Veterinary Microbiology, 27 ( 1991 ) 49-62 Elsevier Science Publishers B.V., Amsterdam


Improved protection of swine from pleuropneumonia by vaccination with proteinase K-treated outer membrane of Actinobacillus

(Haemophilus) pleuropneumoniae Yu-Wei Chiang ~, Theresa F. Young, Vicki J. Rapp-Gabrielson 2 and Richard F. R o s s 3

Veterinary Medical Research Institute, College of Veterinary Medicine, Iowa State University Ames, IA 50011, USA (Accepted 27 August 1990)

ABSTRACT Chiang, Y.-W., Young, T.F., Rapp-Gabrielson, V.J. and Ross, R.F., 1991. Improved protection of swine from pleuropneumonia by vaccination with proteinase K-treated outer membrane of Act# nobacillus (Haemophilus) pleuropneumoniae. Vet. Microbiol., 27:49-62 The immunogenic and protective potentials of an outer membrane-enriched fraction (OM) from a serotype 5 strain of Actinobacillus (Haemophilus)pleuropneumoniae (APP) and the same OM degraded with proteinase K or periodate were evaluated in swine. Groups of pigs were vaccinated with two doses of OM, proteinase K-treated OM (P-OM), periodate-treated OM (PI-OM), or placebo vaccine and challenged intranasally with the homologous strain of APP. Results from triplicate experiments indicated that proteinase K treatment of OM resulted in an improved efficacy. This improved efficacy of P-OM vaccine over untreated OM vaccine was evidenced not only by less severe lung lesions in P-OM vaccinated pigs but also by significant reduction (P< 0.05 ) in the number of P-OM vaccinated pigs which developed lung lesions upon challenge with APP. Assessment of sera from vaccinated animals by immunoblotting, complement fixation test, or ELISA indicated that the immunogenicity of some but not all protein or carbohydrate components were reduced (or eliminated) by proteinase K and periodate treatments respectively.


Vaccination of pigs is one of the current methods used in attempts to control swine pleuropneumonia caused by Actinobacillus (Haemophilus) pleuro~Present address: Biological R&D, Boehringer Ingelheim Animal Health, Inc., St. Joseph, MO 64502, USA. 2Present address: Department of Veterinary Science, North Dakota State University, Fargo, ND 58105, USA. 3To whom correspondence should be addressed.


© 1991 - - Elsevier Science Publishers B.V.

pneumoniae (APP). Whole cell bacterins used in the field usually reduce the rate of mortality but fail to prevent infection and the development o[ lesions (Nielsen, 1976: Nielsen, 1979: Rosendal, 1981: Mason et al., 1982: Nielsen, 1984" Higgins et al., 1985: Kume et al., 1985, Nielsen, 1985). Evidence from several studies indicated that capsule (Jensen and Bertram, 1986: Inzana et al., 1988: Jacques et al., 1988), lipopolysaccharide (LPS: Fenwick et al., 1986c: Jensen and Bertram, 1986: Maudsley et al., 1986 ), and cytotoxins/hemolysins (Bendixen et al., 1981: Nakai et al., 1983: Martin et al., 1985: Kume et al., 1986; Rosendal et al., 1988) from APP are important in pathogenesis of pleuropneumonia. Immunization of swine with l:scherichia coil J5 (which contains cross-reacting LPS core antigens: Fenwick et al., 1986a,b) or passively with monospecific serum to capsule ( Inzana et al., 1988 ) protected pigs from lethal infection but not from development of pneumonia. Neutralizing antibodies for APP cytotoxin or hemolysin have been demonstrated in sera from convalescent pigs (Bendixen et al., 1981: Rosendal et al., 1988 and Frey et al., 1989) and from actively as well as passively immunized pigs ( Bosse et al., 1990). Evidence has been presented that vaccine prepared from purified type I APP hemolysin induced protection against mortality after APP challenge but not against development of lesions (van den Bosch et al., 1990). In a previous study (Rapp and Ross, 1988), pigs vaccinated with an outer membrane fraction (OM) from APP strain 200 (serotype 5 ) a d j u v a n t e d either with aluminum hydroxide or with incomplete Freund's adjuvant ( IEA ) had significantly less pneumonia compared to control pigs after intranasal challenge with the homologous strain. In addition, the efficacy of OM vaccine was comparable to that of whole cell bacterin adjuvanted with IFA. In an earlier study (Rapp and Ross, 1986 ), antibody reactions with protein components but not carbohydrate components of OM, detected by immunoblotting with convalescent serum, were eliminated when OM was treated with proteinase K prior to immunoblotting. In contrast, reactions with carbohydrate components but not protein components were eliminated when OM was treated with periodate prior to immunoblotting. These findings suggested that treatment with proteinase K or with periodate might have selectively eliminated protein components or carbohydrate components in OM. Accordingly, by using proteinase K or periodate treatment, this study was originally designed to determine which OM components (proteins or carbohydrates) may be involved in eliciting protection following vaccination. Findings from this study indicated that in contrast to our original assumption, proteinase K or periodate treatment did not completely eliminate the immunogenicity of protein components or carbohydrate components. However, our findings also indicated that, compared to untreated OM, treatment of OM with proteinase K resulted in an improved efficacy.




Bacterium A. pleuropneumoniae ISU strain 200 (APP 200) was used throughout this experimentation. APP forms two colony types (Rapp et al., 1985 ), adherent and smooth, which have the same outer membrane protein profiles as reported previously (Rapp et al., 1986). Because of greater ease in handling, the smooth type was used to prepare antigens whereas the adherent type, which was virulent for swine, was used to prepare challenge inoculum. APP strain 200 of smooth-colony type were inoculated in brain heart infusion broth (BHI; Difco Laboratories) containing NAD (40/lg/ml; Sigma Chemical Co. ), incubated for 6 h in a shaking water bath and harvested by centrifugation (Rapp and Ross, 1986 ). For use in inoculation of swine, adherent-colony type strain 200 was grown in M96 mycoplasma broth for 6 h and diluted in BHI to the desired concentration (Rapp and Ross, 1986).

Bacterial antigens An OM of APP was prepared (Rapp et al., 1986). Briefly, bacteria suspended in 10 m M HEPES buffer (pH 7.4) were sonicated and then centrifuged to remove intact cells and cell debris. The total membrane fraction was sedimented by ultracentrifugation, treated with 1% (w/v) sodium N-lauroyl sarcosinate (Sarkosyl; Sigma) to selectively solubilize the inner membrane and again ultracentrifuged to pellet OM. Protein concentration of OM was determined (Markwell et al., 1978 ), using bovine serum albumin as standard. Proteinase K-treated OM (P-OM) was prepared by adding proteinase K (Sigma) to OM in 10 mM HEPES buffer (3 mg proteinase K to 1 mg protein of OM) and incubating at 60 ° C for 60 min. Periodate-treated OM (PI-OM) was prepared by adding sodium metaperiodate to OM in 10 mM HEPES buffer to a final concentration of 0.05 M, incubating for 6 h at 4 ° C, and then dialyzing against 10 m M HEPES buffer at 4°C overnight with two changes of buffer. Lipopolysaccharide (LPS) was prepared by a modified method of phenolwater extraction (Westphal and Jann, 1965). Briefly, washed bacterial cells were subjected to hot phenol-water extraction followed by dialysis and lyophilization of the pooled aqueous phase. The dried material was suspended in 0.01 M Tris buffer (pH 7.2), treated with DNase I (10/~g/ml; Sigma) and RNase IIA ( 100/tg/ml; Sigma) for 6 h at 37°C, and centrifuged at low speed to remove insoluble material. The supernate was then centrifuged at 105 000 X g for 4 h at 4 ° C. The resulting pellet was resuspended in water and ultracentrifugation was repeated. The pelleted LPS was resuspended in water, lyophilized, and stored at - 70 ° C. Purified capsule isolated from APP 200 (Jensen and Bertram, 1986) was provided by Mr. A. Jensen, National Animal Disease Center, Ames, IA. An-

other purified capsule which was prepared from serotype 5 APP strain J45 ( Inzana, 1987 ) was provided through the courtesy of Dr. T. Inzana, Virginia Polytechnic Institute and State University, Blacksburg, VA.

Vaccination and experimental infection Cross-bred (Yorkshire-Hampshire-Landrace) male and female pigs were obtained at 9 to 10 weeks of age from the Iowa State University Animal Resource Station. The herd was established by surgical procurement and is barrier maintained. It was free of known respiratory pathogens including Actinobacillus spp. Pigs were randomly allocated to treatment groups on the basis of sex and litter of origin and housed in isolation rooms. Feed consisting of a 16% protein swine grower ration without antibiotics or other growth promotants was provided ad libitum. Two doses of OM or treated OM emulsified ( 1 : 1 ) in incomplete Freund's adjuvant (IFA: Difco) were administered subcutaneously in the neck at 3week intervals. Each dose of vaccine contained 500/~g of protein in 2 ml. Proteinase K-treated OM and PI-OM vaccines contained product derived from the same concentration of OM. Control pigs received only the buffer ( 10 mM HEPES containing 0.037% formalin and 0.01% merthiolate) which was emulsified in IFA. Two weeks after the second dose of vaccine, pigs were given approximately 10 7 CFU of APP in 5 ml of broth intranasally (Rapp and Ross, 1986). Rectal temperatures and character of respiration of pigs were recorded shortly before the challenge and daily afterwards. The individual doing the clinical assessment had no knowledge about the treatment each animal received. Pigs were necropsied 1 week after challenge. During necropsy, samples from the lung, pleura, pericardial fluid, heart blood, bronchial lymph node, liver, kidney, and spleen were collected with cotton tipped swabs for bacteriological examination. Pneumonic lesions on the ventral and dorsal surfaces of lungs were drawn proportionally on a lung sketch (Ross et al., 1984). The lungs were sliced in approximately 1 cm sections and examined for any gross lesions which were not evident on the surface. The total area of lung on the sketch and the total area of lesions were calculated using a digital image analyzer (MOP-3, Carl Zeiss Inc. ).

SDS-PAGE Discontinuous SDS-PAGE was done with a 3.8% stacking gel and a 10% separating gel (Rapp et al., 1986) in order to examine the outcome of proteinase K or periodate treatment of OM. Usually, two gels with identical arrangement of samples were run simultaneously. One gel was stained with Coomassie blue and another was silver-stained according to the procedures which preferentially stain LPS (Tsai and Frasch, 1982; Hitchcock and Brown, 1983).



Serum antibody response Serum samples were collected 1 day before first vaccination, 1 day before challenge, and on the day of necropsy. Serum antibody response to antigenic components of OM was evaluated by modification of an immunoblotting technique described by Rapp and Ross ( 1986 ). Briefly, components in OM were separated by SDS-PAGE and electrophoretically transferred to a nitrocellulose sheet. The nitrocellulose sheet was cut into strips which were then incubated for 1 h in Tris-saline Tween 20 (TST) buffer (0.01 M Tris hydrochloride, 0.15 M NaC1, and 0.05% Tween 20, pH 7.4). Subsequently, strips were incubated with test sera (diluted 1:200 in TST buffer) for 2 h followed by a 30-min wash in TST buffer. Strips were then incubated with peroxidaseconjugated rabbit anti-swine IgG (diluted 1 : 1000 in TST buffer; heavy- and light-chain specific, Cappel Laboratories) for 2 h. Following three 10-min washes in TST buffer, bands were visualized by reacting for 5-10 min with 0.0025% o-dianisidine (Sigma) and 0.01% H202 in Tris-saline buffer containing 0.01 M Tris hydrochloride, 0.15 M NaCI and 0.25% methanol (pH 7.4). Serum antibody titers to APP whole cell antigen were determined by complement fixation (CF) test (Schultz et al., 1982). Serum antibody response to APP LPS was measured by an ELISA procedure (Y.-W. Chiang, Ph.D. Thesis, Iowa State University, Ames, IA, 1986). Briefly, APP LPS (1 /tg/well) in 0.05 M carbonate buffer (pH 9.6 ) was adsorbed to microtitration plates overnight at 37 °C. Following removal of unadsorbed antigen, serum samples diluted 1:200 in phosphate buffered saline solution (PBS, pH 7.2) were added and incubated at 37°C for 20 rain. The plates were washed and incubated for 15 min at 37°C with peroxidase-conjugated rabbit anti-swine IgG diluted 1 : 1000 in PBS. Color development was initiated by adding peroxidase substrate solution (Kirkegaard & Perry Laboratories, Inc. ) and the plates were incubated at 37°C for 10 rain before the reaction was stopped by adding 0.5% hydrofluoric acid. Optical density (OD) was determined at 405 nm with an ELISA reader. Data analysis The differences in rectal temperature, character of respiration, and pneumonic lesions between pooled control and vaccinated groups or between different vaccinated groups were analyzed by an analysis of variance procedure (general linear model) with experiment as a blocking factor. The differences in the number of pigs developing pneumonia between pooled control and vaccinated groups or between different vaccinated groups were analyzed by a chisquare procedure. The paired t-test was used to analyze the differences in serum antibody response as measured by ELISA or by CF test. The statistically significant level was set at P < 0.05.


S D S - PAGE Coomassie blue-stained SDS-PAGE profiles of OM and P-OM (Fig. 1, lane G and D, respectively) indicated that the APP-associated protein components in OM were effectively digested by proteinase K. Proteinase K present in P-OM apparently was responsible for three of the bands shown in the profile of P-OM since these bands were also found in a control profile of proteinase K (Fig. 1, lane E). T r e a t m e n t of OM with periodate also affected protein components as indicated by the profile of PI-OM (Fig. 1, lane F). No discernible protein bands were visualized in the profiles of APP LPS (Fig. 1, lane A) and two APP capsule preparations (Fig. 1, lane B and C). Silverstained S D S - P A G E profiles of APP antigens revealed that components similar or identical to those of LPS were also present in OM. Those components were eliminated ( large molecular weight ) or greatly reduced (small molecular weight) by periodate treatment but not by proteinase K treatment (Fig.

¸¸¸7¸? ¸











Fig. 1. Coomassie blue-stained SDS-PAGE profiles of APP antigens. MW, molecular weight markers; lane A, LPS (501~g) from APP strain 200: lane B, capsule preparation (50 ug) from APP strain 200: lane C, capsule preparation (50/~g) from APP slrain J45, lane D, protcinase K-treated outer membrane fraction (90/zg) from APP strain 200; lane E, proteinase ( 100 ltg) lane F, periodate-treated outer membrane fraction (90/lg) from APP strain 200: lane G, untreated outer membrane fraction ( 90/zg ) from APP strain 200.



94 67

43 30

20.1 14.4 MW







Fig. 2. Silver-stainedSDS-PAGE profiles of APP antigens. The arrangement of samples is the same as described in the legendof Fig. 1. 2 ). Although proteins were also stained by silver nitrate (see molecular weight markers), the staining was rather weak and the color was distinctly different from that of carbohydrate. It is not clear whether OM or P-OM contained components similar to those of capsular antigens since the profiles of both capsule preparations appeared as weakly stained and diffuse smears at the very top of the gel (Fig. 2, lane B and C). The strongly silver-stained band in the profile of capsular preparation from APP strain J45 probably was due to the contaminating LPS components (Fig. 2, lane C).

Serum antibody response A representative immunoblot of OM with pre-vaccination and pre-challenge sera from two animals in each group is shown in Fig. 3. Protein components of OM were efficiently transferred to the nitrocellulose sheet as indicated by India ink staining (Fig. 3, lane 1 ). Enhanced antibody reactions occurred between OM protein components and pre-challenge sera from OM vaccinated animals (Fig. 3, lanes 6-9), P-OM vaccinated animals (Fig. 3, lanes 10-13), and PI-OM vaccinated animals (Fig. 3, lanes 1 4 - 1 7 ) b u t not from control animals (Fig. 3, lanes 2-5 ). However, the intensity of reactions between pre-challenge sera from P-OM and PI-OM vaccinated pigs and cer-









9 10 11 12 13 14 15 16 17

Fig. 3. Representative immunoblot of OM from APP strain 200 with pre-~accinanon and prochallenge sera from control pigs and pigs vaccinated with OM, P-OM, and PI-OM. Lane l. India mk stain: lane 2, pre-vaccination of control l : lane 3, pre-challenge of control l : lane 4, pre\ accination of control 2. lane 5, pre-challenge of control 2: lane 6. pre-~,accmation o1( )M \accinated 1: lane 7. pre-challcnge of OM vaccinated I. lane 8, pre-vaccination ol'OM ~accmatcd 2: lane 9. prc-challenge of OM vaccinated 2: lane 10, pre-vaccination of P-OM vaccinated 1: lane 11, pre-challenge of P-OM vaccinated 1: lanc 12, pre-vaccination of P-OM vaccinated 2: lanc 13, pre-challenge of P-OM vaccinated 2: lane 14, pre-vaccination of I>I-()M vaccinated 1: lane t 5, pre-challenge of PI-OM vaccinated 1: lane 16. pre-vaccination of PI-OM vaccinated 2: lane 17, pre-challenge of PI-OM vaccinated 2. The major outer membrane protems are indicated to the left of the immunoblot.

tain protein components of OM, as indicated by the intensity of staining of the bands, was obviously less than that from OM vaccinated pigs. Moreover, no reactions were detected between protein components with small molecular weights (between 29 000 and 38 500) and pre-challenge sera from P-OM vaccinated animals (Fig. 3, lanes 10-13 ). The CF serum antibody response to APP whole cell antigen increased greatly in both OM and P-OM vaccinated pigs 2 weeks after administration of the second dose of vaccine (Table 1 ). In contrast, 13 of 17 pigs vaccinated with PI-OM had no detectable CF antibodies after vaccination. Mean antibody titers dropped slightly for both OM and P-OM groups 1 week after challenge. In contrast, the CF antibody titers of the PI-OM vaccinated pigs increased rapidly by 1 week after challenge, although they were still significantly lower



TABLE 1 Serum antibody response ~to APP whole cell antigen measured by complement fixation test in control pigs and in pigs vaccinated with OM, P-OM, and PI-OM Treatment

Control OM vaccinated P-OM vaccinated PI-OM vaccinated

Antibody titer z (mean log2 titer +-SD )

(n = (n = (n = (n=

17) 16) 16 ) 17)




1.00 + 0 1.00_+ 0 1.00 +-0 1.00+-0

1.00 +-0 8.25 +- 1.29 a 8.62 +- 2.10 n 1.88+- 1.57 abe

1.29 + 0.85 7.56 -+ 1.21 a 7.94 +- 1.29 a 5.94+ 1.71 abe

~Results were pooled from three experiments. 2Titer is reciprocal of highest dilution of serum in which 30% or less hemolysis occurred. ~The indicated value is significantly different from that of corresponding control, P < 0.05. bThe indicated value is significantly different from that of corresponding OM vaccinated treatment P

Improved protection of swine from pleuropneumonia by vaccination with proteinase K-treated outer membrane of Actinobacillus (Haemophilus) pleuropneumoniae.

The immunogenic and protective potentials of an outer membrane-enriched fraction (OM) from a serotype 5 strain of Actinobacillus (Haemophilus) pleurop...
901KB Sizes 0 Downloads 0 Views