Small-Molecule Inhibitors of Lethal Factor Protease Activity Protect against Anthrax Infection

Abstract

Bacillus anthracis, the causative agent of anthrax, manifests its pathogenesis through the action of two secreted toxins. The bipartite lethal and edema toxins, a combination of lethal factor or edema factor with the protein protective antigen, are important virulence factors for this bacterium. We previously developed small-molecule inhibitors of lethal factor proteolytic activity (LFIs) and demonstrated their in vivo efficacy in a rat lethal toxin challenge model. In this work, we show that these LFIs protect against lethality caused by anthrax infection in mice when combined with subprotective doses of either antibiotics or neutralizing monoclonal antibodies that target edema factor. Significantly, these inhibitors provided protection against lethal infection when administered as a monotherapy. As little as two doses (10 mg/kg) administered at 2 h and 8 h after spore infection was sufficient to provide a significant survival benefit in infected mice. Administration of LFIs early in the infection was found to inhibit dissemination of vegetative bacteria to the organs in the first 32 h following infection. In addition, neutralizing antibodies against edema factor also inhibited bacterial dissemination with similar efficacy. Together, our findings confirm the important roles that both anthrax toxins play in establishing anthrax infection and demonstrate the potential for small-molecule therapeutics targeting these proteins.

INTRODUCTION

Bacillus anthracis requires the action of two toxins to manifest anthrax disease. Lethal toxin (LT) and edema toxin (ET) are comprised of three proteins: protective antigen (PA), lethal factor (LF), and edema factor (EF). PA is a receptor-binding component that translocates LF (a protease) or EF (an adenylate cyclase) into cells (for a review, see reference 1). LF cleaves members of the mitogen-activated protein kinase kinase (MEK) family (2, 3) and the rodent Nlrp1/Nlrp1b inflammasome sensors (4, 5). Cleavage of the MEK proteins leads to inhibition of a wide variety of immune cell functions. Cleavage of Nlrp1 in rodents results in the activation of the inflammasome, macrophage pyroptosis, and induction of interleukin-1 (IL-1) and IL-18 and in an accompanying cytokine storm (for a review, see reference 6). This cytokine response is linked to a protective neutrophil response which is higher in mouse strains harboring toxin-susceptible “sensitive” Nlrp1 alleles (7). Edema factor induction of cyclic AMP (cAMP) also results in a number of consequences for the innate immune response (for reviews, see references 1 and 8) and has recently been shown to play an important role in establishing bacterial infection (9).

Anthrax toxins have been implicated in both early and late stages of anthrax infection. In early stages, the toxins impair the function of innate immune first responders, thus allowing B. anthracis to establish infection. Tissue-specific deletion of the primary anthrax toxin receptor on myeloid cells (including all major cells of the immune system) results in complete resistance to infection, while maintaining full susceptibility to challenge with either LT or ET (10). In late stages of disease, the high levels of the toxins in the blood induce unknown vascular events that are poorly understood (11, 12) but contribute to the death of the host. While the mechanism of LT-induced death is unknown, the available data suggest that the cardiovascular system is a target (13–20). In the case of ET, extensive hemorrhagic events and a shock-like death have been observed in mice, most likely due to cAMP-mediated effects on the vasculature (11).

The use of antibiotics which can clear active infection by B. anthracis is less effective in preventing the death of the host if sufficiently high levels of the toxins have accumulated in cells. Furthermore, we and others have found that unlike PA, LF appears to remain active in cells (21) and in animal tissues (M. Moayeri, unpublished data) for days, as shown by continued cleavage of MEK proteins by the toxin during this time. As a result, postinfection treatment of infected animals with anti-PA antibodies alone is time dependent and can be surprisingly ineffective (22). Thus, the use of inhibitors that can block the enzymatic action of LF within cells is an important approach against this disease in postexposure scenarios. Unlike the case for monoclonal antibodies (MAbs), the use of LF inhibitors (LFIs) would allow the toxin to be targeted during all stages of infection.

We previously described a series of small-molecule LFIs which possessed subnanomolar inhibitor constants (K_i values) with demonstrated in vivo efficacy in the protection of rats against challenge with LT (23–25). In this work, we demonstrate that LFIs provide a significant survival benefit when used as a monotherapy and offer full protection when used in combination with subprotective doses of antibiotic or anti-EF monoclonal antibodies in a murine spore infection model. These results suggest that the use of small-molecule LFIs described in this work offers an important therapeutic approach in the treatment of postexposure anthrax.

MATERIALS AND METHODS

Materials.

LFIs 8541 and 8420 have been previously described (24). The vehicle for LFI delivery was 4% dimethyl sulfoxide (DMSO)–10% polyethylene glycol (PEG) 400 in a phosphate-citrate buffer (0.1 M citric acid, 0.2 M Na₂HPO₄, pH 6.0). Ciprofloxacin was purchased from Hospira Inc., Lakeforest, IL. The neutralizing EF monoclonal antibodies 7F10, 4A6, and 3F2 have been described previously (26).

Spores.

Spores were prepared from the nonencapsulated, toxigenic B. anthracis Ames 35 (A35) strain (27) by growth on NBY sporulation agar at 37°C for 24 h followed by 6 to 8 days at 28°C. Plates were inspected by microscopy to verify >95% sporulation. Spores were purified from plates by four cycles of sterile water washes and centrifugation, followed by heat treatment at 70°C for 30 min to kill any remaining vegetative bacteria. Spore quantification was performed using a Petroff-Hausser counting chamber (Hausser Scientific, Horsham, PA). Spore viability was tested by dilution plating. Spore preparation quality was tested by in vivo subcutaneous (s.c.) challenge of the C57BL/6J mouse strain with 2 times the 100% lethal dose (LD₁₀₀) (2 × 10⁷ spores) and comparison of disease progression with archival data accumulated by our laboratory in mouse challenge studies using similarly prepared spore preparations.

PK studies.

Pharmacokinetic (PK) studies on the LFIs were performed by Covance Laboratories (Madison, WI). Briefly, C57BL/6J mice (n = 9/compound) were injected with LFIs (10 mg/kg, s.c.), and blood was collected (with EDTA as an anticoagulant) from 3 mice at each time point (0.25, 0.5, 1.0, 2.0, 4.0, 8.0, 12.0, and 24.0 h). Samples were extracted with acetonitrile and analyzed by liquid chromatography-tandem mass spectrometry (LC/MS-MS). The concentration-time data were used as the input for calculation of the PK parameters using the software PK Solutions (v 2.0; Summit Research Services, Montrose, CO).

Animal studies.

C57BL/6J mice (female, 8 to 10 weeks old) were obtained from Jackson Laboratories (Bar Harbor, ME). Mice were infected with 2 × 10⁷ A35 spores (s.c., 200 μl, scruff of the neck) in all experiments. Mice were treated with combinations of ciprofloxacin (CIP), a triple cocktail of purified anti-EF mouse monoclonal antibodies (MAbs) 7F10, 4A6 and 3F2 (26), or LFIs using the various schedules described in the figure legends. In experiments involving the use of ciprofloxacin, this compound was administered once daily at a dose of 25 mg/kg (intraperitoneally [i.p.], 400 μl), a dose which was found to provide partial protection during the development of the model. The anti-EF MAbs were prepared in phosphate-buffered saline (PBS) and administered intravenously (i.v.) (200 μl, single bolus,1 h prior to infection). In each experiment, they were administered as a premixed cocktail prepared from equal amounts of each MAb (0.1, 0.25, and 1.25 mg/kg/MAb). The LFIs or vehicle was injected s.c. (100 μl, in the rump) at various times postinfection according to the schedules shown in each figure. Mice were monitored for signs of malaise and survival twice daily for 7 days following infection. Dissemination and survival of vegetative bacteria were assessed in certain experiments by first collecting organs (liver, kidney, heart, and spleen) from individual mice at 32 h postinfection and then homogenizing the pooled organs using a Tissue Tearor (BioSpec Products, OK). Homogenates were adjusted to an equal volume (10 ml) in sterile PBS before dilution plating on LB agar. CFU were calculated for whole organs based on colony counts for 100-μl platings of these homogenates. All studies were carried out in accordance with protocols approved by the NIAID Animal Care and Use Committee.

RESULTS

LFIs provide protection against spore infection given as monotherapy and in combination with antibiotic treatment.

Compounds PT-8420 and PT-8541 (Fig. 1), with K_i values of 0.33 and 0.04 nM, respectively, were previously identified and characterized for their ability to protect animals from death in a rat model of LT intoxication (24). The PK data collected for these compounds after s.c. administration (Table 1) indicate that they have good exposure in the plasma (maximum concentration [C_max], >1 μM) and a reasonable half-life (t_1/2) (t_1/2 at β phase, >3.8 h). In addition, the C_maxs are >19,000-fold (PT-8420) and >31,000-fold (PT-8541) above their respective K_i values, suggesting that good exposure at the target site should be possible with the dosing regimen selected for these studies. We tested these compounds for their ability to protect C57BL/6J mice against challenge with a lethal dose of A35 (Sterne-like) spores both as monotherapy and in combination with ciprofloxacin (CIP). We established that a single daily dosing of CIP at 25 mg/kg (i.p.) consistently resulted in survival of 20 to 60% of mice challenged with 2 × 10⁷ A35 spores (2 times the LD₁₀₀) (data not shown). In these studies, the first dose of CIP was given at 4 h after spore infection, with subsequent doses at 24, 48, 72, and 96 h. Our previous results (7) and those of others (28) showed that spores germinate within the first hour after a subcutaneous (s.c.) inoculation, producing vegetative bacteria that would be susceptible to antibiotic given at 3 h postinfection. Significant edema indicative of ET production by vegetative cells is also measurable in the first 4 h after infection (7). Plating of blood samples at 32 or 48 h showed that this schedule of dosing with CIP killed all bacteria in the animals that succumbed to infection as well as those that survived (data not shown). Thus, bacterial clearance from the blood was not correlated with survival. When the LFIs were administered twice a day (b.i.d.) (at 2 h and 8 h postinfection and twice daily thereafter as shown in Fig. 2), in combination with the subprotective antibiotic treatment described above, all animals survived with little to no signs of malaise. Interestingly, however, we found that for animals not receiving any antibiotic treatment, LFI monotherapy given with the same b.i.d. protocol resulted in a 2- to 3-fold increase in mean survival time (MST) and survival rates of 2/15 for PT-8420 and 6/15 for PT-8541 (Fig. 2). Survival of 40% (PT-8420) and 66% (PT-8541) of the animals was observed at the time when LFI treatment was suspended (gray line at 104 h in Fig. 2) and demonstrates that the LFIs had a considerable therapeutic effect in this study. In addition, animals treated with the LFIs had a later onset and lower severity of clinical symptoms than vehicle-treated controls. These results indicated not only that LFIs were completely successful in providing full protection when combined with subprotective doses of antibiotic therapy but also that they provided a significant survival benefit when given as monotherapy.

Fig 1.

Structures of PT-8420 and PT-8541.

Table 1.

PK data for single s.c. dosing of LFIs at 10 mg/kg (PT-8420) or 5 mg/kg (PT-8541) in female C57BL/6 mice

Compound	Ki (nM)	Molecular massa (g/mol)	Cmaxb		Tmax (h)	t1/2(β)c (h)	AUC∞ (ng · hr/ml)	AUMC∞ (ng · hr2/ml)	MRTd (h)
			ng/ml	μM
PT-8420	0.33	470.98	3,040	6.41	0.25	3.87	3,004.9	5,194.3	1.7
PT-8541	0.042	454.98	600	1.32	0.25	4.62	694.2	1,694.9	2.4

^aMolecular mass of the salt form.

^bC_max is the value observed at T_max and may not be the true maximum concentration attained by the LFI in the plasma.

^cTerminal phase half-life determined using either 2 time points (t = 8 and 12 h) or 3 time points (t = 8, 12, and 24 h).

^dMRT, mean residence time (AUMC_∞/AUC_∞).

Fig 2.

LFI protection against spore infection in combination with antibiotic treatment and as monotherapies. C57BL/6J mice (n = 15/group) were infected with 2 × 10⁷ A35 spores (s.c., scruff of neck) at time zero (red arrow). LFIs (PT-8420 and PT-8541) or vehicle was injected s.c. at a distal location (flank, b.i.d., according to the schedules shown; black arrows). Select groups were treated i.p. with a subprotective dose of CIP alone (25 mg/kg, s.i.d.; purple arrows) or CIP in addition to LFIs (8420+CIP and 8541+CIP). Mice were monitored for signs of malaise and survival twice daily for 7 days. The results shown are pooled from two independent experiments. The gray vertical line indicates the time of the last LFI treatment (104 h). The P values calculated using the log rank (Mantel-Cox) test comparing the vehicle group to the remaining groups are <0.0001. The MST values for groups with nonsurvivors were as follows: vehicle, 59 h; CIP, 103 h; PT-8420, 99 h; PT-8541, 168 h.

LFIs given postinfection provide complete protection when combined with anti-EF MAb therapy.

Recent studies have shown that EF plays an important role in dissemination during anthrax infection (9). Our previous finding that MAbs against EF can provide moderate protection against anthrax in the murine model (26) suggested that they be tested in combination with the LFIs. To determine if the presence of an LFI would provide added benefit in this model, a cocktail of three EF MAbs was given at 1 h prior to spore infection, at 0.1, 0.25, or 1.25 mg/kg/MAb, followed by PT-8541 according to the schedule shown in Fig. 3. Consistent with the previous experiment, PT-8541 provided partial protection when given as monotherapy. A dose-dependent protection by the EF MAbs when given as monotherapy was seen at a significantly higher per-weight efficacy than previously observed with suboptimal preparations of the same antibodies (26). The low-dose (0.1-mg/kg/MAb) treatment group succumbed in a manner similar to that for the vehicle-treated controls. Although there was no difference in final survival outcome (40%) between the mid-dose (0.25-mg/kg/MAb) and high-dose (1.25-mg/kg/MAb) treatment groups, the overall health of the group treated with the higher dose of EF MAbs was significantly better than that of the mid-dose group (data not shown). This finding is reflected in the observation of the first deaths occurring in the high-dose group 2 days later than those in the mid-dose group (Fig. 3). When PT-8541 was combined with the subprotective doses of the EF MAbs (at 0.25 or 1.25 mg/kg/MAb), the mice displayed little to no malaise and were completely protected from death (Fig. 3). Although the combination of the antibody at the lowest dose (0.1 mg/kg/MAb) with PT-8541 resulted in a significant extension in MST and 100% of animals being fully protected at the time the last dose of the LFI was administered (gray line at 104 h in Fig. 3), the final survival rate in this case was 20%. While this combination treatment did not lead to an improved survival outcome relative to treatment with LFI alone, the overall health of the combination therapy treatment groups was significantly better than that of either monotherapy treatment group (data not shown), as reflected by the first deaths being seen only after stopping treatment with the LFI (Fig. 3). In each case where the combination therapy was tested, an added benefit was clearly seen relative to the monotherapy groups.

Fig 3.

Combination LFI and anti-EF monoclonal antibody therapy. C57BL/6J mice (n = 5 to n = 15/group) were infected with 2 × 10⁷ A35 spores (s.c., scruff of neck) at time zero (red arrow). At 1 h prior to infection mice were injected i.v. with a cocktail composed of three anti-EF monoclonal antibodies at a low (LOW) (0.1 mg/kg/MAb), medium (MED) (0.25 mg/kg/MAb), or high (HIGH) (1.25 mg/kg/MAb) dose (blue arrow), followed by treatment with LFI 8541 (10 mg/kg, b.i.d., s.c. in rump; black arrows) or with vehicle (b.i.d. in rump; black arrows). Control groups (n = 20) were infected and treated with LFI 8541 alone (10 mg/kg, b.i.d., s.c., in rump; black arrows) or vehicle alone (b.i.d., s.c., in rump; black arrows). Mice were monitored for signs of malaise and survival twice daily for 7 days. The gray vertical line indicates the time of the last LFI treatment (104 h). The results shown are pooled from three independent experiments. P values, calculated by the log rank (Mantel-Cox) test, comparing the vehicle group to the remaining groups are <0.0001. MST values for groups with nonsurvivors were as follows: vehicle, 53 h; PT-8541, 117 h; EF MAb (LOW) + vehicle, 72 h; EF MAb (MED) + vehicle, 102 h; EF MAb (HIGH) + vehicle, 163 h; EF MAb (LOW) + 8541, 144 h. The P values for vehicle versus all groups are <0.001, except vehicle versus EF MAb (LOW) + vehicle which is 0.0298. The P value for each EF MAb + vehicle group versus the corresponding dose of EF MAb + 8541 group is <0.0001, except for the LOW (0.1 mg/kg/MAb) groups, where the P value is 0.0011. The P value for the group treated with PT-8541 alone versus EF MAb (LOW) + 8541 is 0.0911, those for while PT-8541 versus EF MAb (MED) + 8541 and PT-8541 versus EF MAb (HIGH) + 8541 are <0.0001.

Because both the EF MAbs and PT-8541 were able to provide partial protection when given at early time points in the study (EF MAbs at −1 h and PT-8541 at +2 h) and were fully protective in combination, we examined whether administration of an LFI at a later time point would be beneficial. We had previously found that EF MAbs did not protect against anthrax infection at doses up to 5 mg/kg when given at 18 to 24 h postinfection (data not shown). When PT-8541 was administered as monotherapy starting at 2 h postinfection (schedule 1, black arrows, in Fig. 4), a significant increase in the MST was observed. If the same twice-daily dosing schedule was delayed until 24 h or 32 h (Fig. 4, schedule 2, black arrows) after infection, the survival curve was not different from that for the vehicle-treated control group. In contrast, when PT-8541 was administered following the delayed schedule and combined with a single bolus of EF MAbs given 1 h prior to infection (schedule 2, blue and black arrows), there was a significant protective effect observed relative to vehicle-treated controls (P < 0.001). This protective effect was diminished compared to that when starting the LFI treatment at the earlier time of 2 h postinfection (schedule 1, blue and black arrows). Thus, in this mouse model, both EF and LF appear to play an important role in the early hours following spore infection, and targeting of the toxins at the early stages of the disease provides for better protection than with therapy given at later time points. In addition, these results show that inhibition of both LF and EF provides an added benefit toward protecting the host against anthrax compared to targeting either one alone.

Fig 4.

Early versus delayed LFI therapy against anthrax infection. C57BL/6J mice (n = 5 to n = 6/group) were infected with 2 × 10⁷ A35 spores (s.c., scruff of neck) at time zero (red arrow). Two different therapy schedules were followed. In schedule 1 (SCH1), a cocktail of three EF MAbs or PBS was given 1 h prior to infection (0.1 mg/kg/MAb, i.v.) (blue arrow), followed by twice daily PT-8541 injections starting at 2 h postinfection (s.c. in rump, b.i.d.; black arrows). In schedule 2 (SCH2), the first administration of PT-8541 was delayed by either 24 h or 32 h (s.c., black arrows). Two groups of animals did not get any EF MAb prior to infection (DELAYED 8541-SCH2-LFI 32 h or DELAYED 8541-SCH2-LFI 24 h), and another did (EF MAb + DELAYED 8541-SCH2-LFI 32 h). Mice were monitored for signs of malaise and survival twice daily for 7 days. The gray vertical line indicates the time of the last LFI treatment (104 h). MST values are as follows: PT-8541 (SCH1), 94 h; DELAYED 8541 (SCH2), 48 h; EF MAb + 8541 (SCH1), 120 h; EF MAb + DELAYED 8541 (SCH 2), 72 h; and vehicle only, 48 h. P values were calculated by the log rank (Mantel-Cox) test comparing the vehicle group to all groups and are <0.001, except for the vehicle versus DELAYED 8541 (SCH2), which has a P value of 0.4788. The P value comparing early PT-8541 treatment versus delayed PT-8541 treatment is 0.0016. The P value comparing EF MAb + 8541 versus EF MAb + delayed 8541 is 0.0020.

Early LFI treatment provides a significant survival benefit.

To further investigate the role of LF early in the disease, we tested the protective effects of PT-8541 when given only on the first day at 2 h and 8 h after spore infection (Fig. 5, schedule 2). Surprisingly, this LFI provided a significant shift in MST compared to that of the control group. Although there was no difference in the survival curves resulting from the single-day (schedule 2) versus the five-day (schedule 1) dosing regimens with PT-8541, 40% of the animals in the multiday dosing group were alive at the time the last dose was administered (Fig. 5, gray line), whereas all animals in the single-day dosing group had succumbed by this time point in the study.

Fig 5.

Protective effects of single-day LFI dosing. C57BL/6J mice (n = 5/group) were infected with 2 × 10⁷ A35 spores (s.c., scruff of neck) at time zero (red arrow). Inhibitor was either injected twice daily for 5 days (schedule 1) or only twice on day 1 at 2 and 8 h postinfection (schedule 2). Mice were monitored for signs of malaise and survival twice daily for 7 days. The gray vertical line indicates the time of the last LFI treatment (104 h). MST values are as follows: vehicle, 65 h; PT-8541 SCH1, 94.5 h; PT-8541 SCH2, 96 h. The P values calculated by the log rank (Mantel-Cox) test comparing each treatment group to the vehicle group were <0.05. The P value for comparison of PT-8541 SCH1 versus SCH2 is 0.4669.

We then compared the organ CFU plate counts at 32 h postinfection from animals in the LFI or EF MAb monotherapy group and the combination therapy group. In the LFI monotherapy group, only three injections of LFI were given, at 4 h, 8 h, and 24 h postinfection. In the EF MAb monotherapy group, only a single bolus of the antibody cocktail was given, at 1 h prior to infection. Along with the vehicle control and monotherapy groups, one LFI-treated group also received EF MAb (0.25 mg/kg/MAb) at 1 h preinfection. Interestingly, treatment with both the EF MAbs and PT-8541 as monotherapy led to an equivalent decrease in the organ CFU counts relative to those in vehicle-treated controls at 32 h (Fig. 6). Somewhat surprising was the similar result obtained with the combination therapy group since earlier studies, (Fig. 3) (PT-8541 at 10 mg/kg and EF MAb at 0.25 mg/kg/MAb) showed that complete protection was possible with this treatment regimen. These results indicate that both LF and EF play an important role early in infection, either by supporting the ability of innate immune cells to clear B. anthracis or by inhibiting other factors required for dissemination of the bacterium. Perhaps what is more interesting is the observation that their combined effect on survival late in infection is not simply related to additive effects on early dissemination of the bacterium.

Fig 6.

Bacterial dissemination in LFI- and EF MAb-treated mice. C57BL/6J mice (n = 5/group) were infected with 2 × 10⁷ A35 spores (s.c., scruff of neck) at time zero. Mice were injected with a single dose (i.v.) of EF MAb cocktail (0.25 mg/kg/MAb) or PBS at 1 h prior to infection, followed by PT-8541 (10 mg/kg, s.c.) or vehicle treatment at 2, 8, and 24 h postinfection. Organs were collected at 32 h for plating to calculate vegetative bacterial CFU as a measure of dissemination. P values (by two-tailed Mann Whitney test) comparing each group to the vehicle-treated group are as follows: PT-8541, 0.0079, PT-8541 + EF MAb, 0.0079, EF MAb, 0.0119.

DISCUSSION

It has long been accepted that combining antibiotics (to treat bacteremia) and monoclonal antibodies (to target toxins) would be a logical approach for the treatment of anthrax (for reviews, see references 29, 30, and 31). Although antibiotics are effective at sterilizing the blood of the host, their survival benefit is diminished once a significant amount of the anthrax toxins has been produced and entered the host's cells. Since the anthrax toxins cannot be treated with antibodies after they are in the intracellular environment, small-molecule inhibitors that target the toxins provide an important therapeutic approach.

Many LFIs have shown great promise in vitro in macrophage death neutralization assays but have not been tested in vivo (32–34). When tested in vivo, toxin challenge models are often utilized (24, 35). Alternatively, LFIs have been used in a limited number of spore infection studies in combination with antibiotic treatment. As an example, in a murine DBA/2 Sterne infection challenge model, Forino et al. administered B1-11B3 (5 mg/kg, once a day [s.i.d.]) in combination with a subprotective dose of antibiotic (36). Although an increase in survival rate was seen in the combination treatment group (40%, compared to only 20% observed for the antibiotic-only group), the difference in the MSTs of the treatment groups was not statistically significant (36). One possible explanation for this result is that the DBA/2 mice used in these studies are complement deficient and thus highly susceptible to B. anthracis infection in a manner demonstrated to be somewhat independent of the early effects of LT on immune cells (7). As a result, the use of this mouse strain may have limited the ability of these studies to reveal the beneficial effects of treatment with LFIs. In addition to this experiment, only two prior studies have tested the use of LFIs as monotherapy in the treatment of anthrax infection. In a study of LFIs related to those discussed above (36), the same group used various LFIs (25 mg/kg daily) to treat a Sterne spore infection in complement-deficient A/J mice. They showed that 2/8 challenged mice receiving LFI treatment survived, compared to no survival in controls (37). In a different study, the LFI (2R)-2-[(4-fluoro-3-methylphenyl)sulfonylamino]-N-hydroxy-2-(tetrahydro-2H-pyran-4-yl)acetamide, developed at Merck Research Laboratories, was tested as monotherapy in a Sterne strain vegetative cell murine challenge. Bacterial challenge (i.p.) was performed with 10⁸ CFU in BALB/c mice at 24 h after beginning a continuous infusion of inhibitor, and 50 to 60% of animals were saved (38). The LFI was also tested as a monotherapy in Dutch Belted rabbits using a subcutaneous B. anthracis Ames strain spore challenge (10⁴ spores) where the LFI was given three times daily (100 mg/kg/dose for 7 days). When the spore challenge occurred at the same time or 2 h after the first LFI dose, a significant shift in MST was observed, and 2/4 rabbits survived. Dosing of the LFI at 24 h after spore challenge resulted in a doubling of the MST relative to that for vehicle-treated controls, with only 1/4 rabbits surviving. Late LFI treatment combined with ciprofloxacin therapy (66 h after spore infection) protected 4/4 rabbits in these studies (38).

Our studies use the complement-sufficient C57BL/6J mouse and a challenge dose of 2 times the LD₁₀₀, which allows us to assess the impact of interfering with different steps in the infection process over a five-day window. Our previous work has shown that antibody therapies against PA, LF, or EF administered later than 24 h after infection rarely provide protection in this model (data not shown), and this emphasizes the benefit of targeting the toxin molecules early in infection. Unless LF is inhibited, it disables the cells of the innate immune response which battle the infection, allowing division and dissemination of bacteria (10). Our data indicate that the beneficial impact of inhibiting each toxin (LT or ET) alone is equivalent in preventing bacterial dissemination or survival in organ sites. Yet while monotherapy against each toxin at a given subprotective dose extends the MST, the targeting of both toxins provides full protection. Thus, it is possible that inhibition of bacterial dissemination and survival at the early stages of disease is not the only mechanism by which the LFIs or anti-EF MAbs lead to protection. The simultaneous inhibition of both toxins may also have an impact at later stages of disease in a manner independent of early dissemination.

Our studies demonstrate that the use of LFIs in combination with antibiotics or anti-EF therapeutics offers a promising approach to postexposure treatment of anthrax. Thus, the benefit of being able to target LF after it has entered cells is relevant to survival outcome. It is also clear that a better understanding of the parameters that define the therapeutic window for targeting the toxins during infection is an area which requires further investigation.