Clofazimine Reduces the Survival of Salmonella enterica in Macrophages and Mice

Abstract

Drug resistant pathogens are on the rise, and new treatments are needed for bacterial infections. Efforts toward antimicrobial discovery typically identify compounds that prevent bacterial growth in microbiological media. However, the microenvironments to which pathogens are exposed during infection differ from rich media and alter the biology of the pathogen. We and others have therefore developed screening platforms that identify compounds that disrupt pathogen growth within cultured mammalian cells. Our platform focuses on Gram-negative bacterial pathogens, which are of particular clinical concern. We screened a panel of 707 drugs to identify those with efficacy against Salmonella enterica Typhimurium growth within macrophages. One of the drugs identified, clofazimine (CFZ), is an antibiotic used to treat mycobacterial infections that is not recognized for potency against Gram-negative bacteria. We demonstrated that in macrophages CFZ enabled the killing of S. Typhimurium at single digit micromolar concentrations, and in mice, CFZ reduced tissue colonization. We confirmed that CFZ does not inhibit the growth of S. Typhimurium and E. coli in standard microbiological media. However, CFZ prevents bacterial replication under conditions consistent with the microenvironment of macrophage phagosomes, in which S. Typhimurium resides during infection: low pH, low magnesium and phosphate, and the presence of certain cationic antimicrobial peptides. These observations suggest that in macrophages and mice the efficacy of CFZ against S. Typhimurium is facilitated by multiple aspects of soluble innate immunity. Thus, systematic screens of existing drugs for infection-based potency are likely to identify unexpected opportunities for repurposing drugs to treat difficult pathogens.

Keywords: Gram-negative bacteria; Salmonella enterica serovar Typhimurium; antibiotics; clofazimine (CFZ); macrophage.

Figure 1.

CFZ inhibits S. Typhimurium infection in cell culture macrophages. (A, B) RAW 264.7 macrophage-like cells were infected with S. Typhimurium harboring a chromosomal sif B::gfp insertion. After 2 h, cells were treated with vehicle (DMSO) or CFZ. After 18 h of infection, cells were fixed and imaged. (A) Representative micrographs from cells treated with DMSO or CFZ. (B) GFP⁺ macrophage area quantified from micrographs of cells treated with 2-fold dilutions of CFZ from 21 μM. GFP⁺ macrophage area is defined as the number of GFP⁺ pixels per macrophage divided by the total number of pixels per macrophage, averaged across all macrophages. Mean and standard error of two independent experiments performed in duplicate. (C-E) RAW 264.7 macrophages were infected with S. Typhimurium and treated as in (A). After 2, 4, or 18 h of infection, cells were lysed and plated for enumeration of CFU at the indicated time points (h.p.i., hours post-infection). Mean and SEM of three independent experiments performed in duplicate. (C, D) Data represented as IC50 curves. (E) Comparison of CFU from infected macrophages treated with 21 μM CFZ at indicated hours post-infection (h.p.i.). **, p < 0.0003, Students t test.

Figure 2.

CFZ reduces S. Typhimurium colonization of the spleen in wild-type but not in ob/ob mice. (A, C) C57Bl/6 or ob/ob (C57Bl/6 background) mice were intraperitoneally inoculated with 1 × 10⁴ wild-type bacterial CFU. At 20 min and 24 h post-infection, mice were dosed orally with 10 mg/kg CFZ. At 48 h post-infection, mice were euthanized, and the spleen and liver were homogenized and plated for enumeration of CFU. *, p ≤ 0.05, Mann-Whitney. ns, not significant. (B) Accumulation of pigment (arrow) in the fat pads of ob/ob mice treated with vehicle versus CFZ.

Figure 3.

CFZ activity in MHB is potentiated by polymyxin B or by the loss oflptD. (A-D) Growth curves of bacteria with increasing doses of CFZ. Absorbance (600 nm) was measured over 18 h. (A, B) S. Typhimurium without or with 1 μg/mL PMB. (C) Wild-type S. Typhimurium was grown with 0.75 μg/mL PMB in the presence or absence of 21 μM CFZ. CFU were enumerated at the indicated hours post-treatment (h.p.t.). **, p < 0.0003, Student’s t test. (D, E) E. coli K12 wild-type and lptD4213 mutant strains. Mean and standard deviation of duplicate samples from a representative of three independent experiments. (F-I) S. Typhimurium checkerboard growth assays with the indicated cAMP and CFZ. Absorbance (600 nm) was measured after 18 h of growth and averaged across three independent experiments. (J) PMBM permeabilization of the S. Typhimurium outer membrane as monitored by novobiocin inhibition of growth. OD₆₀₀ after 18 h of growth was averaged across three independent experiments. (K, L) S. Typhimurium outer membrane permeabilization as monitored by nitrocefin (100 μM) access to and hydrolysis in the periplasm in the presence of (K) PMB or (L) PMBN. Data were normalized to buffer control and plotted as the mean and SEM of 1/slope of the linear region (over 60 min) of the A₄₈₆ plots from three independent experiments. *, p ≤ 0.05 compared to vehicle control by one-way ANOVA with Dunnett’s multiple comparison test. (M) Permeabilization of the inner membrane after 30 min of treatment as monitored by increased fluorescence of propidium iodide (Ex/Em of 533 nm/630 nm) upon gaining access to the cytosol and binding DNA. Mean and SEM of three- independent experiments normalized to buffer controls. Positive controls: heat killed (HK) and 0.001% SDS. (N) S. Typhimurium was grown in MHB and treated as indicated with PMB (0.75 μg/mL), CFZ (21 μM), or the iron chelator deferasirox (DFS).

Figure 4.

CFZ inhibits bacterial growth in LPM, a nutrient-poor medium that mimics the phagosome and permeabilizes the outer membrane, unless pH, magnesium, or phosphate are increased. (A) S. Typhimurium or (B) E. coli were grown in LPM medium under the conditions indicated, and absorbance (600 nm) was measured over 18 h. Mean and standard deviation of triplicate samples from a representative of three independent experiments. (C) S. Typhimurium or (D) E. coli were grown in LPM with 21 μM CFZ. Absorbance (600 nm) was measured after 18 h of growth. Mean and SEM across at least three independent experiments. (E) S. Typhimurium outer membrane permeabilization monitored by nitrocefin [100 μM] hydrolysis after growth in the indicated conditions for 18 h. Mean and SEM across three independent experiments. ****, p ≤ 0.0001; **, p ≤ 0.002, by one-way ANOVA with Dunnett’s multiple comparison test.