 24 hours postinoculation. Pulmonary lesions occurred in mice of all five groups, however, 5 x 10(7) CFU/mL was the minimal dose which consistently produced lesions. Focal parenchymal necrosis, suppurative bronchiolitis, and flooding of interalveolar septa and alveoli by edema fluid, fibrin, neutrophils and macrophages, were observed microscopically. We conclude that outbred Swiss Webster mice can be used as a model for the study of selected disease mechanisms of acute lung inflammation and that this model may be used to determine some of the pathogenic mechanisms involved in the development of pulmonary lesions in bovine pneumonic pasteurellosis.