Abstract
Bovine pneumonic pasteurellosis vaccines incorporate various antigens of Mannheimia haemolytica, including the acknowledged virulence factor leukotoxin (Lkt), and Gs60, a surface lipoprotein. To examine the role of antibodies to Gs60 in protection, an enzyme-linked immunosorbent assay (ELISA) was developed for retrospective analysis of serum samples from previous trials in which vaccines containing native or recombinant Gs60 were administered parenterally. The analysis revealed a positive correlation between the titer of antibodies to Gs60 and protection against experimental challenge in both vaccinates and naturally exposed controls. There was a strong correlation between production of IgG antibodies to Gs60 and Lkt neutralizing antibodies. Analysis of the relationship between the serum antibody titers and resistance to experimental challenge using linear statistical models revealed a significant association between prechallenge titers of serum antibodies to Lkt and protection. Further analysis suggested that antibodies against Gs60 were beneficial when Lkt neutralizing antibody titers were low.
Résumé
Les vaccins pour la pneumonie bovine à Pasteurella contiennent divers antigènes de Mannheimia haemolytica incluant la leucotoxine (Lkt), facteur de virulence reconnu, ainsi que Gs60, une lipoprotéine de surface. Afin d’examiner le rôle des anticorps contre Gs60 dans la protection, une épreuve immunoenzymatique (ELISA) a été développée pour analyse rétrospective d’échantillons de sérum provenant d’études antérieures au cours desquelles des vaccins contenant la Gs60 native ou recombinante étaient administrés par voie parentérale. L’analyse a révélé une corrélation positive entre le titre d’anticorps contre Gs60 et la protection contre une infection expérimentale autant chez des animaux vaccinés que des témoins exposés naturellement. Il y avait une forte corrélation entre la production d’anticorps de type IgG contre Gs60 et des anticorps neutralisants Lkt. Une analyse de la relation entre les titres d’anticorps sériques et la résistance à une infection expérimentale utilisant des modèles statistiques linéaires a révélé une association significative entre les titres d’anticorps sériques pré-infection avec Lkt et la protection. Des analyses supplémentaires ont suggéré que les anticorps contre Gs60 étaient bénéfiques lorsque les titres d’anticorps neutralisants anti-Lkt étaient bas.
(Traduit par Docteur Serge Messier)
Introduction
Bovine respiratory disease (BRD) can be caused by viral and bacterial pathogens acting singly or in combination. Although Mannheimia haemolytica is a common bacterium of the upper respiratory tract and nasopharynx of healthy ruminants, it can also act as an opportunistic pathogen, causing huge economic losses in the dairy and beef cattle industries worldwide (1,2). Serotypes A1 and A2 of M. haemolytica reside in the upper respiratory tract of cattle and sheep, but serotype A1 is the most prevalent serotype isolated from the lung of diseased cattle after necropsy. Serotype A2 is more commonly associated with pneumonic pasteurellosis in sheep (3). Serotype A6 strains are antigenically similar to serotype A1 strains; together they account for almost all cases of bovine pneumonic pasteurellosis worldwide (4).
Some factors implicated in the virulence of M. haemolytica, including leukotoxin (Lkt) and Gs60, may contribute to the efficacy of vaccines. The heat-labile protein Lkt is secreted during bacterial growth and plays a critical role in the pathogenesis of pneumonic pasteurellosis subsequent to colonization of bacteria in the lower respiratory tract (5). Induction of antibodies to Lkt after natural or experimental exposure has been linked to protection against pneumonic disease (6,7). Studies of lkt genes have revealed that the various serotypes of M. haemolytica produce different types of Lkt (8) Nevertheless, polyclonal antiserum raised using Lkt of one serotype is capable of cross-neutralization of the Lkt of other serotypes. Homologous neutralization is, however, more efficient (9,10).
Gs60 is a surface antigen of M. haemolytica and a member of the LppC family of bacterial outer membrane lipoproteins with unknown bioactivity (11). Many Pasteurellaceae, including such pathogens as Histophilus somni, Pasteurella multocida, and M. haemolytica, have open reading frames encoding homologues of Gs60 (12). Analysis of the Gs60 gene suggests the presence of an N-terminal signal peptide containing cysteine. Similar to other bacterial lipoproteins, Gs60 could be anchored through this N-terminal cysteine to the membrane of the bacterium (12–14). Antibodies recognizing a partial sequence of Gs60 have been shown to correlate with protection against pneumonia caused by M. haemolytica (14,15). Antibodies to native Gs60 (detected by western blot) were linked with disease resistance by Lo and Mellors in 1995 (11).
A wide variety of vaccination programs have been used against the various organisms involved in BRD, including M. haemolytica vaccine; however, BRD remains a substantial problem. In 1987 Shewen and Wilkie (16) introduced a vaccine derived from cell-free logarithmic-phase culture supernatant of M. haemolytica serotype A1 that contained Lkt. Since vaccinated animals also had high serum titers of agglutinating antibodies against M. haemolytica, comparable to those in animals vaccinated with whole-cell bacterins, it was concluded that the culture-supernatant vaccine also contained soluble surface antigens of M. haemolytica (16). These surface antigens included serotype-specific antigens, such that antibodies produced by exposure to surface antigens of M. haemolytica A1 did not react as effectively in a bacterial agglutination test with serotype A11 (now classified as M. glucosida) and vice versa. In a subsequent trial using recombinant Lkt, Conlon, Shewen, and Lo (17) found that both agglutination and toxin neutralization antibody titers were correlated to protection against disease. They suggested that although toxin neutralization is essential for immunity it is not sufficient alone for strong protection. Optimal protection requires vaccination with a combination of Lkt and surface antigens. Current vaccines for M. haemolytica are based on culture supernatant containing both Lkt and surface antigens, with or without the addition of whole killed bacteria (18). These vaccines include Presponse SQ (Boehringer Ingelheim Vetmedica, St. Joseph, Missouri, USA), which is based on the original Shewen and Wilkie vaccine.
The bioactivity of Gs60 and its role in pathogenesis or in protection against pneumonia caused by M. haemolytica is poorly understood. Although the presence of anti-Gs60 antibodies in the serum of vaccinated calves has been shown by other researchers, justification for its intentional inclusion in a vaccine as a protective antigen is lacking. This research was initiated to examine responses to Gs60 in experimentally vaccinated calves and to evaluate evidence for its association with protection against experimental challenge. Serum samples from previous vaccination and challenge studies using parenterally injected culture-supernatant vaccines were used for this purpose. An anti-Gs60 enzyme-linked immunosorbent assay (ELISA) was developed and used to determine the antigen-specific IgG response in serum.
Materials and methods
Mannheimia haemolytica genes and cultures
Unless stated otherwise, all M. haemolytica genes, vaccines, and other products were laboratory stocks of serotype A1 derived from a culture originally isolated from a pneumonic calf by Ernst L. Biberstein, University of California, Davis, California, USA, and deposited in the American Type Culture Collection (ATCC) in 1986 by one of us (P.E.S.) as ATCC 43270. Escherichia coli DH5α from our laboratory stock was used for the cloning experiments.
Production and purification of His-tagged recombinant Gs60 (rGs60)
Plasmid pGs60-His was constructed by cloning a fragment containing the gene gs60 into the expression plasmid pET-28a. Briefly, primers GS60-F (5′-CAAGCTCGGAATTCTATATT and GS60-R (5′-TGGCGGAATTCATTGAATTT) (EcoRI sites underlined) were used in a polymerase chain reaction (PCR) to amplify a 1.5 kilo base pair fragment of genomic DNA coding for Gs60. This fragment excludes 45 amino acids from the N-terminal that are involved in the secretion and lipidation of Gs60 as well as 13 amino acids from the C-terminal. Pwo polymerase was used in the PCR, which was conducted as follows: 94°C for 2 min; 10 cycles at 94°C for 15 s, then 57°C for 30 s, and then 72°C for 1 min; and then 20 cycles at 94°C for 15 s, then 57°C for 30 s, and then 72°C for 1 min, with a 5-s addition every cycle. The PCR product was purified, digested with EcoRI, and ligated into pET-28a digested with EcoRI and treated with alkaline phosphatase. This produced an in-frame fusion that added 6 histidine residues to Gs60 for purification. The ligated DNA was transformed into E. coli DH5α selecting for kanamycin resistance. Recombinant plasmid was recovered, mapped, and sequenced to confirm correct orientation and proper insertion of the gs60 fragment. rGs60-His was expressed in E. coli strain BLR (DE3) pLysS (Novagen, EMD Biosciences, Darmstadt, Germany).
Two 30-mL cultures were prepared in lauryl tryptose (LT) medium containing 0.3 mg/mL of kanamycin (Sigma-Aldrich, St. Louis, Missouri, USA) by overnight culture at 37°C with shaking. Each was then subcultured into 750 mL of LT broth containing kanamycin and incubated for 1.5 h at 37°C with shaking. To induce the expression of rGs60, 0.18 g of isopropyl β-D-1-thiogalactopyranoside (Sigma-Aldrich) in 1 mL of double-distilled water (ddH2O) was added, to a final concentration of 1 mM. After incubation at 37°C for 1.5 h, the bacteria were spun in a Beckman J2-21 M/E centrifuge with a JA-14 rotor for 10 min at 15 344 × g. The cell pellets were resuspended in 5 mL of lysis solution (50 mM NaH2PO4, 300 mM NaCl, 10 mM imidazole; pH 8.0) per gram of pellet. Lysozyme (100 mg/mL in ddH2O; Sigma-Aldrich) was added, to a final concentration of 1 mg/mL, and the suspension was held on ice for 10 min. The lysate was sonicated on ice by means of a Heat Systems W-385 ultrasonic sonicator (Medsonic, Brooklyn, New York, USA) for 6 cycles of 10 s with a 10-second gap between each burst. The sonicated lysates were centrifuged for 20 min at 15 344 × g and the supernatant was collected. Purification of rGs60-His was achieved by immobilized-metal affinity chromatography with Ni-NTA (nickel-nitrilotriacetic acid) matrices (Qiagen, Mississauga, Ontario). Fractions of 1 mL were collected and examined by western immunoblot with mouse anti-Gs60 monoclonal antibody (Mab) provided by DOW AgroSciences, Zionsville, Indiana, USA (clone 4H9G10) and alkaline phosphatase-conjugated rat Mab X56 against mouse IgG1 (BD Biosciences Pharmingen, San Diego, California, USA), each at a 1/500 dilution (Figure 1). Fractions containing Gs60 were pooled and used as antigen in the Gs60 ELISA.
Figure 1.
Western immunoblot with monoclonal anti-Gs60 antibody to detect rGs60. Secondary antibody was rat monoclonal antibody to mouse IgG1. Lane 1 — molecular mass standards (sizes shown on the left); lane 2 — recombinant Gs60-His expressed in Escherichia coli (expected molecular mass ~ 62 kDa); lane 3 — concentrated culture supernatant of Mannheimia haemolytica containing native Gs60 (expected molecular mass ~ 60.8 kDa).
The amount of rGs60 in the pooled sample was determined by the Lowry protein microassay (19) with comparison to a standard curve developed for bovine serum albumin fraction V (Sigma–Aldrich). An aliquot of rGs60 was submitted to the Biological Mass Spectrometry Facility of the Department of Molecular and Cellular Biology, University of Guelph, Guelph, Ontario, for matrix-assisted laser desorption/ionization/time of flight mass spectrometry (20), which revealed that at least 7 peptide fragments of the Gs60 protein of M. haemolytica were in the sample, with 22% coverage of the Gs60 sequence.
Assay of bovine IgG antibodies to Gs60
An ELISA was designed to quantify IgG antibodies to Gs60 in calf serum. Maxisorp flat-bottom 96-well plates (Nalge Nunc International, Rochester, New York, USA) were coated with anti-Gs60 Mab (clone 4H9G10), 100 μL/well of a 1/500 dilution in phosphate-buffered saline (PBS). Alternate rows were coated with a 1/100 dilution of normal mouse serum in PBS to control for nonspecific binding. The plates were incubated at 37°C for 4 h and then overnight at 4°C. Normal horse serum [5% v/v dilution in wash buffer containing 0.5% fish skin gelatin and 0.05% Tween 20 (Sigma–Aldrich) in PBS] was used to block nonspecific binding. After washing, purified rGs60-His, diluted 1/250 in wash buffer, was added as antigen. Fourfold serial dilutions of each calf serum sample (starting from 1/10) were prepared in wash buffer and added to the coated and blocked ELISA plate.
Alkaline phosphatase-conjugated mouse Mab against bovine IgG (clone BG-18; Sigma-Aldrich) at a concentration of 1/500 in block buffer was used to detect bound bovine IgG. This antibody is not subisotype-specific but recognizes an epitope common to bovine IgG1 and IgG2. ρ-nitrophenyl phosphate (Kirkegaard and Perry Laboratories, Gaithersburg, Maryland, USA) was used as the substrate for color production, and optical densities were read on an ELISA plate reader (Power Wave XS; BioTek Instruments, Winooski, Vermont, USA) at 405 and 630 nm. Antibody titers were determined by comparison to a standard curve, as previously described (21,22). The curve was generated by serial dilution of a positive serum sample obtained from a colostrum-deprived calf that had been vaccinated 4 times with transudate collected from the thorax of a naive colostrum-deprived calf after experimental challenge with M. haemolytica and filtered at 0.45 μm. Titers were expressed as the log2 of the reciprocal dilution.
Leukotoxin neutralization assay
Calves’ serum samples had previously been tested for M. haemolytica Lkt neutralizing antibodies as described elsewhere (16,17,23). Briefly, lyophilized leukotoxic logarithmic-phase culture supernatant of M. haemolytica stabilized with fetal bovine serum (7% of the volume of the supernatant before lyophilization), 3 mg/mL, was incubated for 30 min with 2-fold serial dilutions of each sample of serum. The mixtures were added to bovine lymphoma cells (BL-3 cells, originally provided by Gordon Theilen, University of California, Davis) seeded at 2 × 105 cells/well in a Nunc flat-bottom plate. The titer was expressed as the log2 of the reciprocal of the 50% end-point dilution, the highest dilution that protected at least 50% of cells, measured by uptake of the vital dye neutral red.
Vaccination trials
Serum samples (frozen at −20°C) from several previous trials of vaccination against M. haemolytica (Table I) were used to determine the induction of antibodies to Gs60 after vaccination and the relationship between antibody titer and protection against experimental challenge. These included a 2005 trial in which calves were vaccinated intramuscularly with E. coli-expressed rGs60 and a 2007 trial in which only serum samples from nonvaccinated control calves were examined (the vaccinated calves in this trial received vaccines irrelevant to this study). These trials were not published. Because some calves that received E. coli-expressed rGs60 had anaphylaxis after vaccination, it was decided, as well, to examine serum samples from an earlier trial (1990) in which postvaccination anaphylaxis was observed in several calves that received purified M. haemolytica capsular polysaccharide as a vaccine (23). Since the capsular polysaccharide was prepared by extraction from the bacterial surface (24), contamination with a fragment of Gs60 was possible. This trial included 28 calves that received Presponse vaccine alone or in combination with other antigens. In all the trials the calves were challenged by intrabronchial instillation of 25 mL of a log-phase culture of M. haemolytica A1 at the dose shown in Table I. The percentage of lung tissue that was pneumonic was evaluated at necropsy 5 or 6 d after challenge (or sooner if the calf was euthanized for humane reasons). The percentage in each lobe was estimated and then multiplied by the percentage contribution of the respective lobe to the overall lung weight as described previously (25). The final value was the sum of all weighted percentages.
Table I.
Experimental challenge trials used as the source of serum in this study
| Trial | Group | Vaccine | Number of calves | Challenge day and dose, CFU/mL |
|---|---|---|---|---|
| A, 2005 | 1 | PBS only (3 mL), aluminum hydroxide (1 mL); total 4 mL/dose IM | 2 | Day 29; 1.25 × 109 |
| 2 | Escherichia coli expressed recombinant Gs60-His (rGs60-His), 3 mL (300 μg), adjuvanted with aluminum hydroxide (1 mL); total 4 mL/dose IM | 3 | ||
| B, 2007 | Unvaccinated control calves | 10 | Day 37; 2 × 109 | |
| C, 1990 (23) | 1 | PBS, 4 mL/dose IM; no adjuvant | 5 | Day 32; 1 × 1012 |
| 2 | Presponse (Wyeth Animal Health, Guelph, Ontario), 2 mL/dose IM; Mannheimia haemolytica A1 culture supernatant adjuvanted with QuilA and aluminum hydroxide | 5 | ||
| 3 | CPS of M. haemolytica A1, 1 mg in 2 mL of PBS/dose IM | 5 | ||
| 4 | Presponse plus CPS, 4 mL/dose | 5 | ||
| 5 | CPS plus E. coli-expressed truncated rLkt of M. haemolytica (rLkt60), 6.4 mg in 2 mL of PBS; total 4 mL/dose IM | 5 | ||
| 6 | Presponse plus CPS and rLkt60 in 2 mL of PBS; total 4 mL/dose IM | 4 | ||
| 7 | Presponse plus CPS and mock E. coli control in 2 mL of PBS; total 4 mL/dose IM | 4 |
CFU — colony-forming units; PBS — phosphate-buffered saline; IM — intramuscularly; CPS — capsular polysaccharide; rLkt — recombinant leukotoxin.
Statistical analysis of the data
The necropsy and serologic data were statistically analyzed to compare the percentage of lung tissue that was pneumonic in the control and vaccinated groups and to determine the relationship of the Gs60 and Lkt antibody titers to resistance to challenge. SAS software (SAS, Cary, North Carolina, USA) was used for all calculations. Correlations were calculated by Pearson’s method. General linear modeling was used to evaluate the effects of anti-Gs60 IgG antibodies and Lkt neutralizing antibodies on the percentage of pneumonic lung tissue. In all tests a value of P ≤ 0.05 was considered significant.
Results
Trial A, conducted in 2005, demonstrated the immunogenicity of rGs60. Owing to the small sample size (3 vaccinates and 2 controls), statistical analysis was not practical. The increase in anti-Gs60 antibody levels before challenge in 2 of the 3 vaccinated calves but in none of the controls confirmed that rGs60-His was immunogenic (Figure 2).
Figure 2.
Serum IgG antibody titers (−log2) to Gs60 in the 3 calves vaccinated in trial A with E. coli-expressed rGs60 on days 0 and 14 and challenged on day 29; top 3 lines. The titers in the 2 control calves (overlapping bottom lines) were below the limit of detection for the assay.
In trial B, conducted in 2007, a significant correlation was found between the serum IgG antibody titer to Gs60 on the day of challenge and the percentage of lung tissue that was pneumonic after challenge in the 10 unvaccinated control calves (P = 0.02, r2 = 0.50) (Figure 3A). There was a trend (P = 0.11, r2 = 0.27) of decreasing percent pneumonic tissue as the Lkt neutralizing titer increased (Figure 3B) and a significant positive correlation (P = 0.04, r2 = 0.42) between the Lkt neutralizing titer and the serum IgG titer to Gs60 (Figure 3C).
Figure 3.
Correlations between prechallenge antibody titers to Gs60 and protection against pneumonic lesions in trial B; linear regression lines of the best fit are presented. The antibodies were presumed to be due to natural exposure in these 10 unvaccinated calves. A — Correlation between percentage of pneumonic tissue and serum IgG titer (P = 0.02, r2 = 0.50). B — Correlation between percentage of pneumonic tissue and serum leukotoxin (Lkt) neutralizing antibody titer (P = 0.11, r2 = 0.27). C — Correlation between the 2 titers (P = 0.04, r2 = 0.42).
In trial C, conducted in 1990, the prechallenge serum IgG antibody titers to Gs60 in all groups receiving the commercial vaccine (Presponse) with or without additional antigens were significantly higher than the titers in the groups that did not receive Presponse (P < 0.01), but the titers of Lkt neutralizing antibodies and the percentages of lung tissue that was pneumonic after challenge were not significantly different among the 7 study groups (Table II). There was a significant correlation between the serum IgG titer of antibody to Gs60 before challenge and the percentage of pneumonic tissue (P = 0.001, r2 = 0.30): the slope of the regression line suggested that calves with a higher titer before challenge had a lower percentage of pneumonic tissue after challenge (Figure 4A). This analysis included all 33 calves, regardless of vaccination status. When the calves were segregated by vaccine group it was evident that those receiving Presponse alone or in combination with other antigens had higher antibody titers (P < 0.001) and lower pneumonic scores (P < 0.05) than those not receiving Presponse (Figure 4B). In addition, there was a significant correlation between the Lkt neutralizing titer before challenge and the percentage of pneumonic tissue (P < 0.001, r2 = 0.45): the slope of the regression line suggested that calves with a higher titer before challenge had a lower percentage of pneumonic tissue after challenge (Figure 4C). Again, when the calves were segregated by vaccine group it was evident that those receiving Presponse had higher antibody titers (P = 0.01) and lower pneumonic scores (P = 0.05) (Figure 4D).
Table II.
Serum antibody activity before challenge and percentage of lung tissue that was pneumonic after challenge for calves in trial C (23)
| Vaccine group | Lkt neutralizing antibody titer, −log2 (reciprocal of 50% end-point) | IgG titer to Gs60, −log2 (based on standard curve) | Percentage of pneumonic tissue |
|---|---|---|---|
| PBS | 5 | 4.2 | 34 |
| 4 | 3.6 | 22 | |
| 7 | 9.1 | 12 | |
| 12 | 7.3 | 2.5 | |
| 9 | 6.4 | 2.5 | |
| Mean ± SD | 7.4 ± 3.2 | 6.12 ± 2.5 | 14.6 ± 13.5 |
| Presponse | 12 | 8.2 | 1.5 |
| 11 | 9.2 | 3.5 | |
| 8 | 6.7 | 6.5 | |
| 8 | 10.9 | 8.5 | |
| 10 | 11.0 | 1.0 | |
| Mean ± SD | 9.8 ± 1.7 | 9.2 ± 1.8 | 4.2 ± 4.2 |
| CPS | 4 | 4.0 | 41.5 |
| 8 | 6.9 | 14.0 | |
| 7 | 7.4 | 1.0 | |
| 9 | 6.6 | 1.0 | |
| 11 | 10.9 | 1.5 | |
| Mean ± SD | 7.8 ± 2.5 | 7.1 ± 2.4 | 11.8 ± 17.5 |
| CPS + Presponse | 9 | 10.4 | 1.5 |
| 7 | 11.3 | 6.5 | |
| 10 | 7.9 | 5.5 | |
| 9 | 8.5 | 1.5 | |
| 13 | 11.7 | 3.5 | |
| Mean ± SD | 9.6 ± 2.1 | 9.9 ± 1.6 | 3.7 ± 2.2 |
| CPS + rLkt | 2 | 3.5 | 44 |
| 5 | 5.5 | 42 | |
| 6 | 8.1 | 4.5 | |
| 7 | 9.4 | 6.5 | |
| 11 | 8.4 | 1.0 | |
| Mean ± SD | 6.2 ± 3.2 | 6.9 ± 2.4 | 19.6 ± 21.4 |
| Presponse + rLkt + CPS | 8 | 9.0 | 4.0 |
| 10 | 11.2 | 1.0 | |
| 13 | 8.7 | 1.5 | |
| 11 | 10.4 | 6.0 | |
| Mean ± SD | 10.5 ± 2 | 9.8 ± 1.1 | 3.1 ± 2.3 |
| Presponse + mock + CPS | 9 | 11.9 | 8 |
| 8 | 11.9 | 1 | |
| 9 | 11.4 | 40 | |
| 7 | 9.9 | 0.5 | |
| Mean ± SD | 8.2 ± 0.9 | 11.2 ± 0.9 | 12.3 ± 18.7 |
SD — standard deviation; mock — E. coli mock recombinant.
Figure 4.
Correlations between prechallenge antibody titers and protection against pneumonic lesions in trial C. Panels A and C represent all 33 calves, regardless of vaccination status. Panels B and D represent calves according to vaccination group. A — Correlation between percentage of pneumonic tissue and serum IgG titer to Gs60 (P = 0.001, r2 = 0.30). B — Percentage of pneumonic tissue versus serum IgG titer to Gs60 by vaccination group. C — Correlation between percentage of pneumonic tissue and serum Lkt neutralizing antibody titer (P < 0.001, r2 = 0.45). D — Percentage of pneumonic tissue versus serum Lkt neutralizing antibody titer by vaccination group. PBS — phosphate-buffered saline; CPS — capsular polysaccharide.
In trial C a significant positive correlation was also observed between the serum IgG titers to Gs60 and the Lkt neutralizing antibody titers of the calves before challenge (P < 0.001 and r2 = 0.34). Therefore, the calves that responded to M. haemolytica naturally or by vaccination tended to recognize both Gs60 and Lkt (Figure 5A). The relationship between the antibody titers at the time of challenge is demonstrated according to vaccine group in Figure 5B. The general linear model that included serum titers for both antibodies before challenge revealed that the Lkt neutralizing antibodies had a stronger association with protection against lung lesions than did the antibodies to Gs60. In this model a significant interaction was found between the 2 types of antibodies (P < 0.001), which implied that the impact of serum IgG antibodies to Gs60 on the percentage of pneumonic tissue differed according to whether the titer of Lkt neutralizing antibodies was high or low. The nature of the interaction is illustrated in Table III, in which the calves were divided into 4 groups on the basis of high or low titers of antibody to Lkt and Gs60. In the calves with higher titers of Lkt neutralizing antibodies the percentage of pneumonic lung tissue was not associated with the titer of antibody to Gs60. However, in the calves with lower titers of Lkt neutralizing antibodies the percentage of pneumonic lung tissue was high (23%) when the titers of antibody to Gs60 were low and low (5.5%) when the titers of antibody to Gs60 were high. Thus, the response to Gs60 was associated with lung protection when the titers of Lkt neutralizing antibodies were low (P < 0.001, r2 = 0.64).
Figure 5.
Relationship between prechallenge serum antibody titers to Gs60 and Lkt and protection against pneumonic lesions in trial C. A — Correlation between serum IgG titer to Gs60 and serum Lkt neutralizing antibody titer regardless of vaccination status (P < 0.001, r2 = 0.34). B — Serum IgG titers to Gs60 and Lkt neutralizing antibody titers by vaccination group. C — Correlation between serum IgG titers to Gs60 and percentage of pneumonic tissue for calves with Lkt neutralizing antibody titers ≤ 8 at challenge (P < 0.001, r2 = 0.64). D — Correlation between serum IgG titers to Gs60 and percentage of pneumonic tissue for calves with Lkt neutralizing antibody titers > 8 (P = 0.18, r2 = 0.11). There was a significant correlation between high serum antibody titers to Gs60 and low percentage of pneumonic tissue only in the calves with low antibody titers to Lkt.
Table III.
Percentage of pneumonic tissue after challenge in groups of calves from trial C with different categories of serum antibody titers
| Serum IgG titer to Gs60 (−log2) | Serum Lkt neutralizing antibody titer (−log2); mean values ± SD for each group of calves
|
|
|---|---|---|
| > 8 | ≤ 8 | |
| > 9 | Group 1, n = 9 | Group 2, n = 7 |
| Percentage of pneumonic tissue, 7.3% | Percentage of pneumonic tissue, 5.5% | |
| Lkt neutralizing antibody titer, 10.33 ± 1.32 | Lkt neutralizing antibody titer, 7.42 ± 0.53 | |
| Anti-Gs60 IgG titer, 10.95 ± 0.80 | Anti-Gs60 IgG titer, 10.24 ± 1.17 | |
| ≤ 9 | Group 3, n = 8 | Group 4, n = 9 |
| Percentage of pneumonic tissue, 2.1% | Percentage of pneumonic tissue, 23% | |
| Lkt neutralizing antibody titer, 10.62 ± 1.59 | Lkt neutralizing antibody titer, 5.44 ± 2.00 | |
| Anti-Gs60 IgG titer, 7.78 ± 0.92 | Anti-Gs60 IgG titer, 5.60 ± 1.78 | |
Discussion
Vaccines derived from M. haemolytica culture supernatant have been in use for more than 2 decades to protect feedlot calves against pneumonic pasteurellosis. These vaccines contain many soluble antigens of M. haemolytica, including Lkt and surface antigens such as Gs60. The current study was conducted to obtain a better perspective on the possible role of Gs60 in protection against pneumonia caused by M. haemolytica. Serologic and necropsy data from 3 previous experimental challenge trials were analyzed. The 2005 trial (A) showed that intramuscular injection of E. coli-expressed rGs60 induced an IgG antibody response in 2 of 3 calves; a 2009 trial confirmed the immunogenicity of rGs60 expressed by transgenic alfalfa when injected intramuscularly (Hodgins DC, Lee RWH, Ziauddin A, Strommer JN, Lo RYC, Shewen PE unpublished data). Together the serologic results from the 3 trials suggested that a serum IgG antibody response to Gs60 can be induced in calves by natural exposure to bacteria and by parenteral vaccination. Similarly, a serum Lkt neutralizing antibody response can be induced by natural exposure or by vaccination. The antibody responses to Gs60 and Lkt increased simultaneously, whether induced by natural exposure or by parenteral vaccination with Presponse.
For trials B and C a significant negative correlation was found between the serum IgG titer to Gs60 before challenge and the percentage of pneumonic tissue in the calves’ lungs after challenge. Similarly, Weldon et al (15) reported recognition of a partial Gs60 molecule (SW20C) in serum samples from animals that were either susceptible or resistant to experimental challenge with M. haemolytica and a significant correlation between the intensity of the response and protection. These results are also consistent with data reported by Lo and Mellors (11), who showed recognition of Gs60 by serum from calves resistant to pneumonic pasteurellosis after vaccination with Presponse. In trials B and C, as the IgG antibody titer to Gs60 increased, the percentage of pneumonic tissue decreased, implying that protection could be linked to anti-Gs60 IgG.
Because of the accepted role of Lkt in pathogenesis and protection, it was important to explore the relative roles of anti-Gs60 IgG and Lkt neutralizing antibodies in protection against pneumonia. Although there were significant correlations between either type of antibody and the percentage of pneumonic tissue, simultaneous inclusion of titers for both types of antigen in a general linear model revealed that the Lkt neutralizing antibody titers were more strongly associated with protection. However, further analysis of serum samples from trial C revealed a significant interaction between the prechallenge anti-Gs60 IgG antibody titer and the Lkt neutralizing antibody titer: in calves with low titers of Lkt neutralizing antibodies, higher titers of antibodies to Gs60 were associated with lower percentages of pneumonic tissue. Thus, in situations in which Lkt neutralizing antibody titers are low, IgG antibodies to Gs60 are especially important for protection.
The observations described here highlight again the complexity of protective immunity in M. haemolytica pneumonia, both bacterial virulence and resistance to pneumonia depending on the interplay of several factors. Nevertheless, these findings suggest that inclusion of Gs60 as an antigen in future vaccines against pneumonic pasteurellosis would be warranted.
Acknowledgments
We appreciate the generosity of DOW AgroSciences, Zionsville, Indiana, USA, in providing the mouse anti-Gs60 monoclonal antibody. This study was funded by the Natural Sciences and Engineering Research Council of Canada and the Ontario Cattlemen’s Association through the Agricultural Adaptation Council, with previous support also from the Ontario Ministry of Agriculture, Food and Rural Affairs.
References
- 1.Loneragan GH, Dargatz DA, Morley PS, Smith MA. Trends in mortality ratios among cattle in US feedlots. J Am Vet Med Assoc. 2001;219:1122–1127. doi: 10.2460/javma.2001.219.1122. [DOI] [PubMed] [Google Scholar]
- 2.DeRosa DC, Mechor GD, Staats JJ, Chengappa MM, Shryock TR. Comparison of Pasteurella spp. simultaneously isolated from nasal and transtracheal swabs from cattle with clinical signs of bovine respiratory disease. J Clin Microbiol. 2000;38:327–332. doi: 10.1128/jcm.38.1.327-332.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Rowe HA, Poxton IR, Donachie W. Survival of Mannheimia (Pasteurella) haemolytica in tracheobronchial washings of sheep and cattle. Vet Microbiol. 2001;81:305–314. doi: 10.1016/s0378-1135(01)00361-3. [DOI] [PubMed] [Google Scholar]
- 4.Rice JA, Carrasco-Medina L, Hodgins DC, Shewen PE. Mannheimia haemolytica and bovine respiratory disease. Anim Health Res Rev. 2007;8:117–128. doi: 10.1017/S1466252307001375. [DOI] [PubMed] [Google Scholar]
- 5.Shewen PE, Wilkie BN. Evidence for the Pasteurella haemolytica cytotoxin as a product of actively growing bacteria. Am J Vet Res. 1985;46:1212–1214. [PubMed] [Google Scholar]
- 6.Shewen PE, Wilkie BN. Pasteurella haemolytica cytotoxin neutralizing activity in sera from Ontario beef cattle. Can J Comp Med. 1983;47:497–498. [PMC free article] [PubMed] [Google Scholar]
- 7.Gentry MJ, Confer AW, Panciera RJ. Serum neutralization of cytotoxin from Pasteurella haemolytica, serotype 1 and resistance to experimental bovine pneumonic pasteurellosis. Vet Immunol Immunopathol. 1985;9:239–250. doi: 10.1016/0165-2427(85)90074-1. [DOI] [PubMed] [Google Scholar]
- 8.Davies RL, Whittam TS, Selander RK. Sequence diversity and molecular evolution of the leukotoxin (lktA) gene in bovine and ovine strains of Mannheimia (Pasteurella) haemolytica. J Bacteriol. 2001;183:1394–1404. doi: 10.1128/JB.183.4.1394-1404.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Shewen PE, Wilkie BN. Pasteurella haemolytica cytotoxin: Production by recognized serotypes and neutralization by type-specific rabbit antisera. Am J Vet Res. 1983;44:715–719. [PubMed] [Google Scholar]
- 10.Lainson FA, Murray J, Davies RC, Donachie W. Characterization of epitopes involved in the neutralization of Pasteurella haemolytica serotype A1 leukotoxin. Microbiology. 1996;142(Pt 9):2499–2507. doi: 10.1099/00221287-142-9-2499. [DOI] [PubMed] [Google Scholar]
- 11.Lo RYC, Mellors A. The isolation of recombinant plasmids expressing secreted antigens of Pasteurella haemolytica A1 and the characterization of an immunogenic 60 kDa antigen. Vet Microbiol. 1996;51:381–391. doi: 10.1016/0378-1135(96)00052-1. [DOI] [PubMed] [Google Scholar]
- 12.Madan Babu M, Sankaran K. DOLOP — database of bacterial lipoproteins. Bioinformatics. 2002;18:641–643. doi: 10.1093/bioinformatics/18.4.641. [DOI] [PubMed] [Google Scholar]
- 13.Juncker AS, Willenbrock H, Von Heijne G, Brunak S, Nielsen H, Krogh A. Prediction of lipoprotein signal peptides in Gram-negative bacteria. Protein Sci. 2003;12:1652–1662. doi: 10.1110/ps.0303703. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Narita S, Matsuyama S, Tokuda H. Lipoprotein trafficking in Escherichia coli. Arch Microbiol. 2004;182:1–6. doi: 10.1007/s00203-004-0682-4. [DOI] [PubMed] [Google Scholar]
- 15.Weldon SK, Mosier DA, Simons KR, Craven RC, Confer AW. Identification of a potentially important antigen of Pasteurella haemolytica. Vet Microbiol. 1994;40:283–291. doi: 10.1016/0378-1135(94)90117-1. [DOI] [PubMed] [Google Scholar]
- 16.Shewen PE, Wilkie BN. Vaccination of calves with leukotoxic culture supernatant from Pasteurella haemolytica. Can J Vet Res. 1988;52:30–36. [PMC free article] [PubMed] [Google Scholar]
- 17.Conlon JA, Shewen PE, Lo RYC. Efficacy of recombinant leukotoxin (rLKT) in protection against pneumonic challenge with live Pasteurella haemolytica A1. Infect Immun. 1991;59:587–591. doi: 10.1128/iai.59.2.587-591.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Bowland SL, Shewen PE. Bovine respiratory disease: Commercial vaccines currently available in Canada. Can Vet J. 2000;41:33–48. [PMC free article] [PubMed] [Google Scholar]
- 19.Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem. 1951;193:265–275. [PubMed] [Google Scholar]
- 20.Hillenkamp F, Karas M, Beavis RC, Chait BT. Matrix-assisted laser desorption/ionization mass spectrometry of biopolymers. Anal Chem. 1991;63:1193A–1203A. doi: 10.1021/ac00024a002. [DOI] [PubMed] [Google Scholar]
- 21.Leinikki PO, Passila S. Quantitative, semi-automated, enzyme-linked immunosorbent assay for viral antibodies. J Infect Dis. 1977;136(Suppl):S294–S299. doi: 10.1093/infdis/136.supplement_2.s294. [DOI] [PubMed] [Google Scholar]
- 22.Sacks JM, Gillette KG, Frank GH. Development and evaluation of an enzyme-linked immunosorbent assay for bovine antibody to Pasteurella haemolytica: Constructing an enzyme-linked immunosorbent assay titer. Am J Vet Res. 1988;49:38–41. [PubMed] [Google Scholar]
- 23.Conlon JA, Shewen PE. Clinical and serological evaluation of a Pasteurella haemolytica A1 capsular polysaccharide vaccine. Vaccine. 1993;11:767–772. doi: 10.1016/0264-410x(93)90263-w. [DOI] [PubMed] [Google Scholar]
- 24.Adlam C, Knights JM, Mugridge A, et al. Purification, characterization and immunological properties of the serotype-specific capsular polysaccharide of Pasteurella haemolytica (serotype A1) organisms. J Gen Microbiol. 1984;130:2415–2426. doi: 10.1099/00221287-130-9-2415. [DOI] [PubMed] [Google Scholar]
- 25.Jericho KW, Langford EV. Aerosol vaccination of calves with Pasteurella haemolytica against respiratory disease. Can J Comp Med. 1982;46:287–292. [PMC free article] [PubMed] [Google Scholar]