Human monoclonal antibodies against anthrax lethal factor and protective antigen act independently to protect against Bacillus anthracis infection and enhance endogenous immunity to anthrax

Abstract

The unpredictable nature of bioterrorism and the absence of real-time detection systems have highlighted the need for an efficient postexposure therapy for Bacillus anthracis infection. One approach is passive immunization through the administration of antibodies that mitigate the biological action of anthrax toxin. We isolated and characterized two protective fully human monoclonal antibodies with specificity for protective antigen (PA) and lethal factor (LF). These antibodies, designated IQNPA (anti-PA) and IQNLF (anti-LF), were developed as hybridomas from individuals immunized with licensed anthrax vaccine. The effective concentration of IQNPA that neutralized 50% of the toxin in anthrax toxin neutralization assays was 0.3 nM, while 0.1 nM IQNLF neutralized the same amount of toxin. When combined, the antibodies had additive neutralization efficacy. IQNPA binds to domain IV of PA containing the host cell receptor binding site, while IQNLF recognizes domain I containing the PA binding region in LF. A single 180-mug dose of either antibody given to A/J mice 2.5 h before challenge conferred 100% protection against a lethal intraperitoneal spore challenge with 24 50% lethal doses [LD50s] of B. anthracis Sterne and against rechallenge on day 20 with a more aggressive challenge dose of 41 LD50s. Mice treated with either antibody and infected with B. anthracis Sterne developed detectable murine anti-PA and anti-LF immunoglobulin G antibody responses by day 17 that were dependent on which antibody the mice had received. Based on these results, IQNPA and IQNLF act independently during prophylactic anthrax treatment and do not interfere with the establishment of endogenous immunity.

FIG. 1.

Toxin neutralization by IQNPA and IQNLF. The abilities of both antibodies to neutralize a given amount of toxin were evaluated using the TNA assay (▵ and □). In order to determine if combining IQNPA and IQNLF had a positive or negative effect on neutralization, the two MAbs were mixed in various ratios prior to use in the assay (▴ and ▪). The results are expressed as the effective concentration of antibody that protected 50% of the cells in the assay (EC₅₀).

FIG. 2.

rLF domain recognition by IQNLF. (A and B) SDS-denatured and native Western blots probed with IQNLF. (C) Silver-stained SDS-denaturing gel showing the approximate size of each domain. The gels were loaded with markers (left and right lanes) and the same concentration (50 ng) of each of the following constructs: full-length rLF (lane 1), LF domain I (lane 2), LF domain II (lane 3), LF domain III (lane 4), LF domain IV (lane 5), and LF domains II to IV (lane 6).

FIG. 3.

Protection of mice from an i.p. anthrax spore challenge by prechallenge injection of IQNPA and IQNLF. Mice were inoculated i.p. with 180 μg of either IQNPA (n = 10) or IQNLF (n = 10) 2.5 h prior to an i.p. spore challenge consisting of 24 LD₅₀s. The MTD ± SD for the untreated controls was 87 ± 27 h. Twenty days later surviving mice were rechallenged with 41 LD₅₀s. In this second challenge the MTD ± SD for the controls was 69 ± 18 h.

FIG. 4.

Mouse specific anti-toxin IgG titers after B. anthracis Sterne strain spore challenge. Anti-PA and anti-LF IgG titers were measured in five sample bleeds taken during the course of the study. Solid symbols (▴ and •) indicate GMTs obtained for mice treated with IQNPA, while open symbols (▵ and ○) indicate GMTs obtained for mice treated with IQNLF. Mice were challenged on days 0 and 20, as indicated by the vertical dashed lines. Mice produced a modest amount of anti-toxin IgG 10 days after the first challenge, which increased following the second challenge on day 20. Mice treated with IQNPA produced a greater anti-PA IgG response than anti-LF IgG response, while the converse was true for mice receiving IQNLF.

FIG. 5.

Clearance of IQNPA and IQNLF in vivo. The levels of IQNPA and IQNLF asymptotically decreased by 77.5 and 91.3%, respectively, over the course of the 34-day study. The half-life (t_1/2) of each antibody was calculated using the formula: t_1/2=(t₂−t₁)×ln(2)/ln([C_t1/C_t2)], where t₁ and t₂ are times 1 and 2, respectively, and C_t1and C_t2are the MAb titers at t₁ and t₂, respectively. Using this equation, IQNPA had a half-life of 11.2 days, while IQNLF had a half-life of 6.8 days.