Clofazimine Analogs with Efficacy against Experimental Tuberculosis and Reduced Potential for Accumulation^▿

Compounds.

RIF and isoniazid (INH) were purchased from Sigma, and CFM was provided by Nanjing Liye (China). Novel riminophenazine analogs were synthesized by the Institute of Materia Medica (Yin and Huang, unpublished).

Bacterial strains.

The M. tuberculosis strains used in these studies included the laboratory strain H37Rv (ATCC 27294; American Type Culture Collection, Rockville, MD) and both drug-susceptible and drug-resistant clinical isolates. All isolates were obtained from the State Laboratory of Tuberculosis Reference of China.

Determinations of solubility and logP.

To determine the logP, water (2 ml) was added to 1 ml of a 4-mg/ml solution of analyte in octan-1-ol. The mixture was shaken for 6 h at room temperature. After centrifugation and manual phase separation, the aqueous phase was analyzed by high-pressure liquid chromatography (HPLC). The logP was calculated as follows: logP = log ([analyte]_octan-1-ol/[analyte]_water). To determine aqueous solubility, 10 mg of compound was dissolved in a pH 7.0 buffer solution and in a pH 1.0 buffer solution. Equilibration of test compound was achieved after 30 min by vigorous shaking for 30 s at a time over 5-min intervals at 20 ± 5°C.

MICs.

MICs against replicating M. tuberculosis were determined by the microplate Alamar blue assay (MABA) (11). RIF, INH, and CFM were included as positive controls. Riminophenazine analog stock solutions and the range of final testing concentrations were 64 μg/ml and 32 to 0.5 μg/ml, respectively. For the most active compounds, the stock concentration and final testing concentration range were lowered to 3.2 μg/ml and 2 to 0.015 μg/ml, respectively. M. tuberculosis H37Rv or a clinical isolate was grown to late log phase (70 to 100 Klett units) in Difco Middlebrook 7H9 Broth (catalog no. 271310) supplemented with 0.2% (vol/vol) glycerol, 0.05% Tween 80, and 10% (vol/vol) albumin-dextrose-catalase (BBL Middlebrook ADC Enrichment, catalog no. 212352) (7H9-ADC-TG). Cultures were centrifuged, washed twice, and then resuspended in phosphate-buffered saline. Suspensions were then passed through an 8 μm-pore-size filter to remove clumps, and aliquots were frozen at −80°C. Twofold dilutions of riminophenazine analogs were prepared in 7H9-ADC-TG in a volume of 100 μl in 96-well, black, clear-bottom microplates (BD Biosciences, Franklin Lakes, NJ). M. tuberculosis (100 μl containing 2 × 10⁵ CFU) was added, yielding a final testing volume of 200 μl. The plates were incubated at 37°C; on day 7 of incubation, 12.5 μl of 20% Tween 80 and 20 μl of Alamar blue were added to all wells. After incubation at 37°C for 16 to 24 h, the fluorescence was read at an excitation of 530 nm and an emission of 590 nm. The MIC was defined as the lowest concentration effecting a reduction in fluorescence of ≥90% relative to the mean of replicate bacterium-only controls. MICs against nonreplicating M. tuberculosis were determined using a low-oxygen-recovery assay (LORA) as described previously (10).

Cytotoxicity.

Vero cells were cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS). The cells were incubated in a humidified atmosphere of 5% CO₂ at 37°C. Stocks of cells were cultured in 25-cm² tissue culture flasks and subcultured two to three times per week. Cytotoxicity testing was performed in a transparent 96-well microplate. Outer perimeter wells were filled with sterile water to prevent dehydration in experimental wells. The cells were incubated at 37°C under 5% CO₂ until confluent and then diluted with culture medium to 4 × 10⁵ cells/ml. Threefold serial dilutions of the stock solutions resulted in final concentrations of 64 to 0.26 μg/ml in a final volume of 100 μl. After incubation at 37°C for 48 h, the medium was removed, and the monolayers were washed twice with 100 μl of warm Hanks balanced salt solution (HBSS). Warm medium (100 μl) and 10 μl of freshly made methyl-thiazolyldiphenyl-tetrazolium bromide (MTT) were added to each well, and then the plates were incubated for 4 h, after which the absorbance was determined at 492 nm.

Macrophage assay.

J774A.1 cells were grown to confluence in 75-cm² cell culture flasks in Dulbecco modified Eagle medium (DMEM) containing 10% FBS. The cells were detached by using a cell scraper and centrifuged at 200 × g for 5 min at room temperature, and the pellet was resuspended to a final concentration of 3 × 10⁵ to 6 × 10⁵ cells/ml. Aliquots (1 ml) of cell suspension were distributed into 24-well plates (Corning/Costar), and the plates were incubated at 37°C in a 5% CO₂ incubator for 16 h. M. tuberculosis H37Rv frozen cultures were thawed, sonicated for 15 s, and diluted to a final concentration of 3 × 10⁷ to 6 × 10⁷ CFU/ml with DMEM, and 100 μl of the dilution was dispensed into each well to give multiplicity of infection of ∼10 bacteria/cell. The plates were incubated at 37°C for 4 h to allow for phagocytosis. The supernatant was aspirated, and the cells were washed three times with DMEM without FBS to remove the extracellular mycobacteria. The contents of the wells were replaced with DMEM containing 1% FBS and different concentrations of the compounds (1.0, 0.5, and 0.25 μg/ml). Control wells received drug-free medium. The medium with or without drugs was replaced every day. After 3 days of incubation, monolayers were visually inspected under the microscope to ensure they remained intact, and then the medium was removed and the macrophages were lysed with 200 μl of 0.1% sodium dodecyl sulfate. After 10 min of incubation at 37°C, 800 μl of fresh medium was added. Portions (20 μl) of the lysates were inoculated into 96-well, black, clear-bottom microplates, and 7H9-ADC-TG was added, yielding a final testing volume of 200 μl. The plates were incubated at 37°C; on day 7 of incubation, 12.5 μl of 20% Tween 80 and 20 μl of Alamar blue were added to all wells. After incubation at 37°C for 16 to 24 h, the fluorescence of the wells was read at an excitation of 530 nm and an emission of 590 nm. Infected macrophages incubated in drug-free control wells were designated 100% for H37Rv viability, while 7H9 medium controls were used to determine background fluorescence. The intracellular mycobacterial growth inhibition activities of the compounds were calculated as the percent reduction in background-corrected fluorescence compared to the 100% H37Rv viability positive control fluorescence (Flu) as follows: % inhibition = [1 − (Flu_drug − Flu_background)/(Flu_control − Flu_background)] × 100.

Pharmacokinetics.

Specific-pathogen-free (SPF) male BALB/c mice weighing 18 to 22 g were used with three mice per time point. Feed and water were freely given during acclimation with a 16-h fasting period prior to dose administration. Compounds were prepared at 2 mg/ml in 0.5% carboxymethyl cellulose (CMC), and a 0.2-ml suspension was administrated orally to each mouse to achieve a 20-mg/kg (body weight) dose. Blood samples were collected via the fossa orbitalis vein at 0, 0.5, 1, 2, 4, 8, and 24 h after administration and centrifuged at 2,500 × g for 10 min to obtain serum which was then stored at −20°C. Portions (100 μl) of the serum was added to 200 μl of acetonitrile, and the resulting solution was vortex mixed for 2 min and then centrifuged at 2,500 × g for 10 min to remove protein. The supernatant was removed to an Eppendorf tube and centrifuged at 11,000 × g for 5 min. The supernatant was moved to a sample bottle, and 40 μl was injected for HPLC analysis (Waters 600 Controller, 2487 dual λ absorbance detector, 717 plus Autosampler and Waters Empower chromatogram workstation). The analytical column was a Waters SymmetryShield RP₁₈ (3.9 by 150 mm, 5 μm). The mobile phase was prepared by mixing acetic sodium buffer solution and methanol in various proportions determined in an earlier experiment. The flow rate was 1 ml/min, with an λ_max of 495 nm. All chromatograms obtained were evaluated by the peak area. Pharmacokinetic (PK) analyses of the plasma concentration-time relationships for the riminophenazines were performed using Drug and StatisticsVersion 3.0 (Bontz, Inc., Beijing, China). A noncompartmental library model was used to calculate PK parameters, including the maximum concentration of drug in plasma (C_max), the elimination half-life, and the area under the concentration-time curve from 0 to 24 h (AUC₀₋₂₄). The PK data presented in Table 6 are reported as means ± standard deviations (SD).

Aerosol model of acute infection.

SPF BALB/c male mice, weighing 18 to 20 g, were used in the present study. Mice were infected via aerosol with a suspension of 5 × 10⁶ CFU of M. tuberculosis H37Rv/ml using a Glas-Col inhalation system to deposit 50 to 100 bacilli into the lungs of each animal. The course of infection was followed by plating homogenates of harvested organs (obtained from three animals) on 7H11 agar plates (7H11 plates containing 10% oleic acid-albumin-dextrose-catalase [OADC] enrichment and 50 μg of cycloheximide, 200 U of polymyxin B, 50 μg of carbenicillin, and 20 μg of trimethoprim/ml) and determining the CFU on days 3, 10, and 30 postinfection. The drugs and compounds were dissolved or suspended in 0.5% CMC and administered by oral gavage in a maximum volume of 200 μl such that a dose of 20 mg/kg (body weight) was achieved. Mice were treated 5 days per week during the acute phase of infection from day 10 until day 30. Each treated group included five or six mice, while the control group, which received only CMC, was composed of seven to ten mice. Mice were sacrificed the day after the last day of treatment; the organ weights were determined, and the lungs were removed, homogenized, and serially diluted in 10-fold steps in HBSS. Portions (100 μl) were spread onto 7H11 agar in duplicate. The plates were incubated at 37°C for 2 to 3 weeks. The data are expressed as the log₁₀ (and as the log₁₀ reduction) provided by a given dose of the compound against the growth of the organism in the untreated control group. Mean log₁₀ values were calculated from bacterial burden counts. A Student t test was used to compare means between the test and control groups. A P value of ≤0.05 was considered significant.

Riminophenazine analogs with enhanced potency compared to CFM against drug-sensitive and drug-resistant replicating M. tuberculosis.

The 12 riminophenazine analogs possess MICs ranging from 0.011 to 0.038 μg/ml, with the most active exhibiting potency 10-fold better than CFM (MIC = 0.12 μg/ml). In addition, the compounds have acceptable cytotoxicity against Vero cells with selectivity indices ranging from 18 to >5,818 (Table 1). To further explore the in vitro activities of the compounds compared to CFM, MICs were determined using a microplate Alamar blue assay (MABA) against a panel of 10 drug-sensitive M. tuberculosis clinical isolates and against 10 clinical isolates with varying drug resistance profiles, including both MDR and XDR strains. The MICs for this set of riminophenazine analogs against the drug-sensitive isolates range from 0.015 to 0.116 μg/ml, and most compounds are more potent than CFM (0.044 to 0.24 μg/ml) (Table 2). Likewise, the new compounds demonstrated enhanced activity compared to CFM against the drug-resistant panel, with MICs ranging from 0.015 to 0.121 μg/ml compared to 0.108 to 0.24 μg/ml for CFM (Table 3). This result indicates the potential utility of these compounds against both MDR and XDR TB strains, including strains resistant to both fluoroquinolones and aminoglycosides. In addition, these data suggest a novel mode of action for this class compared to TB drugs in current use.

Riminophenazine analogs active against nonreplicating M. tuberculosis and intracellular M. tuberculosis.

The activity of CFM against intracellular M. tuberculosis has been well documented (20, 37, 44). It is presumed to be related to the tendency of this antibacterial compound to accumulate in phagocytic cells (4, 14). We investigated the activity of the new riminophenazine analogs against M. tuberculosis replicating within J774A.1 macrophages by using a recovery assay. We compared the extracellular recovery of bacteria following 3 days of intracellular exposure to compounds at various concentrations. After 7 days of recovery, the bacterial titers for compound exposed wells, as detected via Alamar blue, were lower than for those of the no compound control wells. Dose-dependent activity was observed for all riminophenazine analogs and was at least equipotent to CFM (Table 4). Intracellular bacterial growth for the drug free control during the assay period was determined by CFU enumeration to be ∼1 log₁₀ (6.07 ± 0.04 log₁₀ CFU on day 0 and 7.16 ± 0.11 log₁₀ CFU on day 3). Therefore, the diminished bacterial recovery observed following exposure to the riminophenazine analogs versus the no drug control reflects either bacteriostatic or bactericidal activity exerted during the intracellular phase of the assay. Further studies directly enumerating intracellular bacteria versus duration of compound exposure will be useful to distinguish growth inhibition from bactericidal activity for compounds that are progressed.

We also determined the activities of the new compounds against low-oxygen-induced nonreplicating M. tuberculosis using the LORA assay. These bacteria may reflect a subpopulation of drug-tolerant bacteria that contributes to the persistence of a TB infection in the face of prolonged chemotherapy. The majority of the riminophenazine analogs demonstrated activity in this assay with the MICs of TBI-161, TBI-678, and TBI-688 being in the range of that of CFM (Table 5). Of interest, these activities do not correlate with potency against replicating bacteria or intracellular bacteria, suggesting a complex mode of action for this class.

Less lipophilic riminophenazine analogs with decreased plasma half-lives compared to CFM.

Riminophenazine analogs with lower lipophilicity and shorter half-lives compared to CFM can be expected to accumulate less in tissues, including those of the MPS. We hypothesize that a consequence of reduced tissue accumulation and crystallization will be reduced skin discoloration. Therefore, compounds of interest for further investigation will demonstrate shorter half-lives and reduced lipophilicity compared to CFM, while affording the exposure necessary to efficacy when dosed orally. The logP values of the 12 new compounds are lower than those of CFM (logP = 5.34) and range from 3.51 to 4.82. No compound has aqueous solubility above the limit of detection (0.01 g/100 ml) at pH 7, but solubility at pH 1 is enhanced for the new compounds (ranging from 0.2 to 10.4 g/100 ml) compared to CFM, which exhibits solubility below the detection limit (Table 1). Through previous work with CFM, we demonstrated that the mouse is a suitable animal model for evaluation of the potential of riminophenazine analogs to accumulate and discolor tissue (28). We therefore determined primary PK parameters for each riminophenazine analog in mice, dosed once by the oral route (Table 6). All but one of the new compounds possesses a shorter plasma half-life and increased C_max and AUC compared to CFM. Shorter serum half-lives in combination with lower lipophilicity may indicate a decreased potential for the new compounds to produce adverse effects due to drug accumulation.

Riminophenazine analogs that are efficacious against acute TB infection in mice.

A murine model of acute infection with M. tuberculosis H37Rv was used to assess the efficacy of the new riminophenazine analogs in comparison that of to CFM. Prior to conducting efficacy testing, a preliminary tolerability study was carried out in mice for all 12 compounds. Through this, it was determined that three of the compounds (TBI-161, TBI-678, and TBI-688) have 50% lethal doses lower than 600 mg/kg. The remaining nine compounds were therefore prioritized for efficacy evaluation. The nine compounds demonstrated a range of efficacies against murine acute M. tuberculosis infection that were, in all cases, superior to that of CFM when administered at an equivalent dose for 20 days (Table 7). In this model of acute infection, the lung bacterial burden of the untreated control group increased from 3.17 ± 0.39 log₁₀ CFU to 8.53 ± 0.32 log₁₀ CFU over the 20-day course of chemotherapy. At the end of the treatment period, the mean lung CFU counts for all riminophenazine-treated groups (including CFM) were 3 to 5 log units lower than those for the untreated control group (P < 0.001). Treatment with any of the new riminophenazine analogs resulted in mean lung CFU counts lower than those of the CFM-treated group (P < 0.001). The lowest lung burdens were observed after treatment with TBI-449, TBI-450, and TBI-416. Although the new riminophenazine analogs demonstrated efficacy superior to CFM in this model and at this dose, none of these compounds reduced the lung burden below that observed pretreatment in this experiment. We conclude that, under the conditions selected, we observed primarily bacteriostatic rather than bactericidal activity for the riminophenazine analogs and CFM. Further studies are planned to evaluate the maximal efficacy achievable and bactericidal potential of selected compounds against murine TB. We have previously observed that CFM treatment of mice leads to discoloration of their internal organs, ears, and feet in a dose- and duration-dependent manner (28). In the present study we observed discoloration of ears, internal organs, and fat as seen previously, following dosing of CFM at 20 mg/kg for 20 days. However, the new riminophenazine analogs, administered at the same dose for the same length of time, caused a range of discoloration effects. Of interest, two of the most efficacious compounds, TBI-416 and TBI-449, appeared to cause the least discoloration of tissues (data not shown) in this preliminary assessment.