Immunization of Mice with Anthrax Protective Antigen Limits Cardiotoxicity but Not Hepatotoxicity Following Lethal Toxin Challenge

Abstract

Protective immunity against anthrax is inferred from measurement of vaccine antigen-specific neutralizing antibody titers in serum samples. In animal models, in vivo challenges with toxin and/or spores can also be performed. However, neither of these approaches considers toxin-induced damage to specific organ systems. It is therefore important to determine to what extent anthrax vaccines and existing or candidate adjuvants can provide organ-specific protection against intoxication. We therefore compared the ability of Alum, CpG DNA and the CD1d ligand α-galactosylceramide (αGC) to enhance protective antigen-specific antibody titers, to protect mice against challenge with lethal toxin, and to block cardiotoxicity and hepatotoxicity. By measurement of serum cardiac Troponin I (cTnI), and hepatic alanine aminotransferase (ALT), and aspartate aminotransferase (AST), it was apparent that neither vaccine modality prevented hepatic intoxication, despite high Ab titers and ultimate survival of the subject. In contrast, cardiotoxicity was greatly diminished by prior immunization. This shows that a vaccine that confers survival following toxin exposure may still have an associated morbidity. We propose that organ-specific intoxication should be monitored routinely during research into new vaccine modalities.

Keywords: Bacillus anthracis; lethal toxin; neutralizing antibody; protective antigen; troponin.

Figure 1

Anthrax toxin leads to elevation of serum hepatic AST and ALT and cardiac Troponin I in naïve mice. (A) C57Bl/6 mice were treated with LT in PBS or PBS alone and blood samples collected after 48 h. AST, ALT and Troponin I concentrations were then measured. Data show the mean ± SEM enzyme concentration (n = 10 per group); (B) Mice were treated with either LT in PBS or PBS alone. Mice were observed for times indicated and survival monitored (n = 10 per group); (C) Mice were treated with the doses of LT indicated and rested for 48 h before collecting blood samples and measuring AST, ALT and cTnI. Each data point represents an individual mouse; (D) Mice were treated with LT and blood samples were collected at times shown. Data shows the mean ± SEM serum AST, ALT and cTnI enzyme concentrations (n = 3 per group); (E) Mice were treated with PA, LF, mutant lethal factor (mLF), mLT or LT and blood samples were collected at 48 h. AST, ALT and cTnI were then measured. (F) Survival was monitored (n = 5 per group). Statistical significance in (A) and (D) were determined by Mann Whitney U-test, and in (E) by one-way ANOVA and Bonferroni’s multiple comparisons post-test. Statistical significance in (B) and (F) was determined by performing Kaplan-Meier analysis in conjunction with a log rank test. The p values indicate significant differences between survival of the PBS- and the LT-treated groups in (B), and between all groups and the LT-treated group in (F).

Figure 1

Anthrax toxin leads to elevation of serum hepatic AST and ALT and cardiac Troponin I in naïve mice. (A) C57Bl/6 mice were treated with LT in PBS or PBS alone and blood samples collected after 48 h. AST, ALT and Troponin I concentrations were then measured. Data show the mean ± SEM enzyme concentration (n = 10 per group); (B) Mice were treated with either LT in PBS or PBS alone. Mice were observed for times indicated and survival monitored (n = 10 per group); (C) Mice were treated with the doses of LT indicated and rested for 48 h before collecting blood samples and measuring AST, ALT and cTnI. Each data point represents an individual mouse; (D) Mice were treated with LT and blood samples were collected at times shown. Data shows the mean ± SEM serum AST, ALT and cTnI enzyme concentrations (n = 3 per group); (E) Mice were treated with PA, LF, mutant lethal factor (mLF), mLT or LT and blood samples were collected at 48 h. AST, ALT and cTnI were then measured. (F) Survival was monitored (n = 5 per group). Statistical significance in (A) and (D) were determined by Mann Whitney U-test, and in (E) by one-way ANOVA and Bonferroni’s multiple comparisons post-test. Statistical significance in (B) and (F) was determined by performing Kaplan-Meier analysis in conjunction with a log rank test. The p values indicate significant differences between survival of the PBS- and the LT-treated groups in (B), and between all groups and the LT-treated group in (F).

Figure 2

Immunization with PA induces PA-specific Ab and inhibits cardio- but not hepato-toxicity. (A) C57Bl/6 mice were immunized with 8 µg of PA in PBS on day 0 and boosted with 5 µg of PA on day 28. Mice were bled on days indicated. Data show PA-specific IgG1 (left), IgG2b (center) and IgG2c (right) Ab titers. Each data point represents an individual mouse; (B) Naïve and PA-immunized mice were treated with LT and blood samples collected after 48 h. AST, ALT and Troponin I were then measured. Data show the mean ±SEM serum enzyme concentrations (n = 5 for naïve group, n = 10 for immunized group). Two outliers (non-responders with value = 0) were removed from the naïve challenged group in the cTn1 assay; (C) Liver sections were prepared from parallel groups of mice and stained with H&E; (D) Slides were scored for leukocyte infiltration and coagulative necrosis (n = 5 per group). Statistical significance in (A,B) and (D) was determined by one-way ANOVA and Bonferroni’s multiple comparisons post-test.

Figure 3

Effects of adjuvant inclusion on PA-specific Ab titers and LT-induced toxicity. Mice were immunized and bled according to schedule in Figure 2 using the formulations indicated (A) Shows PA-specific IgG1 titers; (B) Shows PA-specific IgG2b and IgG2c titers. 9–10 mice per group were immunized. Naïve and immunized mice were bled and then treated with high dose LT split over two doses (200 µg PA plus LF followed by 100 µg PA plus LF 24 h later). Statistical significance in (A) and (B) was determined by one-way ANOVA and Bonferroni’s multiple comparisons post-test; (C) ALT and AST were measured as described (n = 3–5 mice per group). One-way ANOVA did not reveal significant differences between groups in (C). The p value shown in (C) results from Mann-Whitney U test comparison of the naïve/unchallenged and naïve/LT-challenged groups; (D) Troponin I concentrations were then measured as described (n = 3–5 mice per group). One way ANOVA revealed significant differences between the means and p value ranges indicate results of post-test with Bonferroni’s correction applied; (E) Shows survival of naïve and immunized mice (n = 4–5 mice per group). Statistical significance was determined by performing Kaplan-Meier analysis in conjunction with a log rank test. The p values indicate significant differences between survival of the indicated groups and the naïve group.

Figure 3

Effects of adjuvant inclusion on PA-specific Ab titers and LT-induced toxicity. Mice were immunized and bled according to schedule in Figure 2 using the formulations indicated (A) Shows PA-specific IgG1 titers; (B) Shows PA-specific IgG2b and IgG2c titers. 9–10 mice per group were immunized. Naïve and immunized mice were bled and then treated with high dose LT split over two doses (200 µg PA plus LF followed by 100 µg PA plus LF 24 h later). Statistical significance in (A) and (B) was determined by one-way ANOVA and Bonferroni’s multiple comparisons post-test; (C) ALT and AST were measured as described (n = 3–5 mice per group). One-way ANOVA did not reveal significant differences between groups in (C). The p value shown in (C) results from Mann-Whitney U test comparison of the naïve/unchallenged and naïve/LT-challenged groups; (D) Troponin I concentrations were then measured as described (n = 3–5 mice per group). One way ANOVA revealed significant differences between the means and p value ranges indicate results of post-test with Bonferroni’s correction applied; (E) Shows survival of naïve and immunized mice (n = 4–5 mice per group). Statistical significance was determined by performing Kaplan-Meier analysis in conjunction with a log rank test. The p values indicate significant differences between survival of the indicated groups and the naïve group.