Fully Virulent Bacillus anthracis Does Not Require the Immunodominant Protein BclA for Pathogenesis

Abstract

The BclA protein is the immunodominant epitope on the surface of Bacillus anthracis spores; however, its roles in pathogenesis are unclear. We constructed a BclA deletion mutant (bclA) of the fully virulent Ames strain. This derivative retained full virulence in several small-animal models of infection despite the bclA deletion.

Bacillus anthracis is the causative agent of anthrax (6, 9, 18). The spore of B. anthracis is the infectious particle for all forms of the disease, including inhalational anthrax (6, 9, 18). There is significant support for the concept that spore antigens of B. anthracis contribute to protective immunity (1, 3, 5-7, 12, 15, 16, 28, 29).

The exosporium, the outermost spore structure, is a loose-fitting layer that envelops the spore (11). Approximately 20 exosporium-associated protein and glycoprotein species have been identified from analyses of B. anthracis and Bacillus cereus (2, 4, 21, 23-27). The exosporium membrane projects “hair-like” fibers (11), of which the major component is the BclA glycoprotein (23, 25). Upon identifying the immunodominant BclA protein, Sylvestre et al. observed no appreciable differences in virulence between the nonencapsulated toxinogenic Sterne vaccine strain and a bclA mutant derivative when spores were administered to mice subcutaneously (25). To further characterize the role of the BclA protein in B. anthracis pathogenicity, the bclA gene was deleted by allelic exchange and replaced with the Ω kan-2 fragment (20) from the chromosome of the fully virulent Ames strain of B. anthracis. The deletion was confirmed by PCR and Western blotting analyses, and as demonstrated by transmission electron and immunofluorescence microscopy, the Ames bclA mutant did not express the hair-like fibers projecting from the exosporium membrane present on wild-type Ames spores (data not shown).

We analyzed differences in the ability of spores of the bclA mutant to germinate, as it was previously shown that the exosporium contains enzymes that may affect spore germination (14, 21, 23, 27). To monitor both early and later germination events, we assayed the increase in fluorescence and decrease in absorbance of spores with times of incubation in media that induce germination (19, 30). We found no significant difference (P > 0.05) in germination between the strains by either assay (Fig. 1A and B).

FIG. 1.

Effect of the BclA protein on the germination rates of B. anthracis spores in vitro. (A) Microtiter spectrofluorometric assay of germination of spores in the presence of a germination medium with alanine, adenosine, and Casamino Acids (AAC), as described previously (30). The germination of spores of the wild-type Ames strain (○) and bclA strain (Δ) was monitored every minute for 1 h. The data (percent increase in relative fluorescence units [RFU]) are the relative increase in RFU at a given time point compared to the RFU at time zero, multiplied by 100. Representative data are presented, and similar results were obtained in at least two additional experiments. (B) Germination was also measured by absorbance readings as previously described (19). The A₆₀₀ of each sample was measured at various times and is plotted as the percentage of the initial A₆₀₀ [A₆₀₀ (init)] at time zero that is represented by the A₆₀₀ at a given time point [A₆₀₀ (t)]. Absorbance data are from three independent experiments, and standard errors of the means are depicted.

We examined the potential role of the BclA protein in pathogenesis using guinea pig intramuscular (i.m.) (13) and mouse intranasal (i.n.) (17) infection models. Guinea pigs were challenged i.m. with either approximately 980 spores of wild-type Ames or 860 spores of the bclA mutant. All guinea pigs succumbed to the infection by the second day after challenge (Fig. 2A). We also employed a more sensitive assay that utilizes in vivo competition in the guinea pig model. Guinea pigs were coinfected i.m. with approximately 1,000 spores. The spores delivered i.m. were an approximately equal mixture of wild-type Ames (57%) and bclA mutant (43%) spores. Two days later, moribund animals were euthanized and their spleens were harvested. Levels of bacterial load were determined within the spleens. Wild-type Ames and bclA bacteria were recovered from the spleen as measured by bacterial counts from Luria-Bertani (LB) agar plates (total number of bacteria) or LB agar plates containing kanamycin (number of bclA mutant bacteria only). Nearly identical relative percent recoveries were obtained when the recovered percent of splenic bacterial CFU was divided by the percent of spores in the challenge inoculum, suggesting that spores from both strains germinated, survived, and replicated at the same rate in vivo (Fig. 2B).

FIG. 2.

Effect of the BclA protein on virulence in the guinea pig infection model. (A) Female Hartley guinea pigs were challenged i.m. with spores of the Ames wild-type strain (○) or spores of the bclA strain (Δ). (B) The effect of the bclA mutation on in vivo fitness in a competitive guinea pig infection model. Guinea pigs were coinfected i.m. with spores of both the Ames wild-type strain and the bclA strain. The error bars represent standard errors of the means. Wt, wild type.

We also used a mouse model of i.n. infection in an attempt to identify differences in virulence associated with the bclA mutation. BALB/c mice were challenged i.n. with either 9.25 × 10⁵ spores of the Ames wild-type strain or 7.85 × 10⁵ spores of the bclA mutant. There were no statistically significant differences observed in either survival rate or mean time to death between the two strains (Fig. 3A).

FIG. 3.

Effect of the BclA protein on virulence in the mouse infection model. (A) BALB/c mice were challenged intranasally with spores of the Ames wild-type strain (○) or the bclA strain (Δ). (B) The effect of the bclA mutation on the rate of spore clearance from mouse lungs. Mice were exposed to an aerosolized inoculum of spores of both wild-type Ames and bclA mutant. The number of heat-resistant (65°C for 30 min) spores retained within the lungs at designated time points was determined by bacterial counts on selective and nonselective media. The error bars represent standard errors of the means. *, P = 0.03; **, P = 0.0005.

To further resolve the role of BclA in virulence, we developed a second in vivo competition assay to examine the rate of clearance of spores from the lungs of mice infected with aerosolized spores. Mice were exposed to an aerosol containing approximately equal concentrations of both the wild-type Ames (48%) and the bclA mutant (52%) spores; a calculated total inhaled dose of 4 × 10⁵ spores was delivered (8). Mice were euthanized at various times postinfection, and bronchoalveolar lavages (BAL) were performed. BAL fluid samples (both heated and unheated aliquots) were plated on LB agar plates or LB agar plates containing kanamycin. There were no differences noted between the results obtained from the heated (CFU counts representing heat-resistant ungerminated spores) (Fig. 3B) and unheated BAL fluid (data not shown), confirming that spore germination is negligible within the lungs themselves (8, 11a). Overall, these data suggested that both the wild-type Ames and the bclA mutant spores were cleared from the mouse lungs at similar rates (Fig. 3B; Table 1). However, on day 2 and day 4 there were statistically significant differences between the amounts of wild-type Ames and bclA mutant spores recovered from the BAL fluid (Fig. 3B). The implications of these findings for pathogenesis are unclear, as there were no significant differences observed at any other time point and the bclA mutant appeared to be of virulence equal to that of wild-type Ames by the pulmonary route (Fig. 3 and data not shown). However, these data suggest that the bclA mutant spores might be cleared slightly less efficiently from the lungs than wild-type spores, possibly because the mutant spores were able to bind better to host cells, such as the epithelial cells lining the lung and/or airway.

TABLE 1.

Comparative rates of retention of the wild-type Ames and bclA mutant spores in the lungs of mice exposed to aerosol

Day	% Retention of straina:
	Ames	bclA
0	100	100
1	38.5 ± 4.1	47.4 ± 8.5
2	10.0 ± 1.0	14.2 ± 2.3
3	4.8 ± 2.5	8.8 ± 3.2
4	4.6 ± 0.9	12.1 ± 1.4

^aMean percent retention in BAL fluids (±standard errors of the means) was calculated as the percentage of the initial recovery at time zero.

The exact roles of the BclA protein and the exosporium of B. anthracis in pathogenesis of and host protection against anthrax still remain to be determined. Several other species of Bacillus also possess an exosporium structure; however, except for B. cereus, these bacteria are not normally pathogens of mammals and are common environmental saprophytes. It was recently shown that spores of B. anthracis are able to germinate on and around roots of plants in a grass plant-soil model system (22). Perhaps the entire exosporium or the exosporial fibers are required for the interaction between spores and roots of the plants or serve in some other role for survival in the environment.

Studies such as the one reported here should contribute in several ways to the identification of novel vaccine candidates as well as targets for detector systems. First, a better understanding of how the immune response recognizes B. anthracis spores is necessary, particularly in relation to immunodominant antigens, such as BclA. More importantly, it is vital to characterize the roles of candidate spore antigens in bacterial pathogenesis and the role of the immune response in protective immunity. Our studies with a bclA mutant of the Ames strain revealed that BclA was not required for the pathogenesis of fully virulent B. anthracis in two animal species by different challenge methods. While BclA is immunodominant and accordingly an attractive target for detection technologies, this protein could easily be removed from a B. anthracis variant engineered to evade detection systems, and yet these altered spores would retain full virulence.