Actinobacillus pleuropneumoniae for Use in Live Vaccines

Vol. 61, No. 5

INFECrION AND IMMUNrIY, May 1993, p. 1682-1686

0019-9567/93/051682-05$02.00/O

Copyright © 1993, American Society for Microbiology

Safety, Stability, and Efficacy of Noncapsulated Mutants of Actinobacillus pleuropneumoniae for Use in Live Vaccines THOMAS J. INZANA,* JEANNE TODD, AND HUGO P. VEIT

Veterinary Microbiology Research Laboratories, Virginia-Maryland Regional College of Veterinary Medicine, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061 Received 2 November 1992/Accepted 29 January 1993

Clonal, noniridescent mutants of ActinobaciUlus pleuropneumoniae serotypes 1 and 5 were isolated following chemical mutagenesis with ethyl methanesulfonate. The absence of any detectable capsule was confirmed by inhibition radioimmunoassay. There were no differences between the parent and mutant strains in lipopolysaccharide or protein electrophoretic profiles or in hemolytic activity. There was no detectable reversion to the encapsulated phenotype in vitro after passage in mice or pigs or in microporous capsules that were implanted subcutaneously in pigs for 6 weeks. The mutants were able to survive for more than 1 week in pigs following subcutaneous inoculation, which resulted in a strong immune response to whole cells and Apx toxins I and II. Intratracheal challenge of pigs with the serotype 5 mutant at a dose 1 log greater than the 50% lethal dose for the parent resulted in no clinical disease or lesions except in one pig that had slight pneumonia and pleuritis. Twenty-four hours after challenge, A. pleuropneumoniae could not be recovered from the respiratory tracts of any of the challenged pigs except for the one infected pig; this isolate remained noncapsulated. Immunization of pigs with one or both serotypes of noncapsulated mutants protected all pigs against clinical disease following intratracheal challenge with the virulent homologous or heterologous serotype. Nonimmunized control pigs and pigs immunized with a commercial bacterin died or had to be euthanized within 24 h of challenge. Thus, live noncapsulated mutants of A. pleuropneumoniae may provide safe and cost-effective protection against swine pleuropneumonia. These observations support the possibility that noncapsulated mutants of other encapsulated, toxin-producing bacteria may also prove to be efficacious live-vaccine candidates.

by homologous and heterologous serotypes seems to occur following natural infection with A. pleuropneumoniae (17). Our objective was to develop live, stable, attenuated strains of A. pleuropneumoniae that would be safe to use in nonimmune pigs, would provide improved protection against clinical disease and development of lesions, and would be practical for commercial development.

Actinobacillus (Haemophilus) pleuropneumoniae is the etiologic agent of porcine pleuropneumonia, which is responsible for large economic losses worldwide in the swine industry. Swine infected with A. pleuropneumoniae may be asymptomatic or may present with chronic to acute disease. Acute disease in a nonimmune herd is often associated with high mortality. In contrast, chronic disease may be less obvious but still responsible for substantial economic losses due to poor weight gain and failure to thrive. Introduction of asymptomatic carriers into a nonimmune herd is an important cause of pleuropneumonia outbreaks (22). Acute pleuropneumonia is characterized by fibrinous adhesions in the pleural cavity and necrotic and hemorrhagic lesions in the lungs (19). Chronic lesions involve considerable fibrosis and purulent inflammation, while asymptomatic infections have few gross lesions other than convalescent chronic lesions (i.e., fibrosis with pleural adhesions). The bacterial factors that are responsible for virulence in A. pleuropneumoniae include capsular polysaccharide, lipopolysaccharide (LPS) or endotoxin, and the hemolytic and cytotoxic RTX (Apx) toxins (6). Heat-labile Apx toxins are thought to be particularly important in the development of severe lesions and for induction of protective immunity in animals; nonhemolytic mutants are avirulent and cannot induce protective immunity in animals (11). In addition, commercial vaccines consisting of killed bacteria that lack the Apx toxins do not provide adequate protection (5, 16, 18). Protection by antibody to capsule or LPS is also inadequate, but the capsule is a required virulence factor and prevents opsonization and clearance of bacteria in the lung by the host (8). Optimal protection of pigs against infection

*

MATERLILS AND METHODS Bacterial strains and mutagenesis. The bacterial strains used in this study were A. pleuropneumoniae J45 serotype 5, A. pleuropneumoniae 4074 serotype 1, and the noncapsulated mutants of these strains. The source of the parent strains and growth conditions have been described previously (9). Bacteria were treated with ethyl methanesulfonic acid by modification of a procedure described by Murchison et al. (15). Briefly, bacteria in brain heart infusion broth containing NAD (BHI-N) were grown to mid-log phase, 0.015 ml of ethyl methanesulfonic acid per ml of culture was added, and the bacteria were incubated for an additional 3 h. The bacteria were then washed in 0.01 M phosphate-buffered saline (PBS), pH 7.4, and resuspended in BHI-N. Undiluted and diluted suspensions of the mutagenized bacteria were spread on BHI-N plates and incubated at 37°C for 24 to 48 h. Clonal isolates of noniridescent colonies that were confirmed to beA. pleuropneumoniae (by hemolytic activity and failure to grow on blood agar) were obtained in pure culture and further characterized. Phenotypic characterization and stability. Total capsule content was determined by inhibition radioimmunoassay of mid-log-phase bacterial cultures (including bacteria and medium) as described elsewhere (7). Electrophoretic profiles of LPS (one dimension) and total proteins (one and two dimen-

Corresponding author. 1682

VOL. 61, 1993

NONCAPSULATED A. PLEUROPNEUMONIAE LIVE VACCINES

sions) were evaluated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) followed by silver staining (11) or immunoblotting (13). Hemolytic activity was determined by a quantitative hemolysis assay using rabbit erythrocytes (14). Quantitative determination of Apx toxin II from culture supernatants of strains J45 and 4074 and their noncapsulated mutants was accomplished by immunoblotting of culture supernatants with monoclonal antibody 8C2 specific for Apx toxin II (14). To determine the stability of the noncapsulated phenotype in vitro, strain J45-C was grown to 109 CFU/ml (determined spectrophotometrically), and 1 ml was centrifuged at 10,000 x g for 15 min, suspended in 25 RI of PBS, and tested for extracellular capsule by inhibition radioimmunoassay (7). In addition, noncapsulated mutants were passaged in vitro on BHI-N and screened for iridescence and capsule production by latex agglutination (10). Briefly, 100 ml of each noncapsulated mutant was grown to 2 x 109 CFU/ml, harvested, and suspended in 1 ml of PBS, and 25 RI (5 x 109 CFU) was tested for capsule by latex agglutination. In addition, the noncapsulated mutants were recovered from mice and pigs, which were challenged and immunized with the mutants to determine whether reversion occurred in vivo. Virulence and immunoprotection. Pigs 6 to 8 weeks of age were obtained from a local commercial herd free of clinical pleuropneumonia. Sera from sows and piglets used in the study were assayed for antibodies to serotype 1 and 5 capsules by radioimmunoassay (7) and for antibodies to Apx toxins I and II by enzyme-linked immunosorbent assay (ELISA) (13). All pigs used in this study had antibody titers for these antigens less than or within the negative range (7, 13). Neutralizing hemolytic antibody titers were kindly determined by Bradley Fenwick, Kansas State University, Manhattan. Several variables were examined to determine the optimum dose, route, and method of immunization of the noncapsulated mutants. Bacteria were administered to pigs subcutaneously, intradermally, or intranasally for one or two immunizations. Bacteria (2 x 109 CFU in 5 ml) were suspended in 7% mucin, microagarose beads, Freund's incomplete adjuvant, or PBS for subcutaneous and intradermal immunizations, and 108 CFU were suspended in 2.5 ml of PBS per nostril for intranasal immunizations. A. pleuropneumoniae was isolated from the immunization sites on a selective medium, and the amount of growth was scored as 0 (no A. pleuropneumoniae isolated), 1 + (< 10 colonies), 2+ (between 10 and 50 colonies), 3+ (between 51 and 100 colonies), or 4+ (> 100 colonies). The time the bacteria could survive following immunization at each site and in each medium and the immune response were determined. In addition, some pigs were immunized subcutaneously with mutants in capsule implants (Filtertek, Inc., Hebron, Ill.) containing approximately 4 x 107 CFU in 35 ,u of PBS. For all challenge experiments, the bacteria were grown in supplemented Columbia broth to mid-log phase, centrifuged once, and suspended to 109 CFU/ml in PBS. Tenfold dilutions of the bacterial suspension were made prior to challenge. Pigs were challenged intratracheally with 5 ml of bacterial suspension (5 x 107 CFU total) following mild sedation with Stresnil (Pittman-Moore, Inc., Washington Crossing, N.J.). This dose is 10 times the 50% lethal dose (LD50) in pigs, which was determined in a large separate study using 40 pigs (data not shown). Control pigs were immunized subcutaneously with bacterial diluent, 2 x 109 CFU of the live parent, or a commercial A. pleuropneumoniae killed vaccine (bacterin) containing a standardized

1683

suspension of serotypes 1, 5, and 7 with aluminum hydroxide as adjuvant. The pigs were immunized subcutaneously with 2 ml of the commercial vaccine at the same time and for the same number of doses as the pigs receiving the live vaccines. Pigs were challenged 3 weeks after the final immunization as described above. Serum for specific antibody determination was obtained prior to immunization and again immediately prior to challenge. Antibody titers were determined for noncapsulated whole cells by ELISA (9), for Apx toxins I and II by indirect ELISA (13), and for capsule by ELISA (9) or radioimmunoassay (7). Pigs were necropsied as soon as possible after death or immediately after euthanasia with sodium pentobarbital. Lung lesions were scored using the following criteria: 0, healthy lung (no gross lesions noted); 1 +, < 10% lung involvement with localized congestion only; 2+, 25% lung involvement with consolidation or congestion only and no necrosis or pleural lesions; 3+, 50% lung involvement with some consolidation, congestion, mild pleural lesions, and congestion; 4+, 75 to 100% lung involvement with pneumonia and consolidation, extensive pleural adhesions, and necrotic lesions. Statistics. The two-tailed P value for significance was determined by Fisher's exact test for 2-by-2 contingency tables. RESULTS characterization and stability. Noncapsulated Phenotypic mutants of strain J45 serotype 5 and strain 4074 serotype 1 that contained less than 4 ng of capsule per 109 CFU (limit of detection of the radioimmunoassay) were selected. The mutants (>5 x 1010 CFU/ml) were also not capable of agglutinating latex beads covalently conjugated to affinitypurified antibody to the homologous capsule (sensitivity, 10 ng/ml, or about 2.5 x 104 CFU). In contrast, strain J45 contained 130 ,ug and strain 4074 contained 165 ,ug of capsule per 109 CFU. The one-dimensional electrophoretic profiles of total membrane proteins and LPS for strains J45 and J45-C were identical; the two-dimensional protein electrophoretic profiles for the parent and mutant were very similar, and the few differences noted in the minor proteins were not consistent between preparations (similar electrophoretic profiles of strain J45 are shown in reference 11). The hemolytic activities of the parents and mutants were identical (50% hemolytic titer, 1:512). In addition, culture supernatants of the parents and mutants each contained a single band of equivalent intensity when analyzed by SDS-PAGE. In immunoblots, monoclonal antibody 8C2, which is specific for Apx toxin II, reacted equally well with culture supernatants from the mutants and the parents (data not shown). A suspension of 109 CFU of the noncapsulated mutants was nonreactive by inhibition radioimmunoassay. Repeated culture of colonies of the mutants remained noniridescent, and suspensions of 2 x 1011 CFU/ml did not agglutinate latex beads coated with affinity-purified antibody to capsule, indicating that the mutations were stable. Nonetheless, since in vitro tests for reversion are limited in sensitivity, mice and pigs were challenged with the noncapsulated mutants. The LD50 of the noncapsulated mutants is known to be similar to the LD50 of the parents, probably because the mice succumb to the lethal effects of the toxins (8). However, encapsulated revertants have never been isolated from mice lethally challenged with the noncapsulated mutants described here (8; data not shown). In most cases, the noncapsulated mutants could not be recovered from the lungs of pigs

1684

INFEcr. IMMUN.

INZANA ET AL.

bacteria recovered were still noncapsulated. In all other pigs challenged intranasally, the bacteria could not be recovered from the lungs, even 24 h after challenge (Table 1). However, the bacteria were capable of surviving for at least 2 to 3 weeks subcutaneously or intradermally in PBS, Freund's incomplete adjuvant, 7% mucin, or agarose beads (Table 2). Intradermal immunizations resulted in more severe inflammation and necrosis at the injection site than occurred with subcutaneous immunizations. The strongest immune response to the Apx toxins and the greatest protection against challenge occurred when the pigs were immunized subcutaneously with J45-C in PBS or Freund's incomplete adjuvant (Table 2). The immune response to whole cells was similar to or greater than the response to the Apx toxins. Neutralizing hemolysin titers paralleled the hemolysin ELISA titers. Pigs with less than an eightfold increase in hemolysin ELISA titer did not have significant neutralizing titers, whereas pigs with an eightfold or greater increase in titer did have neutralizing titers (data not shown). The optimum protective dose was 2 x 109 CFU/ml in PBS given in two immunizations 2 to 3 weeks apart. No adjuvant was required to enhance protective immunity. Each immunization site examined either was normal or, at the highest doses, had a small abscess noticeable only following incision (Table 3). Strains J45-C and 4074-C were highly effective in protectpigs against a severe intratracheal challenge dose (5 x ini 10 CFU) of the virulent homologous or heterologous serotype. Twenty-four hours after challenge, all of the pigs immunized with the live noncapsulated mutants or the parents were either clinically healthy or slightly lethargic; by 48 h after challenge, all of these pigs were clinically healthy. In contrast, a bacterin vaccine provided no protection against this challenge dose: all challenged pigs injected with the bacterin died or were euthanized because of severe necrotizing pleuropneumonia within 36 h. The vaccine was significantly more protective against mortality (P < 0.00001) and development of lesions (P = 0.00215; calculated as presence or absence of lesions, not as severity of lesions) than was the bacterin or PBS in control pigs. Protection

TABLE 1. Virulence of noncapsulated A. pleuropneumoniae J45C for pigs challenged intranasally and intratracheally No. positive/total no. tested

Challenge route

Postchallenge time (days)a

Mortality

Intratracheal Intranasalb Intranasalb Intranasalb

5 3 7 1

0/3 0/2 0/2 0/3

Recovery of

Lung

lesions

A. pleuroA luo pneumoniae

0/3 0/2 12C 0/3

0/3 0/2 1/2C 0/3

Number of days from challenge to necropsy. Nasal swabs and tonsil, trachea, and lung samples were obtained from all pigs challenged intranasally. c One pig had slight pneumonia and pleuritis, and noncapsulated A. pleuropneumoniae was recovered from its lungs. a

b

challenged intranasally or intratracheally 24 h after infection. However, A. pleuropneumoniae was recovered from the lungs of one pig challenged intranasally with J45-C, and the mutant was still noncapsulated (Table 1). None of the mutants recovered from immunizations of the pigs ever reverted to the encapsulated form, even though 4+ growth was usually recovered from the immunization sites. In addition, microporous capsules containing J45-C were implanted subcutaneously in pigs and recovered after 6 weeks; the cultured bacteria were still viable, and there was no evidence of any reversion to the encapsulated phenotype, as determined by iridescence and inhibition radioimmunoassay of pooled colonies (Table 2). Virulence and immunoprotection. Noncapsulated mutant J45-C was avirulent and caused no mortality in 10 pigs by intranasal exposure or intratracheal injection at a dose 20 or 10 times greater, respectively, than the LD50 of the parent. The three pigs challenged intratracheally were necropsied 5 days postchallenge; all the pigs' lungs were healthy, and A. pleuropneumoniae was not recovered. The pigs challenged intranasally were necropsied 1, 3, and 7 days postchallenge; only one pig examined 7 days after intranasal challenge had pneumonia (about 35% of the total lung mass), but the

TABLE 2. Survival, immunogenicity, and protective efficacy of J45-C in various suspending media Apx toxin titer

Mediuma

PBS

Incubation conditions or site

Survival timeb

In vitro Intradermal

>4 days >7 days 7 days >4 days >7 days >7 days >4 days >7 days >7 days >4 days >6 wk

Intranasale FIA 7% Mucin Microagarose beads Capsule implants

Subcutaneous In vitro Intradermal Subcutaneous In vitro Intradermal Subcutaneous Subcutaneous Subcutaneous

No. positive/total no.

Prechallenge

Postchallenge

Mortality

Pneumonia

score Lesion rangec

80-160

>1,280

0/3

1/3

0-1

80-160

>1,280

0/3

2/3

0-2

80-160 160 160

>1,280 320 320-640

0/9 3/3 0/2

5/9 3/3 2/2

1-4 2-4 2-4

a FIA, Freund's incomplete adjuvant. b Some pigs were necropsied 1, 3, and 7 days after immunization, and the immunization sites were cultured for bacteria. A. pleuropneumoniae was isolated in 3+ or 4+ quantity from each site. Pigs immunized subcutaneously were challenged 2 weeks after a second immunization. c See Materials and Methods for definition of lesion scores. d Moderate to severe pneumonia determined by gross necropsy examination. Bacteria were recovered from all pneumonic lungs. Pigs with slight pneumonia were clinically normal. e Of seven pigs immunized intranasally, bacteria were recovered from the lungs of one pig 7 days after immunization (same pig as in Table 1) but not from the lungs or trachea of any pigs 1 or 3 days after immunization.

VOL. 61, 1993

NONCAPSULATED A. PLEUROPNEUMONAE LIVE VACCINES

1685

TABLE 3. Protective efficacy of J45-C in PBS in relation to dose and number of subcutaneous immunizations Dose

(CFU) 8 x 4x 2x 5 x 5 x 1 x 1x 2x

107 108 109 108 108 109 109

109

No. of immunizations'

Mortality (no. dead/

1 1 1 1 2 1 2 2

1/1 1/1 0/1 3/3 0/3 3/3 0/3 0/3

no.

Immunization

Lesion score

siteb

treated)

or

Normal Normal Small abscessd ND ND ND ND Small abscessd

rangec

3+ 3+ 1+ 3+ 3+ 2-3+ 2-4+ 0_3+e

Fold rise in titer Apx toxin Cells

None 2 8 2 4 None 8 .16

8 4 >8 4-8 4-8 2 8-32 .16

a Pigs given one immunization were challenged 4 weeks later. Pigs given two immunizations were challenged 2 weeks after the last immunization. Subcutaneous immunization site was examined at necropsy and evaluated for the extent of abscess formation. ND, not determined. c Described as the percentage of lung area with pneumonic lesions. d 12,800 >12,800 .6,400

>12,800 .3,200 >3,200 >3,200 .6,400