TEXT

Top Abstract Text References

The gram-negative bacterium Actinobacillus pleuropneumoniae causes porcine pleuropneumonia, a highly contagious infection of pigs that is responsible for substantial economic losses worldwide. First described in 1957, A. pleuropneumoniae infection has become more frequent in recent years, particularly where animals are subjected to intensive breeding conditions (29, 33). Infection is highly host specific, and organisms spread from one animal to another via respiratory droplets (27, 33). The disease is characterized by increasingly severe pulmonary distress leading to death after 24 to 48 h or to chronic infection resulting in a failure to thrive (14, 33). No effective vaccine strategy has yet been devised to prevent infection. In the process of elucidating the sequence of critical interactions between organism and host that lead to severe disease, capsule, lipopolysaccharide, RTX-like toxins (ApxI to ApxIVA), lipoproteins, secreted proteases, and superoxide dismutase (SOD) have been identified as potential determinants of virulence (9, 12, 18, 26, 28, 32). This defines at once the potential of purified gene products or knockout mutant A. pleuropneumoniae strains for use as subunit or live-attenuated vaccines, respectively, and the recent development of genetic systems for A. pleuropneumoniae (15, 25, 37, 40) now allows these possibilities to be formally investigated.

The gene sodC, which encodes the copper- and zinc-cofactored SOD ([Cu,Zn]-SOD) of A. pleuropneumoniae, was previously identified, cloned, and sequenced in this laboratory (17, 18). SODs are metalloproteins that catalyze the dismutation of the highly reactive superoxide radical anion to hydrogen peroxide and oxygen (22). This is the first step in a series of reactions which prevent the accumulation of cytotoxic oxygen free radicals and nitrogen free radicals generated by the reduction of molecular oxygen. SODs are classified according to the metalloprotein at their active site, and three major forms have been identified in bacteria. Iron- and manganese-cofactored SODs are located in the cytoplasm and serve to remove superoxide anions generated during the course of aerobic respiration (10). In contrast, in all cases where location has been determined experimentally, bacterial [Cu,Zn]-SODs have been shown to be periplasmic (18; reference 31 and references therein). As superoxide is unable to cross the cytoplasmic membrane (13), it has been suggested that [Cu,Zn]-SOD provides protection from superoxide generated outside the cell. During the course of infection, many bacterial pathogens are exposed to high levels of activated oxygen species generated by inflammatory cells during the respiratory burst. The theoretical capacity of bacterial [Cu,Zn]-SODs to dismutate exogenous superoxide suggests a role for these enzymes in pathogenesis. In support of this hypothesis, Neisseria meningitidis and Salmonella enterica serovar Typhimurium strains deficient in [Cu,Zn]-SOD production are attenuated in the mouse model of infection (5, 8, 41). However, the picture is not clear-cut. For example, two groups have presented conflicting evidence for the role of [Cu,Zn]-SOD in the growth and multiplication of Brucella abortus in BALB/c mice (19, 38). To investigate the role of [Cu,Zn]-SOD in the pathogenesis of A. pleuropneumoniae infection and to assess the live-vaccine potential of a [Cu,Zn]-SOD-deficient mutant, we have constructed a sodC mutant of A. pleuropneumoniae strain 4074 by allele exchange. We investigated the impact of this mutation on virulence in experimental infections of the natural host, the pig.

Construction of A. pleuropneumoniae sodC mutant. The cloned sodC gene was interrupted by the insertion of a kanamycin resistance cassette (Kan^r) at the unique EcoNI site located 144 bp downstream from the sodC initiation codon. Briefly, a 3.4-kb EcoRI-EcoRV fragment containing A. pleuropneumoniae sodC was excised from pJSK150 (18) and cloned into pBluescript to generate plasmid pJSK155. A 1.3-kb HincII fragment containing Kan^r from plasmid pB16 (pB16 is a precursor to pB51 [35] and contains the Tn903 Kan^r gene flanked on either side by a polylinker containing various restriction sites including the HincII site) was then cloned into pJSK155 that had been linearized with EcoNI and Klenow treated to generate blunt ends. The resultant plasmid, designated pJSK333, was transferred into A. pleuropneumoniae strain 4074 (serotype 1) by electroporation (15, 24). As the ColE1 origin of plasmid pJSK333 does not function in A. pleuropneumoniae, kanamycin-resistant transformants can only arise following recombination between the mutant sodC allele and wild-type sodC sequences by single or double crossover. After 48 h of growth, three kanamycin-resistant transformants were isolated on selective brain heart infusion plates supplemented with 10% Levinthal's base (1). Southern blot and PCR analysis with sodC-specific oligonucleotide primers demonstrated that in all cases the mutant sodC::Kan allele had replaced the wild-type sodC gene on the chromosome in a double crossover event (data not shown). As sodC is expressed as a monocistronic unit (21), insertion of the antibiotic cassette would not be expected to exert a polar effect on the activity of neighboring genes. One isolate, SK564, was used in subsequent experiments.

SOD expression in wild type and sodC mutants of A. pleuropneumoniae. To verify that [Cu,Zn]-SOD production was abrogated in strain SK564, wild-type and SK564 mutant strains were grown aerobically in liquid culture until the stationary phase of growth. Whole-cell sonicates were separated electrophoretically on nondenaturing gels and stained for SOD activity as described previously (18). Two bands of SOD activity were observed when wild-type A. pleuropneumoniae was grown under nutrient-replete, aerobic conditions (Fig. 1) (17). Mn-SOD activity, manifested as a diffuse, achromatic zone at a position of relatively high mobility, was present in both wild-type and SK564 mutant extracts (Fig. 1, lanes 1 and 2). In contrast, [Cu,Zn]-SOD activity was present only in wild-type extracts (the prominent achromatic band in Fig. 1, lane 1) and was absent from extracts derived from the sodC::Kan strain, SK564 (Fig. 1, lane 2).

View larger version (83K): [in this window] [in a new window]   FIG. 1.   SOD activity of whole-cell sonicates of A. pleuropneumoniae strains 4074 (wild type, lane 1) and SK564 (sodC::Kan, lane 2). Each lane contained approximately 40 µg of total protein. [Cu,Zn]-SOD activity in the wild-type extract is indicated by the arrow. The more diffuse Mn-SOD activity, present in both lanes, is indicated by the dashed line.

Effect of sodC mutation on susceptibility to exogenous superoxide in vitro. The A. pleuropneumoniae wild type and sodC mutant strain SK564 were exposed to exogenous superoxide generated by the oxidation of xanthine as described previously (8). Production of superoxide in solution was initiated by the addition of 0.1 U of xanthine oxidase/ml, and bacterial survival was monitored by estimation of viable counts at hourly intervals over a 3-h period. The viability of wild-type A. pleuropneumoniae was unaffected by superoxide in vitro (100% survival or greater) (Fig. 2). In contrast, sodC mutant bacteria were highly susceptible to the microbicidal action of superoxide; 99% of mutant bacteria were killed after 3 h of incubation in the presence of xanthine oxidase (Fig. 2). Thus, the A. pleuropneumoniae [Cu,Zn]-SOD plays a crucial role in the detoxification of exogenously generated superoxide, allowing bacteria to survive challenge with superoxide in vitro. These results are comparable to those obtained with other organisms; Salmonella serovar Typhimurium and N. meningitidis mutants lacking [Cu,Zn]-SOD were strikingly more susceptible than otherwise isogenic wild-type bacteria to superoxide generated in solution by the xanthine-xanthine oxidase system (5, 8, 41).

View larger version (14K): [in this window] [in a new window]   FIG. 2.   Effect of the sodC mutation on survival of A. pleuropneumoniae exposed to superoxide in vitro. Survival of wild type bacteria (squares) and sodC mutant bacteria (circles) in phosphate-buffered saline with 0.1 mM xanthine (open symbols) and in phosphate-buffered saline with 0.1 mM xanthine and 0.1 U of xanthine oxidase/ml (filled symbols) is shown. Data (means and ranges for duplicate experiments) are percentages of the starting cell density (range, 1 × 108 to 2.5 × 108 CFU/ml).

Virulence of SodC-defective A. pleuropneumoniae. In the cases of Salmonella serovar Typhimurium and N. meningitidis sodC mutants, susceptibility to exogenous superoxide in vitro correlates with susceptibility to killing by superoxide-mediated host defense mechanisms and to attenuation of virulence in mouse models of infection (5, 8, 41). As challenge with A. pleuropneumoniae is known to elicit a vigorous respiratory burst of oxygen free radical production from porcine neutrophils (7), we investigated the effect of sodC mutation on the course of A. pleuropneumoniae infection in the natural (porcine) host. Two groups of seven 4-week-old, conventionally weaned piglets determined to be A. pleuropneumoniae free by bacteriological and serological analyses were inoculated intratracheally (30) with a range of doses of either strain 4074 (group B) or strain SK564 (group A), as detailed in Table 1. No significant difference was seen at any dose. Both animals receiving the high dose (ca. 2 × 10⁷ CFU) and one of two animals given the intermediate dose (ca. 2 × 10⁵ CFU) but no animal given the low dose (ca. 4 × 10³ CFU) in group A or B developed clinical signs of respiratory disease within 36 h (Table 1). Surviving animals were euthanized after 42 h.